Datasets:

DanCip

/

github-issues-top50-repos

like 0

The CentOS builds on GitLab have started failing a week or two ago, all with the same pattern: (see https://gitlab.com/D4N/exiv2/-/jobs/151300428): ``` ln: failed to create symbolic link '/usr/bin/python3': File exists ``` I'll have to adapt the script `ci/install_dependencies.sh`.

CentOS builds are consistently failing on GitLab

https://api.github.com/repos/Exiv2/exiv2/issues/666/comments

https://github.com/Exiv2/exiv2/issues/666

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : CentOS builds are consistently failing on GitLab Issue date: --- start body --- The CentOS builds on GitLab have started failing a week or two ago, all with the same pattern: (see https://gitlab.com/D4N/exiv2/-/jobs/151300428): ``` ln: failed to create symbolic link '/usr/bin/python3': File exists ``` I'll have to adapt the script `ci/install_dependencies.sh`. --- end body --- comments : --- start comments --- --- end comments ---

This is the one I noticed, but there could be others: http://dev.exiv2.org/users/9462 If it's no longer in use, maybe new signups should be disabled?

Old issue tracker is being spammed.

https://api.github.com/repos/Exiv2/exiv2/issues/664/comments

https://github.com/Exiv2/exiv2/issues/664

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : Old issue tracker is being spammed. Issue date: --- start body --- This is the one I noticed, but there could be others: http://dev.exiv2.org/users/9462 If it's no longer in use, maybe new signups should be disabled? --- end body --- comments : --- start comments --- --- end comments ---

Required steps for MinGW cross compilation support: - [x] Fix the windows build in #645 and remove the binary files from that PR - [x] Add a MinGW build to the GitLab CI pipeline. Fedora's MinGW can already build exiv2 as a static library, dynamic linking is still broken though. Todo @D4N - [ ] Remove the templating of `BasicError` and drop `WError`, instead convert the `wchar_t` string to a `char_t` string in the few cases where this class is actually used. Then we can drop the awkward function `BasicError<charT>::setMsg()` which is exported from a template function. This hopefully fixes dynamic linking on MinGW. See: #685. - [ ] Run the test suite for MinGW with wine on Linux. Notes: - Wine requires X to be running, so we'll probably have to use something like Xvfb. - set `export WINEDEBUG=-all` otherwise wine logs to stderr - The MinGW dlls are by default not known by wine, they have to be added to `WINEPATH`. On Fedora this results in: `export WINEPATH=/usr/x86_64-w64-mingw32/sys-root/mingw/bin/` - Wine seems to change the line ending, at least the python test complain about whitespace changes at the end of all lines. This can be probably fixed by a simple search & replace when running under wine.

MinGW cross compilation support

https://api.github.com/repos/Exiv2/exiv2/issues/663/comments

https://github.com/Exiv2/exiv2/issues/663

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : MinGW cross compilation support Issue date: --- start body --- Required steps for MinGW cross compilation support: - [x] Fix the windows build in #645 and remove the binary files from that PR - [x] Add a MinGW build to the GitLab CI pipeline. Fedora's MinGW can already build exiv2 as a static library, dynamic linking is still broken though. Todo @D4N - [ ] Remove the templating of `BasicError` and drop `WError`, instead convert the `wchar_t` string to a `char_t` string in the few cases where this class is actually used. Then we can drop the awkward function `BasicError<charT>::setMsg()` which is exported from a template function. This hopefully fixes dynamic linking on MinGW. See: #685. - [ ] Run the test suite for MinGW with wine on Linux. Notes: - Wine requires X to be running, so we'll probably have to use something like Xvfb. - set `export WINEDEBUG=-all` otherwise wine logs to stderr - The MinGW dlls are by default not known by wine, they have to be added to `WINEPATH`. On Fedora this results in: `export WINEPATH=/usr/x86_64-w64-mingw32/sys-root/mingw/bin/` - Wine seems to change the line ending, at least the python test complain about whitespace changes at the end of all lines. This can be probably fixed by a simple search & replace when running under wine. --- end body --- comments : --- start comments --- --- end comments ---

In case I want the tool to rename all files in a directory accordingly to the timestamps (but the question would apply to all cases, IHMO), how does the tool behaves? If I have more photos with the same EXIF timestamps, will they be all overwritten but the last one?

How does the rename (mv) work in case of name clash?

https://api.github.com/repos/Exiv2/exiv2/issues/662/comments

https://github.com/Exiv2/exiv2/issues/662

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : How does the rename (mv) work in case of name clash? Issue date: --- start body --- In case I want the tool to rename all files in a directory accordingly to the timestamps (but the question would apply to all cases, IHMO), how does the tool behaves? If I have more photos with the same EXIF timestamps, will they be all overwritten but the last one? --- end body --- comments : --- start comments --- --- end comments ---

The build error after merging #617 was not caught on my side because the flatpak-based builder keeps on checking out the tag 0.27 instead of the branch 0.27. For 0.28 it would be nice if the branch would be 0.28 and the initial version tagged with 0.28.0 or something like that

Request: For future milestone releases, please make it possible to easily distinguish between initial release (eg. 0.27) and a branch for that version(eg. 0.27)

https://api.github.com/repos/Exiv2/exiv2/issues/661/comments

https://github.com/Exiv2/exiv2/issues/661

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : Request: For future milestone releases, please make it possible to easily distinguish between initial release (eg. 0.27) and a branch for that version(eg. 0.27) Issue date: --- start body --- The build error after merging #617 was not caught on my side because the flatpak-based builder keeps on checking out the tag 0.27 instead of the branch 0.27. For 0.28 it would be nice if the branch would be 0.28 and the initial version tagged with 0.28.0 or something like that --- end body --- comments : --- start comments --- --- end comments ---

The following email has arrived from Google: ---------- Hi all, If your open source organization is interested in being a mentoring organization for Google Summer of Code 2019 please be sure to submit your organization's application via g.co/gsoc before February 6 at 20:00 UTC. Organizations chosen for GSoC 2019 will be announced on February 26th. Best, Google Open Source Programs Office ------------ I am interested in mentoring a student in 2019. I've discussed this by email with Dan and Luis and they enthusiastically support the idea. Due to their commitments, the work to recruit and mentor a student will fall on my shoulders and they will support on a part-time basis. In 2012 and 2013, Exiv2 was treated as a KDE project. KDE were awarded about 60 "slots" (from 1200) which were split into about 10 "sub-projects". The Chairman of KDE/Graphics/GSoC was @cgilles. Gilles: Are you still in that role and able to help us, or should I apply directly to Google? I will publish a project description and students will prepare proposals. I expect those proposals to be detailed and involve prototyping work. This is a test to ensure that students are willing to invest effort to ensure they are chosen and add this to their CV. I hope we'll recruit a European student who can visit/stay with me in England for a week at the out-set of their work. I'd like to call the project something like: "Product Strengthening". Exiv2 is a mature open-source library and used by many applications. We have three current initiatives. 1) The "dots". Exiv2 v0.27 will have quarterly updates in 2019 and 2020. _**Robin**_ and _**Luis**_ 2) "Modernisation" for Exiv2 v0.28. _**Luis**_ and _**Dan**_ 3) Dealing with users. _**Robin**_ All this activity leaves little time to look around and strengthen the code. Topics we have discussed are: > 1) Refactor the bash scripts into the python framework > 2) Support CTest > 3) Extended testing (to be run on the buildserver) > - I have a copy of the ExifTool test files on the build server > - I have the RAW images from the Swiss web-site on the build server > - Trawl camera manufacturers web site for more raw samples > 4) Better lens recognition (discussed with @frli8848) > 5) Various enhancements to our test suite. For example: more unit tests and localisation > 6) Fuzzing https://github.com/google/oss-fuzz > 7) Performance testing and optimisation. For example #530 > 8) Better web documentation concerning makernotes For example #646 > 9) Static Analysis with cppcheck and PVS-Studio > 10) Building for mobile (iOS and Android) We'll think of more topics between now and May. How much can be done will depend on the ability of the student and the depth to which we dig on any topic. CTest is a well defined topic and is either done or not. However the other tasks are vague. For example running exiv2 on every Exif Tool test file could be a single command to run exiv2 on 6500 files and compare the output with a reference. Or we could go manually through the files and analyse ExifTool vs Exiv2 to discover metadata which we don't report. So, that could a one-day project, or a life-time of discoveries and fixes. GSoC is 12 weeks. Let's aim for 8 topics in which something useful can be achieved in about one week. We can revisit more interesting/fruitful topics towards the end of the project time. Why am I interested in this? This will encourage us to do a "heads up" and deal with important, yet non-urgent, topics. The student will have a good experience and Exiv2 will be stronger. This is a win-win for everyone. I don't expect GSoC to impact progress towards Exiv2 v0.28. Working on this may delay a "dot" release. I am confident that everybody involved in Exiv2 will give the student a great welcome and encouragement. Comments welcome: @fgeek @FreddieWitherden @frli8848 @piponazo @cryptomilk @Kicer86 @nkbj @tbeu @cgilles @D4N @boardhead

Google Summer of Code 2019

https://api.github.com/repos/Exiv2/exiv2/issues/659/comments

https://github.com/Exiv2/exiv2/issues/659

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : Google Summer of Code 2019 Issue date: --- start body --- The following email has arrived from Google: ---------- Hi all, If your open source organization is interested in being a mentoring organization for Google Summer of Code 2019 please be sure to submit your organization's application via g.co/gsoc before February 6 at 20:00 UTC. Organizations chosen for GSoC 2019 will be announced on February 26th. Best, Google Open Source Programs Office ------------ I am interested in mentoring a student in 2019. I've discussed this by email with Dan and Luis and they enthusiastically support the idea. Due to their commitments, the work to recruit and mentor a student will fall on my shoulders and they will support on a part-time basis. In 2012 and 2013, Exiv2 was treated as a KDE project. KDE were awarded about 60 "slots" (from 1200) which were split into about 10 "sub-projects". The Chairman of KDE/Graphics/GSoC was @cgilles. Gilles: Are you still in that role and able to help us, or should I apply directly to Google? I will publish a project description and students will prepare proposals. I expect those proposals to be detailed and involve prototyping work. This is a test to ensure that students are willing to invest effort to ensure they are chosen and add this to their CV. I hope we'll recruit a European student who can visit/stay with me in England for a week at the out-set of their work. I'd like to call the project something like: "Product Strengthening". Exiv2 is a mature open-source library and used by many applications. We have three current initiatives. 1) The "dots". Exiv2 v0.27 will have quarterly updates in 2019 and 2020. _**Robin**_ and _**Luis**_ 2) "Modernisation" for Exiv2 v0.28. _**Luis**_ and _**Dan**_ 3) Dealing with users. _**Robin**_ All this activity leaves little time to look around and strengthen the code. Topics we have discussed are: > 1) Refactor the bash scripts into the python framework > 2) Support CTest > 3) Extended testing (to be run on the buildserver) > - I have a copy of the ExifTool test files on the build server > - I have the RAW images from the Swiss web-site on the build server > - Trawl camera manufacturers web site for more raw samples > 4) Better lens recognition (discussed with @frli8848) > 5) Various enhancements to our test suite. For example: more unit tests and localisation > 6) Fuzzing https://github.com/google/oss-fuzz > 7) Performance testing and optimisation. For example #530 > 8) Better web documentation concerning makernotes For example #646 > 9) Static Analysis with cppcheck and PVS-Studio > 10) Building for mobile (iOS and Android) We'll think of more topics between now and May. How much can be done will depend on the ability of the student and the depth to which we dig on any topic. CTest is a well defined topic and is either done or not. However the other tasks are vague. For example running exiv2 on every Exif Tool test file could be a single command to run exiv2 on 6500 files and compare the output with a reference. Or we could go manually through the files and analyse ExifTool vs Exiv2 to discover metadata which we don't report. So, that could a one-day project, or a life-time of discoveries and fixes. GSoC is 12 weeks. Let's aim for 8 topics in which something useful can be achieved in about one week. We can revisit more interesting/fruitful topics towards the end of the project time. Why am I interested in this? This will encourage us to do a "heads up" and deal with important, yet non-urgent, topics. The student will have a good experience and Exiv2 will be stronger. This is a win-win for everyone. I don't expect GSoC to impact progress towards Exiv2 v0.28. Working on this may delay a "dot" release. I am confident that everybody involved in Exiv2 will give the student a great welcome and encouragement. Comments welcome: @fgeek @FreddieWitherden @frli8848 @piponazo @cryptomilk @Kicer86 @nkbj @tbeu @cgilles @D4N @boardhead --- end body --- comments : --- start comments --- --- end comments ---

"False" start with F. It needs to check if word start with f or a number start with f.

std::istream& operator>>(std::istream& is, Rational& r) fails to convert string False

https://api.github.com/repos/Exiv2/exiv2/issues/658/comments

https://github.com/Exiv2/exiv2/issues/658

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : std::istream& operator>>(std::istream& is, Rational& r) fails to convert string False Issue date: --- start body --- "False" start with F. It needs to check if word start with f or a number start with f. --- end body --- comments : --- start comments --- --- end comments ---

Compiling exiv2-0.27.0a works fine, but then packages linking against it using `pkg-config` can fail, because the exiv2.pc generation does not determine the correct `libdir`, but instead sets it to [`${prefix}/lib`](https://github.com/Exiv2/exiv2/blob/master/cmake/exiv2.pc.in#L3). Distributions (such as Gentoo) that use `/usr/lib64` for example will either not be able to find, or end up using libraries for the wrong architecture. The previous version used `@libdir@` to ensure the correct library directory and that successfully allows other programs to be built against it.

exiv2-0.27.0a hardcodes pkgconfig/exiv2.pc libdir with ${prefix}/lib

https://api.github.com/repos/Exiv2/exiv2/issues/651/comments

https://github.com/Exiv2/exiv2/issues/651

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : exiv2-0.27.0a hardcodes pkgconfig/exiv2.pc libdir with ${prefix}/lib Issue date: --- start body --- Compiling exiv2-0.27.0a works fine, but then packages linking against it using `pkg-config` can fail, because the exiv2.pc generation does not determine the correct `libdir`, but instead sets it to [`${prefix}/lib`](https://github.com/Exiv2/exiv2/blob/master/cmake/exiv2.pc.in#L3). Distributions (such as Gentoo) that use `/usr/lib64` for example will either not be able to find, or end up using libraries for the wrong architecture. The previous version used `@libdir@` to ensure the correct library directory and that successfully allows other programs to be built against it. --- end body --- comments : --- start comments --- --- end comments ---

Fixing 644 produces linker warnings on MacOS-X. 644 is a build-breaker for Suse. The linking message is ugly and probably benign. ``` [ 50%] Linking CXX shared library ../lib/libexiv2.dylib ld: warning: direct access in function 'Exiv2::Internal::CiffEntry::doAdd(std::__1::auto_ptr<Exiv2::Internal::CiffComponent>)' from file 'CMakeFiles/exiv2lib_int.dir/crwimage_int.cpp.o' to global weak symbol 'typeinfo for Exiv2::BasicError<char>' from file 'CMakeFiles/exiv2lib.dir/basicio.cpp.o' means the weak symbol cannot be overridden at runtime. This was likely caused by different translation units being compiled with different visibility settings. ``` We've seen this message before (around the time of Exiv2 v0.27 RC3) and didn't know what it meant.

Linker warning on MacOS-X (after fixing 644)

https://api.github.com/repos/Exiv2/exiv2/issues/648/comments

https://github.com/Exiv2/exiv2/issues/648

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : Linker warning on MacOS-X (after fixing 644) Issue date: --- start body --- Fixing 644 produces linker warnings on MacOS-X. 644 is a build-breaker for Suse. The linking message is ugly and probably benign. ``` [ 50%] Linking CXX shared library ../lib/libexiv2.dylib ld: warning: direct access in function 'Exiv2::Internal::CiffEntry::doAdd(std::__1::auto_ptr<Exiv2::Internal::CiffComponent>)' from file 'CMakeFiles/exiv2lib_int.dir/crwimage_int.cpp.o' to global weak symbol 'typeinfo for Exiv2::BasicError<char>' from file 'CMakeFiles/exiv2lib.dir/basicio.cpp.o' means the weak symbol cannot be overridden at runtime. This was likely caused by different translation units being compiled with different visibility settings. ``` We've seen this message before (around the time of Exiv2 v0.27 RC3) and didn't know what it meant. --- end body --- comments : --- start comments --- --- end comments ---

I'm looking for the location and size metadata of the contrast-detect AF point in my Nikon D850 raw files. Such tags include `Exif.NikonAf2.AFAreaXPosition`, `Exif.NikonAf2.AFAreaYPosition`, `Exif.NikonAf2.AFAreaWidth`, etc. Exiv2 finds the correct values for the D810 but not for the D850. Exiftool does return this information for the D850, so I checked its source. Apparently the file offset of this metadata depends on the value of `Exif.NikonAf2.Version`. For example, compare https://github.com/exiftool/exiftool/blob/9e0ce916748abb56dbb8593d2184445080f678d0/lib/Image/ExifTool/Nikon.pm#L3519-L3525 with https://github.com/exiftool/exiftool/blob/9e0ce916748abb56dbb8593d2184445080f678d0/lib/Image/ExifTool/Nikon.pm#L3627-L3633. Note specifically the different `Condition` keys.

Exiv2 does not find certain AF tags for the Nikon D850

https://api.github.com/repos/Exiv2/exiv2/issues/646/comments

https://github.com/Exiv2/exiv2/issues/646

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : Exiv2 does not find certain AF tags for the Nikon D850 Issue date: --- start body --- I'm looking for the location and size metadata of the contrast-detect AF point in my Nikon D850 raw files. Such tags include `Exif.NikonAf2.AFAreaXPosition`, `Exif.NikonAf2.AFAreaYPosition`, `Exif.NikonAf2.AFAreaWidth`, etc. Exiv2 finds the correct values for the D810 but not for the D850. Exiftool does return this information for the D850, so I checked its source. Apparently the file offset of this metadata depends on the value of `Exif.NikonAf2.Version`. For example, compare https://github.com/exiftool/exiftool/blob/9e0ce916748abb56dbb8593d2184445080f678d0/lib/Image/ExifTool/Nikon.pm#L3519-L3525 with https://github.com/exiftool/exiftool/blob/9e0ce916748abb56dbb8593d2184445080f678d0/lib/Image/ExifTool/Nikon.pm#L3627-L3633. Note specifically the different `Condition` keys. --- end body --- comments : --- start comments --- --- end comments ---

Hi, After upgrade from 0.26 to 0.27 on darwin i am getting following error ``` libc++abi.dylib: terminating with uncaught exception of type Exiv2::BasicError<char>: No namespace info available for XMP prefix `DICOM' ``` This is part of the trace, issue is triggered when app is calling gexiv2_metadata_register_xmp_namespace ``` * frame #0: 0x00007fff60261b86 libsystem_kernel.dylib`__pthread_kill + 10 frame #1: 0x00007fff60317c50 libsystem_pthread.dylib`pthread_kill + 285 frame #2: 0x00007fff601cb1c9 libsystem_c.dylib`abort + 127 frame #3: 0x00007fff5d84d231 libc++abi.dylib`abort_message + 231 frame #4: 0x00007fff5d84d3b5 libc++abi.dylib`default_terminate_handler() + 241 frame #5: 0x00007fff5f056c8f libobjc.A.dylib`_objc_terminate() + 105 frame #6: 0x00007fff5d858dfe libc++abi.dylib`std::__terminate(void (*)()) + 8 frame #7: 0x00007fff5d858be2 libc++abi.dylib`__cxxabiv1::failed_throw(__cxxabiv1::__cxa_exception*) + 27 frame #8: 0x00007fff5d84bbb9 libc++abi.dylib`__cxa_throw + 113 frame #9: 0x000000010127cfe4 libexiv2.27.dylib`Exiv2::XmpProperties::nsInfoUnsafe(std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> > const&) + 260 frame #10: 0x000000010127cdfc libexiv2.27.dylib`Exiv2::XmpProperties::ns(std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> > const&) + 300 frame #11: 0x0000000100fe973c libgexiv2.2.dylib`gexiv2_metadata_register_xmp_namespace + 172 ``` Reverting to 0.26 fixing the issue

exiv2 0.27+ on darwin fails with Exiv2::BasicError<char> error

https://api.github.com/repos/Exiv2/exiv2/issues/644/comments

https://github.com/Exiv2/exiv2/issues/644

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : exiv2 0.27+ on darwin fails with Exiv2::BasicError<char> error Issue date: --- start body --- Hi, After upgrade from 0.26 to 0.27 on darwin i am getting following error ``` libc++abi.dylib: terminating with uncaught exception of type Exiv2::BasicError<char>: No namespace info available for XMP prefix `DICOM' ``` This is part of the trace, issue is triggered when app is calling gexiv2_metadata_register_xmp_namespace ``` * frame #0: 0x00007fff60261b86 libsystem_kernel.dylib`__pthread_kill + 10 frame #1: 0x00007fff60317c50 libsystem_pthread.dylib`pthread_kill + 285 frame #2: 0x00007fff601cb1c9 libsystem_c.dylib`abort + 127 frame #3: 0x00007fff5d84d231 libc++abi.dylib`abort_message + 231 frame #4: 0x00007fff5d84d3b5 libc++abi.dylib`default_terminate_handler() + 241 frame #5: 0x00007fff5f056c8f libobjc.A.dylib`_objc_terminate() + 105 frame #6: 0x00007fff5d858dfe libc++abi.dylib`std::__terminate(void (*)()) + 8 frame #7: 0x00007fff5d858be2 libc++abi.dylib`__cxxabiv1::failed_throw(__cxxabiv1::__cxa_exception*) + 27 frame #8: 0x00007fff5d84bbb9 libc++abi.dylib`__cxa_throw + 113 frame #9: 0x000000010127cfe4 libexiv2.27.dylib`Exiv2::XmpProperties::nsInfoUnsafe(std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> > const&) + 260 frame #10: 0x000000010127cdfc libexiv2.27.dylib`Exiv2::XmpProperties::ns(std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> > const&) + 300 frame #11: 0x0000000100fe973c libgexiv2.2.dylib`gexiv2_metadata_register_xmp_namespace + 172 ``` Reverting to 0.26 fixing the issue --- end body --- comments : --- start comments --- --- end comments ---

The `convert.hpp` file should include `exiv2lib_export.h`. Otherwise, this occurs: ``` In file included from /builds/mardy/imaginario/src/metadata.cpp:26:0: /builds/mardy/imaginario/staging/include/exiv2/convert.hpp:53:5: error: 'EXIV2API' does not name a type EXIV2API void copyExifToXmp(const ExifData& exifData, XmpData& xmpData); ^ /builds/mardy/imaginario/staging/include/exiv2/convert.hpp:55:5: error: 'EXIV2API' does not name a type EXIV2API void moveExifToXmp(ExifData& exifData, XmpData& xmpData); ^ /builds/mardy/imaginario/staging/include/exiv2/convert.hpp:58:5: error: 'EXIV2API' does not name a type EXIV2API void copyXmpToExif(const XmpData& xmpData, ExifData& exifData); ^ ... ```

error: 'EXIV2API' does not name a type

https://api.github.com/repos/Exiv2/exiv2/issues/636/comments

https://github.com/Exiv2/exiv2/issues/636

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : error: 'EXIV2API' does not name a type Issue date: --- start body --- The `convert.hpp` file should include `exiv2lib_export.h`. Otherwise, this occurs: ``` In file included from /builds/mardy/imaginario/src/metadata.cpp:26:0: /builds/mardy/imaginario/staging/include/exiv2/convert.hpp:53:5: error: 'EXIV2API' does not name a type EXIV2API void copyExifToXmp(const ExifData& exifData, XmpData& xmpData); ^ /builds/mardy/imaginario/staging/include/exiv2/convert.hpp:55:5: error: 'EXIV2API' does not name a type EXIV2API void moveExifToXmp(ExifData& exifData, XmpData& xmpData); ^ /builds/mardy/imaginario/staging/include/exiv2/convert.hpp:58:5: error: 'EXIV2API' does not name a type EXIV2API void copyXmpToExif(const XmpData& xmpData, ExifData& exifData); ^ ... ``` --- end body --- comments : --- start comments --- --- end comments ---

I've discussed this with ncopa. I loaded an Alpine VM on the build server. I have tested the following patch on MinGW/msys2, msvc, Linux, Cygwin and MacOS. It works on Alpine Linux. Can this patch be applied to both 0.27 and 'master' ``` diff --git a/include/exiv2/futils.hpp b/include/exiv2/futils.hpp index 4d0ea6d5..322a9566 100644 --- a/include/exiv2/futils.hpp +++ b/include/exiv2/futils.hpp @@ -31,7 +31,19 @@ #include "config.h" #include "exiv2lib_export.h" +// https://stackoverflow.com/questions/3052041/how-to-get-posix-strerror-r-instead-of-gnu-version +#ifdef _GNU_SOURCE +#define _GNU_SOURCE_RESTORE _GNU_SOURCE +#undef _GNU_SOURCE +#endif #include <string> +#ifdef _GNU_SOURCE +#define STRERROR_RETURNS_CHAR_PTR +#endif +#ifdef _GNU_SOURCE_RESTORE +#define _GNU_SOURCE _GNU_SOURCE_RESTORE +#undef _GNU_SOURCE_RESTORE +#endif // namespace extensions namespace Exiv2 diff --git a/src/futils.cpp b/src/futils.cpp index 00e71172..ebe05e50 100644 --- a/src/futils.cpp +++ b/src/futils.cpp @@ -53,14 +53,6 @@ #include <algorithm> #include <stdexcept> -#ifdef EXV_HAVE_STRERROR_R -#if defined(__GLIBC__) && defined(_GNU_SOURCE) -extern char *strerror_r(int errnum, char *buf, size_t n); -#else -extern int strerror_r(int errnum, char *buf, size_t n); -#endif -#endif - namespace Exiv2 { const char* ENVARDEF[] = {"/exiv2.php", "40"}; //!< @brief default URL for http exiv2 handler and time-out const char* ENVARKEY[] = {"EXIV2_HTTP_POST", "EXIV2_TIMEOUT"}; //!< @brief request keys for http exiv2 handler and time-out @@ -348,7 +340,7 @@ namespace Exiv2 { std::ostringstream os; #ifdef EXV_HAVE_STRERROR_R const size_t n = 1024; -#if defined(__GLIBC__) && defined(_GNU_SOURCE) +#if defined(STRERROR_RETURNS_CHAR_PTR) char *buf = 0; char buf2[n]; std::memset(buf2, 0x0, n); ```

Fix for build breaker in #616

https://api.github.com/repos/Exiv2/exiv2/issues/631/comments

https://github.com/Exiv2/exiv2/issues/631

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : Fix for build breaker in #616 Issue date: --- start body --- I've discussed this with ncopa. I loaded an Alpine VM on the build server. I have tested the following patch on MinGW/msys2, msvc, Linux, Cygwin and MacOS. It works on Alpine Linux. Can this patch be applied to both 0.27 and 'master' ``` diff --git a/include/exiv2/futils.hpp b/include/exiv2/futils.hpp index 4d0ea6d5..322a9566 100644 --- a/include/exiv2/futils.hpp +++ b/include/exiv2/futils.hpp @@ -31,7 +31,19 @@ #include "config.h" #include "exiv2lib_export.h" +// https://stackoverflow.com/questions/3052041/how-to-get-posix-strerror-r-instead-of-gnu-version +#ifdef _GNU_SOURCE +#define _GNU_SOURCE_RESTORE _GNU_SOURCE +#undef _GNU_SOURCE +#endif #include <string> +#ifdef _GNU_SOURCE +#define STRERROR_RETURNS_CHAR_PTR +#endif +#ifdef _GNU_SOURCE_RESTORE +#define _GNU_SOURCE _GNU_SOURCE_RESTORE +#undef _GNU_SOURCE_RESTORE +#endif // namespace extensions namespace Exiv2 diff --git a/src/futils.cpp b/src/futils.cpp index 00e71172..ebe05e50 100644 --- a/src/futils.cpp +++ b/src/futils.cpp @@ -53,14 +53,6 @@ #include <algorithm> #include <stdexcept> -#ifdef EXV_HAVE_STRERROR_R -#if defined(__GLIBC__) && defined(_GNU_SOURCE) -extern char *strerror_r(int errnum, char *buf, size_t n); -#else -extern int strerror_r(int errnum, char *buf, size_t n); -#endif -#endif - namespace Exiv2 { const char* ENVARDEF[] = {"/exiv2.php", "40"}; //!< @brief default URL for http exiv2 handler and time-out const char* ENVARKEY[] = {"EXIV2_HTTP_POST", "EXIV2_TIMEOUT"}; //!< @brief request keys for http exiv2 handler and time-out @@ -348,7 +340,7 @@ namespace Exiv2 { std::ostringstream os; #ifdef EXV_HAVE_STRERROR_R const size_t n = 1024; -#if defined(__GLIBC__) && defined(_GNU_SOURCE) +#if defined(STRERROR_RETURNS_CHAR_PTR) char *buf = 0; char buf2[n]; std::memset(buf2, 0x0, n); ``` --- end body --- comments : --- start comments --- --- end comments ---

Hi, the tarball http://exiv2.org/builds/exiv2-0.27.0-Source.tar.gz contains hidden MacOSX files. As you're using globbing to install the install headers the macosx files get installed too. a) Do not use globbing b) Do not add hidden macosx files to tarballs

[0.27.0] The tarball on exiv2.org contains macosx files which get installed

https://api.github.com/repos/Exiv2/exiv2/issues/627/comments

https://github.com/Exiv2/exiv2/issues/627

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : [0.27.0] The tarball on exiv2.org contains macosx files which get installed Issue date: --- start body --- Hi, the tarball http://exiv2.org/builds/exiv2-0.27.0-Source.tar.gz contains hidden MacOSX files. As you're using globbing to install the install headers the macosx files get installed too. a) Do not use globbing b) Do not add hidden macosx files to tarballs --- end body --- comments : --- start comments --- --- end comments ---

The library name libxmp confilicts with the xmp project. See https://github.com/cmatsuoka/libxmp and https://xmp.sf.net/ Maybe rename it to `libexiv2-xmp`'?

[libxmp] Name conflict with xmp project

https://api.github.com/repos/Exiv2/exiv2/issues/624/comments

https://github.com/Exiv2/exiv2/issues/624

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : [libxmp] Name conflict with xmp project Issue date: --- start body --- The library name libxmp confilicts with the xmp project. See https://github.com/cmatsuoka/libxmp and https://xmp.sf.net/ Maybe rename it to `libexiv2-xmp`'? --- end body --- comments : --- start comments --- --- end comments ---

Please use GNUInstallDirs to install files so it is possible to overwrite those path for correct installation on distributions. See `cmake --help-module GNUInstallDirs`.

[cmake] Use GNUInstallDirs for doc and cmake config installation

https://api.github.com/repos/Exiv2/exiv2/issues/623/comments

https://github.com/Exiv2/exiv2/issues/623

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : [cmake] Use GNUInstallDirs for doc and cmake config installation Issue date: --- start body --- Please use GNUInstallDirs to install files so it is possible to overwrite those path for correct installation on distributions. See `cmake --help-module GNUInstallDirs`. --- end body --- comments : --- start comments --- --- end comments ---

Hi, I am trying to understand the NEF format. So in the Maker Note, there are some encrypted blocks (e.g. 0x97 = 151 = Color Balance), which is an array of bytes. It starts with a four-letter ASCII like "0205" and at some offset (4 or more, you specify it in "nikonArrayIdx" in makernote_int.cpp) the encrypted data starts (which we can decrypt). Does this data (after decryption) have any regular structure? [Yo describe it](http://www.exiv2.org/tags-nikon.html) as list of IFDs. In my case, I have these bytes for Color Balance: [1, 0, 1, 0, 1, 0, 1, 0, 0, 0, 0, 0, 0, 0, 2, 26, 1, 0, 1, 0, 1, 123, 3, 3, 0, 1, 0, 0, 0, 0, 0, 0, 1, 0, 3, 31, 3, 50, 5, 36, 4, 14, 4, 63, 2, 191, 4, 128, 4, 3, 4, 66, 2, 69, 0, 76, 0, 151, 0, 29, 2, 57, 2, 78, 0, 135, 0, 144, 2, 116, 2, 28, 65, 66, 68, 66, 67, 66, 68, 66, 3, 131, 0, 255, 0, 0, 0, 166, 32, 16, 3, 66, 33, 0, 3, 51, 0, 115, 0, 0, ... I don't see an IFD structure in it :( what these data mean?

NEF - encrypted data blocks

https://api.github.com/repos/Exiv2/exiv2/issues/621/comments

https://github.com/Exiv2/exiv2/issues/621

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : NEF - encrypted data blocks Issue date: --- start body --- Hi, I am trying to understand the NEF format. So in the Maker Note, there are some encrypted blocks (e.g. 0x97 = 151 = Color Balance), which is an array of bytes. It starts with a four-letter ASCII like "0205" and at some offset (4 or more, you specify it in "nikonArrayIdx" in makernote_int.cpp) the encrypted data starts (which we can decrypt). Does this data (after decryption) have any regular structure? [Yo describe it](http://www.exiv2.org/tags-nikon.html) as list of IFDs. In my case, I have these bytes for Color Balance: [1, 0, 1, 0, 1, 0, 1, 0, 0, 0, 0, 0, 0, 0, 2, 26, 1, 0, 1, 0, 1, 123, 3, 3, 0, 1, 0, 0, 0, 0, 0, 0, 1, 0, 3, 31, 3, 50, 5, 36, 4, 14, 4, 63, 2, 191, 4, 128, 4, 3, 4, 66, 2, 69, 0, 76, 0, 151, 0, 29, 2, 57, 2, 78, 0, 135, 0, 144, 2, 116, 2, 28, 65, 66, 68, 66, 67, 66, 68, 66, 3, 131, 0, 255, 0, 0, 0, 166, 32, 16, 3, 66, 33, 0, 3, 51, 0, 115, 0, 0, ... I don't see an IFD structure in it :( what these data mean? --- end body --- comments : --- start comments --- --- end comments ---

``` find exiv2-0.27.0-Source -name ".*" | wc -l 1335 ``` Probably none of those should end up in a release tarball, but worse, at least in case of include directory they also end up being installed.

0.27 tarball contains cruft

https://api.github.com/repos/Exiv2/exiv2/issues/620/comments

https://github.com/Exiv2/exiv2/issues/620

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : 0.27 tarball contains cruft Issue date: --- start body --- ``` find exiv2-0.27.0-Source -name ".*" | wc -l 1335 ``` Probably none of those should end up in a release tarball, but worse, at least in case of include directory they also end up being installed. --- end body --- comments : --- start comments --- --- end comments ---

Sony DSC-RX100M5A is a new compact camera (non-interchangeable lens). Seems like it can be detected by `Exif.Sony2.SonyModelID` = 369. Relevant output from `exiv2 -pt`: ``` Exif.Image.Make Ascii 5 SONY Exif.Image.Model Ascii 13 DSC-RX100M5A Exif.Image.Software Ascii 19 DSC-RX100M5A v1.00 Exif.Sony2.SonyModelID Short 1 (369) ``` File: 

Support Sony DSC-RX100M5A

https://api.github.com/repos/Exiv2/exiv2/issues/612/comments

https://github.com/Exiv2/exiv2/issues/612

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : Support Sony DSC-RX100M5A Issue date: --- start body --- Sony DSC-RX100M5A is a new compact camera (non-interchangeable lens). Seems like it can be detected by `Exif.Sony2.SonyModelID` = 369. Relevant output from `exiv2 -pt`: ``` Exif.Image.Make Ascii 5 SONY Exif.Image.Model Ascii 13 DSC-RX100M5A Exif.Image.Software Ascii 19 DSC-RX100M5A v1.00 Exif.Sony2.SonyModelID Short 1 (369) ``` File:  --- end body --- comments : --- start comments --- --- end comments ---

With https://github.com/Exiv2/exiv2/pull/601/commits/427bbad060f044f5857416af6633f3595af198ba, Exiv::Image::AutoPtr got replaced with Exiv::Image::UniquePtr. It is very hard to follow stable and "next" versions if you do that. Also, from semantics, I don't think that there is even a need to do the rename on Exiv2 level. It's still an automatic pointer for Exiv2::Image. So either, please make it throw a deprecation warning in 0.27 and remove it in 0.28 or re-add the type as a typedef to ::UniquePtr and mark it deprecated in 0.28, remving the typedef in 0.29

Please do not remove Public API without proper deprecation cycle

https://api.github.com/repos/Exiv2/exiv2/issues/611/comments

https://github.com/Exiv2/exiv2/issues/611

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : Please do not remove Public API without proper deprecation cycle Issue date: --- start body --- With https://github.com/Exiv2/exiv2/pull/601/commits/427bbad060f044f5857416af6633f3595af198ba, Exiv::Image::AutoPtr got replaced with Exiv::Image::UniquePtr. It is very hard to follow stable and "next" versions if you do that. Also, from semantics, I don't think that there is even a need to do the rename on Exiv2 level. It's still an automatic pointer for Exiv2::Image. So either, please make it throw a deprecation warning in 0.27 and remove it in 0.28 or re-add the type as a typedef to ::UniquePtr and mark it deprecated in 0.28, remving the typedef in 0.29 --- end body --- comments : --- start comments --- --- end comments ---

compilation trace : /mnt/devel/GIT/6.x/project/bundles/mxe/temp.build/ext_exiv2/ext_exiv2-prefix/src/ext_exiv2/src/futils.cpp: In function 'std::__cxx11::string Exiv2::getProcessPath()': /mnt/devel/GIT/6.x/project/bundles/mxe/temp.build/ext_exiv2/ext_exiv2-prefix/src/ext_exiv2/src/futils.cpp:467:9: error: 'HANDLE' was not declared in this scope HANDLE processHandle = NULL; ^ /mnt/devel/GIT/6.x/project/bundles/mxe/temp.build/ext_exiv2/ext_exiv2-prefix/src/ext_exiv2/src/futils.cpp:468:9: error: 'processHandle' was not declared in this scope processHandle = OpenProcess(PROCESS_QUERY_INFORMATION | PROCESS_VM_READ, FALSE, GetCurrentProcessId()); ^ /mnt/devel/GIT/6.x/project/bundles/mxe/temp.build/ext_exiv2/ext_exiv2-prefix/src/ext_exiv2/src/futils.cpp:468:37: error: 'PROCESS_QUERY_INFORMATION' was not declared in this scope processHandle = OpenProcess(PROCESS_QUERY_INFORMATION | PROCESS_VM_READ, FALSE, GetCurrentProcessId()); ^ /mnt/devel/GIT/6.x/project/bundles/mxe/temp.build/ext_exiv2/ext_exiv2-prefix/src/ext_exiv2/src/futils.cpp:468:65: error: 'PROCESS_VM_READ' was not declared in this scope processHandle = OpenProcess(PROCESS_QUERY_INFORMATION | PROCESS_VM_READ, FALSE, GetCurrentProcessId()); ^ /mnt/devel/GIT/6.x/project/bundles/mxe/temp.build/ext_exiv2/ext_exiv2-prefix/src/ext_exiv2/src/futils.cpp:468:82: error: 'FALSE' was not declared in this scope processHandle = OpenProcess(PROCESS_QUERY_INFORMATION | PROCESS_VM_READ, FALSE, GetCurrentProcessId()); ^ /mnt/devel/GIT/6.x/project/bundles/mxe/temp.build/ext_exiv2/ext_exiv2-prefix/src/ext_exiv2/src/futils.cpp:468:109: error: 'GetCurrentProcessId' was not declared in this scope processHandle = OpenProcess(PROCESS_QUERY_INFORMATION | PROCESS_VM_READ, FALSE, GetCurrentProcessId()); ^ /mnt/devel/GIT/6.x/project/bundles/mxe/temp.build/ext_exiv2/ext_exiv2-prefix/src/ext_exiv2/src/futils.cpp:468:110: error: 'OpenProcess' was not declared in this scope processHandle = OpenProcess(PROCESS_QUERY_INFORMATION | PROCESS_VM_READ, FALSE, GetCurrentProcessId()); ^ /mnt/devel/GIT/6.x/project/bundles/mxe/temp.build/ext_exiv2/ext_exiv2-prefix/src/ext_exiv2/src/futils.cpp:470:13: error: 'TCHAR' was not declared in this scope TCHAR filename[MAX_PATH]; ^ /mnt/devel/GIT/6.x/project/bundles/mxe/temp.build/ext_exiv2/ext_exiv2-prefix/src/ext_exiv2/src/futils.cpp:471:58: error: 'filename' was not declared in this scope if (GetModuleFileNameEx(processHandle, NULL, filename, MAX_PATH) != 0) { ^ /mnt/devel/GIT/6.x/project/bundles/mxe/temp.build/ext_exiv2/ext_exiv2-prefix/src/ext_exiv2/src/futils.cpp:471:68: error: 'MAX_PATH' was not declared in this scope if (GetModuleFileNameEx(processHandle, NULL, filename, MAX_PATH) != 0) { ^ /mnt/devel/GIT/6.x/project/bundles/mxe/temp.build/ext_exiv2/ext_exiv2-prefix/src/ext_exiv2/src/futils.cpp:471:76: error: 'GetModuleFileNameEx' was not declared in this scope if (GetModuleFileNameEx(processHandle, NULL, filename, MAX_PATH) != 0) { ^ /mnt/devel/GIT/6.x/project/bundles/mxe/temp.build/ext_exiv2/ext_exiv2-prefix/src/ext_exiv2/src/futils.cpp:474:38: error: 'CloseHandle' was not declared in this scope CloseHandle(processHandle); ^ src/CMakeFiles/exiv2lib.dir/build.make:363 : la recette pour la cible « src/CMakeFiles/exiv2lib.dir/futils.cpp.obj » a échouée gmake[6]: *** [src/CMakeFiles/exiv2lib.dir/futils.cpp.obj] Erreur 1 gmake[6]: *** Attente des tâches non terminées.... /mnt/devel/GIT/6.x/project/bundles/mxe/temp.build/ext_exiv2/ext_exiv2-prefix/src/ext_exiv2/src/convert.cpp:1510:10: warning: 'bool {anonymous}::convertStringCharsetWindows(std::__cxx11::string&, const char*, const char*)' defined but not used [-Wunused-function] bool convertStringCharsetWindows(std::string& str, const char* from, const char* to) ^ CMakeFiles/Makefile2:212 : la recette pour la cible « src/CMakeFiles/exiv2lib.dir/all » a échouée gmake[5]: *** [src/CMakeFiles/exiv2lib.dir/all] Erreur 2 Makefile:129 : la recette pour la cible « all » a échouée gmake[4]: *** [all] Erreur 2 ext_exiv2/CMakeFiles/ext_exiv2.dir/build.make:112 : la recette pour la cible « ext_exiv2/ext_exiv2-prefix/src/ext_exiv2-stamp/ext_exiv2-build » a échouée gmake[3]: *** [ext_exiv2/ext_exiv2-prefix/src/ext_exiv2-stamp/ext_exiv2-build] Erreur 2 CMakeFiles/Makefile2:415 : la recette pour la cible « ext_exiv2/CMakeFiles/ext_exiv2.dir/all » a échouée gmake[2]: *** [ext_exiv2/CMakeFiles/ext_exiv2.dir/all] Erreur 2 CMakeFiles/Makefile2:427 : la recette pour la cible « ext_exiv2/CMakeFiles/ext_exiv2.dir/rule » a échouée gmake[1]: *** [ext_exiv2/CMakeFiles/ext_exiv2.dir/rule] Erreur 2 Makefile:196 : la recette pour la cible « ext_exiv2 » a échouée gmake: *** [ext_exiv2] Erreur 2 FAILED COMMAND: ${MXE_BUILD_TARGETS}-cmake --build . --config RelWithDebInfo --target ext_exiv2 -- -j$CPU_CORES [gilles@localhost mxe]$ In fact, in futils.cpp there are 2 problems in getProcessPath() : std::string getProcessPath() { std::string ret("unknown"); #if defined(WIN32) <======= Must be _MSC_VER HANDLE processHandle = NULL; processHandle = OpenProcess(PROCESS_QUERY_INFORMATION | PROCESS_VM_READ, FALSE, GetCurrentProcessId()); if (processHandle != NULL) { TCHAR filename[MAX_PATH]; if (GetModuleFileNameEx(processHandle, NULL, filename, MAX_PATH) != 0) { ret = filename; } CloseHandle(processHandle); } #elif defined(__APPLE__) const int pid = getpid(); char pathbuf[PROC_PIDPATHINFO_MAXSIZE]; if (proc_pidpath (pid, pathbuf, sizeof(pathbuf)) > 0) { ret = pathbuf; } #elif defined(__linux__) || defined(__CYGWIN__) || defined(__MINGW__) // http://stackoverflow.com/questions/606041/how-do-i-get-the-path-of-a-process-in-unix-linux char proc[100]; char path[500]; sprintf(proc,"/proc/%d/exe", getpid()); ssize_t l = readlink (proc, path,sizeof(path)-1); << =========== Do not exists under Mingw if (l>0) { path[l]=0; ret = path; } #endif #if defined(WIN32) const size_t idxLastSeparator = ret.find_last_of('\\'); #else const size_t idxLastSeparator = ret.find_last_of('/'); #endif return ret.substr(0, idxLastSeparator); } // --------------------------------------- I take a look under Qt5 code which is fully portable and support very well MinGW, the readlink() call is wrapped with access() Gilles Caulier

0.27 : compilation is broken with NLS option enabled under MinGW

https://api.github.com/repos/Exiv2/exiv2/issues/610/comments

https://github.com/Exiv2/exiv2/issues/610

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : 0.27 : compilation is broken with NLS option enabled under MinGW Issue date: --- start body --- compilation trace : /mnt/devel/GIT/6.x/project/bundles/mxe/temp.build/ext_exiv2/ext_exiv2-prefix/src/ext_exiv2/src/futils.cpp: In function 'std::__cxx11::string Exiv2::getProcessPath()': /mnt/devel/GIT/6.x/project/bundles/mxe/temp.build/ext_exiv2/ext_exiv2-prefix/src/ext_exiv2/src/futils.cpp:467:9: error: 'HANDLE' was not declared in this scope HANDLE processHandle = NULL; ^ /mnt/devel/GIT/6.x/project/bundles/mxe/temp.build/ext_exiv2/ext_exiv2-prefix/src/ext_exiv2/src/futils.cpp:468:9: error: 'processHandle' was not declared in this scope processHandle = OpenProcess(PROCESS_QUERY_INFORMATION | PROCESS_VM_READ, FALSE, GetCurrentProcessId()); ^ /mnt/devel/GIT/6.x/project/bundles/mxe/temp.build/ext_exiv2/ext_exiv2-prefix/src/ext_exiv2/src/futils.cpp:468:37: error: 'PROCESS_QUERY_INFORMATION' was not declared in this scope processHandle = OpenProcess(PROCESS_QUERY_INFORMATION | PROCESS_VM_READ, FALSE, GetCurrentProcessId()); ^ /mnt/devel/GIT/6.x/project/bundles/mxe/temp.build/ext_exiv2/ext_exiv2-prefix/src/ext_exiv2/src/futils.cpp:468:65: error: 'PROCESS_VM_READ' was not declared in this scope processHandle = OpenProcess(PROCESS_QUERY_INFORMATION | PROCESS_VM_READ, FALSE, GetCurrentProcessId()); ^ /mnt/devel/GIT/6.x/project/bundles/mxe/temp.build/ext_exiv2/ext_exiv2-prefix/src/ext_exiv2/src/futils.cpp:468:82: error: 'FALSE' was not declared in this scope processHandle = OpenProcess(PROCESS_QUERY_INFORMATION | PROCESS_VM_READ, FALSE, GetCurrentProcessId()); ^ /mnt/devel/GIT/6.x/project/bundles/mxe/temp.build/ext_exiv2/ext_exiv2-prefix/src/ext_exiv2/src/futils.cpp:468:109: error: 'GetCurrentProcessId' was not declared in this scope processHandle = OpenProcess(PROCESS_QUERY_INFORMATION | PROCESS_VM_READ, FALSE, GetCurrentProcessId()); ^ /mnt/devel/GIT/6.x/project/bundles/mxe/temp.build/ext_exiv2/ext_exiv2-prefix/src/ext_exiv2/src/futils.cpp:468:110: error: 'OpenProcess' was not declared in this scope processHandle = OpenProcess(PROCESS_QUERY_INFORMATION | PROCESS_VM_READ, FALSE, GetCurrentProcessId()); ^ /mnt/devel/GIT/6.x/project/bundles/mxe/temp.build/ext_exiv2/ext_exiv2-prefix/src/ext_exiv2/src/futils.cpp:470:13: error: 'TCHAR' was not declared in this scope TCHAR filename[MAX_PATH]; ^ /mnt/devel/GIT/6.x/project/bundles/mxe/temp.build/ext_exiv2/ext_exiv2-prefix/src/ext_exiv2/src/futils.cpp:471:58: error: 'filename' was not declared in this scope if (GetModuleFileNameEx(processHandle, NULL, filename, MAX_PATH) != 0) { ^ /mnt/devel/GIT/6.x/project/bundles/mxe/temp.build/ext_exiv2/ext_exiv2-prefix/src/ext_exiv2/src/futils.cpp:471:68: error: 'MAX_PATH' was not declared in this scope if (GetModuleFileNameEx(processHandle, NULL, filename, MAX_PATH) != 0) { ^ /mnt/devel/GIT/6.x/project/bundles/mxe/temp.build/ext_exiv2/ext_exiv2-prefix/src/ext_exiv2/src/futils.cpp:471:76: error: 'GetModuleFileNameEx' was not declared in this scope if (GetModuleFileNameEx(processHandle, NULL, filename, MAX_PATH) != 0) { ^ /mnt/devel/GIT/6.x/project/bundles/mxe/temp.build/ext_exiv2/ext_exiv2-prefix/src/ext_exiv2/src/futils.cpp:474:38: error: 'CloseHandle' was not declared in this scope CloseHandle(processHandle); ^ src/CMakeFiles/exiv2lib.dir/build.make:363 : la recette pour la cible « src/CMakeFiles/exiv2lib.dir/futils.cpp.obj » a échouée gmake[6]: *** [src/CMakeFiles/exiv2lib.dir/futils.cpp.obj] Erreur 1 gmake[6]: *** Attente des tâches non terminées.... /mnt/devel/GIT/6.x/project/bundles/mxe/temp.build/ext_exiv2/ext_exiv2-prefix/src/ext_exiv2/src/convert.cpp:1510:10: warning: 'bool {anonymous}::convertStringCharsetWindows(std::__cxx11::string&, const char*, const char*)' defined but not used [-Wunused-function] bool convertStringCharsetWindows(std::string& str, const char* from, const char* to) ^ CMakeFiles/Makefile2:212 : la recette pour la cible « src/CMakeFiles/exiv2lib.dir/all » a échouée gmake[5]: *** [src/CMakeFiles/exiv2lib.dir/all] Erreur 2 Makefile:129 : la recette pour la cible « all » a échouée gmake[4]: *** [all] Erreur 2 ext_exiv2/CMakeFiles/ext_exiv2.dir/build.make:112 : la recette pour la cible « ext_exiv2/ext_exiv2-prefix/src/ext_exiv2-stamp/ext_exiv2-build » a échouée gmake[3]: *** [ext_exiv2/ext_exiv2-prefix/src/ext_exiv2-stamp/ext_exiv2-build] Erreur 2 CMakeFiles/Makefile2:415 : la recette pour la cible « ext_exiv2/CMakeFiles/ext_exiv2.dir/all » a échouée gmake[2]: *** [ext_exiv2/CMakeFiles/ext_exiv2.dir/all] Erreur 2 CMakeFiles/Makefile2:427 : la recette pour la cible « ext_exiv2/CMakeFiles/ext_exiv2.dir/rule » a échouée gmake[1]: *** [ext_exiv2/CMakeFiles/ext_exiv2.dir/rule] Erreur 2 Makefile:196 : la recette pour la cible « ext_exiv2 » a échouée gmake: *** [ext_exiv2] Erreur 2 FAILED COMMAND: ${MXE_BUILD_TARGETS}-cmake --build . --config RelWithDebInfo --target ext_exiv2 -- -j$CPU_CORES [gilles@localhost mxe]$ In fact, in futils.cpp there are 2 problems in getProcessPath() : std::string getProcessPath() { std::string ret("unknown"); #if defined(WIN32) <======= Must be _MSC_VER HANDLE processHandle = NULL; processHandle = OpenProcess(PROCESS_QUERY_INFORMATION | PROCESS_VM_READ, FALSE, GetCurrentProcessId()); if (processHandle != NULL) { TCHAR filename[MAX_PATH]; if (GetModuleFileNameEx(processHandle, NULL, filename, MAX_PATH) != 0) { ret = filename; } CloseHandle(processHandle); } #elif defined(__APPLE__) const int pid = getpid(); char pathbuf[PROC_PIDPATHINFO_MAXSIZE]; if (proc_pidpath (pid, pathbuf, sizeof(pathbuf)) > 0) { ret = pathbuf; } #elif defined(__linux__) || defined(__CYGWIN__) || defined(__MINGW__) // http://stackoverflow.com/questions/606041/how-do-i-get-the-path-of-a-process-in-unix-linux char proc[100]; char path[500]; sprintf(proc,"/proc/%d/exe", getpid()); ssize_t l = readlink (proc, path,sizeof(path)-1); << =========== Do not exists under Mingw if (l>0) { path[l]=0; ret = path; } #endif #if defined(WIN32) const size_t idxLastSeparator = ret.find_last_of('\\'); #else const size_t idxLastSeparator = ret.find_last_of('/'); #endif return ret.substr(0, idxLastSeparator); } // --------------------------------------- I take a look under Qt5 code which is fully portable and support very well MinGW, the readlink() call is wrapped with access() Gilles Caulier --- end body --- comments : --- start comments --- --- end comments ---

Hi, I am a beginner and I have no clear idea how to proceed with the installation of exiv2. Following the instructions: I download the zip exiv2-master and I unzip it I open the powershell of Windows and I open the directory in which power exiv2-master in this way: cd D:\Download\exiv2-master Then I create the folder build and I open it in this way: mkdir build cd build Then I run this command: cmake .. -G "Unix Makefiles" and it returns me this message: -- The CXX compiler identification is unknown -- The C compiler identification is unknown CMake Error at CMakeLists.txt:3 (project): No CMAKE_CXX_COMPILER could be found. Tell CMake where to find the compiler by setting either the environment variable "CXX" or the CMake cache entry CMAKE_CXX_COMPILER to the full path to the compiler, or to the compiler name if it is in the PATH. -- Check for working C compiler: C:/cygwin64/bin/cc -- Check for working C compiler: C:/cygwin64/bin/cc -- works -- Detecting C compiler ABI info -- Detecting C compiler ABI info - done -- Configuring incomplete, errors occurred! See also "D:/Download/exiv2-master/build/CMakeFiles/CMakeOutput.log". See also "D:/Download/exiv2-master/build/CMakeFiles/CMakeError.log". I have no idea of what I have to do now to solve this problem. I tried to find something on google about the C++11 compiler but I didn't undestand what I have to do. Someone can help me please? Thank you very much for your help!

install problem

https://api.github.com/repos/Exiv2/exiv2/issues/609/comments

https://github.com/Exiv2/exiv2/issues/609

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : install problem Issue date: --- start body --- Hi, I am a beginner and I have no clear idea how to proceed with the installation of exiv2. Following the instructions: I download the zip exiv2-master and I unzip it I open the powershell of Windows and I open the directory in which power exiv2-master in this way: cd D:\Download\exiv2-master Then I create the folder build and I open it in this way: mkdir build cd build Then I run this command: cmake .. -G "Unix Makefiles" and it returns me this message: -- The CXX compiler identification is unknown -- The C compiler identification is unknown CMake Error at CMakeLists.txt:3 (project): No CMAKE_CXX_COMPILER could be found. Tell CMake where to find the compiler by setting either the environment variable "CXX" or the CMake cache entry CMAKE_CXX_COMPILER to the full path to the compiler, or to the compiler name if it is in the PATH. -- Check for working C compiler: C:/cygwin64/bin/cc -- Check for working C compiler: C:/cygwin64/bin/cc -- works -- Detecting C compiler ABI info -- Detecting C compiler ABI info - done -- Configuring incomplete, errors occurred! See also "D:/Download/exiv2-master/build/CMakeFiles/CMakeOutput.log". See also "D:/Download/exiv2-master/build/CMakeFiles/CMakeError.log". I have no idea of what I have to do now to solve this problem. I tried to find something on google about the C++11 compiler but I didn't undestand what I have to do. Someone can help me please? Thank you very much for your help! --- end body --- comments : --- start comments --- --- end comments ---

There are several issues with this: 1) We need to use #include <curl/curl.h> when EXIV2_ENABLE_CURL is set 2) We need to link CURL_LIBRARIES to conntest when EXIV2_ENABLE_CURL is set 3) There's a typo in the comment at the top of samples/conntest.cpp The work-around is to not use CURL and we'll fix this in v0.27.1. ``` 605 rmills@rmillsmm:~/gnu/github/exiv2/exiv2/build $ git diff diff --git a/samples/CMakeLists.txt b/samples/CMakeLists.txt index a3e491e..46aaa53 100644 --- a/samples/CMakeLists.txt +++ b/samples/CMakeLists.txt @@ -90,6 +90,9 @@ endif() # ****************************************************************************** # connection test application add_executable(conntest conntest.cpp) +if ( EXIV2_ENABLE_CURL ) +target_link_libraries( conntest PRIVATE ${CURL_LIBRARIES} ) +endif() list(APPEND APPLICATIONS conntest) # ****************************************************************************** diff --git a/samples/conntest.cpp b/samples/conntest.cpp index 22fce2c..bcef8d1 100644 --- a/samples/conntest.cpp +++ b/samples/conntest.cpp @@ -1,5 +1,5 @@ // ***************************************************************** -*- C++ -*- -// con-test.cpp +// conntest.cpp // Tester application for testing the http/https/ftp/ssh/sftp connection #include <exiv2/exiv2.hpp> @@ -7,6 +7,10 @@ #include <iostream> #include <cstdlib> +#ifdef EXV_USE_CURL +#include <curl/curl.h> +#endif + void httpcon(const std::string& url, bool useHttp1_0 = false) { Exiv2::Dictionary response; Exiv2::Dictionary request; 606 rmills@rmillsmm:~/gnu/github/exiv2/exiv2/build $ ```

samples/conntest.cpp and -DEXIV2_ENABLE_CURL=1 is broken

https://api.github.com/repos/Exiv2/exiv2/issues/604/comments

https://github.com/Exiv2/exiv2/issues/604

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : samples/conntest.cpp and -DEXIV2_ENABLE_CURL=1 is broken Issue date: --- start body --- There are several issues with this: 1) We need to use #include <curl/curl.h> when EXIV2_ENABLE_CURL is set 2) We need to link CURL_LIBRARIES to conntest when EXIV2_ENABLE_CURL is set 3) There's a typo in the comment at the top of samples/conntest.cpp The work-around is to not use CURL and we'll fix this in v0.27.1. ``` 605 rmills@rmillsmm:~/gnu/github/exiv2/exiv2/build $ git diff diff --git a/samples/CMakeLists.txt b/samples/CMakeLists.txt index a3e491e..46aaa53 100644 --- a/samples/CMakeLists.txt +++ b/samples/CMakeLists.txt @@ -90,6 +90,9 @@ endif() # ****************************************************************************** # connection test application add_executable(conntest conntest.cpp) +if ( EXIV2_ENABLE_CURL ) +target_link_libraries( conntest PRIVATE ${CURL_LIBRARIES} ) +endif() list(APPEND APPLICATIONS conntest) # ****************************************************************************** diff --git a/samples/conntest.cpp b/samples/conntest.cpp index 22fce2c..bcef8d1 100644 --- a/samples/conntest.cpp +++ b/samples/conntest.cpp @@ -1,5 +1,5 @@ // ***************************************************************** -*- C++ -*- -// con-test.cpp +// conntest.cpp // Tester application for testing the http/https/ftp/ssh/sftp connection #include <exiv2/exiv2.hpp> @@ -7,6 +7,10 @@ #include <iostream> #include <cstdlib> +#ifdef EXV_USE_CURL +#include <curl/curl.h> +#endif + void httpcon(const std::string& url, bool useHttp1_0 = false) { Exiv2::Dictionary response; Exiv2::Dictionary request; 606 rmills@rmillsmm:~/gnu/github/exiv2/exiv2/build $ ``` --- end body --- comments : --- start comments --- --- end comments ---

cmake/FindSSH.cmake reports being unable to find libssh. There doesn't seem to be any code in conanfile.py to find libssh. It's working OK on MacOS-X because I build using "pure" CMake (no conan) and libssh is install in /usr/local/ The work-around is to not request this feature in v0.27 We can fix this in 0.27.1 ``` ... -- Found ZLIB: C:/Users/rmills/.conan/data/zlib/1.2.11/conan/stable/package/6cc50b139b9c3d27b3e9042d5f5372d327b3a9f7/lib/zlib.lib (found version "1.2.11") -- Found CURL: C:/Users/rmills/.conan/data/libcurl/7.60.0/bincrafters/stable/package/ae6191bc1194317ff7cdbc84d4d302ab49975618/lib/libcurl.lib (found version "7.60.0") CMake Error at C:/Program Files/CMake/share/cmake-3.12/Modules/FindPackageHandleStandardArgs.cmake:137 (message): Could NOT find SSH (missing: SSH_LIBRARY SSH_INCLUDE_DIR) Call Stack (most recent call first): C:/Program Files/CMake/share/cmake-3.12/Modules/FindPackageHandleStandardArgs.cmake:378 (_FPHSA_FAILURE_MESSAGE) cmake/FindSSH.cmake:20 (FIND_PACKAGE_HANDLE_STANDARD_ARGS) cmake/findDependencies.cmake:23 (find_package) CMakeLists.txt:61 (include) -- Configuring incomplete, errors occurred! See also "C:/Users/rmills/gnu/github/exiv2/buildserver/build/CMakeFiles/CMakeOutput.log". See also "C:/Users/rmills/gnu/github/exiv2/buildserver/build/CMakeFiles/CMakeError.log". ```

cmake+conan with -DEXIV2_ENABLE_LIBSSH=On is broken on msvc

https://api.github.com/repos/Exiv2/exiv2/issues/603/comments

https://github.com/Exiv2/exiv2/issues/603

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : cmake+conan with -DEXIV2_ENABLE_LIBSSH=On is broken on msvc Issue date: --- start body --- cmake/FindSSH.cmake reports being unable to find libssh. There doesn't seem to be any code in conanfile.py to find libssh. It's working OK on MacOS-X because I build using "pure" CMake (no conan) and libssh is install in /usr/local/ The work-around is to not request this feature in v0.27 We can fix this in 0.27.1 ``` ... -- Found ZLIB: C:/Users/rmills/.conan/data/zlib/1.2.11/conan/stable/package/6cc50b139b9c3d27b3e9042d5f5372d327b3a9f7/lib/zlib.lib (found version "1.2.11") -- Found CURL: C:/Users/rmills/.conan/data/libcurl/7.60.0/bincrafters/stable/package/ae6191bc1194317ff7cdbc84d4d302ab49975618/lib/libcurl.lib (found version "7.60.0") CMake Error at C:/Program Files/CMake/share/cmake-3.12/Modules/FindPackageHandleStandardArgs.cmake:137 (message): Could NOT find SSH (missing: SSH_LIBRARY SSH_INCLUDE_DIR) Call Stack (most recent call first): C:/Program Files/CMake/share/cmake-3.12/Modules/FindPackageHandleStandardArgs.cmake:378 (_FPHSA_FAILURE_MESSAGE) cmake/FindSSH.cmake:20 (FIND_PACKAGE_HANDLE_STANDARD_ARGS) cmake/findDependencies.cmake:23 (find_package) CMakeLists.txt:61 (include) -- Configuring incomplete, errors occurred! See also "C:/Users/rmills/gnu/github/exiv2/buildserver/build/CMakeFiles/CMakeOutput.log". See also "C:/Users/rmills/gnu/github/exiv2/buildserver/build/CMakeFiles/CMakeError.log". ``` --- end body --- comments : --- start comments --- --- end comments ---

There are ugly compiler warnings coming from msvc (2017 and 2015, possibly others) from quicktimevideo.cpp and riffvideo.cpp The built code is OK and passes the test suite. ``` riffvideo.cpp C:\Program Files (x86)\Microsoft Visual Studio\2017\Community\VC\Tools\MSVC\14.10.25017\include\xutility(3161): warning C4244: 'argument': conversion from 'const int64_t' to 'long', possible loss of data (compiling source file C:\Users\rmills\gnu\github\exiv2\buildserver\src\quicktimevideo.cpp) [C:\Users\rmills\gnu\github\exiv2\buildserver\build\src\exiv2lib.vcxproj] C:\Program Files (x86)\Microsoft Visual Studio\2017\Community\VC\Tools\MSVC\14.10.25017\include\xutility(3180): note: see reference to function template instantiation '_InIt std::_Find_unchecked1<_InIt,_Ty>(_InIt,_InIt,const _Ty &,std::false_type)' being compiled with [ _InIt=const Exiv2::Internal::TagDetails *, _Ty=int64_t ] (compiling source file C:\Users\rmills\gnu\github\exiv2\buildserver\src\quicktimevideo.cpp) ``` One fix is to cast `(uint32_t)` in quicktimevideo.cpp#1333 ``` switch(i) { case GraphicsMode: td = find(graphicsModetags, (uint32_t) returnBufValue(buf,2)); if(td) xmpData_["Xmp.video.GraphicsMode"] = exvGettext(td->label_); break; case OpColor: xmpData_["Xmp.video.OpColor"] = returnBufValue(buf,2); break; default: break; } ``` and in riffvideo.cpp#1269 ``` const char* RiffVideo::printAudioEncoding(uint64_t i) { const TagDetails* td; td = find(audioEncodingValues ,(uint32_t) i); if(td) return exvGettext(td->label_); return "Undefined"; } // RiffVideo::printAudioEncoding ```

msvc compiler warning with -DEXIV2_ENABLE_VIDEO=On

https://api.github.com/repos/Exiv2/exiv2/issues/602/comments

https://github.com/Exiv2/exiv2/issues/602

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : msvc compiler warning with -DEXIV2_ENABLE_VIDEO=On Issue date: --- start body --- There are ugly compiler warnings coming from msvc (2017 and 2015, possibly others) from quicktimevideo.cpp and riffvideo.cpp The built code is OK and passes the test suite. ``` riffvideo.cpp C:\Program Files (x86)\Microsoft Visual Studio\2017\Community\VC\Tools\MSVC\14.10.25017\include\xutility(3161): warning C4244: 'argument': conversion from 'const int64_t' to 'long', possible loss of data (compiling source file C:\Users\rmills\gnu\github\exiv2\buildserver\src\quicktimevideo.cpp) [C:\Users\rmills\gnu\github\exiv2\buildserver\build\src\exiv2lib.vcxproj] C:\Program Files (x86)\Microsoft Visual Studio\2017\Community\VC\Tools\MSVC\14.10.25017\include\xutility(3180): note: see reference to function template instantiation '_InIt std::_Find_unchecked1<_InIt,_Ty>(_InIt,_InIt,const _Ty &,std::false_type)' being compiled with [ _InIt=const Exiv2::Internal::TagDetails *, _Ty=int64_t ] (compiling source file C:\Users\rmills\gnu\github\exiv2\buildserver\src\quicktimevideo.cpp) ``` One fix is to cast `(uint32_t)` in quicktimevideo.cpp#1333 ``` switch(i) { case GraphicsMode: td = find(graphicsModetags, (uint32_t) returnBufValue(buf,2)); if(td) xmpData_["Xmp.video.GraphicsMode"] = exvGettext(td->label_); break; case OpColor: xmpData_["Xmp.video.OpColor"] = returnBufValue(buf,2); break; default: break; } ``` and in riffvideo.cpp#1269 ``` const char* RiffVideo::printAudioEncoding(uint64_t i) { const TagDetails* td; td = find(audioEncodingValues ,(uint32_t) i); if(td) return exvGettext(td->label_); return "Undefined"; } // RiffVideo::printAudioEncoding ``` --- end body --- comments : --- start comments --- --- end comments ---

**Program version**: `exiv2 0.27.0.3` **Program output with sample picture**: ``` exiv2 -pt DSC_0069.JPG | grep -ai lens Exif.Nikon3.LensType Byte 1 D G VR Exif.Nikon3.Lens Rational 4 70-300mm F4-5.6 Exif.Nikon3.LensFStops Undefined 4 5.91667 Exif.NikonLd3.LensIDNumber Byte 1 207 Exif.NikonLd3.LensFStops Byte 1 F5.9 ``` **Sample picture (JPG + NEF) link**: https://mega.nz/#!my51lI7D!iNzUKqCAm6KExGX8ztHyhBOeF7f6vpLHQMwAGSSxtwY **Sample picture (JPG) (taken with lens cap on)**:  **Lens model url**: https://www.tamron.jp/en/product/lenses/a030.html **Lens model picture**: 

Tamron 70-300mm (Model: A030) (Nikon) not recognized

https://api.github.com/repos/Exiv2/exiv2/issues/598/comments

https://github.com/Exiv2/exiv2/issues/598

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : Tamron 70-300mm (Model: A030) (Nikon) not recognized Issue date: --- start body --- **Program version**: `exiv2 0.27.0.3` **Program output with sample picture**: ``` exiv2 -pt DSC_0069.JPG | grep -ai lens Exif.Nikon3.LensType Byte 1 D G VR Exif.Nikon3.Lens Rational 4 70-300mm F4-5.6 Exif.Nikon3.LensFStops Undefined 4 5.91667 Exif.NikonLd3.LensIDNumber Byte 1 207 Exif.NikonLd3.LensFStops Byte 1 F5.9 ``` **Sample picture (JPG + NEF) link**: https://mega.nz/#!my51lI7D!iNzUKqCAm6KExGX8ztHyhBOeF7f6vpLHQMwAGSSxtwY **Sample picture (JPG) (taken with lens cap on)**:  **Lens model url**: https://www.tamron.jp/en/product/lenses/a030.html **Lens model picture**:  --- end body --- comments : --- start comments --- --- end comments ---

We have been notified of a problem with digiKam writing metadata to PNG files, after which the ICC profile is broken. https://bugs.kde.org/show_bug.cgi?id=401970 Exiv2 uses a fixed profile name "ICC PROFILE" with 11 characters + zero separator. The profile name can vary between 1-79 characters. Also a warning of the libpng with wrong ICC profile CRC checksum fixes this patch. Maik [pngIccProfileName.txt](https://github.com/Exiv2/exiv2/files/2683451/pngIccProfileName.txt)

Patch for broken ICC profile in PNG files

https://api.github.com/repos/Exiv2/exiv2/issues/597/comments

https://github.com/Exiv2/exiv2/issues/597

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : Patch for broken ICC profile in PNG files Issue date: --- start body --- We have been notified of a problem with digiKam writing metadata to PNG files, after which the ICC profile is broken. https://bugs.kde.org/show_bug.cgi?id=401970 Exiv2 uses a fixed profile name "ICC PROFILE" with 11 characters + zero separator. The profile name can vary between 1-79 characters. Also a warning of the libpng with wrong ICC profile CRC checksum fixes this patch. Maik [pngIccProfileName.txt](https://github.com/Exiv2/exiv2/files/2683451/pngIccProfileName.txt) --- end body --- comments : --- start comments --- --- end comments ---

https://github.com/Exiv2/exiv2/blob/af7ca268095258abe7df20b40ec1dc4b5c3d1c46/src/asfvideo.cpp#L362 [3_1__6557887998130913604972615.wmv.zip](https://github.com/Exiv2/exiv2/files/2671313/3_1__6557887998130913604972615.wmv.zip)

Attached a wmv file for the ASFVIDEO.cpp the while loop continues and never stops

https://api.github.com/repos/Exiv2/exiv2/issues/593/comments

https://github.com/Exiv2/exiv2/issues/593

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : Attached a wmv file for the ASFVIDEO.cpp the while loop continues and never stops Issue date: --- start body --- https://github.com/Exiv2/exiv2/blob/af7ca268095258abe7df20b40ec1dc4b5c3d1c46/src/asfvideo.cpp#L362 [3_1__6557887998130913604972615.wmv.zip](https://github.com/Exiv2/exiv2/files/2671313/3_1__6557887998130913604972615.wmv.zip) --- end body --- comments : --- start comments --- --- end comments ---

In dealing with ASAN issues on MacOS-X, I see there are warning from the linker. I don't understand them. They were present in RC2. I have not searched further back. They might be benign. ``` [ 94%] Linking CXX shared library ../lib/libexiv2.dylib ld: warning: direct access in function 'Exiv2::XmpProperties::printProperty(std::__1::basic_ostream<char, std::__1::char_traits<char> >&, std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> > const&, Exiv2::Value const&)' from file 'CMakeFiles/exiv2lib.dir/properties.cpp.o' to global weak symbol 'typeinfo for std::__1::basic_ostream<char, std::__1::char_traits<char> >& (std::__1::basic_ostream<char, std::__1::char_traits<char> >&, Exiv2::Value const&, Exiv2::ExifData const*)' from file 'CMakeFiles/exiv2lib.dir/exif.cpp.o' means the weak symbol cannot be overridden at runtime. This was likely caused by different translation units being compiled with different visibility settings. ld: warning: direct access in function 'Exiv2::Internal::CanonMakerNote::printCsLensType(std::__1::basic_ostream<char, std::__1::char_traits<char> >&, Exiv2::Value const&, Exiv2::ExifData const*)' from file 'CMakeFiles/exiv2lib_int.dir/canonmn_int.cpp.o' to global weak symbol 'typeinfo for std::__1::basic_ostream<char, std::__1::char_traits<char> >& (std::__1::basic_ostream<char, std::__1::char_traits<char> >&, Exiv2::Value const&, Exiv2::ExifData const*)' from file 'CMakeFiles/exiv2lib.dir/exif.cpp.o' means the weak symbol cannot be overridden at runtime. This was likely caused by different translation units being compiled with different visibility settings. ld: warning: direct access in function 'Exiv2::Internal::printMinoltaSonyLensID(std::__1::basic_ostream<char, std::__1::char_traits<char> >&, Exiv2::Value const&, Exiv2::ExifData const*)' from file 'CMakeFiles/exiv2lib_int.dir/minoltamn_int.cpp.o' to global weak symbol 'typeinfo for std::__1::basic_ostream<char, std::__1::char_traits<char> >& (std::__1::basic_ostream<char, std::__1::char_traits<char> >&, Exiv2::Value const&, Exiv2::ExifData const*)' from file 'CMakeFiles/exiv2lib.dir/exif.cpp.o' means the weak symbol cannot be overridden at runtime. This was likely caused by different translation units being compiled with different visibility settings. ld: warning: direct access in function 'Exiv2::Internal::printLensType(std::__1::basic_ostream<char, std::__1::char_traits<char> >&, Exiv2::Value const&, Exiv2::ExifData const*)' from file 'CMakeFiles/exiv2lib_int.dir/pentaxmn_int.cpp.o' to global weak symbol 'typeinfo for std::__1::basic_ostream<char, std::__1::char_traits<char> >& (std::__1::basic_ostream<char, std::__1::char_traits<char> >&, Exiv2::Value const&, Exiv2::ExifData const*)' from file 'CMakeFiles/exiv2lib.dir/exif.cpp.o' means the weak symbol cannot be overridden at runtime. This was likely caused by different translation units being compiled with different visibility settings. [ 94%] Built target exiv2lib Scanning dependencies of target exiv2 [ 95%] Building CXX object src/CMakeFiles/exiv2.dir/exiv2.cpp.o [ 96%] Building CXX object src/CMakeFiles/exiv2.dir/actions.cpp.o [ 97%] Building CXX object src/CMakeFiles/exiv2.dir/getopt.cpp.o [ 98%] Building CXX object src/CMakeFiles/exiv2.dir/utils.cpp.o [100%] Linking CXX executable ../bin/exiv2 [100%] Built target exiv2 686 rmills@rmillsmm:~/Downloads/foo/build $ cmake .. -DEXIV2_TEAM_USE_SANITIZERS=1 -DEXIV2_BUILD_SAMPLES=0 ; make ``` I've tried to make sense of the first of the 4 errors and it becomes: ``` ld: warning: direct access in function 'Exiv2::XmpProperties::printProperty(basic_ostream, const&, Exiv2::Value const&)' from file 'CMakeFiles/exiv2lib.dir/properties.cpp.o' to global weak symbol 'typeinfo for std::__1::basic_ostream&, Exiv2::Value const&, Exiv2::ExifData const*)' from file 'CMakeFiles/exiv2lib.dir/exif.cpp.o' means the weak means the weak symbol cannot be overridden at runtime. This was likely caused by different translation units being compiled with different visibility settings. ``` I don't know that this means. What I can see is that in properties.cpp, we have: ``` std::ostream& XmpProperties::printProperty(std::ostream& os, const std::string& key, const Value& value) { PrintFct fct = printValue; if (value.count() != 0) { const XmpPrintInfo* info = find(xmpPrintInfo, key); if (info) fct = info->printFct_; } return fct(os, value, 0); } ``` I think he's complaining about **`fct`** (no pun intended).

MacOS-X Linker warning when building with -DEXIV2_TEAM_USE_SANITIZERS=On

https://api.github.com/repos/Exiv2/exiv2/issues/591/comments

https://github.com/Exiv2/exiv2/issues/591

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : MacOS-X Linker warning when building with -DEXIV2_TEAM_USE_SANITIZERS=On Issue date: --- start body --- In dealing with ASAN issues on MacOS-X, I see there are warning from the linker. I don't understand them. They were present in RC2. I have not searched further back. They might be benign. ``` [ 94%] Linking CXX shared library ../lib/libexiv2.dylib ld: warning: direct access in function 'Exiv2::XmpProperties::printProperty(std::__1::basic_ostream<char, std::__1::char_traits<char> >&, std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> > const&, Exiv2::Value const&)' from file 'CMakeFiles/exiv2lib.dir/properties.cpp.o' to global weak symbol 'typeinfo for std::__1::basic_ostream<char, std::__1::char_traits<char> >& (std::__1::basic_ostream<char, std::__1::char_traits<char> >&, Exiv2::Value const&, Exiv2::ExifData const*)' from file 'CMakeFiles/exiv2lib.dir/exif.cpp.o' means the weak symbol cannot be overridden at runtime. This was likely caused by different translation units being compiled with different visibility settings. ld: warning: direct access in function 'Exiv2::Internal::CanonMakerNote::printCsLensType(std::__1::basic_ostream<char, std::__1::char_traits<char> >&, Exiv2::Value const&, Exiv2::ExifData const*)' from file 'CMakeFiles/exiv2lib_int.dir/canonmn_int.cpp.o' to global weak symbol 'typeinfo for std::__1::basic_ostream<char, std::__1::char_traits<char> >& (std::__1::basic_ostream<char, std::__1::char_traits<char> >&, Exiv2::Value const&, Exiv2::ExifData const*)' from file 'CMakeFiles/exiv2lib.dir/exif.cpp.o' means the weak symbol cannot be overridden at runtime. This was likely caused by different translation units being compiled with different visibility settings. ld: warning: direct access in function 'Exiv2::Internal::printMinoltaSonyLensID(std::__1::basic_ostream<char, std::__1::char_traits<char> >&, Exiv2::Value const&, Exiv2::ExifData const*)' from file 'CMakeFiles/exiv2lib_int.dir/minoltamn_int.cpp.o' to global weak symbol 'typeinfo for std::__1::basic_ostream<char, std::__1::char_traits<char> >& (std::__1::basic_ostream<char, std::__1::char_traits<char> >&, Exiv2::Value const&, Exiv2::ExifData const*)' from file 'CMakeFiles/exiv2lib.dir/exif.cpp.o' means the weak symbol cannot be overridden at runtime. This was likely caused by different translation units being compiled with different visibility settings. ld: warning: direct access in function 'Exiv2::Internal::printLensType(std::__1::basic_ostream<char, std::__1::char_traits<char> >&, Exiv2::Value const&, Exiv2::ExifData const*)' from file 'CMakeFiles/exiv2lib_int.dir/pentaxmn_int.cpp.o' to global weak symbol 'typeinfo for std::__1::basic_ostream<char, std::__1::char_traits<char> >& (std::__1::basic_ostream<char, std::__1::char_traits<char> >&, Exiv2::Value const&, Exiv2::ExifData const*)' from file 'CMakeFiles/exiv2lib.dir/exif.cpp.o' means the weak symbol cannot be overridden at runtime. This was likely caused by different translation units being compiled with different visibility settings. [ 94%] Built target exiv2lib Scanning dependencies of target exiv2 [ 95%] Building CXX object src/CMakeFiles/exiv2.dir/exiv2.cpp.o [ 96%] Building CXX object src/CMakeFiles/exiv2.dir/actions.cpp.o [ 97%] Building CXX object src/CMakeFiles/exiv2.dir/getopt.cpp.o [ 98%] Building CXX object src/CMakeFiles/exiv2.dir/utils.cpp.o [100%] Linking CXX executable ../bin/exiv2 [100%] Built target exiv2 686 rmills@rmillsmm:~/Downloads/foo/build $ cmake .. -DEXIV2_TEAM_USE_SANITIZERS=1 -DEXIV2_BUILD_SAMPLES=0 ; make ``` I've tried to make sense of the first of the 4 errors and it becomes: ``` ld: warning: direct access in function 'Exiv2::XmpProperties::printProperty(basic_ostream, const&, Exiv2::Value const&)' from file 'CMakeFiles/exiv2lib.dir/properties.cpp.o' to global weak symbol 'typeinfo for std::__1::basic_ostream&, Exiv2::Value const&, Exiv2::ExifData const*)' from file 'CMakeFiles/exiv2lib.dir/exif.cpp.o' means the weak means the weak symbol cannot be overridden at runtime. This was likely caused by different translation units being compiled with different visibility settings. ``` I don't know that this means. What I can see is that in properties.cpp, we have: ``` std::ostream& XmpProperties::printProperty(std::ostream& os, const std::string& key, const Value& value) { PrintFct fct = printValue; if (value.count() != 0) { const XmpPrintInfo* info = find(xmpPrintInfo, key); if (info) fct = info->printFct_; } return fct(os, value, 0); } ``` I think he's complaining about **`fct`** (no pun intended). --- end body --- comments : --- start comments --- --- end comments ---

I have tested them in latest version 0.27-RC3. 1. ./exiv2 -pR [pngimage-heap-bof-poc-1](https://github.com/TeamSeri0us/pocs/blob/master/exiv2/20181206/pngimage-heap-bof-poc-1) 2. ./exiv2 -Y 2011 -O 02 -D 29 adjust [tiffimage_int-out-of-bound-read-poc-2](https://github.com/TeamSeri0us/pocs/blob/master/exiv2/20181206/tiffimage_int-out-of-bound-read-poc-2). 3. ./exiv2 -M'set Xmp.dc.title lang="de-DE" Euros' [jp2image-heap-bof-poc-3](https://github.com/TeamSeri0us/pocs/blob/master/exiv2/20181206/jp2image-heap-bof-poc-3). 4. ./iptcprint [abort-poc-4](https://github.com/TeamSeri0us/pocs/blob/master/exiv2/20181206/abort-poc-4) 5. ./exiv2 insert [jp2image-infiniteloop-poc-5](https://github.com/TeamSeri0us/pocs/blob/master/exiv2/20181206/jp2image-infiniteloop-poc-5) And, more info can be found at https://github.com/TeamSeri0us/pocs/tree/master/exiv2/20181206.

Several bugs in exiv2 0.27-rc3

https://api.github.com/repos/Exiv2/exiv2/issues/590/comments

https://github.com/Exiv2/exiv2/issues/590

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : Several bugs in exiv2 0.27-rc3 Issue date: --- start body --- I have tested them in latest version 0.27-RC3. 1. ./exiv2 -pR [pngimage-heap-bof-poc-1](https://github.com/TeamSeri0us/pocs/blob/master/exiv2/20181206/pngimage-heap-bof-poc-1) 2. ./exiv2 -Y 2011 -O 02 -D 29 adjust [tiffimage_int-out-of-bound-read-poc-2](https://github.com/TeamSeri0us/pocs/blob/master/exiv2/20181206/tiffimage_int-out-of-bound-read-poc-2). 3. ./exiv2 -M'set Xmp.dc.title lang="de-DE" Euros' [jp2image-heap-bof-poc-3](https://github.com/TeamSeri0us/pocs/blob/master/exiv2/20181206/jp2image-heap-bof-poc-3). 4. ./iptcprint [abort-poc-4](https://github.com/TeamSeri0us/pocs/blob/master/exiv2/20181206/abort-poc-4) 5. ./exiv2 insert [jp2image-infiniteloop-poc-5](https://github.com/TeamSeri0us/pocs/blob/master/exiv2/20181206/jp2image-infiniteloop-poc-5) And, more info can be found at https://github.com/TeamSeri0us/pocs/tree/master/exiv2/20181206. --- end body --- comments : --- start comments --- --- end comments ---

http://dev.exiv2.org/boards/3/topics/3249?r=3250

Sidecar files only update on the first modification

https://api.github.com/repos/Exiv2/exiv2/issues/589/comments

https://github.com/Exiv2/exiv2/issues/589

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : Sidecar files only update on the first modification Issue date: --- start body --- http://dev.exiv2.org/boards/3/topics/3249?r=3250 --- end body --- comments : --- start comments --- --- end comments ---

The `Image` object has `protected` members `pixelHeight_` and `pixelWidth_` (as `int`) and corresponding (overrideable) `public` accessors. However none of the video interfaces overwrite or set these values, instead they all, except `riffvideo` have their own `uint64_t` variables to track height/width. A user would very well expect that an `Image` ref to a `*Video` object would provide correct width/height values (returns `0` as defaulted in `Image` ctr) due to the above: instead these values are in the `XMP.video` tags. This is the situation on on the `master` and the `0.27-RC3` branch. Fix is relatively simple - please incl for 0.27 general release. ``` diff --git a/src/asfvideo.cpp b/src/asfvideo.cpp index b385088c..d24413b6 100644 --- a/src/asfvideo.cpp +++ b/src/asfvideo.cpp @@ -516,7 +516,7 @@ namespace Exiv2 { if(stream == 2) { xmpData_["Xmp.video.Width"] = temp; - width_ = temp; + pixelWidth_ = width_ = temp; } else if(stream == 1) { xmpData_["Xmp.audio.Codec"] = test->printAudioEncoding(temp); @@ -532,7 +532,7 @@ namespace Exiv2 { if(stream == 2) { xmpData_["Xmp.video.Height"] = temp; - height_ = temp; + pixelHeight_ = height_ = temp; } else if(stream == 1) { xmpData_["Xmp.audio.SampleRate"] = temp; diff --git a/src/matroskavideo.cpp b/src/matroskavideo.cpp index 531d2f8c..10de9525 100644 --- a/src/matroskavideo.cpp +++ b/src/matroskavideo.cpp @@ -592,10 +592,10 @@ namespace Exiv2 { xmpData_[mt->label_] = returnValue(buf, size); if (mt->val_ == 0x0030 || mt->val_ == 0x14b0) { - width_ = returnValue(buf, size); + pixelWidth_ = width_ = returnValue(buf, size); } else if (mt->val_ == 0x003a || mt->val_ == 0x14ba) { - height_ = returnValue(buf, size); + pixelHeight_ = height_ = returnValue(buf, size); } break; diff --git a/src/quicktimevideo.cpp b/src/quicktimevideo.cpp index c063d17c..e4d068a4 100644 --- a/src/quicktimevideo.cpp +++ b/src/quicktimevideo.cpp @@ -1537,14 +1537,14 @@ namespace Exiv2 { if(currentStream_ == Video) { temp = returnBufValue(buf, 2) + static_cast<int64_t>((buf.pData_[2] * 256 + buf.pData_[3]) * 0.01); xmpData_["Xmp.video.Width"] = temp; - width_ = temp; + pixelWidth_ = width_ = temp; } break; case ImageHeight: if(currentStream_ == Video) { temp = returnBufValue(buf, 2) + static_cast<int64_t>((buf.pData_[2] * 256 + buf.pData_[3]) * 0.01); xmpData_["Xmp.video.Height"] = temp; - height_ = temp; + pixelHeight_ = height_ = temp; } break; default: diff --git a/src/riffvideo.cpp b/src/riffvideo.cpp index 16610a19..107e5a88 100644 --- a/src/riffvideo.cpp +++ b/src/riffvideo.cpp @@ -1082,10 +1082,12 @@ namespace Exiv2 { break; case imageWidth_h: width = Exiv2::getULong(buf.pData_, littleEndian); + pixelWidth_ = width; xmpData_["Xmp.video.Width"] = width; break; case imageHeight_h: height = Exiv2::getULong(buf.pData_, littleEndian); + pixelHeight_ = height; xmpData_["Xmp.video.Height"] = height; break; } ```

video image height/width not correct via Image::pixel{Height,Width}()

https://api.github.com/repos/Exiv2/exiv2/issues/586/comments

https://github.com/Exiv2/exiv2/issues/586

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : video image height/width not correct via Image::pixel{Height,Width}() Issue date: --- start body --- The `Image` object has `protected` members `pixelHeight_` and `pixelWidth_` (as `int`) and corresponding (overrideable) `public` accessors. However none of the video interfaces overwrite or set these values, instead they all, except `riffvideo` have their own `uint64_t` variables to track height/width. A user would very well expect that an `Image` ref to a `*Video` object would provide correct width/height values (returns `0` as defaulted in `Image` ctr) due to the above: instead these values are in the `XMP.video` tags. This is the situation on on the `master` and the `0.27-RC3` branch. Fix is relatively simple - please incl for 0.27 general release. ``` diff --git a/src/asfvideo.cpp b/src/asfvideo.cpp index b385088c..d24413b6 100644 --- a/src/asfvideo.cpp +++ b/src/asfvideo.cpp @@ -516,7 +516,7 @@ namespace Exiv2 { if(stream == 2) { xmpData_["Xmp.video.Width"] = temp; - width_ = temp; + pixelWidth_ = width_ = temp; } else if(stream == 1) { xmpData_["Xmp.audio.Codec"] = test->printAudioEncoding(temp); @@ -532,7 +532,7 @@ namespace Exiv2 { if(stream == 2) { xmpData_["Xmp.video.Height"] = temp; - height_ = temp; + pixelHeight_ = height_ = temp; } else if(stream == 1) { xmpData_["Xmp.audio.SampleRate"] = temp; diff --git a/src/matroskavideo.cpp b/src/matroskavideo.cpp index 531d2f8c..10de9525 100644 --- a/src/matroskavideo.cpp +++ b/src/matroskavideo.cpp @@ -592,10 +592,10 @@ namespace Exiv2 { xmpData_[mt->label_] = returnValue(buf, size); if (mt->val_ == 0x0030 || mt->val_ == 0x14b0) { - width_ = returnValue(buf, size); + pixelWidth_ = width_ = returnValue(buf, size); } else if (mt->val_ == 0x003a || mt->val_ == 0x14ba) { - height_ = returnValue(buf, size); + pixelHeight_ = height_ = returnValue(buf, size); } break; diff --git a/src/quicktimevideo.cpp b/src/quicktimevideo.cpp index c063d17c..e4d068a4 100644 --- a/src/quicktimevideo.cpp +++ b/src/quicktimevideo.cpp @@ -1537,14 +1537,14 @@ namespace Exiv2 { if(currentStream_ == Video) { temp = returnBufValue(buf, 2) + static_cast<int64_t>((buf.pData_[2] * 256 + buf.pData_[3]) * 0.01); xmpData_["Xmp.video.Width"] = temp; - width_ = temp; + pixelWidth_ = width_ = temp; } break; case ImageHeight: if(currentStream_ == Video) { temp = returnBufValue(buf, 2) + static_cast<int64_t>((buf.pData_[2] * 256 + buf.pData_[3]) * 0.01); xmpData_["Xmp.video.Height"] = temp; - height_ = temp; + pixelHeight_ = height_ = temp; } break; default: diff --git a/src/riffvideo.cpp b/src/riffvideo.cpp index 16610a19..107e5a88 100644 --- a/src/riffvideo.cpp +++ b/src/riffvideo.cpp @@ -1082,10 +1082,12 @@ namespace Exiv2 { break; case imageWidth_h: width = Exiv2::getULong(buf.pData_, littleEndian); + pixelWidth_ = width; xmpData_["Xmp.video.Width"] = width; break; case imageHeight_h: height = Exiv2::getULong(buf.pData_, littleEndian); + pixelHeight_ = height; xmpData_["Xmp.video.Height"] = height; break; } ``` --- end body --- comments : --- start comments --- --- end comments ---

``` C:\Users\rmills\gnu\github\exiv2\exiv2\src\bigtiffimage.cpp(364): warning C4244: 'argument': conversion from 'const uint64_t' to 'const unsign ed int', possible loss of data [C:\Users\rmills\gnu\github\exiv2\exiv2\build\src\exiv2lib.vcxproj] C:\Users\rmills\gnu\github\exiv2\exiv2\src\bigtiffimage.cpp(385): warning C4244: 'argument': conversion from 'uint64_t' to 'const unsigned int ', possible loss of data [C:\Users\rmills\gnu\github\exiv2\exiv2\build\src\exiv2lib.vcxproj] ```

Ugly compiler warning from src/bigtiffimage.cpp on 32 bit Visual Studio

https://api.github.com/repos/Exiv2/exiv2/issues/585/comments

https://github.com/Exiv2/exiv2/issues/585

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : Ugly compiler warning from src/bigtiffimage.cpp on 32 bit Visual Studio Issue date: --- start body --- ``` C:\Users\rmills\gnu\github\exiv2\exiv2\src\bigtiffimage.cpp(364): warning C4244: 'argument': conversion from 'const uint64_t' to 'const unsign ed int', possible loss of data [C:\Users\rmills\gnu\github\exiv2\exiv2\build\src\exiv2lib.vcxproj] C:\Users\rmills\gnu\github\exiv2\exiv2\src\bigtiffimage.cpp(385): warning C4244: 'argument': conversion from 'uint64_t' to 'const unsigned int ', possible loss of data [C:\Users\rmills\gnu\github\exiv2\exiv2\build\src\exiv2lib.vcxproj] ``` --- end body --- comments : --- start comments --- --- end comments ---

I am building the latest master (d8a15e1d132e260dd333ebc5af1e6929257a6d0f) with video support but it appears this functionality is not working. I believe the build config is good (see below) but `exiv2 <some mp4 file>` fails complaining for GOpro fiels and iphone files. The same problem persists for the samples from the exiv2 site: [http://dev.exiv2.org/attachments/385/MVI_2229.MOV] ``` $ exiv2 MVI_2229.MOV Exiv2 exception in print action for file MVI_2229.MOV: MVI_2229.MOV: The file contains data of an unknown image type $ exiv2 -vV | grep video enable_video=1 xmlns=video:http://www.video/ ``` Is this is regression or user error? The build configuration: ``` mkdir build; cd build cmake -DEXIV2_ENABLE_VIDEO=ON -DCMAKE_INSTALL_PREFIX=/usr -DEXIV2_BUILD_SAMPLES=OFF .. ... -- ------------------------------------------------------------------ -- Building shared library: YES -- Building PNG support: YES -- XMP metadata support: YES -- Native language support: YES -- Conversion of Windows XP tags: YES -- Nikon lens database: YES -- Building video support: YES -- Building webready support: NO -- Building exiv2 command: YES -- Building samples: NO -- Building PO files: NO -- Building unit tests: NO -- Building doc: NO -- Building with coverage flags: NO -- Using ccache: NO -- ------------------------------------------------------------------ -- Configuring done -- Generating done -- Build files have been written to: /home/ray/dev/exiv2/build ```

master build - NO video supprt even compiling w/video support

https://api.github.com/repos/Exiv2/exiv2/issues/584/comments

https://github.com/Exiv2/exiv2/issues/584

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : master build - NO video supprt even compiling w/video support Issue date: --- start body --- I am building the latest master (d8a15e1d132e260dd333ebc5af1e6929257a6d0f) with video support but it appears this functionality is not working. I believe the build config is good (see below) but `exiv2 <some mp4 file>` fails complaining for GOpro fiels and iphone files. The same problem persists for the samples from the exiv2 site: [http://dev.exiv2.org/attachments/385/MVI_2229.MOV] ``` $ exiv2 MVI_2229.MOV Exiv2 exception in print action for file MVI_2229.MOV: MVI_2229.MOV: The file contains data of an unknown image type $ exiv2 -vV | grep video enable_video=1 xmlns=video:http://www.video/ ``` Is this is regression or user error? The build configuration: ``` mkdir build; cd build cmake -DEXIV2_ENABLE_VIDEO=ON -DCMAKE_INSTALL_PREFIX=/usr -DEXIV2_BUILD_SAMPLES=OFF .. ... -- ------------------------------------------------------------------ -- Building shared library: YES -- Building PNG support: YES -- XMP metadata support: YES -- Native language support: YES -- Conversion of Windows XP tags: YES -- Nikon lens database: YES -- Building video support: YES -- Building webready support: NO -- Building exiv2 command: YES -- Building samples: NO -- Building PO files: NO -- Building unit tests: NO -- Building doc: NO -- Building with coverage flags: NO -- Using ccache: NO -- ------------------------------------------------------------------ -- Configuring done -- Generating done -- Build files have been written to: /home/ray/dev/exiv2/build ``` --- end body --- comments : --- start comments --- --- end comments ---

Please add support for reading the FocusPositon from ARW files. This should be in Tag9402 ``` exiftool 2018-10-16__7M34855.ARW | grep "Positon" Focus Position 2 : 207 ``` I can't read them with exiv2, it is needed to calculate the FocusDistance the correct vignetting for lensfun. If you need an example file: https://xor.cryptomilk.org/darktable/ILCE-7M3/ColorCheckerSG/2018-10-17__7M34864.ARW [UPDATE] Change from FocusDistance to FocusPosition which is a real tag.

Add support for FocusPosition (Tag9402) in Sony RAW files (ARW)

https://api.github.com/repos/Exiv2/exiv2/issues/582/comments

https://github.com/Exiv2/exiv2/issues/582

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : Add support for FocusPosition (Tag9402) in Sony RAW files (ARW) Issue date: --- start body --- Please add support for reading the FocusPositon from ARW files. This should be in Tag9402 ``` exiftool 2018-10-16__7M34855.ARW | grep "Positon" Focus Position 2 : 207 ``` I can't read them with exiv2, it is needed to calculate the FocusDistance the correct vignetting for lensfun. If you need an example file: https://xor.cryptomilk.org/darktable/ILCE-7M3/ColorCheckerSG/2018-10-17__7M34864.ARW [UPDATE] Change from FocusDistance to FocusPosition which is a real tag. --- end body --- comments : --- start comments --- --- end comments ---

This script is recommended in README.md, so it's important that it works smoothly. Perhaps we should have a ReadMe.txt in the ci/ directory to explain what the different script do. I've found `sudo ./install.sh` is great for installing conan. However I have to `$ sudo cp conan/bin/conan /usr/local/bin` to get conan on the path. Anyway, let's focus on install_dependencies.sh ``` $ cd <exiv2dir>/ci $ sudo ./install_dependencies.sh Processing triggers for libc-bin (2.19-0ubuntu6.9) ... ~/gnu/github/exiv2/exiv2/ci ~/gnu/github/exiv2/exiv2/ci CMake Error: The source directory "/usr/src/googletest/googletest" does not exist. Specify --help for usage, or press the help button on the CMake GUI. rmills@ubuntu:~/gnu/github/exiv2/exiv2/ci$ ``` Indeed, /usr/src/googletest/googltest does not exist: ``` rmills@ubuntu:~/gnu/github/exiv2/exiv2/ci$ ls -l /usr/src total 20 drwxr-xr-x 4 root root 4096 Dec 3 11:35 gmock drwxr-xr-x 4 root root 4096 Dec 3 11:35 gtest drwxr-xr-x 27 root root 4096 Aug 3 2016 linux-headers-4.4.0-31 drwxr-xr-x 7 root root 4096 Aug 3 2016 linux-headers-4.4.0-31-generic drwxr-xr-x 6 root root 4096 Dec 3 11:07 parallels-tools-14.1.0.45387 rmills@ubuntu:~/gnu/github/exiv2/exiv2/ci$ ``` The reference to googletest is from **`contrib/vms/utils.source`**: ``` rmills@ubuntu:~/gnu/github/exiv2/exiv2/ci$ grep -Hn googletest ../contrib/vms/* ../contrib/vms/utils.source:8:# Debian & derivatives don't provide binary packages of googletest ../contrib/vms/utils.source:11:# This script builds a shared library of googletest (not googlemock!) inside ../contrib/vms/utils.source:17: cmake -DBUILD_SHARED_LIBS=1 /usr/src/googletest/googletest rmills@ubuntu:~/gnu/github/exiv2/exiv2/ci$ cat ../contrib/vms/utils.source #!/bin/sh function get_distro_id() { local distro_id=$(grep '^ID=' /etc/os-release|awk -F = '{print $2}'|sed 's/\"//g') echo "$distro_id" } # Debian & derivatives don't provide binary packages of googletest # => have to build them ourselves # # This script builds a shared library of googletest (not googlemock!) inside # gtest_build and copies it to /usr/lib/ function debian_build_gtest() { pushd . [ -d gtest_build ] || mkdir gtest_build cd gtest_build cmake -DBUILD_SHARED_LIBS=1 /usr/src/googletest/googletest make cp libgtest* /usr/lib/ popd } rmills@ubuntu:~/gnu/github/exiv2/exiv2/ci$ ``` I hacked the code to work. I suspect we need a more robust test such as `if [ -d /usr/scr/googletest/googlest ] ..` ``` rmills@ubuntu:~/gnu/github/exiv2/exiv2$ git diff contrib/vms/utils.source diff --git a/contrib/vms/utils.source b/contrib/vms/utils.source index ddeef60..916b258 100644 --- a/contrib/vms/utils.source +++ b/contrib/vms/utils.source @@ -14,7 +14,7 @@ function debian_build_gtest() { pushd . [ -d gtest_build ] || mkdir gtest_build cd gtest_build - cmake -DBUILD_SHARED_LIBS=1 /usr/src/googletest/googletest + cmake -DBUILD_SHARED_LIBS=1 /usr/src/gtest make cp libgtest* /usr/lib/ popd rmills@ubuntu:~/gnu/github/exiv2/exiv2$ ```

ci/install_dependencies.sh ubuntu 14.04.5 error

https://api.github.com/repos/Exiv2/exiv2/issues/581/comments

https://github.com/Exiv2/exiv2/issues/581

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : ci/install_dependencies.sh ubuntu 14.04.5 error Issue date: --- start body --- This script is recommended in README.md, so it's important that it works smoothly. Perhaps we should have a ReadMe.txt in the ci/ directory to explain what the different script do. I've found `sudo ./install.sh` is great for installing conan. However I have to `$ sudo cp conan/bin/conan /usr/local/bin` to get conan on the path. Anyway, let's focus on install_dependencies.sh ``` $ cd <exiv2dir>/ci $ sudo ./install_dependencies.sh Processing triggers for libc-bin (2.19-0ubuntu6.9) ... ~/gnu/github/exiv2/exiv2/ci ~/gnu/github/exiv2/exiv2/ci CMake Error: The source directory "/usr/src/googletest/googletest" does not exist. Specify --help for usage, or press the help button on the CMake GUI. rmills@ubuntu:~/gnu/github/exiv2/exiv2/ci$ ``` Indeed, /usr/src/googletest/googltest does not exist: ``` rmills@ubuntu:~/gnu/github/exiv2/exiv2/ci$ ls -l /usr/src total 20 drwxr-xr-x 4 root root 4096 Dec 3 11:35 gmock drwxr-xr-x 4 root root 4096 Dec 3 11:35 gtest drwxr-xr-x 27 root root 4096 Aug 3 2016 linux-headers-4.4.0-31 drwxr-xr-x 7 root root 4096 Aug 3 2016 linux-headers-4.4.0-31-generic drwxr-xr-x 6 root root 4096 Dec 3 11:07 parallels-tools-14.1.0.45387 rmills@ubuntu:~/gnu/github/exiv2/exiv2/ci$ ``` The reference to googletest is from **`contrib/vms/utils.source`**: ``` rmills@ubuntu:~/gnu/github/exiv2/exiv2/ci$ grep -Hn googletest ../contrib/vms/* ../contrib/vms/utils.source:8:# Debian & derivatives don't provide binary packages of googletest ../contrib/vms/utils.source:11:# This script builds a shared library of googletest (not googlemock!) inside ../contrib/vms/utils.source:17: cmake -DBUILD_SHARED_LIBS=1 /usr/src/googletest/googletest rmills@ubuntu:~/gnu/github/exiv2/exiv2/ci$ cat ../contrib/vms/utils.source #!/bin/sh function get_distro_id() { local distro_id=$(grep '^ID=' /etc/os-release|awk -F = '{print $2}'|sed 's/\"//g') echo "$distro_id" } # Debian & derivatives don't provide binary packages of googletest # => have to build them ourselves # # This script builds a shared library of googletest (not googlemock!) inside # gtest_build and copies it to /usr/lib/ function debian_build_gtest() { pushd . [ -d gtest_build ] || mkdir gtest_build cd gtest_build cmake -DBUILD_SHARED_LIBS=1 /usr/src/googletest/googletest make cp libgtest* /usr/lib/ popd } rmills@ubuntu:~/gnu/github/exiv2/exiv2/ci$ ``` I hacked the code to work. I suspect we need a more robust test such as `if [ -d /usr/scr/googletest/googlest ] ..` ``` rmills@ubuntu:~/gnu/github/exiv2/exiv2$ git diff contrib/vms/utils.source diff --git a/contrib/vms/utils.source b/contrib/vms/utils.source index ddeef60..916b258 100644 --- a/contrib/vms/utils.source +++ b/contrib/vms/utils.source @@ -14,7 +14,7 @@ function debian_build_gtest() { pushd . [ -d gtest_build ] || mkdir gtest_build cd gtest_build - cmake -DBUILD_SHARED_LIBS=1 /usr/src/googletest/googletest + cmake -DBUILD_SHARED_LIBS=1 /usr/src/gtest make cp libgtest* /usr/lib/ popd rmills@ubuntu:~/gnu/github/exiv2/exiv2$ ``` --- end body --- comments : --- start comments --- --- end comments ---

https://travis-ci.org/Exiv2/exiv2/jobs/462279634

Exiv2 0.27-RC3 fails on Travis on Ubuntu 14.04.5 LTS with Clang 5.0.0.

https://api.github.com/repos/Exiv2/exiv2/issues/580/comments

https://github.com/Exiv2/exiv2/issues/580

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : Exiv2 0.27-RC3 fails on Travis on Ubuntu 14.04.5 LTS with Clang 5.0.0. Issue date: --- start body --- https://travis-ci.org/Exiv2/exiv2/jobs/462279634 --- end body --- comments : --- start comments --- --- end comments ---

I have a JPG about which exiftool states, amongst other EXIF data: ```` File Modification Date/Time : 2016:07:26 17:36:03+02:00 ```` yet exiv2 states: ```` % exiv2 rename -T *.jpg test.jpg: No Exif data found in the file % ```` I can provide the file by private email if need be.

Some EXIF datetime sources are ignored

https://api.github.com/repos/Exiv2/exiv2/issues/579/comments

https://github.com/Exiv2/exiv2/issues/579

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : Some EXIF datetime sources are ignored Issue date: --- start body --- I have a JPG about which exiftool states, amongst other EXIF data: ```` File Modification Date/Time : 2016:07:26 17:36:03+02:00 ```` yet exiv2 states: ```` % exiv2 rename -T *.jpg test.jpg: No Exif data found in the file % ```` I can provide the file by private email if need be. --- end body --- comments : --- start comments --- --- end comments ---

I need to restore local mtime from EXIF regularly, but maintain original file names. `exiv2 rename` is a destructive operation, `exiv2 rename -T` is benign. Would you be willing to either introduce a new keyword just for c/mtime editing, or allow `exiv2 adjust -T`?

[FR] `-T` without rename

https://api.github.com/repos/Exiv2/exiv2/issues/578/comments

https://github.com/Exiv2/exiv2/issues/578

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : [FR] `-T` without rename Issue date: --- start body --- I need to restore local mtime from EXIF regularly, but maintain original file names. `exiv2 rename` is a destructive operation, `exiv2 rename -T` is benign. Would you be willing to either introduce a new keyword just for c/mtime editing, or allow `exiv2 adjust -T`? --- end body --- comments : --- start comments --- --- end comments ---

``` [ 49%] Linking CXX executable ../bin/unit_tests ld: warning: direct access in function 'Exiv2::Internal::tokenize(std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> > const&, std::__1::vector<std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> >, std::__1::allocator<std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> > > >&, std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> > const&)' from file '../src/CMakeFiles/exiv2lib_int.dir/minoltamn_int.cpp.o' to global weak symbol 'std::__1::char_traits<char>::eq(char, char)' from file '/usr/local/lib/libgtest.a(gtest-all.cc.o)' means the weak symbol cannot be overridden at runtime. This was likely caused by different translation units being compiled with different visibility settings. ld: warning: direct access in function 'Exiv2::Internal::tokenize(std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> > const&, std::__1::vector<std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> >, std::__1::allocator<std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> > > >&, std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> > const&)' from file '../src/CMakeFiles/exiv2lib_int.dir/minoltamn_int.cpp.o' to global weak symbol 'std::__1::char_traits<char>::eq(char, char)' from file '/usr/local/lib/libgtest.a(gtest-all.cc.o)' means the weak symbol cannot be overridden at runtime. This was likely caused by different translation units being compiled with different visibility settings. [ 56%] Built target unit_tests ``` This is coming from a static in src/minotamn_int.cpp. Odd, there are lots of static functions in that file and there's nothing exceptional about **tokenize()**

Ugly warning on MacOS-X when linking unit_tests

https://api.github.com/repos/Exiv2/exiv2/issues/577/comments

https://github.com/Exiv2/exiv2/issues/577

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : Ugly warning on MacOS-X when linking unit_tests Issue date: --- start body --- ``` [ 49%] Linking CXX executable ../bin/unit_tests ld: warning: direct access in function 'Exiv2::Internal::tokenize(std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> > const&, std::__1::vector<std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> >, std::__1::allocator<std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> > > >&, std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> > const&)' from file '../src/CMakeFiles/exiv2lib_int.dir/minoltamn_int.cpp.o' to global weak symbol 'std::__1::char_traits<char>::eq(char, char)' from file '/usr/local/lib/libgtest.a(gtest-all.cc.o)' means the weak symbol cannot be overridden at runtime. This was likely caused by different translation units being compiled with different visibility settings. ld: warning: direct access in function 'Exiv2::Internal::tokenize(std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> > const&, std::__1::vector<std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> >, std::__1::allocator<std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> > > >&, std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> > const&)' from file '../src/CMakeFiles/exiv2lib_int.dir/minoltamn_int.cpp.o' to global weak symbol 'std::__1::char_traits<char>::eq(char, char)' from file '/usr/local/lib/libgtest.a(gtest-all.cc.o)' means the weak symbol cannot be overridden at runtime. This was likely caused by different translation units being compiled with different visibility settings. [ 56%] Built target unit_tests ``` This is coming from a static in src/minotamn_int.cpp. Odd, there are lots of static functions in that file and there's nothing exceptional about **tokenize()** --- end body --- comments : --- start comments --- --- end comments ---

Fails to link: ``` ../src/CMakeFiles/exiv2lib_int.dir/tags_int.cpp.o:tags_int.cpp:(.text+0x486): undefined reference to `Exiv2::GroupInfo::GroupName::GroupName(std::string const&)' ../src/CMakeFiles/exiv2lib_int.dir/tags_int.cpp.o:tags_int.cpp:(.text+0x486): relocation truncated to fit: R_X86_64_PC32 against undefined symbol `Exiv2::GroupInfo::GroupName::GroupName(std::string const&)' ../src/CMakeFiles/exiv2lib_int.dir/tags_int.cpp.o:tags_int.cpp:(.text+0x31d4): undefined reference to `Exiv2::GroupInfo::GroupName::GroupName(std::string const&)' ../src/CMakeFiles/exiv2lib_int.dir/tags_int.cpp.o:tags_int.cpp:(.text+0x31d4): relocation truncated to fit: R_X86_64_PC32 against undefined symbol `Exiv2::GroupInfo::GroupName::GroupName(std::string const&)' collect2: error: ld returned 1 exit status make[2]: *** [unitTests/CMakeFiles/unit_tests.dir/build.make:316: bin/unit_tests.exe] Error 1 make[1]: *** [CMakeFiles/Makefile2:384: unitTests/CMakeFiles/unit_tests.dir/all] Error 2 make: *** [Makefile:130: all] Error 2 405 CYGWIN64:rmills@rmillsmbp-w7:~/gnu/github/exiv2/exiv2/build $ ``` Curious! GroupName::GroupName is a recent addition to Exiv2 (can't remember exactly, but it caused issues with Xcode 10.0). Issue is caused by not declaring GroupInfo::GroupName as EXIV2API in include/exiv2/tags.hpp ``` 576 rmills@rmillsmm:~/gnu/github/exiv2/exiv2 $ git diff include/exiv2/tags.hpp diff --git a/include/exiv2/tags.hpp b/include/exiv2/tags.hpp index 8ca8d61a..7844f7b4 100644 --- a/include/exiv2/tags.hpp +++ b/include/exiv2/tags.hpp @@ -72,7 +72,7 @@ namespace Exiv2 { }; //! Search key to find a GroupInfo by its group name. - struct GroupInfo::GroupName { + struct EXIV2API GroupInfo::GroupName { explicit GroupName(const std::string& groupName); std::string g_; //!< Group name }; 577 rmills@rmillsmm:~/gnu/github/exiv2/exiv2 $ ```

Building unit_tests on MinGW and Cygwin fails to link

https://api.github.com/repos/Exiv2/exiv2/issues/576/comments

https://github.com/Exiv2/exiv2/issues/576

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : Building unit_tests on MinGW and Cygwin fails to link Issue date: --- start body --- Fails to link: ``` ../src/CMakeFiles/exiv2lib_int.dir/tags_int.cpp.o:tags_int.cpp:(.text+0x486): undefined reference to `Exiv2::GroupInfo::GroupName::GroupName(std::string const&)' ../src/CMakeFiles/exiv2lib_int.dir/tags_int.cpp.o:tags_int.cpp:(.text+0x486): relocation truncated to fit: R_X86_64_PC32 against undefined symbol `Exiv2::GroupInfo::GroupName::GroupName(std::string const&)' ../src/CMakeFiles/exiv2lib_int.dir/tags_int.cpp.o:tags_int.cpp:(.text+0x31d4): undefined reference to `Exiv2::GroupInfo::GroupName::GroupName(std::string const&)' ../src/CMakeFiles/exiv2lib_int.dir/tags_int.cpp.o:tags_int.cpp:(.text+0x31d4): relocation truncated to fit: R_X86_64_PC32 against undefined symbol `Exiv2::GroupInfo::GroupName::GroupName(std::string const&)' collect2: error: ld returned 1 exit status make[2]: *** [unitTests/CMakeFiles/unit_tests.dir/build.make:316: bin/unit_tests.exe] Error 1 make[1]: *** [CMakeFiles/Makefile2:384: unitTests/CMakeFiles/unit_tests.dir/all] Error 2 make: *** [Makefile:130: all] Error 2 405 CYGWIN64:rmills@rmillsmbp-w7:~/gnu/github/exiv2/exiv2/build $ ``` Curious! GroupName::GroupName is a recent addition to Exiv2 (can't remember exactly, but it caused issues with Xcode 10.0). Issue is caused by not declaring GroupInfo::GroupName as EXIV2API in include/exiv2/tags.hpp ``` 576 rmills@rmillsmm:~/gnu/github/exiv2/exiv2 $ git diff include/exiv2/tags.hpp diff --git a/include/exiv2/tags.hpp b/include/exiv2/tags.hpp index 8ca8d61a..7844f7b4 100644 --- a/include/exiv2/tags.hpp +++ b/include/exiv2/tags.hpp @@ -72,7 +72,7 @@ namespace Exiv2 { }; //! Search key to find a GroupInfo by its group name. - struct GroupInfo::GroupName { + struct EXIV2API GroupInfo::GroupName { explicit GroupName(const std::string& groupName); std::string g_; //!< Group name }; 577 rmills@rmillsmm:~/gnu/github/exiv2/exiv2 $ ``` --- end body --- comments : --- start comments --- --- end comments ---

I'm putting these notes for my benefit and this issue will be closed in a couple days. I'm very pleased that we can build and run the unit_tests on msvc builds thanks to conan. And they build OK on MacOS-X without using conan. I've been unable to build Exiv2 with conan on Cygwin and/or MinGW and I've said in the docs that we don't support this. Thanks to Dan's wonderful script ci/install_dependencies.sh, setting up on Linux is easy and is documented in README.md. 1) Building and installing shared gtest libraries from https://github.com/google/googletest/archive/release-1.8.0.tar.gz ``` $ mkdir -p ~/gnu/gtest $ cp release-1.8.0.tar.gz ~/gnu/gtest $ cd ~/gnu/gtest $ tar xzf release-1.8.0.tar.gz $ cd googletest-release-1.8.0 $ mkdir build $ cd build $ cmake .. -DBUILD_SHARED_LIBS=1 $ make $ sudo make install ``` 2) Building exiv2 with unit_tests ``` $ mkdir -p <exiv2dir>/build $ cd <exiv2dir>/build $ cmake .. -DEXIV2_BUILD_UNIT_TESTS=1 $ make $ bin/unit_tests.exe ```

Investigation into building unit_tests without conan

https://api.github.com/repos/Exiv2/exiv2/issues/575/comments

https://github.com/Exiv2/exiv2/issues/575

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : Investigation into building unit_tests without conan Issue date: --- start body --- I'm putting these notes for my benefit and this issue will be closed in a couple days. I'm very pleased that we can build and run the unit_tests on msvc builds thanks to conan. And they build OK on MacOS-X without using conan. I've been unable to build Exiv2 with conan on Cygwin and/or MinGW and I've said in the docs that we don't support this. Thanks to Dan's wonderful script ci/install_dependencies.sh, setting up on Linux is easy and is documented in README.md. 1) Building and installing shared gtest libraries from https://github.com/google/googletest/archive/release-1.8.0.tar.gz ``` $ mkdir -p ~/gnu/gtest $ cp release-1.8.0.tar.gz ~/gnu/gtest $ cd ~/gnu/gtest $ tar xzf release-1.8.0.tar.gz $ cd googletest-release-1.8.0 $ mkdir build $ cd build $ cmake .. -DBUILD_SHARED_LIBS=1 $ make $ sudo make install ``` 2) Building exiv2 with unit_tests ``` $ mkdir -p <exiv2dir>/build $ cd <exiv2dir>/build $ cmake .. -DEXIV2_BUILD_UNIT_TESTS=1 $ make $ bin/unit_tests.exe ``` --- end body --- comments : --- start comments --- --- end comments ---

I think we should only support clang >= 5 on Linux. I've given up trying to get clang to work on Cygwin/MinGW/DOS. And clang-4 on ubuntu-14.04. It can't compile hello.cpp: ``` #include <iostream> int main(int argc,const char* argv[]) { std::cout << "hello world" << std::endl; for ( int i = 0 ; i < argc ; i++ ) { std::cout << argv[i] << std::endl; } return 0; } ``` I've encountered issues with Clang 3.4 on Ubuntu 14.04 with -DEXIV2_TEAM_ENABLE_SANITERS=On which uses the option -fno-sanitize-recover=all. This option was added in GCC5, I believe.

Issues with clang

https://api.github.com/repos/Exiv2/exiv2/issues/574/comments

https://github.com/Exiv2/exiv2/issues/574

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : Issues with clang Issue date: --- start body --- I think we should only support clang >= 5 on Linux. I've given up trying to get clang to work on Cygwin/MinGW/DOS. And clang-4 on ubuntu-14.04. It can't compile hello.cpp: ``` #include <iostream> int main(int argc,const char* argv[]) { std::cout << "hello world" << std::endl; for ( int i = 0 ; i < argc ; i++ ) { std::cout << argv[i] << std::endl; } return 0; } ``` I've encountered issues with Clang 3.4 on Ubuntu 14.04 with -DEXIV2_TEAM_ENABLE_SANITERS=On which uses the option -fno-sanitize-recover=all. This option was added in GCC5, I believe. --- end body --- comments : --- start comments --- --- end comments ---

Hi, ``` exiv2 -g Exif.Photo.ISOSpeedRatings -Pt /home/asn/Documents/Download/noise_profiles/Panasonic_G3/P1300439.RW2 RW2 IMAGE 160 ``` I've just asked to get the ISO value but in addition it prints the string `RW2 IMAGE` and it seems to do that only for that RAW format. It doesn't happen for e.g. ARW: ``` exiv2 -g Exif.Photo.ISOSpeedRatings -Pt /home/asn/Pictures/2018/2018-07_Island/2018-07_RAW_Island/2018-07-10__7M32947.ARW 200 ```

exiv2 prints file type for RW2 files if you only request the ISO

https://api.github.com/repos/Exiv2/exiv2/issues/573/comments

https://github.com/Exiv2/exiv2/issues/573

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : exiv2 prints file type for RW2 files if you only request the ISO Issue date: --- start body --- Hi, ``` exiv2 -g Exif.Photo.ISOSpeedRatings -Pt /home/asn/Documents/Download/noise_profiles/Panasonic_G3/P1300439.RW2 RW2 IMAGE 160 ``` I've just asked to get the ISO value but in addition it prints the string `RW2 IMAGE` and it seems to do that only for that RAW format. It doesn't happen for e.g. ARW: ``` exiv2 -g Exif.Photo.ISOSpeedRatings -Pt /home/asn/Pictures/2018/2018-07_Island/2018-07_RAW_Island/2018-07-10__7M32947.ARW 200 ``` --- end body --- comments : --- start comments --- --- end comments ---

I've never seen this until last night and it happened three times on Jenkins. It doesn't happen when I build manually! ``` exiv2.lib(futils.obj) : error LNK2019: unresolved external symbol GetModuleFileNameExA referenced in function "class std::basic_string<char,struct std::char_traits<char>,class std::allocator<char> > __cdecl Exiv2::getProcessPath(void)" (?getProcessPath@Exiv2@@YA?AV?$basic_string@DU?$char_traits@D@std@@V?$allocator@D@2@@std@@XZ) [C:\Users\rmills\gnu\github\exiv2\buildserver\build\samples\iotest.vcxproj] C:\Users\rmills\gnu\github\exiv2\buildserver\build\bin\iotest.exe : fatal error LNK1120: 1 unresolved externals [C:\Users\rmills\gnu\github\exiv2\buildserver\build\samples\iotest.vcxproj] Done Building Project "C:\Users\rmills\gnu\github\exiv2\buildserver\build\samples\iotest.vcxproj" (default targets) -- FAILED. Project "C:\Users\rmills\gnu\github\exiv2\buildserver\build\ALL_BUILD.vcxproj" (1) is building "C:\Users\rmills\gnu\github\exiv2\buildserver\build\samples\iptceasy.vcxproj" (21) on node 1 (default targets). ``` Remarks: 1) It's coming from Exiv2::getProcessPath() calling GetModuleFileNameEx() 2) GetModuleFileNameEx() is in psapi which is linked PRIVATE 3) Microsoft recommend using GetModuleFileName() as it's more efficient.

MSVC iotest.cpp link failure

https://api.github.com/repos/Exiv2/exiv2/issues/572/comments

https://github.com/Exiv2/exiv2/issues/572

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : MSVC iotest.cpp link failure Issue date: --- start body --- I've never seen this until last night and it happened three times on Jenkins. It doesn't happen when I build manually! ``` exiv2.lib(futils.obj) : error LNK2019: unresolved external symbol GetModuleFileNameExA referenced in function "class std::basic_string<char,struct std::char_traits<char>,class std::allocator<char> > __cdecl Exiv2::getProcessPath(void)" (?getProcessPath@Exiv2@@YA?AV?$basic_string@DU?$char_traits@D@std@@V?$allocator@D@2@@std@@XZ) [C:\Users\rmills\gnu\github\exiv2\buildserver\build\samples\iotest.vcxproj] C:\Users\rmills\gnu\github\exiv2\buildserver\build\bin\iotest.exe : fatal error LNK1120: 1 unresolved externals [C:\Users\rmills\gnu\github\exiv2\buildserver\build\samples\iotest.vcxproj] Done Building Project "C:\Users\rmills\gnu\github\exiv2\buildserver\build\samples\iotest.vcxproj" (default targets) -- FAILED. Project "C:\Users\rmills\gnu\github\exiv2\buildserver\build\ALL_BUILD.vcxproj" (1) is building "C:\Users\rmills\gnu\github\exiv2\buildserver\build\samples\iptceasy.vcxproj" (21) on node 1 (default targets). ``` Remarks: 1) It's coming from Exiv2::getProcessPath() calling GetModuleFileNameEx() 2) GetModuleFileNameEx() is in psapi which is linked PRIVATE 3) Microsoft recommend using GetModuleFileName() as it's more efficient. --- end body --- comments : --- start comments --- --- end comments ---

``` 514 MSYS64:rmills@rmillsmbp-w7:~/gnu/github/exiv2/exiv2/build $ make [ 18%] Built target xmp Scanning dependencies of target exiv2lib_int [ 20%] Building C object src/CMakeFiles/exiv2lib_int.dir/localtime.c.o /c/msys64/home/rmills/gnu/github/exiv2/exiv2/src/localtime.c: In function 'tzload': /c/msys64/home/rmills/gnu/github/exiv2/exiv2/src/localtime.c:268:19: warning: implicit declaration of function 'access' [-Wimplicit-function-declaration] if (doaccess && access(name, R_OK) != 0) ^~~~~~ /c/msys64/home/rmills/gnu/github/exiv2/exiv2/src/localtime.c:282:7: warning: implicit declaration of function 'read'; did you mean 'fread'? [-Wimplicit-function-declaration] i = read(fid, u.buf, sizeof u.buf); ^~~~ fread /c/msys64/home/rmills/gnu/github/exiv2/exiv2/src/localtime.c:283:7: warning: implicit declaration of function 'close'; did you mean 'pclose'? [-Wimplicit-function-declaration] if (close(fid) != 0) ^~~~~ pclose [ 45%] Built target exiv2lib_int [ 46%] Linking CXX shared library ../bin/msys-exiv2-27.dll make[2]: *** Deleting file 'bin/msys-exiv2-27.dll' make[2]: *** [src/CMakeFiles/exiv2lib.dir/build.make:1240: bin/msys-exiv2-27.dll] Interrupt make[1]: *** [CMakeFiles/Makefile2:178: src/CMakeFiles/exiv2lib.dir/all] Interrupt make: *** [Makefile:130: all] Interrupt 515 MSYS64:rmills@rmillsmbp-w7:~/gnu/github/exiv2/exiv2/build $ ``` Fix is easy: ``` 512 MSYS64:rmills@rmillsmbp-w7:~/gnu/github/exiv2/exiv2/build $ git diff diff --git a/src/localtime.c b/src/localtime.c index 3ec0925c..55ad84ff 100644 --- a/src/localtime.c +++ b/src/localtime.c @@ -5,6 +5,9 @@ #include "config.h" #include "timegm.h" +#ifdef EXV_HAVE_UNISTD_H +#include "unistd.h" +#endif /* ** This file is in the public domain, so clarified as of 513 MSYS64:rmills@rmillsmbp-w7:~/gnu/github/exiv2/exiv2/build $ ```

Ugly compiler warning on MinGW/msys2 from src/localhost.c

https://api.github.com/repos/Exiv2/exiv2/issues/571/comments

https://github.com/Exiv2/exiv2/issues/571

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : Ugly compiler warning on MinGW/msys2 from src/localhost.c Issue date: --- start body --- ``` 514 MSYS64:rmills@rmillsmbp-w7:~/gnu/github/exiv2/exiv2/build $ make [ 18%] Built target xmp Scanning dependencies of target exiv2lib_int [ 20%] Building C object src/CMakeFiles/exiv2lib_int.dir/localtime.c.o /c/msys64/home/rmills/gnu/github/exiv2/exiv2/src/localtime.c: In function 'tzload': /c/msys64/home/rmills/gnu/github/exiv2/exiv2/src/localtime.c:268:19: warning: implicit declaration of function 'access' [-Wimplicit-function-declaration] if (doaccess && access(name, R_OK) != 0) ^~~~~~ /c/msys64/home/rmills/gnu/github/exiv2/exiv2/src/localtime.c:282:7: warning: implicit declaration of function 'read'; did you mean 'fread'? [-Wimplicit-function-declaration] i = read(fid, u.buf, sizeof u.buf); ^~~~ fread /c/msys64/home/rmills/gnu/github/exiv2/exiv2/src/localtime.c:283:7: warning: implicit declaration of function 'close'; did you mean 'pclose'? [-Wimplicit-function-declaration] if (close(fid) != 0) ^~~~~ pclose [ 45%] Built target exiv2lib_int [ 46%] Linking CXX shared library ../bin/msys-exiv2-27.dll make[2]: *** Deleting file 'bin/msys-exiv2-27.dll' make[2]: *** [src/CMakeFiles/exiv2lib.dir/build.make:1240: bin/msys-exiv2-27.dll] Interrupt make[1]: *** [CMakeFiles/Makefile2:178: src/CMakeFiles/exiv2lib.dir/all] Interrupt make: *** [Makefile:130: all] Interrupt 515 MSYS64:rmills@rmillsmbp-w7:~/gnu/github/exiv2/exiv2/build $ ``` Fix is easy: ``` 512 MSYS64:rmills@rmillsmbp-w7:~/gnu/github/exiv2/exiv2/build $ git diff diff --git a/src/localtime.c b/src/localtime.c index 3ec0925c..55ad84ff 100644 --- a/src/localtime.c +++ b/src/localtime.c @@ -5,6 +5,9 @@ #include "config.h" #include "timegm.h" +#ifdef EXV_HAVE_UNISTD_H +#include "unistd.h" +#endif /* ** This file is in the public domain, so clarified as of 513 MSYS64:rmills@rmillsmbp-w7:~/gnu/github/exiv2/exiv2/build $ ``` --- end body --- comments : --- start comments --- --- end comments ---

This appeared about July 2017. It's surprising as the code is about 10 years and never been touched. ``` 784 rmills@rmillsmm:~/gnu/github/exiv2/exiv2/build $ make Scanning dependencies of target xmp [ 1%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/XMPMeta.cpp.o /Users/rmills/gnu/github/exiv2/exiv2/xmpsdk/src/XMPMeta.cpp:237:3: warning: format specifies type 'unsigned long' but the argument has type 'XMP_OptionBits' (aka 'unsigned int') [-Wformat] OutProcHexInt ( options ); ^~~~~~~~~~~~~~~~~~~~~~~~~ /Users/rmills/gnu/github/exiv2/exiv2/xmpsdk/src/XMPMeta.cpp:87:72: note: expanded from macro 'OutProcHexInt' #define OutProcHexInt(num) { snprintf ( buffer, sizeof(buffer), "%lX", (num) ); /* AUDIT: Using sizeof for snprintf length is safe */ \ ~~~ ^~~~~ /Users/rmills/gnu/github/exiv2/exiv2/xmpsdk/src/XMPMeta.cpp:742:24: warning: implicit conversion from 'unsigned long' to 'XMP_OptionBits' (aka 'unsigned int') changes value from 18446744073709551615 to 4294967295 [-Wconstant-conversion] XMP_OptionBits flag = ~0UL; ~~~~ ^~~~ 2 warnings generated. ``` The fix is also ugly! ``` 781 rmills@rmillsmm:~/gnu/github/exiv2/exiv2/build $ git diff diff --git a/xmpsdk/src/XMPMeta.cpp b/xmpsdk/src/XMPMeta.cpp index 2720cf25..f3e2c627 100644 --- a/xmpsdk/src/XMPMeta.cpp +++ b/xmpsdk/src/XMPMeta.cpp @@ -84,8 +84,13 @@ static const char * kTenSpaces = " "; #define OutProcULong(num) { snprintf ( buffer, sizeof(buffer), "%lu", (num) ); /* AUDIT: Using sizeof for snprintf length is safe */ \ status = (*outProc) ( refCon, buffer, strlen(buffer) ); if ( status != 0 ) goto EXIT; } +#ifdef __APPLE__ +#define OutProcHexInt(num) { snprintf ( buffer, sizeof(buffer), "%X", (num) ); /* AUDIT: Using sizeof for snprintf length is safe */ \ + status = (*outProc) ( refCon, buffer, strlen(buffer) ); if ( status != 0 ) goto EXIT; } +#else #define OutProcHexInt(num) { snprintf ( buffer, sizeof(buffer), "%lX", (num) ); /* AUDIT: Using sizeof for snprintf length is safe */ \ status = (*outProc) ( refCon, buffer, strlen(buffer) ); if ( status != 0 ) goto EXIT; } +#endif #define OutProcHexByte(num) { snprintf ( buffer, sizeof(buffer), "%.2X", (num) ); /* AUDIT: Using sizeof for snprintf length is safe */ \ status = (*outProc) ( refCon, buffer, strlen(buffer) ); if ( status != 0 ) goto EXIT; } @@ -739,7 +744,7 @@ XMPMeta::Initialize() XMP_Assert ( sizeof(XMP_Uns64) == 8 ); XMP_Assert ( sizeof(XMP_OptionBits) == 4 ); // Check that option masking work on all 32 bits. - XMP_OptionBits flag = ~0UL; + XMP_OptionBits flag = ~0; XMP_Assert ( flag == (XMP_OptionBits)(-1L) ); XMP_Assert ( (flag ^ kXMP_PropHasLang) == 0xFFFFFFBFUL ); 782 rmills@rmillsmm:~/gnu/github/exiv2/exiv2/build $ ```

Ugly compiler warnings on MacOS-X from xmpsdk/src/XMPMeta.cpp

https://api.github.com/repos/Exiv2/exiv2/issues/570/comments

https://github.com/Exiv2/exiv2/issues/570

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : Ugly compiler warnings on MacOS-X from xmpsdk/src/XMPMeta.cpp Issue date: --- start body --- This appeared about July 2017. It's surprising as the code is about 10 years and never been touched. ``` 784 rmills@rmillsmm:~/gnu/github/exiv2/exiv2/build $ make Scanning dependencies of target xmp [ 1%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/XMPMeta.cpp.o /Users/rmills/gnu/github/exiv2/exiv2/xmpsdk/src/XMPMeta.cpp:237:3: warning: format specifies type 'unsigned long' but the argument has type 'XMP_OptionBits' (aka 'unsigned int') [-Wformat] OutProcHexInt ( options ); ^~~~~~~~~~~~~~~~~~~~~~~~~ /Users/rmills/gnu/github/exiv2/exiv2/xmpsdk/src/XMPMeta.cpp:87:72: note: expanded from macro 'OutProcHexInt' #define OutProcHexInt(num) { snprintf ( buffer, sizeof(buffer), "%lX", (num) ); /* AUDIT: Using sizeof for snprintf length is safe */ \ ~~~ ^~~~~ /Users/rmills/gnu/github/exiv2/exiv2/xmpsdk/src/XMPMeta.cpp:742:24: warning: implicit conversion from 'unsigned long' to 'XMP_OptionBits' (aka 'unsigned int') changes value from 18446744073709551615 to 4294967295 [-Wconstant-conversion] XMP_OptionBits flag = ~0UL; ~~~~ ^~~~ 2 warnings generated. ``` The fix is also ugly! ``` 781 rmills@rmillsmm:~/gnu/github/exiv2/exiv2/build $ git diff diff --git a/xmpsdk/src/XMPMeta.cpp b/xmpsdk/src/XMPMeta.cpp index 2720cf25..f3e2c627 100644 --- a/xmpsdk/src/XMPMeta.cpp +++ b/xmpsdk/src/XMPMeta.cpp @@ -84,8 +84,13 @@ static const char * kTenSpaces = " "; #define OutProcULong(num) { snprintf ( buffer, sizeof(buffer), "%lu", (num) ); /* AUDIT: Using sizeof for snprintf length is safe */ \ status = (*outProc) ( refCon, buffer, strlen(buffer) ); if ( status != 0 ) goto EXIT; } +#ifdef __APPLE__ +#define OutProcHexInt(num) { snprintf ( buffer, sizeof(buffer), "%X", (num) ); /* AUDIT: Using sizeof for snprintf length is safe */ \ + status = (*outProc) ( refCon, buffer, strlen(buffer) ); if ( status != 0 ) goto EXIT; } +#else #define OutProcHexInt(num) { snprintf ( buffer, sizeof(buffer), "%lX", (num) ); /* AUDIT: Using sizeof for snprintf length is safe */ \ status = (*outProc) ( refCon, buffer, strlen(buffer) ); if ( status != 0 ) goto EXIT; } +#endif #define OutProcHexByte(num) { snprintf ( buffer, sizeof(buffer), "%.2X", (num) ); /* AUDIT: Using sizeof for snprintf length is safe */ \ status = (*outProc) ( refCon, buffer, strlen(buffer) ); if ( status != 0 ) goto EXIT; } @@ -739,7 +744,7 @@ XMPMeta::Initialize() XMP_Assert ( sizeof(XMP_Uns64) == 8 ); XMP_Assert ( sizeof(XMP_OptionBits) == 4 ); // Check that option masking work on all 32 bits. - XMP_OptionBits flag = ~0UL; + XMP_OptionBits flag = ~0; XMP_Assert ( flag == (XMP_OptionBits)(-1L) ); XMP_Assert ( (flag ^ kXMP_PropHasLang) == 0xFFFFFFBFUL ); 782 rmills@rmillsmm:~/gnu/github/exiv2/exiv2/build $ ``` --- end body --- comments : --- start comments --- --- end comments ---

Once 0.27 is released and we move to c++11, we can update the gtest dependency to 1.8.1. I tried to do it now, but the new version is not compatible anymore with old versions of MSVC: https://ci.appveyor.com/project/piponazo/exiv2-wutfp/builds/20577044/job/82wsoafryecm3ctq Continue the development in the branch **updateGtest**.

Update to gtest 1.8.1

https://api.github.com/repos/Exiv2/exiv2/issues/568/comments

https://github.com/Exiv2/exiv2/issues/568

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : Update to gtest 1.8.1 Issue date: --- start body --- Once 0.27 is released and we move to c++11, we can update the gtest dependency to 1.8.1. I tried to do it now, but the new version is not compatible anymore with old versions of MSVC: https://ci.appveyor.com/project/piponazo/exiv2-wutfp/builds/20577044/job/82wsoafryecm3ctq Continue the development in the branch **updateGtest**. --- end body --- comments : --- start comments --- --- end comments ---

XMP (and also custom XMP attributes) and IPTC

Question: is it possible to extract metadata from INDD and PDF file using this library?

https://api.github.com/repos/Exiv2/exiv2/issues/567/comments

https://github.com/Exiv2/exiv2/issues/567

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : Question: is it possible to extract metadata from INDD and PDF file using this library? Issue date: --- start body --- XMP (and also custom XMP attributes) and IPTC --- end body --- comments : --- start comments --- --- end comments ---

In #559 @skal65535 contributed a patch to fix the alpha bit detection in VP8L chunks. He provided a reproduction case that we should add in the python test suite.

Test for alpha bit detection in VP8L chunks

https://api.github.com/repos/Exiv2/exiv2/issues/565/comments

https://github.com/Exiv2/exiv2/issues/565

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : Test for alpha bit detection in VP8L chunks Issue date: --- start body --- In #559 @skal65535 contributed a patch to fix the alpha bit detection in VP8L chunks. He provided a reproduction case that we should add in the python test suite. --- end body --- comments : --- start comments --- --- end comments ---

I suspect this is because the reference output reports 64bit integers. ``` ====================================================================== FAIL: test_run (bugfixes.github.test_CVE_2017_14859.TestCvePoC) ---------------------------------------------------------------------- Traceback (most recent call last): File "/home/rmills/gnu/github/exiv2/exiv2/tests/system_tests.py", line 632, in test_run self.compare_stderr(i, command, processed_stderr, stderr) File "/home/rmills/gnu/github/exiv2/exiv2/tests/bugfixes/github/test_CVE_2017_14859.py", line 20, in compare_stderr self.assertIn(expected_stderr, got_stderr) AssertionError: 'Exiv2 exception in print action for file /home/rmills/gnu/github/exiv2/exiv2/test/data/005-invalid-mem:\ncorrupted image metadata\n' not found in 'Warning: Directory Image, entry 0x011a has unknown Exif (TIFF) type 64772; setting type size 1.\nError: Upper boundary of data for directory Image, entry 0x011b is out of bounds: Offset = 0x00000030, size = 1073741832, exceeds buffer size by 1073734073 Bytes; truncating the entry\nError: Upper boundary of data for directory Photo, entry 0x9003 is out of bounds: Offset = 0x000001f8, size = 3538992, exceeds buffer size by 3531689 Bytes; truncating the entry\nWarning: Directory Nikon3 has an unexpected next pointer; ignored.\nExiv2 exception in print action for file /home/rmills/gnu/github/exiv2/exiv2/test/data/005-invalid-mem:\nArithmetic operation overflow\n' ====================================================================== FAIL: test_run (bugfixes.github.test_CVE_2017_14862.TestCvePoC) ---------------------------------------------------------------------- Traceback (most recent call last): File "/home/rmills/gnu/github/exiv2/exiv2/tests/system_tests.py", line 632, in test_run self.compare_stderr(i, command, processed_stderr, stderr) File "/home/rmills/gnu/github/exiv2/exiv2/tests/bugfixes/github/test_CVE_2017_14862.py", line 23, in compare_stderr self.assertIn(expected_stderr, got_stderr) AssertionError: 'Exiv2 exception in print action for file /home/rmills/gnu/github/exiv2/exiv2/test/data/008-invalid-mem:\ncorrupted image metadata\n' not found in 'Warning: Directory Image, entry 0xff13 has unknown Exif (TIFF) type 65535; setting type size 1.\nError: Offset of directory Image, entry 0xff13 is out of bounds: Offset = 0x30303030; truncating the entry\nWarning: Directory Photo has an unexpected next pointer; ignored.\nError: Offset of directory Photo, entry 0x8827 is out of bounds: Offset = 0x30303030; truncating the entry\nError: Directory Photo, entry 0x9204 has invalid size 4286513153*8; skipping entry.\nWarning: Directory Nikon3 has an unexpected next pointer; ignored.\nError: Upper boundary of data for directory Nikon3, entry 0x0004 is out of bounds: Offset = 0x00000170, size = 1376264, exceeds buffer size by 1369403 Bytes; truncating the entry\nError: Offset of directory Nikon3, entry 0x0006 is out of bounds: Offset = 0x0000e803; truncating the entry\nError: Directory NikonPreview with 12336 entries considered invalid; not read.\nWarning: Directory Nikon3, entry 0x0095 has unknown Exif (TIFF) type 2562; setting type size 1.\nError: Offset of directory Nikon3, entry 0x009c is out of bounds: Offset = 0x000ffff8; truncating the entry\nExiv2 exception in print action for file /home/rmills/gnu/github/exiv2/exiv2/test/data/008-invalid-mem:\nArithmetic operation overflow\n' ====================================================================== FAIL: test_run (bugfixes.github.test_CVE_2017_14864.TestCvePoC) ---------------------------------------------------------------------- Traceback (most recent call last): File "/home/rmills/gnu/github/exiv2/exiv2/tests/system_tests.py", line 632, in test_run self.compare_stderr(i, command, processed_stderr, stderr) File "/home/rmills/gnu/github/exiv2/exiv2/tests/system_tests.py", line 755, in compare_stderr msg="Standard error does not match" File "/home/rmills/gnu/github/exiv2/exiv2/tests/system_tests.py", line 726, in _compare_output expected, got, msg=msg AssertionError: 'Exiv[73 chars]test/data/02-Invalid-mem-def:\nArithmetic operation overflow\n' != 'Exiv[73 chars]test/data/02-Invalid-mem-def:\ncorrupted image metadata\n' Exiv2 exception in print action for file /home/rmills/gnu/github/exiv2/exiv2/test/data/02-Invalid-mem-def: - Arithmetic operation overflow + corrupted image metadata : Standard error does not match ====================================================================== FAIL: test_run (bugfixes.github.test_issue_262.DivByZeroInPrintIFD) ---------------------------------------------------------------------- Traceback (most recent call last): File "/home/rmills/gnu/github/exiv2/exiv2/tests/system_tests.py", line 632, in test_run self.compare_stderr(i, command, processed_stderr, stderr) File "/home/rmills/gnu/github/exiv2/exiv2/tests/system_tests.py", line 755, in compare_stderr msg="Standard error does not match" File "/home/rmills/gnu/github/exiv2/exiv2/tests/system_tests.py", line 726, in _compare_output expected, got, msg=msg AssertionError: '' != 'Exiv2 exception in print action for file [93 chars]ta\n' + Exiv2 exception in print action for file /home/rmills/gnu/github/exiv2/exiv2/test/data/7-printIFD-divbyzero-1: + corrupted image metadata : Standard error does not match ---------------------------------------------------------------------- Ran 135 tests in 2.877s FAILED (failures=4, skipped=5) Makefile:115: recipe for target 'newtests' failed make[5]: *** [newtests] Error 1 Makefile:102: recipe for target 'test' failed make[4]: *** [test] Error 2 CMakeFiles/tests.dir/build.make:57: recipe for target 'CMakeFiles/tests' failed make[3]: *** [CMakeFiles/tests] Error 2 CMakeFiles/Makefile2:136: recipe for target 'CMakeFiles/tests.dir/all' failed make[2]: *** [CMakeFiles/tests.dir/all] Error 2 CMakeFiles/Makefile2:143: recipe for target 'CMakeFiles/tests.dir/rule' failed make[1]: *** [CMakeFiles/tests.dir/rule] Error 2 Makefile:175: recipe for target 'tests' failed make: *** [tests] Error 2 rmills@ubuntu:~/gnu/github/exiv2/exiv2/build$ rmills@ubuntu:~/gnu/github/exiv2/exiv2/build$ rmills@ubuntu:~/gnu/github/exiv2/exiv2/build$ rmills@ubuntu:~/gnu/github/exiv2/exiv2/build$ rmills@ubuntu:~/gnu/github/exiv2/exiv2/build$ ```

Python test suite is failing on 32 bit builds (MinGW32, Cygwin32, Linux)

https://api.github.com/repos/Exiv2/exiv2/issues/562/comments

https://github.com/Exiv2/exiv2/issues/562

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : Python test suite is failing on 32 bit builds (MinGW32, Cygwin32, Linux) Issue date: --- start body --- I suspect this is because the reference output reports 64bit integers. ``` ====================================================================== FAIL: test_run (bugfixes.github.test_CVE_2017_14859.TestCvePoC) ---------------------------------------------------------------------- Traceback (most recent call last): File "/home/rmills/gnu/github/exiv2/exiv2/tests/system_tests.py", line 632, in test_run self.compare_stderr(i, command, processed_stderr, stderr) File "/home/rmills/gnu/github/exiv2/exiv2/tests/bugfixes/github/test_CVE_2017_14859.py", line 20, in compare_stderr self.assertIn(expected_stderr, got_stderr) AssertionError: 'Exiv2 exception in print action for file /home/rmills/gnu/github/exiv2/exiv2/test/data/005-invalid-mem:\ncorrupted image metadata\n' not found in 'Warning: Directory Image, entry 0x011a has unknown Exif (TIFF) type 64772; setting type size 1.\nError: Upper boundary of data for directory Image, entry 0x011b is out of bounds: Offset = 0x00000030, size = 1073741832, exceeds buffer size by 1073734073 Bytes; truncating the entry\nError: Upper boundary of data for directory Photo, entry 0x9003 is out of bounds: Offset = 0x000001f8, size = 3538992, exceeds buffer size by 3531689 Bytes; truncating the entry\nWarning: Directory Nikon3 has an unexpected next pointer; ignored.\nExiv2 exception in print action for file /home/rmills/gnu/github/exiv2/exiv2/test/data/005-invalid-mem:\nArithmetic operation overflow\n' ====================================================================== FAIL: test_run (bugfixes.github.test_CVE_2017_14862.TestCvePoC) ---------------------------------------------------------------------- Traceback (most recent call last): File "/home/rmills/gnu/github/exiv2/exiv2/tests/system_tests.py", line 632, in test_run self.compare_stderr(i, command, processed_stderr, stderr) File "/home/rmills/gnu/github/exiv2/exiv2/tests/bugfixes/github/test_CVE_2017_14862.py", line 23, in compare_stderr self.assertIn(expected_stderr, got_stderr) AssertionError: 'Exiv2 exception in print action for file /home/rmills/gnu/github/exiv2/exiv2/test/data/008-invalid-mem:\ncorrupted image metadata\n' not found in 'Warning: Directory Image, entry 0xff13 has unknown Exif (TIFF) type 65535; setting type size 1.\nError: Offset of directory Image, entry 0xff13 is out of bounds: Offset = 0x30303030; truncating the entry\nWarning: Directory Photo has an unexpected next pointer; ignored.\nError: Offset of directory Photo, entry 0x8827 is out of bounds: Offset = 0x30303030; truncating the entry\nError: Directory Photo, entry 0x9204 has invalid size 4286513153*8; skipping entry.\nWarning: Directory Nikon3 has an unexpected next pointer; ignored.\nError: Upper boundary of data for directory Nikon3, entry 0x0004 is out of bounds: Offset = 0x00000170, size = 1376264, exceeds buffer size by 1369403 Bytes; truncating the entry\nError: Offset of directory Nikon3, entry 0x0006 is out of bounds: Offset = 0x0000e803; truncating the entry\nError: Directory NikonPreview with 12336 entries considered invalid; not read.\nWarning: Directory Nikon3, entry 0x0095 has unknown Exif (TIFF) type 2562; setting type size 1.\nError: Offset of directory Nikon3, entry 0x009c is out of bounds: Offset = 0x000ffff8; truncating the entry\nExiv2 exception in print action for file /home/rmills/gnu/github/exiv2/exiv2/test/data/008-invalid-mem:\nArithmetic operation overflow\n' ====================================================================== FAIL: test_run (bugfixes.github.test_CVE_2017_14864.TestCvePoC) ---------------------------------------------------------------------- Traceback (most recent call last): File "/home/rmills/gnu/github/exiv2/exiv2/tests/system_tests.py", line 632, in test_run self.compare_stderr(i, command, processed_stderr, stderr) File "/home/rmills/gnu/github/exiv2/exiv2/tests/system_tests.py", line 755, in compare_stderr msg="Standard error does not match" File "/home/rmills/gnu/github/exiv2/exiv2/tests/system_tests.py", line 726, in _compare_output expected, got, msg=msg AssertionError: 'Exiv[73 chars]test/data/02-Invalid-mem-def:\nArithmetic operation overflow\n' != 'Exiv[73 chars]test/data/02-Invalid-mem-def:\ncorrupted image metadata\n' Exiv2 exception in print action for file /home/rmills/gnu/github/exiv2/exiv2/test/data/02-Invalid-mem-def: - Arithmetic operation overflow + corrupted image metadata : Standard error does not match ====================================================================== FAIL: test_run (bugfixes.github.test_issue_262.DivByZeroInPrintIFD) ---------------------------------------------------------------------- Traceback (most recent call last): File "/home/rmills/gnu/github/exiv2/exiv2/tests/system_tests.py", line 632, in test_run self.compare_stderr(i, command, processed_stderr, stderr) File "/home/rmills/gnu/github/exiv2/exiv2/tests/system_tests.py", line 755, in compare_stderr msg="Standard error does not match" File "/home/rmills/gnu/github/exiv2/exiv2/tests/system_tests.py", line 726, in _compare_output expected, got, msg=msg AssertionError: '' != 'Exiv2 exception in print action for file [93 chars]ta\n' + Exiv2 exception in print action for file /home/rmills/gnu/github/exiv2/exiv2/test/data/7-printIFD-divbyzero-1: + corrupted image metadata : Standard error does not match ---------------------------------------------------------------------- Ran 135 tests in 2.877s FAILED (failures=4, skipped=5) Makefile:115: recipe for target 'newtests' failed make[5]: *** [newtests] Error 1 Makefile:102: recipe for target 'test' failed make[4]: *** [test] Error 2 CMakeFiles/tests.dir/build.make:57: recipe for target 'CMakeFiles/tests' failed make[3]: *** [CMakeFiles/tests] Error 2 CMakeFiles/Makefile2:136: recipe for target 'CMakeFiles/tests.dir/all' failed make[2]: *** [CMakeFiles/tests.dir/all] Error 2 CMakeFiles/Makefile2:143: recipe for target 'CMakeFiles/tests.dir/rule' failed make[1]: *** [CMakeFiles/tests.dir/rule] Error 2 Makefile:175: recipe for target 'tests' failed make: *** [tests] Error 2 rmills@ubuntu:~/gnu/github/exiv2/exiv2/build$ rmills@ubuntu:~/gnu/github/exiv2/exiv2/build$ rmills@ubuntu:~/gnu/github/exiv2/exiv2/build$ rmills@ubuntu:~/gnu/github/exiv2/exiv2/build$ rmills@ubuntu:~/gnu/github/exiv2/exiv2/build$ ``` --- end body --- comments : --- start comments --- --- end comments ---

Tested in ubuntu 16.04 64bit, exiv2 (master 3f2e0de && 0.27-RC2) `$ exiv2 $POC` [POC file](https://github.com/Marsman1996/pocs/blob/master/exiv2/CVE-2018-19607/poc9-isoSpeed) gdb info: ``` Program received signal SIGSEGV, Segmentation fault. __strcmp_sse2_unaligned () at ../sysdeps/x86_64/multiarch/strcmp-sse2-unaligned.S:31 31 ../sysdeps/x86_64/multiarch/strcmp-sse2-unaligned.S: No such file or directory. (gdb) bt #0 __strcmp_sse2_unaligned () at ../sysdeps/x86_64/multiarch/strcmp-sse2-unaligned.S:31 #1 0x00007ffff794b9ff in Exiv2::isoSpeed (ed=...) at /home/marsman/Desktop/crashana/exiv2/exiv2-3f2e0de/src/easyaccess.cpp:178 #2 0x0000000000425aef in Action::Print::printTag (this=0x65dde0, exifData=..., easyAccessFct=0x412390 <Exiv2::isoSpeed(Exiv2::ExifData const&)@plt>, label="ISO speed") at /home/marsman/Desktop/crashana/exiv2/exiv2-3f2e0de/src/actions.cpp:500 #3 0x0000000000424038 in Action::Print::printSummary (this=0x65dde0) at /home/marsman/Desktop/crashana/exiv2/exiv2-3f2e0de/src/actions.cpp:387 #4 0x0000000000422901 in Action::Print::run (this=0x65dde0, path="/home/marsman/Desktop/poc9-isoSpeed") at /home/marsman/Desktop/crashana/exiv2/exiv2-3f2e0de/src/actions.cpp:246 #5 0x0000000000412e39 in main (argc=2, argv=0x7fffffffdda8) at /home/marsman/Desktop/crashana/exiv2/exiv2-3f2e0de/src/exiv2.cpp:169 ``` ASAN info: ``` ==91500==ERROR: AddressSanitizer: SEGV on unknown address 0x000000000000 (pc 0x7f186e2da05a bp 0x7fff4e8fe260 sp 0x7fff4e8fd9f0 T0) #0 0x7f186e2da059 (/usr/lib/x86_64-linux-gnu/libasan.so.2+0x47059) #1 0x7f186db38b0c in Exiv2::isoSpeed(Exiv2::ExifData const&) /home/marsman/Desktop/crashana/exiv2/exiv2-3f2e0de/src/easyaccess.cpp:178 #2 0x42f9fd in Action::Print::printTag(Exiv2::ExifData const&, std::_List_const_iterator<Exiv2::Exifdatum> (*)(Exiv2::ExifData const&), std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&) const /home/marsman/Desktop/crashana/exiv2/exiv2-3f2e0de/src/actions.cpp:500 #3 0x42d6c2 in Action::Print::printSummary() /home/marsman/Desktop/crashana/exiv2/exiv2-3f2e0de/src/actions.cpp:387 #4 0x42b9db in Action::Print::run(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&) /home/marsman/Desktop/crashana/exiv2/exiv2-3f2e0de/src/actions.cpp:246 #5 0x4139a6 in main /home/marsman/Desktop/crashana/exiv2/exiv2-3f2e0de/src/exiv2.cpp:169 #6 0x7f186cf9a82f in __libc_start_main (/lib/x86_64-linux-gnu/libc.so.6+0x2082f) #7 0x413308 in _start (/home/marsman/Desktop/crashana/exiv2/exiv2-3f2e0de/build_asan/bin/exiv2+0x413308) AddressSanitizer can not provide additional info. SUMMARY: AddressSanitizer: SEGV ??:0 ?? ``` Addition: This bug was found with mem-AFL, which is based on AFL. Mem-AFL is developed by Yanhao(unfuzzable123@gmail.com) & Marsman1996(lqliuyuwei@outlook.com)

CVE-2018-19607: SEGV in Exiv2::isoSpeed at easyaccess.cpp:178

https://api.github.com/repos/Exiv2/exiv2/issues/561/comments

https://github.com/Exiv2/exiv2/issues/561

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : CVE-2018-19607: SEGV in Exiv2::isoSpeed at easyaccess.cpp:178 Issue date: --- start body --- Tested in ubuntu 16.04 64bit, exiv2 (master 3f2e0de && 0.27-RC2) `$ exiv2 $POC` [POC file](https://github.com/Marsman1996/pocs/blob/master/exiv2/CVE-2018-19607/poc9-isoSpeed) gdb info: ``` Program received signal SIGSEGV, Segmentation fault. __strcmp_sse2_unaligned () at ../sysdeps/x86_64/multiarch/strcmp-sse2-unaligned.S:31 31 ../sysdeps/x86_64/multiarch/strcmp-sse2-unaligned.S: No such file or directory. (gdb) bt #0 __strcmp_sse2_unaligned () at ../sysdeps/x86_64/multiarch/strcmp-sse2-unaligned.S:31 #1 0x00007ffff794b9ff in Exiv2::isoSpeed (ed=...) at /home/marsman/Desktop/crashana/exiv2/exiv2-3f2e0de/src/easyaccess.cpp:178 #2 0x0000000000425aef in Action::Print::printTag (this=0x65dde0, exifData=..., easyAccessFct=0x412390 <Exiv2::isoSpeed(Exiv2::ExifData const&)@plt>, label="ISO speed") at /home/marsman/Desktop/crashana/exiv2/exiv2-3f2e0de/src/actions.cpp:500 #3 0x0000000000424038 in Action::Print::printSummary (this=0x65dde0) at /home/marsman/Desktop/crashana/exiv2/exiv2-3f2e0de/src/actions.cpp:387 #4 0x0000000000422901 in Action::Print::run (this=0x65dde0, path="/home/marsman/Desktop/poc9-isoSpeed") at /home/marsman/Desktop/crashana/exiv2/exiv2-3f2e0de/src/actions.cpp:246 #5 0x0000000000412e39 in main (argc=2, argv=0x7fffffffdda8) at /home/marsman/Desktop/crashana/exiv2/exiv2-3f2e0de/src/exiv2.cpp:169 ``` ASAN info: ``` ==91500==ERROR: AddressSanitizer: SEGV on unknown address 0x000000000000 (pc 0x7f186e2da05a bp 0x7fff4e8fe260 sp 0x7fff4e8fd9f0 T0) #0 0x7f186e2da059 (/usr/lib/x86_64-linux-gnu/libasan.so.2+0x47059) #1 0x7f186db38b0c in Exiv2::isoSpeed(Exiv2::ExifData const&) /home/marsman/Desktop/crashana/exiv2/exiv2-3f2e0de/src/easyaccess.cpp:178 #2 0x42f9fd in Action::Print::printTag(Exiv2::ExifData const&, std::_List_const_iterator<Exiv2::Exifdatum> (*)(Exiv2::ExifData const&), std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&) const /home/marsman/Desktop/crashana/exiv2/exiv2-3f2e0de/src/actions.cpp:500 #3 0x42d6c2 in Action::Print::printSummary() /home/marsman/Desktop/crashana/exiv2/exiv2-3f2e0de/src/actions.cpp:387 #4 0x42b9db in Action::Print::run(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&) /home/marsman/Desktop/crashana/exiv2/exiv2-3f2e0de/src/actions.cpp:246 #5 0x4139a6 in main /home/marsman/Desktop/crashana/exiv2/exiv2-3f2e0de/src/exiv2.cpp:169 #6 0x7f186cf9a82f in __libc_start_main (/lib/x86_64-linux-gnu/libc.so.6+0x2082f) #7 0x413308 in _start (/home/marsman/Desktop/crashana/exiv2/exiv2-3f2e0de/build_asan/bin/exiv2+0x413308) AddressSanitizer can not provide additional info. SUMMARY: AddressSanitizer: SEGV ??:0 ?? ``` Addition: This bug was found with mem-AFL, which is based on AFL. Mem-AFL is developed by Yanhao(unfuzzable123@gmail.com) & Marsman1996(lqliuyuwei@outlook.com) --- end body --- comments : --- start comments --- --- end comments ---

Three issues: 1 test harness crashes 2 incorrect version of zlib libraries 3 linker issues involving zlib and expat **1 test harness crashes** I will investigate the crash in conversions.sh as it appears to be coming from yesterday's change in #557 (eraseFamily()).  The message is "Debug Assertion Failed!". Program: MSVCP140D.dll .... include\vector line:240 Expression: vector iterators incompatible. **2 zlib detection** There's an issue with the zlib being used by Conan. It finds and uses the 32 bit libraries installed in `c:\Program Files\zlib`. The work-around is to delete `c:\Program Files\zlib` ``` "C:\Users\rmills\gnu\github\exiv2\exiv2\build\samples\addmoddel.vcxproj" (default target) (3) -> "C:\Users\rmills\gnu\github\exiv2\exiv2\build\src\exiv2lib.vcxproj" (default target) (4) -> (Link target) -> C:\Program Files\zlib\lib\zlibd.lib : warning LNK4272: library machine type 'X86' conflicts with target machine type 'x64' [C:\Users\rmills\ gnu\github\exiv2\exiv2\build\src\exiv2lib.vcxproj] ``` I experienced difficulties with zlib when developing _(the now dead)_ _**cmakeBuild.cmd**_. zlib has different names for different builds of their library such as `zlib1.dll` and `zlibstatic.lib` **3 expat and zlib link warnings** This is ugly. I believe it's benign. The test suite runs well (with the exception of conversions.sh) ``` "C:\Users\rmills\gnu\github\exiv2\exiv2\build\ALL_BUILD.vcxproj" (default target) (1) -> "C:\Users\rmills\gnu\github\exiv2\exiv2\build\samples\addmoddel.vcxproj" (default target) (3) -> "C:\Users\rmills\gnu\github\exiv2\exiv2\build\src\exiv2lib.vcxproj" (default target) (4) -> (Link target) -> zlib.lib(adler32.obj) : warning LNK4099: PDB 'zlibstatic.pdb' was not found with 'zlib.lib(adler32.obj)' or at 'C:\Users\rmills\gnu\github\e xiv2\exiv2\build\src\Debug\zlibstatic.pdb'; linking object as if no debug info [C:\Users\rmills\gnu\github\exiv2\exiv2\build\src\exiv2lib.vcxp roj] zlib.lib(compress.obj) : warning LNK4099: PDB 'zlibstatic.pdb' was not found with 'zlib.lib(compress.obj)' or at 'C:\Users\rmills\gnu\github \exiv2\exiv2\build\src\Debug\zlibstatic.pdb'; linking object as if no debug info [C:\Users\rmills\gnu\github\exiv2\exiv2\build\src\exiv2lib.vc xproj] zlib.lib(crc32.obj) : warning LNK4099: PDB 'zlibstatic.pdb' was not found with 'zlib.lib(crc32.obj)' or at 'C:\Users\rmills\gnu\github\exiv2 \exiv2\build\src\Debug\zlibstatic.pdb'; linking object as if no debug info [C:\Users\rmills\gnu\github\exiv2\exiv2\build\src\exiv2lib.vcxproj] zlib.lib(deflate.obj) : warning LNK4099: PDB 'zlibstatic.pdb' was not found with 'zlib.lib(deflate.obj)' or at 'C:\Users\rmills\gnu\github\e xiv2\exiv2\build\src\Debug\zlibstatic.pdb'; linking object as if no debug info [C:\Users\rmills\gnu\github\exiv2\exiv2\build\src\exiv2lib.vcxp roj] zlib.lib(inflate.obj) : warning LNK4099: PDB 'zlibstatic.pdb' was not found with 'zlib.lib(inflate.obj)' or at 'C:\Users\rmills\gnu\github\e xiv2\exiv2\build\src\Debug\zlibstatic.pdb'; linking object as if no debug info [C:\Users\rmills\gnu\github\exiv2\exiv2\build\src\exiv2lib.vcxp roj] zlib.lib(inftrees.obj) : warning LNK4099: PDB 'zlibstatic.pdb' was not found with 'zlib.lib(inftrees.obj)' or at 'C:\Users\rmills\gnu\github \exiv2\exiv2\build\src\Debug\zlibstatic.pdb'; linking object as if no debug info [C:\Users\rmills\gnu\github\exiv2\exiv2\build\src\exiv2lib.vc xproj] zlib.lib(inffast.obj) : warning LNK4099: PDB 'zlibstatic.pdb' was not found with 'zlib.lib(inffast.obj)' or at 'C:\Users\rmills\gnu\github\e xiv2\exiv2\build\src\Debug\zlibstatic.pdb'; linking object as if no debug info [C:\Users\rmills\gnu\github\exiv2\exiv2\build\src\exiv2lib.vcxp roj] zlib.lib(trees.obj) : warning LNK4099: PDB 'zlibstatic.pdb' was not found with 'zlib.lib(trees.obj)' or at 'C:\Users\rmills\gnu\github\exiv2 \exiv2\build\src\Debug\zlibstatic.pdb'; linking object as if no debug info [C:\Users\rmills\gnu\github\exiv2\exiv2\build\src\exiv2lib.vcxproj] zlib.lib(uncompr.obj) : warning LNK4099: PDB 'zlibstatic.pdb' was not found with 'zlib.lib(uncompr.obj)' or at 'C:\Users\rmills\gnu\github\e xiv2\exiv2\build\src\Debug\zlibstatic.pdb'; linking object as if no debug info [C:\Users\rmills\gnu\github\exiv2\exiv2\build\src\exiv2lib.vcxp roj] zlib.lib(zutil.obj) : warning LNK4099: PDB 'zlibstatic.pdb' was not found with 'zlib.lib(zutil.obj)' or at 'C:\Users\rmills\gnu\github\exiv2 \exiv2\build\src\Debug\zlibstatic.pdb'; linking object as if no debug info [C:\Users\rmills\gnu\github\exiv2\exiv2\build\src\exiv2lib.vcxproj] expatd.lib(loadlibrary.obj) : warning LNK4099: PDB 'expat.pdb' was not found with 'expatd.lib(loadlibrary.obj)' or at 'C:\Users\rmills\gnu\g ithub\exiv2\exiv2\build\src\Debug\expat.pdb'; linking object as if no debug info [C:\Users\rmills\gnu\github\exiv2\exiv2\build\src\exiv2lib.vc xproj] expatd.lib(xmlparse.obj) : warning LNK4099: PDB 'expat.pdb' was not found with 'expatd.lib(xmlparse.obj)' or at 'C:\Users\rmills\gnu\github\ exiv2\exiv2\build\src\Debug\expat.pdb'; linking object as if no debug info [C:\Users\rmills\gnu\github\exiv2\exiv2\build\src\exiv2lib.vcxproj] expatd.lib(xmlrole.obj) : warning LNK4099: PDB 'expat.pdb' was not found with 'expatd.lib(xmlrole.obj)' or at 'C:\Users\rmills\gnu\github\ex iv2\exiv2\build\src\Debug\expat.pdb'; linking object as if no debug info [C:\Users\rmills\gnu\github\exiv2\exiv2\build\src\exiv2lib.vcxproj] expatd.lib(xmltok.obj) : warning LNK4099: PDB 'expat.pdb' was not found with 'expatd.lib(xmltok.obj)' or at 'C:\Users\rmills\gnu\github\exiv 2\exiv2\build\src\Debug\expat.pdb'; linking object as if no debug info [C:\Users\rmills\gnu\github\exiv2\exiv2\build\src\exiv2lib.vcxproj] ```

MSVC debug build issues

https://api.github.com/repos/Exiv2/exiv2/issues/560/comments

https://github.com/Exiv2/exiv2/issues/560

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : MSVC debug build issues Issue date: --- start body --- Three issues: 1 test harness crashes 2 incorrect version of zlib libraries 3 linker issues involving zlib and expat **1 test harness crashes** I will investigate the crash in conversions.sh as it appears to be coming from yesterday's change in #557 (eraseFamily()).  The message is "Debug Assertion Failed!". Program: MSVCP140D.dll .... include\vector line:240 Expression: vector iterators incompatible. **2 zlib detection** There's an issue with the zlib being used by Conan. It finds and uses the 32 bit libraries installed in `c:\Program Files\zlib`. The work-around is to delete `c:\Program Files\zlib` ``` "C:\Users\rmills\gnu\github\exiv2\exiv2\build\samples\addmoddel.vcxproj" (default target) (3) -> "C:\Users\rmills\gnu\github\exiv2\exiv2\build\src\exiv2lib.vcxproj" (default target) (4) -> (Link target) -> C:\Program Files\zlib\lib\zlibd.lib : warning LNK4272: library machine type 'X86' conflicts with target machine type 'x64' [C:\Users\rmills\ gnu\github\exiv2\exiv2\build\src\exiv2lib.vcxproj] ``` I experienced difficulties with zlib when developing _(the now dead)_ _**cmakeBuild.cmd**_. zlib has different names for different builds of their library such as `zlib1.dll` and `zlibstatic.lib` **3 expat and zlib link warnings** This is ugly. I believe it's benign. The test suite runs well (with the exception of conversions.sh) ``` "C:\Users\rmills\gnu\github\exiv2\exiv2\build\ALL_BUILD.vcxproj" (default target) (1) -> "C:\Users\rmills\gnu\github\exiv2\exiv2\build\samples\addmoddel.vcxproj" (default target) (3) -> "C:\Users\rmills\gnu\github\exiv2\exiv2\build\src\exiv2lib.vcxproj" (default target) (4) -> (Link target) -> zlib.lib(adler32.obj) : warning LNK4099: PDB 'zlibstatic.pdb' was not found with 'zlib.lib(adler32.obj)' or at 'C:\Users\rmills\gnu\github\e xiv2\exiv2\build\src\Debug\zlibstatic.pdb'; linking object as if no debug info [C:\Users\rmills\gnu\github\exiv2\exiv2\build\src\exiv2lib.vcxp roj] zlib.lib(compress.obj) : warning LNK4099: PDB 'zlibstatic.pdb' was not found with 'zlib.lib(compress.obj)' or at 'C:\Users\rmills\gnu\github \exiv2\exiv2\build\src\Debug\zlibstatic.pdb'; linking object as if no debug info [C:\Users\rmills\gnu\github\exiv2\exiv2\build\src\exiv2lib.vc xproj] zlib.lib(crc32.obj) : warning LNK4099: PDB 'zlibstatic.pdb' was not found with 'zlib.lib(crc32.obj)' or at 'C:\Users\rmills\gnu\github\exiv2 \exiv2\build\src\Debug\zlibstatic.pdb'; linking object as if no debug info [C:\Users\rmills\gnu\github\exiv2\exiv2\build\src\exiv2lib.vcxproj] zlib.lib(deflate.obj) : warning LNK4099: PDB 'zlibstatic.pdb' was not found with 'zlib.lib(deflate.obj)' or at 'C:\Users\rmills\gnu\github\e xiv2\exiv2\build\src\Debug\zlibstatic.pdb'; linking object as if no debug info [C:\Users\rmills\gnu\github\exiv2\exiv2\build\src\exiv2lib.vcxp roj] zlib.lib(inflate.obj) : warning LNK4099: PDB 'zlibstatic.pdb' was not found with 'zlib.lib(inflate.obj)' or at 'C:\Users\rmills\gnu\github\e xiv2\exiv2\build\src\Debug\zlibstatic.pdb'; linking object as if no debug info [C:\Users\rmills\gnu\github\exiv2\exiv2\build\src\exiv2lib.vcxp roj] zlib.lib(inftrees.obj) : warning LNK4099: PDB 'zlibstatic.pdb' was not found with 'zlib.lib(inftrees.obj)' or at 'C:\Users\rmills\gnu\github \exiv2\exiv2\build\src\Debug\zlibstatic.pdb'; linking object as if no debug info [C:\Users\rmills\gnu\github\exiv2\exiv2\build\src\exiv2lib.vc xproj] zlib.lib(inffast.obj) : warning LNK4099: PDB 'zlibstatic.pdb' was not found with 'zlib.lib(inffast.obj)' or at 'C:\Users\rmills\gnu\github\e xiv2\exiv2\build\src\Debug\zlibstatic.pdb'; linking object as if no debug info [C:\Users\rmills\gnu\github\exiv2\exiv2\build\src\exiv2lib.vcxp roj] zlib.lib(trees.obj) : warning LNK4099: PDB 'zlibstatic.pdb' was not found with 'zlib.lib(trees.obj)' or at 'C:\Users\rmills\gnu\github\exiv2 \exiv2\build\src\Debug\zlibstatic.pdb'; linking object as if no debug info [C:\Users\rmills\gnu\github\exiv2\exiv2\build\src\exiv2lib.vcxproj] zlib.lib(uncompr.obj) : warning LNK4099: PDB 'zlibstatic.pdb' was not found with 'zlib.lib(uncompr.obj)' or at 'C:\Users\rmills\gnu\github\e xiv2\exiv2\build\src\Debug\zlibstatic.pdb'; linking object as if no debug info [C:\Users\rmills\gnu\github\exiv2\exiv2\build\src\exiv2lib.vcxp roj] zlib.lib(zutil.obj) : warning LNK4099: PDB 'zlibstatic.pdb' was not found with 'zlib.lib(zutil.obj)' or at 'C:\Users\rmills\gnu\github\exiv2 \exiv2\build\src\Debug\zlibstatic.pdb'; linking object as if no debug info [C:\Users\rmills\gnu\github\exiv2\exiv2\build\src\exiv2lib.vcxproj] expatd.lib(loadlibrary.obj) : warning LNK4099: PDB 'expat.pdb' was not found with 'expatd.lib(loadlibrary.obj)' or at 'C:\Users\rmills\gnu\g ithub\exiv2\exiv2\build\src\Debug\expat.pdb'; linking object as if no debug info [C:\Users\rmills\gnu\github\exiv2\exiv2\build\src\exiv2lib.vc xproj] expatd.lib(xmlparse.obj) : warning LNK4099: PDB 'expat.pdb' was not found with 'expatd.lib(xmlparse.obj)' or at 'C:\Users\rmills\gnu\github\ exiv2\exiv2\build\src\Debug\expat.pdb'; linking object as if no debug info [C:\Users\rmills\gnu\github\exiv2\exiv2\build\src\exiv2lib.vcxproj] expatd.lib(xmlrole.obj) : warning LNK4099: PDB 'expat.pdb' was not found with 'expatd.lib(xmlrole.obj)' or at 'C:\Users\rmills\gnu\github\ex iv2\exiv2\build\src\Debug\expat.pdb'; linking object as if no debug info [C:\Users\rmills\gnu\github\exiv2\exiv2\build\src\exiv2lib.vcxproj] expatd.lib(xmltok.obj) : warning LNK4099: PDB 'expat.pdb' was not found with 'expatd.lib(xmltok.obj)' or at 'C:\Users\rmills\gnu\github\exiv 2\exiv2\build\src\Debug\expat.pdb'; linking object as if no debug info [C:\Users\rmills\gnu\github\exiv2\exiv2\build\src\exiv2lib.vcxproj] ``` --- end body --- comments : --- start comments --- --- end comments ---

[This commit](https://github.com/Exiv2/exiv2/commit/ab6aac054ade7c01a9cff5a9dbaa095b2e40be5a#diff-8d5bd1c82cc9486b38f5e2e1e5da78b1L131) removes `EXIV2_TEST_VERSION` macro along with deprecated `EXIV2_CHECK_VERSION`, seemingly by error. It brakes `gexiv2`, for example. I'm not at all sure the intention was to leave no way to check the version.

EXIV2_TEST_VERSION is removed in 0.27

https://api.github.com/repos/Exiv2/exiv2/issues/555/comments

https://github.com/Exiv2/exiv2/issues/555

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : EXIV2_TEST_VERSION is removed in 0.27 Issue date: --- start body --- [This commit](https://github.com/Exiv2/exiv2/commit/ab6aac054ade7c01a9cff5a9dbaa095b2e40be5a#diff-8d5bd1c82cc9486b38f5e2e1e5da78b1L131) removes `EXIV2_TEST_VERSION` macro along with deprecated `EXIV2_CHECK_VERSION`, seemingly by error. It brakes `gexiv2`, for example. I'm not at all sure the intention was to leave no way to check the version. --- end body --- comments : --- start comments --- --- end comments ---

http://www.exiv2.org/ says the latest release is 0.26 http://exiv2.dyndns.org says the latest release 0.27 RC2 https://github.com/Exiv2/exiv2/releases says the latest release is 0.27 RC1

exiv2.org out of date compared to exiv2.dyndns.org

https://api.github.com/repos/Exiv2/exiv2/issues/554/comments

https://github.com/Exiv2/exiv2/issues/554

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : exiv2.org out of date compared to exiv2.dyndns.org Issue date: --- start body --- http://www.exiv2.org/ says the latest release is 0.26 http://exiv2.dyndns.org says the latest release 0.27 RC2 https://github.com/Exiv2/exiv2/releases says the latest release is 0.27 RC1 --- end body --- comments : --- start comments --- --- end comments ---

Scanning dependencies of target digikamcore [ 55%] Building CXX object core/app/CMakeFiles/digikamcore.dir/utils/digikam_debug.cpp.o [ 55%] Building CXX object core/app/CMakeFiles/digikamcore.dir/digikamcore_automoc.cpp.o [ 55%] Linking CXX shared library libdigikamcore.so ../libs/metadataengine/CMakeFiles/dmetadata_src.dir/engine/metaengine_exif.cpp.o: dans la fonction « Digikam::MetaEngine::getExifThumbnail(bool) const »: /home/gilles/devel/6.X/core/libs/metadataengine/engine/metaengine_exif.cpp:884: référence indéfinie vers « Exiv2::DataBuf::DataBuf(Exiv2::DataBufRef) » ../libs/metadataengine/CMakeFiles/dmetadata_src.dir/engine/metaengine_exif.cpp.o: dans la fonction « Exiv2::ValueType<unsigned int>::dataArea() const »: /usr/include/exiv2/value.hpp:1740: référence indéfinie vers « Exiv2::DataBuf::DataBuf(Exiv2::DataBufRef) » collect2: erreur : ld a retourné 1 code d'état d'exécution core/app/CMakeFiles/digikamcore.dir/build.make:2109 : la recette pour la cible « core/app/libdigikamcore.so.6.0.0 » a échouée make[2]: *** [core/app/libdigikamcore.so.6.0.0] Erreur 1 CMakeFiles/Makefile2:9006 : la recette pour la cible « core/app/CMakeFiles/digikamcore.dir/all » a échouée make[1]: *** [core/app/CMakeFiles/digikamcore.dir/all] Erreur 2 Makefile:138 : la recette pour la cible « all » a échouée make: *** [all] Erreur 2 Not reproducible with RC2 tag... Gilles Caulier

Exiv2 from git/master : DataBuf symbol not found

https://api.github.com/repos/Exiv2/exiv2/issues/552/comments

https://github.com/Exiv2/exiv2/issues/552

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : Exiv2 from git/master : DataBuf symbol not found Issue date: --- start body --- Scanning dependencies of target digikamcore [ 55%] Building CXX object core/app/CMakeFiles/digikamcore.dir/utils/digikam_debug.cpp.o [ 55%] Building CXX object core/app/CMakeFiles/digikamcore.dir/digikamcore_automoc.cpp.o [ 55%] Linking CXX shared library libdigikamcore.so ../libs/metadataengine/CMakeFiles/dmetadata_src.dir/engine/metaengine_exif.cpp.o: dans la fonction « Digikam::MetaEngine::getExifThumbnail(bool) const »: /home/gilles/devel/6.X/core/libs/metadataengine/engine/metaengine_exif.cpp:884: référence indéfinie vers « Exiv2::DataBuf::DataBuf(Exiv2::DataBufRef) » ../libs/metadataengine/CMakeFiles/dmetadata_src.dir/engine/metaengine_exif.cpp.o: dans la fonction « Exiv2::ValueType<unsigned int>::dataArea() const »: /usr/include/exiv2/value.hpp:1740: référence indéfinie vers « Exiv2::DataBuf::DataBuf(Exiv2::DataBufRef) » collect2: erreur : ld a retourné 1 code d'état d'exécution core/app/CMakeFiles/digikamcore.dir/build.make:2109 : la recette pour la cible « core/app/libdigikamcore.so.6.0.0 » a échouée make[2]: *** [core/app/libdigikamcore.so.6.0.0] Erreur 1 CMakeFiles/Makefile2:9006 : la recette pour la cible « core/app/CMakeFiles/digikamcore.dir/all » a échouée make[1]: *** [core/app/CMakeFiles/digikamcore.dir/all] Erreur 2 Makefile:138 : la recette pour la cible « all » a échouée make: *** [all] Erreur 2 Not reproducible with RC2 tag... Gilles Caulier --- end body --- comments : --- start comments --- --- end comments ---

From git/master today : [gilles@localhost build]$ su Mot de passe : [root@localhost build]# make install/fast Install the project... -- Install configuration: "" -- Installing: /usr/lib64/pkgconfig/exiv2.pc -- Installing: /usr/lib64/libxmp.a -- Installing: /usr/lib64/libexiv2.so.0.27.0.2 -- Installing: /usr/lib64/libexiv2.so.27 -- Up-to-date: /usr/lib64/libexiv2.so -- Set runtime path of "/usr/lib64/libexiv2.so.0.27.0.2" to "$ORIGIN/../lib64" -- Old export file "/usr/share/exiv2/cmake/exiv2Config.cmake" will be replaced. Removing files [/usr/share/exiv2/cmake/exiv2Config-noconfig.cmake]. -- Installing: /usr/share/exiv2/cmake/exiv2Config.cmake -- Installing: /usr/share/exiv2/cmake/exiv2Config-noconfig.cmake -- Installing: /usr/bin/exiv2 -- Set runtime path of "/usr/bin/exiv2" to "$ORIGIN/../lib64" -- Up-to-date: /usr/share/man/man1/exiv2.1 [root@localhost build]# If i switch to git RC1 tag, make install copy the header files as expected. So, it's a regression. Gilles Caulier

Exiv2 from git/master : make install miss header files

https://api.github.com/repos/Exiv2/exiv2/issues/551/comments

https://github.com/Exiv2/exiv2/issues/551

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : Exiv2 from git/master : make install miss header files Issue date: --- start body --- From git/master today : [gilles@localhost build]$ su Mot de passe : [root@localhost build]# make install/fast Install the project... -- Install configuration: "" -- Installing: /usr/lib64/pkgconfig/exiv2.pc -- Installing: /usr/lib64/libxmp.a -- Installing: /usr/lib64/libexiv2.so.0.27.0.2 -- Installing: /usr/lib64/libexiv2.so.27 -- Up-to-date: /usr/lib64/libexiv2.so -- Set runtime path of "/usr/lib64/libexiv2.so.0.27.0.2" to "$ORIGIN/../lib64" -- Old export file "/usr/share/exiv2/cmake/exiv2Config.cmake" will be replaced. Removing files [/usr/share/exiv2/cmake/exiv2Config-noconfig.cmake]. -- Installing: /usr/share/exiv2/cmake/exiv2Config.cmake -- Installing: /usr/share/exiv2/cmake/exiv2Config-noconfig.cmake -- Installing: /usr/bin/exiv2 -- Set runtime path of "/usr/bin/exiv2" to "$ORIGIN/../lib64" -- Up-to-date: /usr/share/man/man1/exiv2.1 [root@localhost build]# If i switch to git RC1 tag, make install copy the header files as expected. So, it's a regression. Gilles Caulier --- end body --- comments : --- start comments --- --- end comments ---

In #527 and #537 we discussed a lot about localisation stuff just before the 0.27 release. The only missing point (that was not critical for 0.27) was to provide some API function to change the language at run-time. While this is trivial in other translation systems like the one provided by Qt, it seems a bit more challenging with **gettext** and **libintl**. The recommended way according to the [gettext documentation](https://www.gnu.org/software/gettext/manual/html_node/gettext-grok.html) is to set the environment variable LANGUAGE to the desired value in the application code. This piece of code seems to work: ```cpp #include <libintl.h> #include <locale.h> #define _(STRING) gettext(STRING) #include <cstdio> #include <cstdlib> int main() { // Setting the i18n environment setlocale (LC_ALL, ""); bindtextdomain ("hello", "/usr/share/locale/"); textdomain ("hello"); // Example of i18n usage printf(_("Hello World\n")); setenv("LANGUAGE", "es_ES", 1); printf(_("Hello World\n")); return EXIT_SUCCESS; } -------- 13:30 $ ./app Hello World Hola Mundo ``` The idea would be to have something like `Exiv2::loadLocalisation()` maybe with a string argument to indicate the language and being able to switch languages at run time.

Localisation for v1.00

https://api.github.com/repos/Exiv2/exiv2/issues/549/comments

https://github.com/Exiv2/exiv2/issues/549

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : Localisation for v1.00 Issue date: --- start body --- In #527 and #537 we discussed a lot about localisation stuff just before the 0.27 release. The only missing point (that was not critical for 0.27) was to provide some API function to change the language at run-time. While this is trivial in other translation systems like the one provided by Qt, it seems a bit more challenging with **gettext** and **libintl**. The recommended way according to the [gettext documentation](https://www.gnu.org/software/gettext/manual/html_node/gettext-grok.html) is to set the environment variable LANGUAGE to the desired value in the application code. This piece of code seems to work: ```cpp #include <libintl.h> #include <locale.h> #define _(STRING) gettext(STRING) #include <cstdio> #include <cstdlib> int main() { // Setting the i18n environment setlocale (LC_ALL, ""); bindtextdomain ("hello", "/usr/share/locale/"); textdomain ("hello"); // Example of i18n usage printf(_("Hello World\n")); setenv("LANGUAGE", "es_ES", 1); printf(_("Hello World\n")); return EXIT_SUCCESS; } -------- 13:30 $ ./app Hello World Hola Mundo ``` The idea would be to have something like `Exiv2::loadLocalisation()` maybe with a string argument to indicate the language and being able to switch languages at run time. --- end body --- comments : --- start comments --- --- end comments ---

I found a bug in function printIFDStructure () in image.cpp:335. In the big do...while loop, there may be more than one IFD structures, namely multiple images. But there is a situation, if the offsets of the second IFD is same as the before, the program would go to infinite loop. This can be reproduced with cmd: `./exiv2 -pR poc` The poc can be found at [here](https://github.com/TeamSeri0us/pocs/blob/master/exiv2/poc).

Insufficient verification(cycle) in function Image::printIFDStructure in image.cpp:335

https://api.github.com/repos/Exiv2/exiv2/issues/547/comments

https://github.com/Exiv2/exiv2/issues/547

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : Insufficient verification(cycle) in function Image::printIFDStructure in image.cpp:335 Issue date: --- start body --- I found a bug in function printIFDStructure () in image.cpp:335. In the big do...while loop, there may be more than one IFD structures, namely multiple images. But there is a situation, if the offsets of the second IFD is same as the before, the program would go to infinite loop. This can be reproduced with cmd: `./exiv2 -pR poc` The poc can be found at [here](https://github.com/TeamSeri0us/pocs/blob/master/exiv2/poc). --- end body --- comments : --- start comments --- --- end comments ---

I got errors from cmake 3.10 on msvc concerning **add_compile_definitions**. I updated CMake to 3.12.4 on the buildserver and everything's working. Comment?

Do we need CMake 3.12 to build?

https://api.github.com/repos/Exiv2/exiv2/issues/543/comments

https://github.com/Exiv2/exiv2/issues/543

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : Do we need CMake 3.12 to build? Issue date: --- start body --- I got errors from cmake 3.10 on msvc concerning **add_compile_definitions**. I updated CMake to 3.12.4 on the buildserver and everything's working. Comment? --- end body --- comments : --- start comments --- --- end comments ---

@piponazo This is a "do you know?" issue. I think you only have to reply and close this. I've documented **unit tests** in README.md Comments welcome. I've run them. On all platforms they are producing a warnings without failing. ``` Warning: Unsupported time format Warning: Unsupported time format Warning: Unsupported time format Warning: Unsupported time format Warning: Unsupported time format Warning: Unsupported time format ``` I suspect this has something to do with localtime.c. I think it's probably better to comment off that test.

Unit tests are giving warning without failing

https://api.github.com/repos/Exiv2/exiv2/issues/542/comments

https://github.com/Exiv2/exiv2/issues/542

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : Unit tests are giving warning without failing Issue date: --- start body --- @piponazo This is a "do you know?" issue. I think you only have to reply and close this. I've documented **unit tests** in README.md Comments welcome. I've run them. On all platforms they are producing a warnings without failing. ``` Warning: Unsupported time format Warning: Unsupported time format Warning: Unsupported time format Warning: Unsupported time format Warning: Unsupported time format Warning: Unsupported time format ``` I suspect this has something to do with localtime.c. I think it's probably better to comment off that test. --- end body --- comments : --- start comments --- --- end comments ---

Fix submitted to 0.27-RC2 ``` 196 rmills@rmillsmbp-w7-64:~/gnu/github/exiv2/exiv2/build $ git diff diff --git a/src/exiv2.cpp b/src/exiv2.cpp index 7c54e401..94bc7219 100644 --- a/src/exiv2.cpp +++ b/src/exiv2.cpp @@ -1018,7 +1018,6 @@ int Params::getopt(int argc, char* const Argv[]) int rc = Util::Getopt::getopt(argc, argv, optstring_); // Further consistency checks if (help_ || version_) { - rc = 0; goto cleanup; } if (action_ == Action::none) { diff --git a/src/getopt.cpp b/src/getopt.cpp index 0d60e849..78a96abc 100644 --- a/src/getopt.cpp +++ b/src/getopt.cpp @@ -103,7 +103,7 @@ namespace Util { progname_ = Util::basename(argv[0]); Util::optind = 0; // reset the Util::Getopt scanner - for (;;) { + for (;!errcnt_;) { int c = Util::getopt(argc, argv, optstring.c_str()); if (c == -1) { break; 197 rmills@rmillsmbp-w7-64:~/gnu/github/exiv2/exiv2/build $ ```

exiv2 -vVg loops

https://api.github.com/repos/Exiv2/exiv2/issues/541/comments

https://github.com/Exiv2/exiv2/issues/541

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : exiv2 -vVg loops Issue date: --- start body --- Fix submitted to 0.27-RC2 ``` 196 rmills@rmillsmbp-w7-64:~/gnu/github/exiv2/exiv2/build $ git diff diff --git a/src/exiv2.cpp b/src/exiv2.cpp index 7c54e401..94bc7219 100644 --- a/src/exiv2.cpp +++ b/src/exiv2.cpp @@ -1018,7 +1018,6 @@ int Params::getopt(int argc, char* const Argv[]) int rc = Util::Getopt::getopt(argc, argv, optstring_); // Further consistency checks if (help_ || version_) { - rc = 0; goto cleanup; } if (action_ == Action::none) { diff --git a/src/getopt.cpp b/src/getopt.cpp index 0d60e849..78a96abc 100644 --- a/src/getopt.cpp +++ b/src/getopt.cpp @@ -103,7 +103,7 @@ namespace Util { progname_ = Util::basename(argv[0]); Util::optind = 0; // reset the Util::Getopt scanner - for (;;) { + for (;!errcnt_;) { int c = Util::getopt(argc, argv, optstring.c_str()); if (c == -1) { break; 197 rmills@rmillsmbp-w7-64:~/gnu/github/exiv2/exiv2/build $ ``` --- end body --- comments : --- start comments --- --- end comments ---

@piponazo I know that localisation support is "work-in-progress". Here are my observations: 1) On MacOS-X, the build fails when linking the exiv2 command line program. I manually fixed this by editing build/src/exiv2.dir/link.dir to mention /usr/local/lib/libexiv2.dylib. ``` [ 39%] Linking CXX executable ../bin/exiv2 Undefined symbols for architecture x86_64: "_libintl_bindtextdomain", referenced from: bindtextdomain(char const*, char const*) in exiv2.cpp.o "_libintl_setlocale", referenced from: _main in exiv2.cpp.o "_libintl_textdomain", referenced from: textdomain(char const*) in exiv2.cpp.o ld: symbol(s) not found for architecture x86_64 clang: error: linker command failed with exit code 1 (use -v to see invocation) make[2]: *** [bin/exiv2] Error 1 make[1]: *** [src/CMakeFiles/exiv2.dir/all] Error 2 make: *** [all] Error 2 565 rmills@rmillsmbp:~/gnu/github/exiv2/master/build $ ``` I believe the fix is in src/CMakeLists.txt ``` 595 rmills@rmillsmbp:~/gnu/github/exiv2/master/build $ git diff diff --git a/src/CMakeLists.txt b/src/CMakeLists.txt index 9ab270d0..d6a2ee8d 100644 --- a/src/CMakeLists.txt +++ b/src/CMakeLists.txt @@ -334,6 +334,9 @@ if(EXIV2_BUILD_EXIV2_COMMAND) ) target_link_libraries( exiv2 PRIVATE exiv2lib ) + if( EXIV2_ENABLE_NLS ) + target_link_libraries( exiv2 PRIVATE ${LIBINTL_LIBRARIES} ) + endif() # modify source lists to suit environment target_include_directories(exiv2 PRIVATE ${CMAKE_SOURCE_DIR}/include/) 596 rmills@rmillsmbp:~/gnu/github/exiv2/master/build $ ``` 2) I have to install the .mo files to use them from the installed exiv2. They are never used by build/bin/exiv2. ``` 583 rmills@rmillsmbp:~/gnu/github/exiv2/master/build $ env LC_ALL=fr_FR bin/exiv2 exiv2: An action must be specified exiv2: At least one file is required Usage: exiv2 [ options ] [ action ] file ... Manipulate the Exif metadata of images. 584 rmills@rmillsmbp:~/gnu/github/exiv2/master/build $ env LC_ALL=fr_FR exiv2 exiv2: Une action doit être spécifié exiv2: Au moins un fichier est nécessaire Utilisation : exiv2 [ options ] [ action ] fichier ... Manipulation des métadonnées EXIF issues des images. 585 rmills@rmillsmbp:~/gnu/github/exiv2/master/build $ ``` 3) I'm not sure that libintl should be linked with libexiv2 as I believe it's only used by applications such as the exiv2 command-line program. 4) Perhaps we should add localisation to every sample application by calling a new API Exiv2::loadLocalisation(). This API would understand when it's called in build/bin and load appropriately. Of course, if we do this we must link libintl with libexiv2. 5) I think we're building the .mo files during install. We should build the .mo files in the build directory and copy them during install. I need them to be built in the build/share directory so they can be copied from a binary build into share. In releasenotes/platform/ReadMe.txt, I've documented the following: ``` To install for use by all users ------------------------------- $ for i in bin include lib ; do sudo mkdir -p /usr/local/$i ; sudo cp -R $i/* /usr/local/$i ; done ``` This can only work if the .mo files are created by the build itself.

Issues with localisation on 'master' (on 2018-11-11)

https://api.github.com/repos/Exiv2/exiv2/issues/537/comments

https://github.com/Exiv2/exiv2/issues/537

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : Issues with localisation on 'master' (on 2018-11-11) Issue date: --- start body --- @piponazo I know that localisation support is "work-in-progress". Here are my observations: 1) On MacOS-X, the build fails when linking the exiv2 command line program. I manually fixed this by editing build/src/exiv2.dir/link.dir to mention /usr/local/lib/libexiv2.dylib. ``` [ 39%] Linking CXX executable ../bin/exiv2 Undefined symbols for architecture x86_64: "_libintl_bindtextdomain", referenced from: bindtextdomain(char const*, char const*) in exiv2.cpp.o "_libintl_setlocale", referenced from: _main in exiv2.cpp.o "_libintl_textdomain", referenced from: textdomain(char const*) in exiv2.cpp.o ld: symbol(s) not found for architecture x86_64 clang: error: linker command failed with exit code 1 (use -v to see invocation) make[2]: *** [bin/exiv2] Error 1 make[1]: *** [src/CMakeFiles/exiv2.dir/all] Error 2 make: *** [all] Error 2 565 rmills@rmillsmbp:~/gnu/github/exiv2/master/build $ ``` I believe the fix is in src/CMakeLists.txt ``` 595 rmills@rmillsmbp:~/gnu/github/exiv2/master/build $ git diff diff --git a/src/CMakeLists.txt b/src/CMakeLists.txt index 9ab270d0..d6a2ee8d 100644 --- a/src/CMakeLists.txt +++ b/src/CMakeLists.txt @@ -334,6 +334,9 @@ if(EXIV2_BUILD_EXIV2_COMMAND) ) target_link_libraries( exiv2 PRIVATE exiv2lib ) + if( EXIV2_ENABLE_NLS ) + target_link_libraries( exiv2 PRIVATE ${LIBINTL_LIBRARIES} ) + endif() # modify source lists to suit environment target_include_directories(exiv2 PRIVATE ${CMAKE_SOURCE_DIR}/include/) 596 rmills@rmillsmbp:~/gnu/github/exiv2/master/build $ ``` 2) I have to install the .mo files to use them from the installed exiv2. They are never used by build/bin/exiv2. ``` 583 rmills@rmillsmbp:~/gnu/github/exiv2/master/build $ env LC_ALL=fr_FR bin/exiv2 exiv2: An action must be specified exiv2: At least one file is required Usage: exiv2 [ options ] [ action ] file ... Manipulate the Exif metadata of images. 584 rmills@rmillsmbp:~/gnu/github/exiv2/master/build $ env LC_ALL=fr_FR exiv2 exiv2: Une action doit être spécifié exiv2: Au moins un fichier est nécessaire Utilisation : exiv2 [ options ] [ action ] fichier ... Manipulation des métadonnées EXIF issues des images. 585 rmills@rmillsmbp:~/gnu/github/exiv2/master/build $ ``` 3) I'm not sure that libintl should be linked with libexiv2 as I believe it's only used by applications such as the exiv2 command-line program. 4) Perhaps we should add localisation to every sample application by calling a new API Exiv2::loadLocalisation(). This API would understand when it's called in build/bin and load appropriately. Of course, if we do this we must link libintl with libexiv2. 5) I think we're building the .mo files during install. We should build the .mo files in the build directory and copy them during install. I need them to be built in the build/share directory so they can be copied from a binary build into share. In releasenotes/platform/ReadMe.txt, I've documented the following: ``` To install for use by all users ------------------------------- $ for i in bin include lib ; do sudo mkdir -p /usr/local/$i ; sudo cp -R $i/* /usr/local/$i ; done ``` This can only work if the .mo files are created by the build itself. --- end body --- comments : --- start comments --- --- end comments ---

This is related with #527. It seems that at some point the file `FindMSGFMT.cmake` was introduced to deal with the .PO localisation files. Now CMake provides `FindGettext` which seems to be doing the same. We should always try to use the module provides by CMake since they are maintained by the community.

CMake: replace custom FindMSGFMT.cmake by FindGettext

https://api.github.com/repos/Exiv2/exiv2/issues/536/comments

https://github.com/Exiv2/exiv2/issues/536

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : CMake: replace custom FindMSGFMT.cmake by FindGettext Issue date: --- start body --- This is related with #527. It seems that at some point the file `FindMSGFMT.cmake` was introduced to deal with the .PO localisation files. Now CMake provides `FindGettext` which seems to be doing the same. We should always try to use the module provides by CMake since they are maintained by the community. --- end body --- comments : --- start comments --- --- end comments ---

http://dev.exiv2.org/issues/1324

Pentax Makernote written twice

https://api.github.com/repos/Exiv2/exiv2/issues/533/comments

https://github.com/Exiv2/exiv2/issues/533

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : Pentax Makernote written twice Issue date: --- start body --- http://dev.exiv2.org/issues/1324 --- end body --- comments : --- start comments --- --- end comments ---

http://dev.exiv2.org/issues/1306

TAMRON SP 15-30mm F/2.8 Di VC USD A012 len recognition

https://api.github.com/repos/Exiv2/exiv2/issues/532/comments

https://github.com/Exiv2/exiv2/issues/532

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : TAMRON SP 15-30mm F/2.8 Di VC USD A012 len recognition Issue date: --- start body --- http://dev.exiv2.org/issues/1306 --- end body --- comments : --- start comments --- --- end comments ---

This matter has arisen from this discussion: https://github.com/Exiv2/exiv2/issues/515 BasicIo::eof() returns BasicIo::tell() == BasicIo::size() and, as explained in #515 has slightly different semantics from `::feof()` As FileIo::size() calls ::stat() this can cause a large number of system calls in a large PNG. In the discussion, I experimented with cacheing within FileIo::size() while reading files. The performance gain seems valuable (about 30%-40% better). However my patch failed the test suite in many places and would require more work before being ready for release. Cacheing FileIo::size() is worthy of more investigation in Exiv2 v0.28.

Cache FileIo::size()

https://api.github.com/repos/Exiv2/exiv2/issues/530/comments

https://github.com/Exiv2/exiv2/issues/530

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : Cache FileIo::size() Issue date: --- start body --- This matter has arisen from this discussion: https://github.com/Exiv2/exiv2/issues/515 BasicIo::eof() returns BasicIo::tell() == BasicIo::size() and, as explained in #515 has slightly different semantics from `::feof()` As FileIo::size() calls ::stat() this can cause a large number of system calls in a large PNG. In the discussion, I experimented with cacheing within FileIo::size() while reading files. The performance gain seems valuable (about 30%-40% better). However my patch failed the test suite in many places and would require more work before being ready for release. Cacheing FileIo::size() is worthy of more investigation in Exiv2 v0.28. --- end body --- comments : --- start comments --- --- end comments ---

This was reported to Apple: https://bugreport.apple.com/web/?problemID=45223696 I documented the issue in releasenotes/releasenotes.txt for v0.27 RC1 I've had a response from Apple. I think they are saying "Can't reproduce". I'll install a fresh MacOS-X VM with Xcode 10 and try again. _Engineering has requested the following information in order to further investigate this issue:_ _Can you please tell at what revision the problem should be reproducible? It looks like there were changes made to work around the problem, so, maybe that’s why it’s not reproducible anymore. In our testing revisions 12d44f8f17bb541d5ce0c05e6020bc5d79a03e34 and afc1a1e5263e894360f80d9316e00d7668760079 were built successfully and resulting ./bin/taglist runs without crashing._

Issue compiling samples/taglist.cpp in Xcode 10

https://api.github.com/repos/Exiv2/exiv2/issues/528/comments

https://github.com/Exiv2/exiv2/issues/528

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : Issue compiling samples/taglist.cpp in Xcode 10 Issue date: --- start body --- This was reported to Apple: https://bugreport.apple.com/web/?problemID=45223696 I documented the issue in releasenotes/releasenotes.txt for v0.27 RC1 I've had a response from Apple. I think they are saying "Can't reproduce". I'll install a fresh MacOS-X VM with Xcode 10 and try again. _Engineering has requested the following information in order to further investigate this issue:_ _Can you please tell at what revision the problem should be reproducible? It looks like there were changes made to work around the problem, so, maybe that’s why it’s not reproducible anymore. In our testing revisions 12d44f8f17bb541d5ce0c05e6020bc5d79a03e34 and afc1a1e5263e894360f80d9316e00d7668760079 were built successfully and resulting ./bin/taglist runs without crashing._ --- end body --- comments : --- start comments --- --- end comments ---

I've had an offer from an engineer in Catalonia to deal with localisation of Exiv2 into Catalonian. I hope he'll deal with that for both Exiv2 v0.26 (for digicam) and v0.27. http://dev.exiv2.org/boards/3/topics/3204 I hope Toni will deal with Localisation for all languages for v0.27. I've never been involved with localisation in the past when it was built with autotools. I'll review po/CMakeLists.txt and ensure that everything is OK. And I'll update README.md and releasenotes/platform/ReadMe.txt to cover this. I hope the localization in v0.27 RC2 will be based on v0.26 (including Toni's Catalonian). By v0.27 RC3, I hope to have Toni's new files in all languages.

Localisation for v0.27

https://api.github.com/repos/Exiv2/exiv2/issues/527/comments

https://github.com/Exiv2/exiv2/issues/527

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : Localisation for v0.27 Issue date: --- start body --- I've had an offer from an engineer in Catalonia to deal with localisation of Exiv2 into Catalonian. I hope he'll deal with that for both Exiv2 v0.26 (for digicam) and v0.27. http://dev.exiv2.org/boards/3/topics/3204 I hope Toni will deal with Localisation for all languages for v0.27. I've never been involved with localisation in the past when it was built with autotools. I'll review po/CMakeLists.txt and ensure that everything is OK. And I'll update README.md and releasenotes/platform/ReadMe.txt to cover this. I hope the localization in v0.27 RC2 will be based on v0.26 (including Toni's Catalonian). By v0.27 RC3, I hope to have Toni's new files in all languages. --- end body --- comments : --- start comments --- --- end comments ---

Once we migrate to C++11 we could remove the existing the file `rwlock.hpp`and use `std::lock_guard` to achieve the same goal.

c++11: Remove rwlock.hpp and use std::lock_guard

https://api.github.com/repos/Exiv2/exiv2/issues/525/comments

https://github.com/Exiv2/exiv2/issues/525

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : c++11: Remove rwlock.hpp and use std::lock_guard Issue date: --- start body --- Once we migrate to C++11 we could remove the existing the file `rwlock.hpp`and use `std::lock_guard` to achieve the same goal. --- end body --- comments : --- start comments --- --- end comments ---

See the JPEG sample face-tagged with Mylio photo manager : https://bugs.kde.org/attachment.cgi?id=114448 The XMP contents is this one : [gilles@localhost Téléchargements]$ exiv2 -px DSC_3079.jpg Xmp.MY.DateConfidence XmpText 4 Full Xmp.exifEX.PhotographicSensitivity XmpText 2 64 Xmp.mwg-rs.Regions XmpText 0 type="Struct" Xmp.mwg-rs.Regions/mwg-rs:AppliedToDimensions XmpText 0 type="Struct" Xmp.mwg-rs.Regions/mwg-rs:AppliedToDimensions/stDim:w XmpText 3 540 Xmp.mwg-rs.Regions/mwg-rs:AppliedToDimensions/stDim:h XmpText 3 960 Xmp.mwg-rs.Regions/mwg-rs:AppliedToDimensions/stDim:unit XmpText 5 pixel Xmp.mwg-rs.Regions/mwg-rs:RegionList XmpText 0 type="Seq" Xmp.mwg-rs.Regions/mwg-rs:RegionList[1] XmpText 0 type="Struct" Xmp.mwg-rs.Regions/mwg-rs:RegionList[1]/mwg-rs:Type XmpText 4 Face Xmp.mwg-rs.Regions/mwg-rs:RegionList[1]/mwg-rs:Name XmpText 7 America Xmp.mwg-rs.Regions/mwg-rs:RegionList[1]/mwg-rs:Area XmpText 0 type="Struct" Xmp.mwg-rs.Regions/mwg-rs:RegionList[1]/mwg-rs:Area/stArea:x XmpText 8 0.825794 Xmp.mwg-rs.Regions/mwg-rs:RegionList[1]/mwg-rs:Area/stArea:y XmpText 8 0.379665 Xmp.mwg-rs.Regions/mwg-rs:RegionList[1]/mwg-rs:Area/stArea:w XmpText 8 0.110064 Xmp.mwg-rs.Regions/mwg-rs:RegionList[1]/mwg-rs:Area/stArea:h XmpText 8 0.077389 Xmp.mwg-rs.Regions/mwg-rs:RegionList[1]/mwg-rs:Area/stArea:unit XmpText 10 normalized Xmp.dc.subject XmpBag 1 America The Xmp.mwg-rs.Regions define the data structure to host regions of image. I want to remove these region data. I tried to remove this tag as well : [gilles@localhost Téléchargements]$ exiv2 -v -M"del Xmp.mwg-rs.Regions" DSC_3079.jpg File 1/1: DSC_3079.jpg Del Xmp.mwg-rs.Regions [gilles@localhost Téléchargements]$ No error, no warning, all sound good, but not in fact : [gilles@localhost Téléchargements]$ exiv2 -px DSC_3079.jpg Xmp.MY.DateConfidence XmpText 4 Full Xmp.exifEX.PhotographicSensitivity XmpText 2 64 Xmp.mwg-rs.Regions XmpText 0 type="Struct" Xmp.mwg-rs.Regions/mwg-rs:AppliedToDimensions XmpText 0 type="Struct" Xmp.mwg-rs.Regions/mwg-rs:AppliedToDimensions/stDim:w XmpText 3 540 Xmp.mwg-rs.Regions/mwg-rs:AppliedToDimensions/stDim:h XmpText 3 960 Xmp.mwg-rs.Regions/mwg-rs:AppliedToDimensions/stDim:unit XmpText 5 pixel Xmp.mwg-rs.Regions/mwg-rs:RegionList XmpText 0 type="Seq" Xmp.mwg-rs.Regions/mwg-rs:RegionList[1] XmpText 0 type="Struct" Xmp.mwg-rs.Regions/mwg-rs:RegionList[1]/mwg-rs:Type XmpText 4 Face Xmp.mwg-rs.Regions/mwg-rs:RegionList[1]/mwg-rs:Name XmpText 7 America Xmp.mwg-rs.Regions/mwg-rs:RegionList[1]/mwg-rs:Area XmpText 0 type="Struct" Xmp.mwg-rs.Regions/mwg-rs:RegionList[1]/mwg-rs:Area/stArea:x XmpText 8 0.825794 Xmp.mwg-rs.Regions/mwg-rs:RegionList[1]/mwg-rs:Area/stArea:y XmpText 8 0.379665 Xmp.mwg-rs.Regions/mwg-rs:RegionList[1]/mwg-rs:Area/stArea:w XmpText 8 0.110064 Xmp.mwg-rs.Regions/mwg-rs:RegionList[1]/mwg-rs:Area/stArea:h XmpText 8 0.077389 Xmp.mwg-rs.Regions/mwg-rs:RegionList[1]/mwg-rs:Area/stArea:unit XmpText 10 normalized Xmp.dc.subject XmpBag 1 America [gilles@localhost Téléchargements]$ exiv2 --version exiv2 0.27.0.1 ... What's wrong ? Gilles Caulier

How to remove XMP Region tag from JPEG ?

https://api.github.com/repos/Exiv2/exiv2/issues/521/comments

https://github.com/Exiv2/exiv2/issues/521

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : How to remove XMP Region tag from JPEG ? Issue date: --- start body --- See the JPEG sample face-tagged with Mylio photo manager : https://bugs.kde.org/attachment.cgi?id=114448 The XMP contents is this one : [gilles@localhost Téléchargements]$ exiv2 -px DSC_3079.jpg Xmp.MY.DateConfidence XmpText 4 Full Xmp.exifEX.PhotographicSensitivity XmpText 2 64 Xmp.mwg-rs.Regions XmpText 0 type="Struct" Xmp.mwg-rs.Regions/mwg-rs:AppliedToDimensions XmpText 0 type="Struct" Xmp.mwg-rs.Regions/mwg-rs:AppliedToDimensions/stDim:w XmpText 3 540 Xmp.mwg-rs.Regions/mwg-rs:AppliedToDimensions/stDim:h XmpText 3 960 Xmp.mwg-rs.Regions/mwg-rs:AppliedToDimensions/stDim:unit XmpText 5 pixel Xmp.mwg-rs.Regions/mwg-rs:RegionList XmpText 0 type="Seq" Xmp.mwg-rs.Regions/mwg-rs:RegionList[1] XmpText 0 type="Struct" Xmp.mwg-rs.Regions/mwg-rs:RegionList[1]/mwg-rs:Type XmpText 4 Face Xmp.mwg-rs.Regions/mwg-rs:RegionList[1]/mwg-rs:Name XmpText 7 America Xmp.mwg-rs.Regions/mwg-rs:RegionList[1]/mwg-rs:Area XmpText 0 type="Struct" Xmp.mwg-rs.Regions/mwg-rs:RegionList[1]/mwg-rs:Area/stArea:x XmpText 8 0.825794 Xmp.mwg-rs.Regions/mwg-rs:RegionList[1]/mwg-rs:Area/stArea:y XmpText 8 0.379665 Xmp.mwg-rs.Regions/mwg-rs:RegionList[1]/mwg-rs:Area/stArea:w XmpText 8 0.110064 Xmp.mwg-rs.Regions/mwg-rs:RegionList[1]/mwg-rs:Area/stArea:h XmpText 8 0.077389 Xmp.mwg-rs.Regions/mwg-rs:RegionList[1]/mwg-rs:Area/stArea:unit XmpText 10 normalized Xmp.dc.subject XmpBag 1 America The Xmp.mwg-rs.Regions define the data structure to host regions of image. I want to remove these region data. I tried to remove this tag as well : [gilles@localhost Téléchargements]$ exiv2 -v -M"del Xmp.mwg-rs.Regions" DSC_3079.jpg File 1/1: DSC_3079.jpg Del Xmp.mwg-rs.Regions [gilles@localhost Téléchargements]$ No error, no warning, all sound good, but not in fact : [gilles@localhost Téléchargements]$ exiv2 -px DSC_3079.jpg Xmp.MY.DateConfidence XmpText 4 Full Xmp.exifEX.PhotographicSensitivity XmpText 2 64 Xmp.mwg-rs.Regions XmpText 0 type="Struct" Xmp.mwg-rs.Regions/mwg-rs:AppliedToDimensions XmpText 0 type="Struct" Xmp.mwg-rs.Regions/mwg-rs:AppliedToDimensions/stDim:w XmpText 3 540 Xmp.mwg-rs.Regions/mwg-rs:AppliedToDimensions/stDim:h XmpText 3 960 Xmp.mwg-rs.Regions/mwg-rs:AppliedToDimensions/stDim:unit XmpText 5 pixel Xmp.mwg-rs.Regions/mwg-rs:RegionList XmpText 0 type="Seq" Xmp.mwg-rs.Regions/mwg-rs:RegionList[1] XmpText 0 type="Struct" Xmp.mwg-rs.Regions/mwg-rs:RegionList[1]/mwg-rs:Type XmpText 4 Face Xmp.mwg-rs.Regions/mwg-rs:RegionList[1]/mwg-rs:Name XmpText 7 America Xmp.mwg-rs.Regions/mwg-rs:RegionList[1]/mwg-rs:Area XmpText 0 type="Struct" Xmp.mwg-rs.Regions/mwg-rs:RegionList[1]/mwg-rs:Area/stArea:x XmpText 8 0.825794 Xmp.mwg-rs.Regions/mwg-rs:RegionList[1]/mwg-rs:Area/stArea:y XmpText 8 0.379665 Xmp.mwg-rs.Regions/mwg-rs:RegionList[1]/mwg-rs:Area/stArea:w XmpText 8 0.110064 Xmp.mwg-rs.Regions/mwg-rs:RegionList[1]/mwg-rs:Area/stArea:h XmpText 8 0.077389 Xmp.mwg-rs.Regions/mwg-rs:RegionList[1]/mwg-rs:Area/stArea:unit XmpText 10 normalized Xmp.dc.subject XmpBag 1 America [gilles@localhost Téléchargements]$ exiv2 --version exiv2 0.27.0.1 ... What's wrong ? Gilles Caulier --- end body --- comments : --- start comments --- --- end comments ---

On the weekend the Debian builds on GitLab started failing, see for instance: for https://gitlab.com/D4N/exiv2/-/jobs/116275047. It complains about the cmake variable `CMAKE_DESCRIPTION_COMPILER_ENV_VAR` being unset. @piponazo have you ever heard of that?

Debian build broken on GitLab

https://api.github.com/repos/Exiv2/exiv2/issues/519/comments

https://github.com/Exiv2/exiv2/issues/519

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : Debian build broken on GitLab Issue date: --- start body --- On the weekend the Debian builds on GitLab started failing, see for instance: for https://gitlab.com/D4N/exiv2/-/jobs/116275047. It complains about the cmake variable `CMAKE_DESCRIPTION_COMPILER_ENV_VAR` being unset. @piponazo have you ever heard of that? --- end body --- comments : --- start comments --- --- end comments ---

I've found that when reading PNG files, (lib)exiv2 calls stat() excessively often: ``` $ strace -e stat exiv2 /path/to/image.png 2>&1|grep stat|wc -l 7265 ``` Yes, the file in question is 114MB, but there's no good reason to call stat() on it over and over since it's unlikely to change in the time it takes to read :-) (also, given the various relatively recent security mitigations, doing syscalls has gotten slower) FWIW, this was a file without any exif data, perhaps if a chunk early in the file had any you could abort?

excessively often calls stat()

https://api.github.com/repos/Exiv2/exiv2/issues/515/comments

https://github.com/Exiv2/exiv2/issues/515

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : excessively often calls stat() Issue date: --- start body --- I've found that when reading PNG files, (lib)exiv2 calls stat() excessively often: ``` $ strace -e stat exiv2 /path/to/image.png 2>&1|grep stat|wc -l 7265 ``` Yes, the file in question is 114MB, but there's no good reason to call stat() on it over and over since it's unlikely to change in the time it takes to read :-) (also, given the various relatively recent security mitigations, doing syscalls has gotten slower) FWIW, this was a file without any exif data, perhaps if a chunk early in the file had any you could abort? --- end body --- comments : --- start comments --- --- end comments ---

In investigating #497 we were requested to support the pkg-config utility by generating the file exiv2.pc @piponazo implemented this in #509 **Action** 1) No action is required in the code. 2) The platform notes releasenotes/platform/ReadMe.txt should be updated appropriately. 3) The "User Manual" README.md should be updated to discuss exiv2.pc. **Analysis** The v0.26 legacy code config/exiv2.in.cmake is: ``` 522 rmills@rmillsmbp:~/gnu/exiv2/v0.26 $ cat config/exiv2.pc.cmake prefix=@CMAKE_INSTALL_PREFIX@ exec_prefix=${prefix} libdir=${prefix}/lib includedir=${prefix}/include Name: exiv2 Description: Exif and IPTC metadata library and tools Version: @PACKAGE_VERSION@ URL: @PACKAGE_URL@ Libs: -L${libdir} -lexiv2 @EXV_LIB_STATIC@ Cflags: -I${includedir} 523 rmills@rmillsmbp:~/gnu/exiv2/v0.26 $ ``` This has morphed in v0.27 into cmake/exiv2.pc.in ``` 524 rmills@rmillsmbp:~/gnu/github/exiv2/exiv2 $ cat cmake/exiv2.pc.in prefix=@CMAKE_INSTALL_PREFIX@ exec_prefix=${prefix} libdir=@CMAKE_INSTALL_FULL_LIBDIR@ includedir=@CMAKE_INSTALL_FULL_INCLUDEDIR@ Name: exiv2 Description: Exif and IPTC metadata library and tools Version: @PACKAGE_VERSION@ URL: @PACKAGE_URL@ Libs: -L${libdir} -lexiv2 Cflags: -I${includedir} 525 rmills@rmillsmbp:~/gnu/github/exiv2/exiv2 $ ``` The legacy file contains the symbol @EXV_LIB_STATIC@ which we believe is a relic of autotools and no longer relevant. I've looked at this in several different ways: 1) Does Exiv2 v0.27 exiv2.pc file work correctly for applications built with static and dynamic libraries. Answer: **Yes** 2) What does exiv2 v0.26 CMake code (CMakeLists.txt and config/*.cmake) do with EXV_LIB_STATIC? Answer: **Nothing** ``` 533 rmills@rmillsmbp:~/gnu/github/exiv2/exiv2 $ cd ../../../exiv2/v0.26 534 rmills@rmillsmbp:~/gnu/exiv2/v0.26 $ find config -name "*" -type f -exec grep -H EXV_LIB_STATIC {} \; config/configure.ac: EXV_LIB_STATIC=$LTLIBICONV config/configure.ac: EXV_LIB_STATIC= config/configure.ac:AC_SUBST(EXV_LIB_STATIC) config/exiv2.pc.cmake:Libs: -L${libdir} -lexiv2 @EXV_LIB_STATIC@ config/autom4te.cache/traces.0:m4trace:configure.ac:310: -1- AC_SUBST([EXV_LIB_STATIC]) config/autom4te.cache/traces.0:m4trace:configure.ac:310: -1- AC_SUBST_TRACE([EXV_LIB_STATIC]) config/autom4te.cache/traces.0:m4trace:configure.ac:310: -1- m4_pattern_allow([^EXV_LIB_STATIC$]) config/autom4te.cache/output.0:EXV_LIB_STATIC config/autom4te.cache/output.0: EXV_LIB_STATIC=$LTLIBICONV config/autom4te.cache/output.0: EXV_LIB_STATIC= config/exiv2.pc.in:Libs: -L${libdir} -lexiv2 @EXV_LIB_STATIC@ 535 rmills@rmillsmbp:~/gnu/exiv2/v0.26 $ find . -name CMakeLists.txt -type f -exec grep -H EXV_LIB_STATIC {} \; 536 rmills@rmillsmbp:~/gnu/exiv2/v0.26 $ ``` 3) What's the difference between the v0.26 CMake exiv2.pc file and v0.27? Answer: **None** ``` 531 rmills@rmillsmbp:~/gnu/github/exiv2/exiv2/build $ diff exiv2.pc ../../../../exiv2/v0.26/build/exiv2.pc 3,4c3,4 < libdir=/usr/local/lib < includedir=/usr/local/include --- > libdir=${prefix}/lib > includedir=${prefix}/include 8,10c8,10 < Version: < URL: < Libs: -L${libdir} -lexiv2 --- > Version: 0.26 > URL: http://www.exiv2.org > Libs: -L${libdir} -lexiv2 532 rmills@rmillsmbp:~/gnu/github/exiv2/exiv2/build $ ``` 4. Is the autotools generated exiv2.pc file different for static/shared builds. Answer: **No** 5. What does v0.26 autotools and CMake generate in exiv2.pc? ``` make config; ./configure --disable-shared --enable-static prefix=/usr/local exec_prefix=${prefix} libdir=${exec_prefix}/lib includedir=${prefix}/include Name: exiv2 Description: Image metadata library and tools Version: 0.26 Libs: -L${libdir} -lexiv2 -liconv Libs.private: -L/usr/local/lib -lintl -liconv -lc -R/usr/local/lib -Wl,-framework -Wl,CoreFoundation -liconv -lz -lexpat Cflags: -I${includedir} 512 rmills@rmillsmbp:~/gnu/exiv2/v0.26/build $ cat exiv2.pc prefix=/usr/local exec_prefix=${prefix} libdir=${prefix}/lib includedir=${prefix}/include Name: exiv2 Description: Exif and IPTC metadata library and tools Version: 0.26 URL: http://www.exiv2.org Libs: -L${libdir} -lexiv2 Cflags: -I${includedir} 513 rmills@rmillsmbp:~/gnu/exiv2/v0.26/build $ ``` 6 Anything else? I'm not happy that in v0.27 we are installing exiv2.pc into **`{prefix}/shared/exiv2.pc`** as that probably requires the user to modify PKG_CONFIG_PATH. I believe it's more conventional (and v0.26 behaviour) to installing exiv2.pc into **`{prefix}/lib/pkgconfig/exiv2.pc`** which the user probably has already set up on their machine.

Generating pkg-config file exiv2.pc from CMake build environment.

https://api.github.com/repos/Exiv2/exiv2/issues/512/comments

https://github.com/Exiv2/exiv2/issues/512

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : Generating pkg-config file exiv2.pc from CMake build environment. Issue date: --- start body --- In investigating #497 we were requested to support the pkg-config utility by generating the file exiv2.pc @piponazo implemented this in #509 **Action** 1) No action is required in the code. 2) The platform notes releasenotes/platform/ReadMe.txt should be updated appropriately. 3) The "User Manual" README.md should be updated to discuss exiv2.pc. **Analysis** The v0.26 legacy code config/exiv2.in.cmake is: ``` 522 rmills@rmillsmbp:~/gnu/exiv2/v0.26 $ cat config/exiv2.pc.cmake prefix=@CMAKE_INSTALL_PREFIX@ exec_prefix=${prefix} libdir=${prefix}/lib includedir=${prefix}/include Name: exiv2 Description: Exif and IPTC metadata library and tools Version: @PACKAGE_VERSION@ URL: @PACKAGE_URL@ Libs: -L${libdir} -lexiv2 @EXV_LIB_STATIC@ Cflags: -I${includedir} 523 rmills@rmillsmbp:~/gnu/exiv2/v0.26 $ ``` This has morphed in v0.27 into cmake/exiv2.pc.in ``` 524 rmills@rmillsmbp:~/gnu/github/exiv2/exiv2 $ cat cmake/exiv2.pc.in prefix=@CMAKE_INSTALL_PREFIX@ exec_prefix=${prefix} libdir=@CMAKE_INSTALL_FULL_LIBDIR@ includedir=@CMAKE_INSTALL_FULL_INCLUDEDIR@ Name: exiv2 Description: Exif and IPTC metadata library and tools Version: @PACKAGE_VERSION@ URL: @PACKAGE_URL@ Libs: -L${libdir} -lexiv2 Cflags: -I${includedir} 525 rmills@rmillsmbp:~/gnu/github/exiv2/exiv2 $ ``` The legacy file contains the symbol @EXV_LIB_STATIC@ which we believe is a relic of autotools and no longer relevant. I've looked at this in several different ways: 1) Does Exiv2 v0.27 exiv2.pc file work correctly for applications built with static and dynamic libraries. Answer: **Yes** 2) What does exiv2 v0.26 CMake code (CMakeLists.txt and config/*.cmake) do with EXV_LIB_STATIC? Answer: **Nothing** ``` 533 rmills@rmillsmbp:~/gnu/github/exiv2/exiv2 $ cd ../../../exiv2/v0.26 534 rmills@rmillsmbp:~/gnu/exiv2/v0.26 $ find config -name "*" -type f -exec grep -H EXV_LIB_STATIC {} \; config/configure.ac: EXV_LIB_STATIC=$LTLIBICONV config/configure.ac: EXV_LIB_STATIC= config/configure.ac:AC_SUBST(EXV_LIB_STATIC) config/exiv2.pc.cmake:Libs: -L${libdir} -lexiv2 @EXV_LIB_STATIC@ config/autom4te.cache/traces.0:m4trace:configure.ac:310: -1- AC_SUBST([EXV_LIB_STATIC]) config/autom4te.cache/traces.0:m4trace:configure.ac:310: -1- AC_SUBST_TRACE([EXV_LIB_STATIC]) config/autom4te.cache/traces.0:m4trace:configure.ac:310: -1- m4_pattern_allow([^EXV_LIB_STATIC$]) config/autom4te.cache/output.0:EXV_LIB_STATIC config/autom4te.cache/output.0: EXV_LIB_STATIC=$LTLIBICONV config/autom4te.cache/output.0: EXV_LIB_STATIC= config/exiv2.pc.in:Libs: -L${libdir} -lexiv2 @EXV_LIB_STATIC@ 535 rmills@rmillsmbp:~/gnu/exiv2/v0.26 $ find . -name CMakeLists.txt -type f -exec grep -H EXV_LIB_STATIC {} \; 536 rmills@rmillsmbp:~/gnu/exiv2/v0.26 $ ``` 3) What's the difference between the v0.26 CMake exiv2.pc file and v0.27? Answer: **None** ``` 531 rmills@rmillsmbp:~/gnu/github/exiv2/exiv2/build $ diff exiv2.pc ../../../../exiv2/v0.26/build/exiv2.pc 3,4c3,4 < libdir=/usr/local/lib < includedir=/usr/local/include --- > libdir=${prefix}/lib > includedir=${prefix}/include 8,10c8,10 < Version: < URL: < Libs: -L${libdir} -lexiv2 --- > Version: 0.26 > URL: http://www.exiv2.org > Libs: -L${libdir} -lexiv2 532 rmills@rmillsmbp:~/gnu/github/exiv2/exiv2/build $ ``` 4. Is the autotools generated exiv2.pc file different for static/shared builds. Answer: **No** 5. What does v0.26 autotools and CMake generate in exiv2.pc? ``` make config; ./configure --disable-shared --enable-static prefix=/usr/local exec_prefix=${prefix} libdir=${exec_prefix}/lib includedir=${prefix}/include Name: exiv2 Description: Image metadata library and tools Version: 0.26 Libs: -L${libdir} -lexiv2 -liconv Libs.private: -L/usr/local/lib -lintl -liconv -lc -R/usr/local/lib -Wl,-framework -Wl,CoreFoundation -liconv -lz -lexpat Cflags: -I${includedir} 512 rmills@rmillsmbp:~/gnu/exiv2/v0.26/build $ cat exiv2.pc prefix=/usr/local exec_prefix=${prefix} libdir=${prefix}/lib includedir=${prefix}/include Name: exiv2 Description: Exif and IPTC metadata library and tools Version: 0.26 URL: http://www.exiv2.org Libs: -L${libdir} -lexiv2 Cflags: -I${includedir} 513 rmills@rmillsmbp:~/gnu/exiv2/v0.26/build $ ``` 6 Anything else? I'm not happy that in v0.27 we are installing exiv2.pc into **`{prefix}/shared/exiv2.pc`** as that probably requires the user to modify PKG_CONFIG_PATH. I believe it's more conventional (and v0.26 behaviour) to installing exiv2.pc into **`{prefix}/lib/pkgconfig/exiv2.pc`** which the user probably has already set up on their machine. --- end body --- comments : --- start comments --- --- end comments ---

The issue may be similar to the issue #51 , with the POC file provided, The vulue of the start will always be `0xfc0000`, so the condition will be true all the time. ./exiv2 -pC [poc_infinite_loop](https://github.com/TeamSeri0us/pocs/blob/master/exiv2/poc_infinite_loop) or the command ./exiv2 -pR [poc_infinite_loop](https://github.com/TeamSeri0us/pocs/blob/master/exiv2/poc_infinite_loop) can trigger the situation: ``` ────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────[ source:/home/fish/Desktop/2018-10-10/exiv2/src/image.cpp+509 ]──── 504 out.write((const char*)buf.pData_,count); 505 } 506 } 507 if ( start ) { 508 io.read(dir.pData_, 4); → 509 start = tooBig ? 0 : byteSwap4(dir,0,bSwap); 510 } 511 } while (start) ; 512 513 if ( bPrint ) { 514 out << Internal::indent(depth) << "END " << io.path() << std::endl; ─────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────[ threads ]──── [#0] Id 1, Name: "exiv2", stopped, reason: BREAKPOINT ───────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────[ trace ]──── [#0] 0x7ffff7979845 → Name: Exiv2::Image::printIFDStructure(this=0x5555557aecb0, io=@0x5555557afae0, out=@0x7ffff7850e40, option=Exiv2::kpsIccProfile, start=0xfc0000, bSwap=0x0, c=0x49, depth=0x0) [#1] 0x7ffff7979c56 → Name: Exiv2::Image::printTiffStructure(this=0x5555557aecb0, io=@0x5555557afae0, out=@0x7ffff7850e40, option=Exiv2::kpsIccProfile, depth=0xffffffff, offset=0x0) [#2] 0x7ffff79b73bb → Name: Exiv2::TiffImage::printStructure(this=0x5555557aecb0, out=@0x7ffff7850e40, option=Exiv2::kpsIccProfile, depth=0x0) [#3] 0x555555586064 → Name: (anonymous namespace)::printStructure(out=@0x7ffff7850e40, option=Exiv2::kpsIccProfile, path="poc_infinite_loop") [#4] 0x555555575d55 → Name: Action::setModeAndPrintStructure(option=Exiv2::kpsIccProfile, path="poc_infinite_loop") [#5] 0x555555575ea3 → Name: Action::Print::run(this=0x5555557aff10, path="poc_infinite_loop") [#6] 0x5555555671ee → Name: main(argc=0x4, argv=0x7fffffffdd58) [#7] 0x7ffff6ee3b97 → Name: __libc_start_main(main=0x555555566f0a <main(int, char* const*)>, argc=0x4, argv=0x7fffffffdd58, init=<optimized out>, fini=<optimized out>, rtld_fini=<optimized out>, stack_end=0x7fffffffdd48) [#8] 0x555555566e2a → Name: _start() ──────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────── Breakpoint 2, Exiv2::Image::printIFDStructure (this=0x5555557aecb0, io=..., out=..., option=Exiv2::kpsIccProfile, start=0xfc0000, bSwap=0x0, c=0x49, depth=0x0) at /home/fish/Desktop/2018-10-10/exiv2/src/image.cpp:509 509 start = tooBig ? 0 : byteSwap4(dir,0,bSwap); gef➤ p start $4 = 0xfc0000 ```

Infinite loop in Exiv2::Image::printIFDStructure function of image.cpp:509

https://api.github.com/repos/Exiv2/exiv2/issues/511/comments

https://github.com/Exiv2/exiv2/issues/511

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : Infinite loop in Exiv2::Image::printIFDStructure function of image.cpp:509 Issue date: --- start body --- The issue may be similar to the issue #51 , with the POC file provided, The vulue of the start will always be `0xfc0000`, so the condition will be true all the time. ./exiv2 -pC [poc_infinite_loop](https://github.com/TeamSeri0us/pocs/blob/master/exiv2/poc_infinite_loop) or the command ./exiv2 -pR [poc_infinite_loop](https://github.com/TeamSeri0us/pocs/blob/master/exiv2/poc_infinite_loop) can trigger the situation: ``` ────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────[ source:/home/fish/Desktop/2018-10-10/exiv2/src/image.cpp+509 ]──── 504 out.write((const char*)buf.pData_,count); 505 } 506 } 507 if ( start ) { 508 io.read(dir.pData_, 4); → 509 start = tooBig ? 0 : byteSwap4(dir,0,bSwap); 510 } 511 } while (start) ; 512 513 if ( bPrint ) { 514 out << Internal::indent(depth) << "END " << io.path() << std::endl; ─────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────[ threads ]──── [#0] Id 1, Name: "exiv2", stopped, reason: BREAKPOINT ───────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────[ trace ]──── [#0] 0x7ffff7979845 → Name: Exiv2::Image::printIFDStructure(this=0x5555557aecb0, io=@0x5555557afae0, out=@0x7ffff7850e40, option=Exiv2::kpsIccProfile, start=0xfc0000, bSwap=0x0, c=0x49, depth=0x0) [#1] 0x7ffff7979c56 → Name: Exiv2::Image::printTiffStructure(this=0x5555557aecb0, io=@0x5555557afae0, out=@0x7ffff7850e40, option=Exiv2::kpsIccProfile, depth=0xffffffff, offset=0x0) [#2] 0x7ffff79b73bb → Name: Exiv2::TiffImage::printStructure(this=0x5555557aecb0, out=@0x7ffff7850e40, option=Exiv2::kpsIccProfile, depth=0x0) [#3] 0x555555586064 → Name: (anonymous namespace)::printStructure(out=@0x7ffff7850e40, option=Exiv2::kpsIccProfile, path="poc_infinite_loop") [#4] 0x555555575d55 → Name: Action::setModeAndPrintStructure(option=Exiv2::kpsIccProfile, path="poc_infinite_loop") [#5] 0x555555575ea3 → Name: Action::Print::run(this=0x5555557aff10, path="poc_infinite_loop") [#6] 0x5555555671ee → Name: main(argc=0x4, argv=0x7fffffffdd58) [#7] 0x7ffff6ee3b97 → Name: __libc_start_main(main=0x555555566f0a <main(int, char* const*)>, argc=0x4, argv=0x7fffffffdd58, init=<optimized out>, fini=<optimized out>, rtld_fini=<optimized out>, stack_end=0x7fffffffdd48) [#8] 0x555555566e2a → Name: _start() ──────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────── Breakpoint 2, Exiv2::Image::printIFDStructure (this=0x5555557aecb0, io=..., out=..., option=Exiv2::kpsIccProfile, start=0xfc0000, bSwap=0x0, c=0x49, depth=0x0) at /home/fish/Desktop/2018-10-10/exiv2/src/image.cpp:509 509 start = tooBig ? 0 : byteSwap4(dir,0,bSwap); gef➤ p start $4 = 0xfc0000 ``` --- end body --- comments : --- start comments --- --- end comments ---

There is a bug in function int getopt(int argc, char * const argv[], const char *optstring) in /src/getopt.c, while the incorrect parameters are processed, the value of optind should plus one, and the program should exit , but not continue to process the other parameters. `./exiv2 -A 20:00:00 -Y 2018 -O 11 -D 01 ad ./test/data/Reagan.tiff` ``` In file: /home/pwd/fuzz/fuzz-exiv2/exiv2/src/getopt.cpp 56 optopt = arg[optpos]; 57 if (!opt) { 58 if (opterr && *optstring != ':') 59 fprintf(stderr, "%s: illegal option: %c\n", argv[0], optopt); 60 return '?'; ► 61 } else if (opt[1] == ':') { 62 if (arg[optpos + 1]) { 63 optarg = (char *)arg + optpos + 1; 64 optind++; 65 optpos = 1; 66 return optopt; ```

infinite loop in getopt.cpp

https://api.github.com/repos/Exiv2/exiv2/issues/510/comments

https://github.com/Exiv2/exiv2/issues/510

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : infinite loop in getopt.cpp Issue date: --- start body --- There is a bug in function int getopt(int argc, char * const argv[], const char *optstring) in /src/getopt.c, while the incorrect parameters are processed, the value of optind should plus one, and the program should exit , but not continue to process the other parameters. `./exiv2 -A 20:00:00 -Y 2018 -O 11 -D 01 ad ./test/data/Reagan.tiff` ``` In file: /home/pwd/fuzz/fuzz-exiv2/exiv2/src/getopt.cpp 56 optopt = arg[optpos]; 57 if (!opt) { 58 if (opterr && *optstring != ':') 59 fprintf(stderr, "%s: illegal option: %c\n", argv[0], optopt); 60 return '?'; ► 61 } else if (opt[1] == ':') { 62 if (arg[optpos + 1]) { 63 optarg = (char *)arg + optpos + 1; 64 optind++; 65 optpos = 1; 66 return optopt; ``` --- end body --- comments : --- start comments --- --- end comments ---

Exiv2 0.26 has SONAME `libexiv2.so.26`, but Exiv2 0.27 has SONAME `libexiv2.so.0`. Is it expected? It looks wrong for the version number to go backward.

Library version in SONAME goes backward

https://api.github.com/repos/Exiv2/exiv2/issues/507/comments

https://github.com/Exiv2/exiv2/issues/507

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : Library version in SONAME goes backward Issue date: --- start body --- Exiv2 0.26 has SONAME `libexiv2.so.26`, but Exiv2 0.27 has SONAME `libexiv2.so.0`. Is it expected? It looks wrong for the version number to go backward. --- end body --- comments : --- start comments --- --- end comments ---

It looks like cmake/packaging.cmake was only intended for cpack runs, however it is included unconditionally from toplevel CMakeLists.txt: ``` include(cmake/packaging.cmake) include(cmake/printSummary.cmake) ``` This causes issues with "install" step because it installs the files for *packing* to %_prefix, which is unintended

[v27-RC1] cmake/packaging.cmake also included for build/install

https://api.github.com/repos/Exiv2/exiv2/issues/505/comments

https://github.com/Exiv2/exiv2/issues/505

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : [v27-RC1] cmake/packaging.cmake also included for build/install Issue date: --- start body --- It looks like cmake/packaging.cmake was only intended for cpack runs, however it is included unconditionally from toplevel CMakeLists.txt: ``` include(cmake/packaging.cmake) include(cmake/printSummary.cmake) ``` This causes issues with "install" step because it installs the files for *packing* to %_prefix, which is unintended --- end body --- comments : --- start comments --- --- end comments ---

Hi there! I'm using libexiv 0.25 on Ubuntu 16.04, and I've noticed a behaviour which is unexpected to me: I've got a Jpeg file on which I call `image.iptcData()`, and then: - if I assign some fields such as `Iptc.Application2.Caption`, this metadata gets properly saved; - if I use the `IptcData::add()` method to add multiple values to the `Iptc.Application2.Keywords` field, then this works *only if I call `image.setIptcData()`* afterwards. It looks like the iptcData pointer in the image changes by itself, in the second case. But by looking at the code of libexiv2, I cannot find any explanation for it. I can try to provide a simple test case, if needed (will require some time, though).

IPTC metadata not saved unless `setIptcData()` is called

https://api.github.com/repos/Exiv2/exiv2/issues/502/comments

https://github.com/Exiv2/exiv2/issues/502

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : IPTC metadata not saved unless `setIptcData()` is called Issue date: --- start body --- Hi there! I'm using libexiv 0.25 on Ubuntu 16.04, and I've noticed a behaviour which is unexpected to me: I've got a Jpeg file on which I call `image.iptcData()`, and then: - if I assign some fields such as `Iptc.Application2.Caption`, this metadata gets properly saved; - if I use the `IptcData::add()` method to add multiple values to the `Iptc.Application2.Keywords` field, then this works *only if I call `image.setIptcData()`* afterwards. It looks like the iptcData pointer in the image changes by itself, in the second case. But by looking at the code of libexiv2, I cannot find any explanation for it. I can try to provide a simple test case, if needed (will require some time, though). --- end body --- comments : --- start comments --- --- end comments ---

make install fails on FreeBSD by looking for the file FreeBSD/ReadMe.txt

https://api.github.com/repos/Exiv2/exiv2/issues/499/comments

https://github.com/Exiv2/exiv2/issues/499

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : make install fails on FreeBSD by looking for the file FreeBSD/ReadMe.txt Issue date: --- start body --- --- end body --- comments : --- start comments --- --- end comments ---

Recently we changed the library name from exiv2 to exiv2lib to overcome some issues detected with the debugging experience on Windows. However, to change the library names might not be the better idea and we are going to restore the original name before releasing v0.27.

Rename exiv2lib to exiv2

https://api.github.com/repos/Exiv2/exiv2/issues/498/comments

https://github.com/Exiv2/exiv2/issues/498

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : Rename exiv2lib to exiv2 Issue date: --- start body --- Recently we changed the library name from exiv2 to exiv2lib to overcome some issues detected with the debugging experience on Windows. However, to change the library names might not be the better idea and we are going to restore the original name before releasing v0.27. --- end body --- comments : --- start comments --- --- end comments ---

Exiv2 currently doesn't compile on FreeBSD: ``` *** Building exiv2 *** [39/228] ninja [24/86] Building CXX object src/CMakeFiles/exiv2.dir/actions.cpp.o FAILED: src/CMakeFiles/exiv2.dir/actions.cpp.o /usr/bin/clang++ -B/home/lantw44/.local/bin -DEXV_HAVE_DLL -DEXV_HAVE_STDINT_H -DEXV_LOCALEDIR=\"share/locale\" -I/home/lantw44/gnome/source/exiv2/src -I/home/lantw44/gnome/source/exiv2/include -I/home/lantw44/gnome/source/exiv2/include/exiv2 -I. -I/home/lantw44/gnome/source/exiv2/xmpsdk/include -I/usr/local/include -Isrc -march=corei7 -B/home/lantw44/.local/bin -pipe -g3 -O0 -gz -fdebug-macro -std=gnu++98 -MD -MT src/CMakeFiles/exiv2.dir/actions.cpp.o -MF src/CMakeFiles/exiv2.dir/actions.cpp.o.d -o src/CMakeFiles/exiv2.dir/actions.cpp.o -c /home/lantw44/gnome/source/exiv2/src/actions.cpp /home/lantw44/gnome/source/exiv2/src/actions.cpp:2063:32: error: use of undeclared identifier 'PTHREAD_RECURSIVE_MUTEX_INITIALIZER_NP' static pthread_mutex_t cs = PTHREAD_RECURSIVE_MUTEX_INITIALIZER_NP; ^ 1 error generated. [29/86] Building CXX object src/CMakeFiles/exiv2lib.dir/convert.cpp.o ninja: build stopped: subcommand failed. ``` The `_NP` suffix of `PTHREAD_RECURSIVE_MUTEX_INITIALIZER_NP` suggests it is a non-portable macro. FreeBSD supports recursive mutex, but it does not define a static initializer for it. FreeBSD pthread.h only defines two mutex initializers, `PTHREAD_MUTEX_INITIALIZER` and `PTHREAD_ADAPTIVE_MUTEX_INITIALIZER_NP`, and neither of them is equivalent to `PTHREAD_RECURSIVE_MUTEX_INITIALIZER_NP` used here. If the recursive behavior is needed, is it possible to use `pthread_mutexattr_settype` instead of allocating it statically?

There is no PTHREAD_RECURSIVE_MUTEX_INITIALIZER_NP on FreeBSD

https://api.github.com/repos/Exiv2/exiv2/issues/497/comments

https://github.com/Exiv2/exiv2/issues/497

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : There is no PTHREAD_RECURSIVE_MUTEX_INITIALIZER_NP on FreeBSD Issue date: --- start body --- Exiv2 currently doesn't compile on FreeBSD: ``` *** Building exiv2 *** [39/228] ninja [24/86] Building CXX object src/CMakeFiles/exiv2.dir/actions.cpp.o FAILED: src/CMakeFiles/exiv2.dir/actions.cpp.o /usr/bin/clang++ -B/home/lantw44/.local/bin -DEXV_HAVE_DLL -DEXV_HAVE_STDINT_H -DEXV_LOCALEDIR=\"share/locale\" -I/home/lantw44/gnome/source/exiv2/src -I/home/lantw44/gnome/source/exiv2/include -I/home/lantw44/gnome/source/exiv2/include/exiv2 -I. -I/home/lantw44/gnome/source/exiv2/xmpsdk/include -I/usr/local/include -Isrc -march=corei7 -B/home/lantw44/.local/bin -pipe -g3 -O0 -gz -fdebug-macro -std=gnu++98 -MD -MT src/CMakeFiles/exiv2.dir/actions.cpp.o -MF src/CMakeFiles/exiv2.dir/actions.cpp.o.d -o src/CMakeFiles/exiv2.dir/actions.cpp.o -c /home/lantw44/gnome/source/exiv2/src/actions.cpp /home/lantw44/gnome/source/exiv2/src/actions.cpp:2063:32: error: use of undeclared identifier 'PTHREAD_RECURSIVE_MUTEX_INITIALIZER_NP' static pthread_mutex_t cs = PTHREAD_RECURSIVE_MUTEX_INITIALIZER_NP; ^ 1 error generated. [29/86] Building CXX object src/CMakeFiles/exiv2lib.dir/convert.cpp.o ninja: build stopped: subcommand failed. ``` The `_NP` suffix of `PTHREAD_RECURSIVE_MUTEX_INITIALIZER_NP` suggests it is a non-portable macro. FreeBSD supports recursive mutex, but it does not define a static initializer for it. FreeBSD pthread.h only defines two mutex initializers, `PTHREAD_MUTEX_INITIALIZER` and `PTHREAD_ADAPTIVE_MUTEX_INITIALIZER_NP`, and neither of them is equivalent to `PTHREAD_RECURSIVE_MUTEX_INITIALIZER_NP` used here. If the recursive behavior is needed, is it possible to use `pthread_mutexattr_settype` instead of allocating it statically? --- end body --- comments : --- start comments --- --- end comments ---

(gdb) r Starting program: /usr/bin/exiv2 -e e ./IMG_2520.CR2 Missing separate debuginfos, use: debuginfo-install glibc-2.22-29.mga6.x86_64 [Thread debugging using libthread_db enabled] Using host libthread_db library "/lib64/libthread_db.so.1". Warning: Exif tag Exif.Photo.MakerNote not encoded Warning: Exif tag Exif.Canon.0x4005 not encoded Program received signal SIGSEGV, Segmentation fault. 0x00007ffff797bde2 in Exiv2::ExifParser::encode(std::vector<unsigned char, std::allocator<unsigned char> >&, unsigned char const*, unsigned int, Exiv2::ByteOrder, Exiv2::ExifData const&) () from /usr/bin/../lib64/libexiv2.so.0 Missing separate debuginfos, use: debuginfo-install lib64expat1-2.2.1-1.mga6.x86_64 lib64zlib1-1.2.11-4.1.mga6.x86_64 libgcc1-5.5.0-1.mga6.x86_64 libstdc++6-5.5.0-1.mga6.x86_64 (gdb) bt #0 0x00007ffff797bde2 in Exiv2::ExifParser::encode(std::vector<unsigned char, std::allocator<unsigned char> >&, unsigned char const*, unsigned int, Exiv2::ByteOrder, Exiv2::ExifData const&) () from /usr/bin/../lib64/libexiv2.so.0 #1 0x00007ffff79a9e89 in Exiv2::JpegBase::doWriteMetadata(Exiv2::BasicIo&) () from /usr/bin/../lib64/libexiv2.so.0 #2 0x00007ffff79a8cda in Exiv2::JpegBase::writeMetadata() () from /usr/bin/../lib64/libexiv2.so.0 #3 0x000000000042e96c in (anonymous namespace)::metacopy(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, int, bool) () #4 0x00000000004277b2 in Action::Extract::run(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&) () #5 0x0000000000412b4e in main () (gdb) CR2 file to test is here : https://cgit.kde.org/digikam.git/tree/core/tests/metadataengine/data/IMG_2520.CR2 Original report from digiKam : https://bugs.kde.org/show_bug.cgi?id=400140 Gilles Caulier

Exiv2 0.27 : CR2 and Exif data extraction crash to ExifParser::encode()

https://api.github.com/repos/Exiv2/exiv2/issues/496/comments

https://github.com/Exiv2/exiv2/issues/496

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : Exiv2 0.27 : CR2 and Exif data extraction crash to ExifParser::encode() Issue date: --- start body --- (gdb) r Starting program: /usr/bin/exiv2 -e e ./IMG_2520.CR2 Missing separate debuginfos, use: debuginfo-install glibc-2.22-29.mga6.x86_64 [Thread debugging using libthread_db enabled] Using host libthread_db library "/lib64/libthread_db.so.1". Warning: Exif tag Exif.Photo.MakerNote not encoded Warning: Exif tag Exif.Canon.0x4005 not encoded Program received signal SIGSEGV, Segmentation fault. 0x00007ffff797bde2 in Exiv2::ExifParser::encode(std::vector<unsigned char, std::allocator<unsigned char> >&, unsigned char const*, unsigned int, Exiv2::ByteOrder, Exiv2::ExifData const&) () from /usr/bin/../lib64/libexiv2.so.0 Missing separate debuginfos, use: debuginfo-install lib64expat1-2.2.1-1.mga6.x86_64 lib64zlib1-1.2.11-4.1.mga6.x86_64 libgcc1-5.5.0-1.mga6.x86_64 libstdc++6-5.5.0-1.mga6.x86_64 (gdb) bt #0 0x00007ffff797bde2 in Exiv2::ExifParser::encode(std::vector<unsigned char, std::allocator<unsigned char> >&, unsigned char const*, unsigned int, Exiv2::ByteOrder, Exiv2::ExifData const&) () from /usr/bin/../lib64/libexiv2.so.0 #1 0x00007ffff79a9e89 in Exiv2::JpegBase::doWriteMetadata(Exiv2::BasicIo&) () from /usr/bin/../lib64/libexiv2.so.0 #2 0x00007ffff79a8cda in Exiv2::JpegBase::writeMetadata() () from /usr/bin/../lib64/libexiv2.so.0 #3 0x000000000042e96c in (anonymous namespace)::metacopy(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, int, bool) () #4 0x00000000004277b2 in Action::Extract::run(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&) () #5 0x0000000000412b4e in main () (gdb) CR2 file to test is here : https://cgit.kde.org/digikam.git/tree/core/tests/metadataengine/data/IMG_2520.CR2 Original report from digiKam : https://bugs.kde.org/show_bug.cgi?id=400140 Gilles Caulier --- end body --- comments : --- start comments --- --- end comments ---

As described in #487, we had to take some actions to fix the compilation on Windows + STATIC mode. The issue is that when using `/MT` (STATIC) on exiv2, expat does not find a symbol: ``` expat.lib(loadlibrary.obj) : error LNK2019: unresolved external symbol __imp__mbspbrk referenced in function _Expat_LoadLibrary [C:\Users\rmills\gnu\github\exiv2\exiv2\build\s ``` It seems that in the expat new release (2.2.6) they have improved their CMake code and now you can control which C Runtime Library to use (by means of the variable `MSVC_USE_STATIC_CRT`). By moving to that version of expat we could simplify more our CMake configuration (And probably get rid off the `EXIV2_ENABLE_DYNAMIC_RUNTIME` variable and just rely in `BUILD_SHARED_LIBS`).

Update expat (to version 2.2.6) and simplify our CMake code

https://api.github.com/repos/Exiv2/exiv2/issues/494/comments

https://github.com/Exiv2/exiv2/issues/494

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : Update expat (to version 2.2.6) and simplify our CMake code Issue date: --- start body --- As described in #487, we had to take some actions to fix the compilation on Windows + STATIC mode. The issue is that when using `/MT` (STATIC) on exiv2, expat does not find a symbol: ``` expat.lib(loadlibrary.obj) : error LNK2019: unresolved external symbol __imp__mbspbrk referenced in function _Expat_LoadLibrary [C:\Users\rmills\gnu\github\exiv2\exiv2\build\s ``` It seems that in the expat new release (2.2.6) they have improved their CMake code and now you can control which C Runtime Library to use (by means of the variable `MSVC_USE_STATIC_CRT`). By moving to that version of expat we could simplify more our CMake configuration (And probably get rid off the `EXIV2_ENABLE_DYNAMIC_RUNTIME` variable and just rely in `BUILD_SHARED_LIBS`). --- end body --- comments : --- start comments --- --- end comments ---

tl;dr; Imho we throw too many exceptions and should instead make the code more robust. Since I have started contributing to exiv2, I have seen a steady increase in the amount of exceptions that we throw from the code (mainly as quick fixes to circumvent issues in fuzzed files) and I am unsure whether that is such a good thing. To illustrate what I mean, take a look at #180, especially at the changes to the test suite. A **lot** of the fuzzed files that were causing exceptions in the pre-#180 code were now no longer throwing exceptions and instead parsed the fuzzed file (kind-of ok) and instead printed warnings to stderr. I am afraid, that if we continue the current trend of throwing exceptions for every problem that we find, then we'll be in trouble in the future: The code will be harder to use by users of the library, as they'll have to catch exceptions and find ways how to report them to the user (which is quite hard for generic errors like `kerCorruptedMetadata`). Also, I have the feeling, that we are currently using exceptions as a "short-cut": instead of fixing the code to be more robust, we just throw an exception and the thing is "fixed". However, that's tactical programming and a bad idea in the long term (unless we want more technical debt). Therefore I propose the following policy for exceptions for discussion: - The code should be as robust as possible and assume sane defaults for parsed values, when the retrieved value is obviously wrong. E.g. if the file ends prematurely, the code should parse as much as possible and return everything that appeared to be valid and **not** throw an exception at the end. - Exceptions should be only thrown in truly exceptional circumstances, i.e.: only as a last resort to prevent a crash (e.g. if an array access would read out of bounds) or when there is **no way** to recover (like when a file cannot be opened). This has imho the following advantages: - Reduced cognitive load for downstream library users. They'll know that Exiv2 will only throw an exception either when there is a bug in the library, or if there is some problem, that they can report to the user (like that the file does not exist). - The code can become more robust, as it'll assume sane values, when the extracted ones are insane. This can be very beneficial, as we are currently mostly testing "unhappy" paths in the test suite, but don't have a lot of tests for files that are slightly corrupted by buggy software. Since most of our users don't live at `HEAD`, we won't be seeing reports from them until a RC/release, thereby increasing our workload. - The library will become a lot more efficient for unhappy paths (see https://ned14.github.io/outcome/faq/#what-kind-of-performance-benefits-will-using-outcome-in-my-code-bring for the performance hit that a thrown exception incurs). The performance hit for the happy paths on the other hand should be manageable.

Reconsider exception throwing policy

https://api.github.com/repos/Exiv2/exiv2/issues/488/comments

https://github.com/Exiv2/exiv2/issues/488

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : Reconsider exception throwing policy Issue date: --- start body --- tl;dr; Imho we throw too many exceptions and should instead make the code more robust. Since I have started contributing to exiv2, I have seen a steady increase in the amount of exceptions that we throw from the code (mainly as quick fixes to circumvent issues in fuzzed files) and I am unsure whether that is such a good thing. To illustrate what I mean, take a look at #180, especially at the changes to the test suite. A **lot** of the fuzzed files that were causing exceptions in the pre-#180 code were now no longer throwing exceptions and instead parsed the fuzzed file (kind-of ok) and instead printed warnings to stderr. I am afraid, that if we continue the current trend of throwing exceptions for every problem that we find, then we'll be in trouble in the future: The code will be harder to use by users of the library, as they'll have to catch exceptions and find ways how to report them to the user (which is quite hard for generic errors like `kerCorruptedMetadata`). Also, I have the feeling, that we are currently using exceptions as a "short-cut": instead of fixing the code to be more robust, we just throw an exception and the thing is "fixed". However, that's tactical programming and a bad idea in the long term (unless we want more technical debt). Therefore I propose the following policy for exceptions for discussion: - The code should be as robust as possible and assume sane defaults for parsed values, when the retrieved value is obviously wrong. E.g. if the file ends prematurely, the code should parse as much as possible and return everything that appeared to be valid and **not** throw an exception at the end. - Exceptions should be only thrown in truly exceptional circumstances, i.e.: only as a last resort to prevent a crash (e.g. if an array access would read out of bounds) or when there is **no way** to recover (like when a file cannot be opened). This has imho the following advantages: - Reduced cognitive load for downstream library users. They'll know that Exiv2 will only throw an exception either when there is a bug in the library, or if there is some problem, that they can report to the user (like that the file does not exist). - The code can become more robust, as it'll assume sane values, when the extracted ones are insane. This can be very beneficial, as we are currently mostly testing "unhappy" paths in the test suite, but don't have a lot of tests for files that are slightly corrupted by buggy software. Since most of our users don't live at `HEAD`, we won't be seeing reports from them until a RC/release, thereby increasing our workload. - The library will become a lot more efficient for unhappy paths (see https://ned14.github.io/outcome/faq/#what-kind-of-performance-benefits-will-using-outcome-in-my-code-bring for the performance hit that a thrown exception incurs). The performance hit for the happy paths on the other hand should be manageable. --- end body --- comments : --- start comments --- --- end comments ---

I noticed few things when trying to debug on Windows + Visual Studio 2017. 1. I can debug the **exiv2** application but I cannot access the exiv2 library symbols when I try to step into a exiv2 function/class. On linux I can debug with gcc/clang without issues. 2. The compilation of the **exiv2** application with exiv2lib as a STATIC library fails (`-DBUILD_SHARED_LIBS=OFF`). **TODO**: Investigate more about these issues and complete the information in this description.

Windows + MSVC issues

https://api.github.com/repos/Exiv2/exiv2/issues/487/comments

https://github.com/Exiv2/exiv2/issues/487

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : Windows + MSVC issues Issue date: --- start body --- I noticed few things when trying to debug on Windows + Visual Studio 2017. 1. I can debug the **exiv2** application but I cannot access the exiv2 library symbols when I try to step into a exiv2 function/class. On linux I can debug with gcc/clang without issues. 2. The compilation of the **exiv2** application with exiv2lib as a STATIC library fails (`-DBUILD_SHARED_LIBS=OFF`). **TODO**: Investigate more about these issues and complete the information in this description. --- end body --- comments : --- start comments --- --- end comments ---

This is occurs on MinGW and MSVC **MSVC** ``` C:\Users\rmills\gnu\github\exiv2\buildserver\build>msys64 157 rmills@rmillsmm-w7-64:~/gnu/github/exiv2/buildserver/build $ cd ../test 158 rmills@rmillsmm-w7-64:~/gnu/github/exiv2/buildserver/test $ PATH="/c/Python36:$PATH" 159 rmills@rmillsmm-w7-64:~/gnu/github/exiv2/buildserver/test $ env EXIV2_BINDIR=${PWD}/../build/bin ./conversions.sh Files /c/Users/rmills/gnu/github/exiv2/buildserver/test/tmp/conversions.out and /c/Users/rmills/gnu/github/exiv2/buildserver/test/data/conversions.out differ 165c165 < Exif.Image.DateTime Ascii 20 2015:04:17 19:10:22 --- > Exif.Image.DateTime Ascii 20 2015:04:18 02:10:22 160 rmills@rmillsmm-w7-64:~/gnu/github/exiv2/buildserver/test $ ``` Here's a way to produce this from the terminal (without using test/conversions.sh) on an MSVC build: ``` C:\Users\rmills\gnu\github\exiv2\buildserver\test>msys64 162 rmills@rmillsmm-w7-64:~/gnu/github/exiv2/buildserver/test $ cd ~/gnu/github/exiv2/buildserver/test 163 rmills@rmillsmm-w7-64:~/gnu/github/exiv2/buildserver/test $ cp data/exiv2-empty.jpg empty.jpg 164 rmills@rmillsmm-w7-64:~/gnu/github/exiv2/buildserver/test $ ../build/bin/exiv2 -eX --verbose --force empty.jpg File 1/1: empty.jpg 165 rmills@rmillsmm-w7-64:~/gnu/github/exiv2/buildserver/test $ ../build/bin/exiv2 -M'set Xmp.xmp.ModifyDate 2015-04-17T18:10:22Z' empty.xmp 166 rmills@rmillsmm-w7-64:~/gnu/github/exiv2/buildserver/test $ TZ=GMT-8 ../build/bin/exiv2 -iX --verbose --force empty.jpg File 1/1: empty.jpg Writing Exif data from .\empty.xmp to empty.jpg Writing XMP data from .\empty.xmp to empty.jpg 168 rmills@rmillsmm-w7-64:~/gnu/github/exiv2/buildserver/test $ ../build/bin/exiv2 -pa --grep date/i empty.jpg Exif.Image.DateTime Ascii 20 2015:04:17 19:10:22 Xmp.xmp.ModifyDate XmpText 20 2015-04-17T18:10:22Z 169 rmills@rmillsmm-w7-64:~/gnu/github/exiv2/buildserver/test $ ``` On MacOS-X, the same commands: ``` 540 rmills@rmillsmm:~/gnu/github/exiv2/buildserver $ cd test 541 rmills@rmillsmm:~/gnu/github/exiv2/buildserver/test $ cp data/exiv2-empty.jpg empty.jpg 542 rmills@rmillsmm:~/gnu/github/exiv2/buildserver/test $ ../build/bin/exiv2 -eX --verbose --force empty.jpg File 1/1: empty.jpg 543 rmills@rmillsmm:~/gnu/github/exiv2/buildserver/test $ ../build/bin/exiv2 -M'set Xmp.xmp.ModifyDate 2015-04-17T18:10:22Z' empty.xmp 544 rmills@rmillsmm:~/gnu/github/exiv2/buildserver/test $ TZ=GMT-8 ../build/bin/exiv2 -iX --verbose --force empty.jpg File 1/1: empty.jpg Writing Exif data from ./empty.xmp to empty.jpg Writing XMP data from ./empty.xmp to empty.jpg 546 rmills@rmillsmm:~/gnu/github/exiv2/buildserver/test $ ../build/bin/exiv2 -pa --grep date/i empty.jpg Exif.Image.DateTime Ascii 20 2015:04:18 02:10:22 Xmp.xmp.ModifyDate XmpText 20 2015-04-17T18:10:22Z 547 rmills@rmillsmm:~/gnu/github/exiv2/buildserver/test $ ``` **Analysis** The command `TZ=GNT-8 exiv2 -iX ....` 1) is correctly copying xmp:ModifyDate to the XMP 2) the convertor on MSVC sets Exif.Image.DateTime to 2015:04:17 19:10:22 and it should be 2015:04:18 02:10:22 on my machine in England (during Daylight saving time). MSVC: ``` Exif.Image.DateTime Ascii 20 2015:04:17 19:10:22 Xmp.xmp.ModifyDate XmpText 20 2015-04-17T18:10:22Z ``` MacOS-X: ``` Exif.Image.DateTime Ascii 20 2015:04:18 02:10:22 Xmp.xmp.ModifyDate XmpText 20 2015-04-17T18:10:22Z ``` **How to fix this** This is hard. Time conversions on v0.26 were performed with src/localtime.c and src/gmtime.h. Gilles asked us to reinstate src/localtime.c because he was getting linking errors on MinGW/msys2. I hoped that fix would resolve this.

test/conversions.sh is failing on TZ test

https://api.github.com/repos/Exiv2/exiv2/issues/485/comments

https://github.com/Exiv2/exiv2/issues/485

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : test/conversions.sh is failing on TZ test Issue date: --- start body --- This is occurs on MinGW and MSVC **MSVC** ``` C:\Users\rmills\gnu\github\exiv2\buildserver\build>msys64 157 rmills@rmillsmm-w7-64:~/gnu/github/exiv2/buildserver/build $ cd ../test 158 rmills@rmillsmm-w7-64:~/gnu/github/exiv2/buildserver/test $ PATH="/c/Python36:$PATH" 159 rmills@rmillsmm-w7-64:~/gnu/github/exiv2/buildserver/test $ env EXIV2_BINDIR=${PWD}/../build/bin ./conversions.sh Files /c/Users/rmills/gnu/github/exiv2/buildserver/test/tmp/conversions.out and /c/Users/rmills/gnu/github/exiv2/buildserver/test/data/conversions.out differ 165c165 < Exif.Image.DateTime Ascii 20 2015:04:17 19:10:22 --- > Exif.Image.DateTime Ascii 20 2015:04:18 02:10:22 160 rmills@rmillsmm-w7-64:~/gnu/github/exiv2/buildserver/test $ ``` Here's a way to produce this from the terminal (without using test/conversions.sh) on an MSVC build: ``` C:\Users\rmills\gnu\github\exiv2\buildserver\test>msys64 162 rmills@rmillsmm-w7-64:~/gnu/github/exiv2/buildserver/test $ cd ~/gnu/github/exiv2/buildserver/test 163 rmills@rmillsmm-w7-64:~/gnu/github/exiv2/buildserver/test $ cp data/exiv2-empty.jpg empty.jpg 164 rmills@rmillsmm-w7-64:~/gnu/github/exiv2/buildserver/test $ ../build/bin/exiv2 -eX --verbose --force empty.jpg File 1/1: empty.jpg 165 rmills@rmillsmm-w7-64:~/gnu/github/exiv2/buildserver/test $ ../build/bin/exiv2 -M'set Xmp.xmp.ModifyDate 2015-04-17T18:10:22Z' empty.xmp 166 rmills@rmillsmm-w7-64:~/gnu/github/exiv2/buildserver/test $ TZ=GMT-8 ../build/bin/exiv2 -iX --verbose --force empty.jpg File 1/1: empty.jpg Writing Exif data from .\empty.xmp to empty.jpg Writing XMP data from .\empty.xmp to empty.jpg 168 rmills@rmillsmm-w7-64:~/gnu/github/exiv2/buildserver/test $ ../build/bin/exiv2 -pa --grep date/i empty.jpg Exif.Image.DateTime Ascii 20 2015:04:17 19:10:22 Xmp.xmp.ModifyDate XmpText 20 2015-04-17T18:10:22Z 169 rmills@rmillsmm-w7-64:~/gnu/github/exiv2/buildserver/test $ ``` On MacOS-X, the same commands: ``` 540 rmills@rmillsmm:~/gnu/github/exiv2/buildserver $ cd test 541 rmills@rmillsmm:~/gnu/github/exiv2/buildserver/test $ cp data/exiv2-empty.jpg empty.jpg 542 rmills@rmillsmm:~/gnu/github/exiv2/buildserver/test $ ../build/bin/exiv2 -eX --verbose --force empty.jpg File 1/1: empty.jpg 543 rmills@rmillsmm:~/gnu/github/exiv2/buildserver/test $ ../build/bin/exiv2 -M'set Xmp.xmp.ModifyDate 2015-04-17T18:10:22Z' empty.xmp 544 rmills@rmillsmm:~/gnu/github/exiv2/buildserver/test $ TZ=GMT-8 ../build/bin/exiv2 -iX --verbose --force empty.jpg File 1/1: empty.jpg Writing Exif data from ./empty.xmp to empty.jpg Writing XMP data from ./empty.xmp to empty.jpg 546 rmills@rmillsmm:~/gnu/github/exiv2/buildserver/test $ ../build/bin/exiv2 -pa --grep date/i empty.jpg Exif.Image.DateTime Ascii 20 2015:04:18 02:10:22 Xmp.xmp.ModifyDate XmpText 20 2015-04-17T18:10:22Z 547 rmills@rmillsmm:~/gnu/github/exiv2/buildserver/test $ ``` **Analysis** The command `TZ=GNT-8 exiv2 -iX ....` 1) is correctly copying xmp:ModifyDate to the XMP 2) the convertor on MSVC sets Exif.Image.DateTime to 2015:04:17 19:10:22 and it should be 2015:04:18 02:10:22 on my machine in England (during Daylight saving time). MSVC: ``` Exif.Image.DateTime Ascii 20 2015:04:17 19:10:22 Xmp.xmp.ModifyDate XmpText 20 2015-04-17T18:10:22Z ``` MacOS-X: ``` Exif.Image.DateTime Ascii 20 2015:04:18 02:10:22 Xmp.xmp.ModifyDate XmpText 20 2015-04-17T18:10:22Z ``` **How to fix this** This is hard. Time conversions on v0.26 were performed with src/localtime.c and src/gmtime.h. Gilles asked us to reinstate src/localtime.c because he was getting linking errors on MinGW/msys2. I hoped that fix would resolve this. --- end body --- comments : --- start comments --- --- end comments ---

Look in this file at line #75 : https://github.com/KDE/digikam/blob/master/core/libs/dmetadata/metaengine_p.h#L75 metadaengine_p.h/cpp is the private implementation of the digiKam Qt Exiv2 wrapper. In this class, all that we need from Exiv2 is called. The rest of digiKam don't know Exiv2. This prevent to make Exiv2 public headers fully opaque from lead digiKam implementation. This improve the compatibility in code with previous Exiv2 versions at compilation time. Other advantage, the C++ exception wrapper is located in the same class and not dispatched everywhere. This simplify the maintenance in time. Lokk now the puzzle at line 75. Before Exiv2 0.27, all library versions information was located in one file : exiv2/exv_conf.h Now, these information are split in a new file exiv2/version.h The break a little bit the compatibility with Exiv2 0.26 and older. Note that in all cases digiKam forace to use 0.26 as minimum Exiv2 version even if previous version can be used. This permit to reduce the crash report already know and fixed in 0.26. I must admit that digiKam was quickly instable with Exiv2 < 0.26. So with the exiv2/version.hpp, we need to wrap with preprocessing rules this header to check if Exiv2 version > 0.26, else the compilation is broken. Note the pre-processor rule, which is a copy that EXIV2_MAKE_VERSION macro defined in... version.hpp. so i cannot use this as well to wrap version.hpp (:=))) Note also the macro that i define at line #85 : https://github.com/KDE/digikam/blob/master/core/libs/dmetadata/metaengine_p.h#L85 This kind of macro must be defined in a Exiv2 header, not in digiKam. I cannot found an equivalent in 0.27. This macro simplify the branching of code with pre-processor rules to preserve compatibility with older version of the library... For digiKam 6.0.0, we can live with this implementation, but it's not the best. This can be improved and simplified in the future... Gilles Caulier

Exiv2 0.27 : new header exiv2/version.hpp have been introduce which make a puzzle with backward version

https://api.github.com/repos/Exiv2/exiv2/issues/483/comments

https://github.com/Exiv2/exiv2/issues/483

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : Exiv2 0.27 : new header exiv2/version.hpp have been introduce which make a puzzle with backward version Issue date: --- start body --- Look in this file at line #75 : https://github.com/KDE/digikam/blob/master/core/libs/dmetadata/metaengine_p.h#L75 metadaengine_p.h/cpp is the private implementation of the digiKam Qt Exiv2 wrapper. In this class, all that we need from Exiv2 is called. The rest of digiKam don't know Exiv2. This prevent to make Exiv2 public headers fully opaque from lead digiKam implementation. This improve the compatibility in code with previous Exiv2 versions at compilation time. Other advantage, the C++ exception wrapper is located in the same class and not dispatched everywhere. This simplify the maintenance in time. Lokk now the puzzle at line 75. Before Exiv2 0.27, all library versions information was located in one file : exiv2/exv_conf.h Now, these information are split in a new file exiv2/version.h The break a little bit the compatibility with Exiv2 0.26 and older. Note that in all cases digiKam forace to use 0.26 as minimum Exiv2 version even if previous version can be used. This permit to reduce the crash report already know and fixed in 0.26. I must admit that digiKam was quickly instable with Exiv2 < 0.26. So with the exiv2/version.hpp, we need to wrap with preprocessing rules this header to check if Exiv2 version > 0.26, else the compilation is broken. Note the pre-processor rule, which is a copy that EXIV2_MAKE_VERSION macro defined in... version.hpp. so i cannot use this as well to wrap version.hpp (:=))) Note also the macro that i define at line #85 : https://github.com/KDE/digikam/blob/master/core/libs/dmetadata/metaengine_p.h#L85 This kind of macro must be defined in a Exiv2 header, not in digiKam. I cannot found an equivalent in 0.27. This macro simplify the branching of code with pre-processor rules to preserve compatibility with older version of the library... For digiKam 6.0.0, we can live with this implementation, but it's not the best. This can be improved and simplified in the future... Gilles Caulier --- end body --- comments : --- start comments --- --- end comments ---

clang scan-build is another very important tool to detect quickly some internal dysfunctions. It's very configurable but slower than cppcheck as it require a code compilation. The HTML report is very easy to review and can give different feedbacks than cppcheck. I use also this static analyzer over whole digiKam implementation (around 1.5 M lines of code), and we clean and stabilize the application with a lots of true errors detected. Gilles Caulier [exiv2-0.27-clang-scan-build-reports.tar.gz](https://github.com/Exiv2/exiv2/files/2467499/exiv2-0.27-clang-scan-build-reports.tar.gz)

Exiv2 0.27 - Clang scan-build static analyzer HTML reports

https://api.github.com/repos/Exiv2/exiv2/issues/475/comments

https://github.com/Exiv2/exiv2/issues/475

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : Exiv2 0.27 - Clang scan-build static analyzer HTML reports Issue date: --- start body --- clang scan-build is another very important tool to detect quickly some internal dysfunctions. It's very configurable but slower than cppcheck as it require a code compilation. The HTML report is very easy to review and can give different feedbacks than cppcheck. I use also this static analyzer over whole digiKam implementation (around 1.5 M lines of code), and we clean and stabilize the application with a lots of true errors detected. Gilles Caulier [exiv2-0.27-clang-scan-build-reports.tar.gz](https://github.com/Exiv2/exiv2/files/2467499/exiv2-0.27-clang-scan-build-reports.tar.gz) --- end body --- comments : --- start comments --- --- end comments ---

cppcheck is a very important tool to detect quickly some internal dysfunctions. It's very configurable and very fast. The HTML report is very easy to review. I use this static analyzer over whole digiKam implementation (around 1.5 M lines of code), and we clean and stabilize the application with a lots of true errors detected. I recommend to review the right light annotated reports from the front page. Gilles Caulier [Exiv2-0.27-cppcheck-reports.tar.gz](https://github.com/Exiv2/exiv2/files/2467417/Exiv2-0.27-cppcheck-reports.tar.gz)

Exiv2 0.27 - cppcheck static analyzer HTML reports

https://api.github.com/repos/Exiv2/exiv2/issues/474/comments

https://github.com/Exiv2/exiv2/issues/474

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : Exiv2 0.27 - cppcheck static analyzer HTML reports Issue date: --- start body --- cppcheck is a very important tool to detect quickly some internal dysfunctions. It's very configurable and very fast. The HTML report is very easy to review. I use this static analyzer over whole digiKam implementation (around 1.5 M lines of code), and we clean and stabilize the application with a lots of true errors detected. I recommend to review the right light annotated reports from the front page. Gilles Caulier [Exiv2-0.27-cppcheck-reports.tar.gz](https://github.com/Exiv2/exiv2/files/2467417/Exiv2-0.27-cppcheck-reports.tar.gz) --- end body --- comments : --- start comments --- --- end comments ---

This problem do no appear with Exiv2 0.26 in 32 or 64 bits target for MinGW. But i remember that i already report this problem with older version of Exiv2 directly to Andreas to support MinGW. With Exiv2 0.27, the MinGW 32 bits target do not link at end : [ 19%] Built target xmp [ 44%] Built target exiv2lib_int [ 45%] Linking CXX shared library ../bin/libexiv2.dll CMakeFiles/exiv2lib.dir/objects.a(crwimage_int.cpp.obj): In function `mkgmtime': /mnt/data/GIT/6.x/project/bundles/mxe/build.win32/usr/i686-w64-mingw32.shared/include/time.h:229: undefined reference to `_mkgmtime32' collect2: error: ld returned 1 exit status This link error do not depend of one configuration option passed to cmake. The linking rules at end used by cmake are mostly the same between 0.26 and 0.27 (dixit "export VERBOSE=1"). In fact there is a rule with timegm() function present in 0.26 and removed to 0.27 which break the MinGW compatibility. With 0.27 : - localtime.c have been removed. - timegm.h was been changed. If i add back localtime.c, restore timegm.h and patch src/CMakeLists.txt to compile localtime.c, this fix the linking problem. See the patch attached, but it's not perfect, as Exiv2 0.26 has some rules to compile localtime.c : # modify source lists to suit environment IF(NOT HAVE_TIMEGM ) SET( LIBEXIV2_SRC ${LIBEXIV2_SRC} localtime.c ) SET( EXIV2_SRC ${EXIV2_SRC} localtime.c ) SET( PATHTEST_SRC ${PATHTEST_SRC} localtime.c ) ENDIF( NOT HAVE_TIMEGM ) The patch add localtime.c as well to compile libexiv2, so some tunes need to be done to complete the job. [Exiv2-0.27-mingw-timegm.patch.txt](https://github.com/Exiv2/exiv2/files/2467357/Exiv2-0.27-mingw-timegm.patch.txt)

MinGW 32 bits : linking failure with mkgmtime()

https://api.github.com/repos/Exiv2/exiv2/issues/473/comments

https://github.com/Exiv2/exiv2/issues/473

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : MinGW 32 bits : linking failure with mkgmtime() Issue date: --- start body --- This problem do no appear with Exiv2 0.26 in 32 or 64 bits target for MinGW. But i remember that i already report this problem with older version of Exiv2 directly to Andreas to support MinGW. With Exiv2 0.27, the MinGW 32 bits target do not link at end : [ 19%] Built target xmp [ 44%] Built target exiv2lib_int [ 45%] Linking CXX shared library ../bin/libexiv2.dll CMakeFiles/exiv2lib.dir/objects.a(crwimage_int.cpp.obj): In function `mkgmtime': /mnt/data/GIT/6.x/project/bundles/mxe/build.win32/usr/i686-w64-mingw32.shared/include/time.h:229: undefined reference to `_mkgmtime32' collect2: error: ld returned 1 exit status This link error do not depend of one configuration option passed to cmake. The linking rules at end used by cmake are mostly the same between 0.26 and 0.27 (dixit "export VERBOSE=1"). In fact there is a rule with timegm() function present in 0.26 and removed to 0.27 which break the MinGW compatibility. With 0.27 : - localtime.c have been removed. - timegm.h was been changed. If i add back localtime.c, restore timegm.h and patch src/CMakeLists.txt to compile localtime.c, this fix the linking problem. See the patch attached, but it's not perfect, as Exiv2 0.26 has some rules to compile localtime.c : # modify source lists to suit environment IF(NOT HAVE_TIMEGM ) SET( LIBEXIV2_SRC ${LIBEXIV2_SRC} localtime.c ) SET( EXIV2_SRC ${EXIV2_SRC} localtime.c ) SET( PATHTEST_SRC ${PATHTEST_SRC} localtime.c ) ENDIF( NOT HAVE_TIMEGM ) The patch add localtime.c as well to compile libexiv2, so some tunes need to be done to complete the job. [Exiv2-0.27-mingw-timegm.patch.txt](https://github.com/Exiv2/exiv2/files/2467357/Exiv2-0.27-mingw-timegm.patch.txt) --- end body --- comments : --- start comments --- --- end comments ---

The procedure to run the test suite is documented in README.md in the "RC1" branch. It works for Cygwin and I've documented a successful approach for Visual Studio Builds (ugly, however it works). I'm running the suite with this command: ``` 564 rmills@rmillsmm-w7:~/gnu/github/exiv2/exiv2/build $ export EXIV2EXT=.exe ; export EXIV2BINDIR=${PWD}/bin ; make tests ``` Using MinGW python and MinGW Bash (and build) results in: ``` Traceback (most recent call last): File "runner.py", line 48, in <module> system_tests.configure_suite(conf_file) File "/home/rmills/gnu/github/exiv2/exiv2/tests/system_tests.py", line 178, in configure_suite rel=rel_path) ValueError: Path replacement for exiv2: /home/rmills/gnu/github/exiv2/exiv2/tests/C:/msys64/home/rmills/gnu/github/exiv2/exiv2/build/bin/exiv2 does not exist (was expanded from C:/msys64/home/rmills/gnu/github/exiv2/exiv2/build/bin/exiv2) make[5]: *** [Makefile:115: newtests] Error 1 ... ``` Using DOS python and MinGW bash (and build) results in: ``` Traceback (most recent call last): File "runner.py", line 48, in <module> system_tests.configure_suite(conf_file) File "/home/rmills/gnu/github/exiv2/exiv2/tests/system_tests.py", line 178, in configure_suite rel=rel_path) ValueError: Path replacement for exiv2: /home/rmills/gnu/github/exiv2/exiv2/tests/C:/msys64/home/rmills/gnu/github/exiv2/exiv2/build/bin/exiv2 does not exist (was expanded from C:/msys64/home/rmills/gnu/github/exiv2/exiv2/build/bin/exiv2) make[5]: *** [Makefile:115: newtests] Error 1 ... $ ``` There's something not correct about how the suite is determining the location of exiv2.exe He's correct that `/home/rmills/gnu/github/exiv2/exiv2/tests/C:/msys64/home/rmills/gnu/github/exiv2/exiv2/build/bin/exiv2` and `/home/rmills/gnu/github/exiv2/exiv2/tests/C:/msys64/home/rmills/gnu/github/exiv2/exiv2/build/bin/exiv2` do not exist. However why is he searching there? One other little _**nit**_. The environment string EXIV2_EXT doesn't feel necessary. If the suite finds a file exiv2.exe, he knows EXIV2_EXT=='.exe'. If he finds an executable file exiv2, he knows EXIV2_EXT='' @D4N I'm happy to run tests on Windows while you make suggestions about changing the python. I'm sure we'll fix this in a few minutes together on Slack.

Running the Test Suite on MinGW builds

https://api.github.com/repos/Exiv2/exiv2/issues/469/comments

https://github.com/Exiv2/exiv2/issues/469

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : Running the Test Suite on MinGW builds Issue date: --- start body --- The procedure to run the test suite is documented in README.md in the "RC1" branch. It works for Cygwin and I've documented a successful approach for Visual Studio Builds (ugly, however it works). I'm running the suite with this command: ``` 564 rmills@rmillsmm-w7:~/gnu/github/exiv2/exiv2/build $ export EXIV2EXT=.exe ; export EXIV2BINDIR=${PWD}/bin ; make tests ``` Using MinGW python and MinGW Bash (and build) results in: ``` Traceback (most recent call last): File "runner.py", line 48, in <module> system_tests.configure_suite(conf_file) File "/home/rmills/gnu/github/exiv2/exiv2/tests/system_tests.py", line 178, in configure_suite rel=rel_path) ValueError: Path replacement for exiv2: /home/rmills/gnu/github/exiv2/exiv2/tests/C:/msys64/home/rmills/gnu/github/exiv2/exiv2/build/bin/exiv2 does not exist (was expanded from C:/msys64/home/rmills/gnu/github/exiv2/exiv2/build/bin/exiv2) make[5]: *** [Makefile:115: newtests] Error 1 ... ``` Using DOS python and MinGW bash (and build) results in: ``` Traceback (most recent call last): File "runner.py", line 48, in <module> system_tests.configure_suite(conf_file) File "/home/rmills/gnu/github/exiv2/exiv2/tests/system_tests.py", line 178, in configure_suite rel=rel_path) ValueError: Path replacement for exiv2: /home/rmills/gnu/github/exiv2/exiv2/tests/C:/msys64/home/rmills/gnu/github/exiv2/exiv2/build/bin/exiv2 does not exist (was expanded from C:/msys64/home/rmills/gnu/github/exiv2/exiv2/build/bin/exiv2) make[5]: *** [Makefile:115: newtests] Error 1 ... $ ``` There's something not correct about how the suite is determining the location of exiv2.exe He's correct that `/home/rmills/gnu/github/exiv2/exiv2/tests/C:/msys64/home/rmills/gnu/github/exiv2/exiv2/build/bin/exiv2` and `/home/rmills/gnu/github/exiv2/exiv2/tests/C:/msys64/home/rmills/gnu/github/exiv2/exiv2/build/bin/exiv2` do not exist. However why is he searching there? One other little _**nit**_. The environment string EXIV2_EXT doesn't feel necessary. If the suite finds a file exiv2.exe, he knows EXIV2_EXT=='.exe'. If he finds an executable file exiv2, he knows EXIV2_EXT='' @D4N I'm happy to run tests on Windows while you make suggestions about changing the python. I'm sure we'll fix this in a few minutes together on Slack. --- end body --- comments : --- start comments --- --- end comments ---

``` [gilles@localhost exiv2 (master)]$ ./bootstrap.mxe /mnt/data/GIT/6.x/project/bundles/mxe/build.win64 MXE target : 64 bits shared Installing to /mnt/data/GIT/6.x/project/bundles/mxe/build.win64 for target x86_64-w64-mingw32.shared with build mode RelWithDebInfo and configure options == Using MXE wrapper: /mnt/data/GIT/6.x/project/bundles/mxe/build.win64/usr/bin/x86_64-w64-mingw32.shared-cmake - cmake version 3.10.0 - warnings for unused CMAKE_POLICY_DEFAULT variables can be ignored == Using MXE toolchain: /mnt/data/GIT/6.x/project/bundles/mxe/build.win64/usr/x86_64-w64-mingw32.shared/share/cmake/mxe-conf.cmake == Using MXE runresult: /mnt/data/GIT/6.x/project/bundles/mxe/build.win64/usr/share/cmake/modules/TryRunResults.cmake loading initial cache file /mnt/data/GIT/6.x/project/bundles/mxe/build.win64/usr/share/cmake/modules/TryRunResults.cmake -- The CXX compiler identification is GNU 5.5.0 -- The C compiler identification is GNU 5.5.0 -- Check for working CXX compiler: /mnt/data/GIT/6.x/project/bundles/mxe/build.win64/usr/bin/x86_64-w64-mingw32.shared-g++ -- Check for working CXX compiler: /mnt/data/GIT/6.x/project/bundles/mxe/build.win64/usr/bin/x86_64-w64-mingw32.shared-g++ -- works -- Detecting CXX compiler ABI info -- Detecting CXX compiler ABI info - done -- Detecting CXX compile features -- Detecting CXX compile features - done -- Check for working C compiler: /mnt/data/GIT/6.x/project/bundles/mxe/build.win64/usr/bin/x86_64-w64-mingw32.shared-gcc -- Check for working C compiler: /mnt/data/GIT/6.x/project/bundles/mxe/build.win64/usr/bin/x86_64-w64-mingw32.shared-gcc -- works -- Detecting C compiler ABI info -- Detecting C compiler ABI info - done -- Detecting C compile features -- Detecting C compile features - done -- Looking for pthread.h -- Looking for pthread.h - found -- Looking for pthread_create -- Looking for pthread_create - not found -- Looking for pthread_create in pthreads -- Looking for pthread_create in pthreads - not found -- Looking for pthread_create in pthread -- Looking for pthread_create in pthread - found -- Found Threads: TRUE -- Found Regex: /mnt/data/GIT/6.x/project/bundles/mxe/build.win64/usr/x86_64-w64-mingw32.shared/include -- Found ZLIB: /mnt/data/GIT/6.x/project/bundles/mxe/build.win64/usr/x86_64-w64-mingw32.shared/lib/libz.dll.a (found version "1.2.11") -- Found EXPAT: /mnt/data/GIT/6.x/project/bundles/mxe/build.win64/usr/x86_64-w64-mingw32.shared/lib/libexpat.dll.a (found version "2.2.6") -- Found Intl: /mnt/data/GIT/6.x/project/bundles/mxe/build.win64/usr/x86_64-w64-mingw32.shared/include -- Found Iconv: /mnt/data/GIT/6.x/project/bundles/mxe/build.win64/usr/x86_64-w64-mingw32.shared/lib/libiconv.dll.a -- ICONV_INCLUDE_DIR : /mnt/data/GIT/6.x/project/bundles/mxe/build.win64/usr/x86_64-w64-mingw32.shared/include -- ICONV_LIBRARIES : /mnt/data/GIT/6.x/project/bundles/mxe/build.win64/usr/x86_64-w64-mingw32.shared/lib/libiconv.dll.a -- Looking for gmtime_r -- Looking for gmtime_r - not found -- Looking for mmap -- Looking for mmap - not found -- Looking for munmap -- Looking for munmap - not found -- Looking for strerror_r -- Looking for strerror_r - not found -- Looking for libintl.h -- Looking for libintl.h - found -- Looking for unistd.h -- Looking for unistd.h - found -- Looking for memory.h -- Looking for memory.h - found -- Looking for process.h -- Looking for process.h - found -- Looking for stdbool.h -- Looking for stdbool.h - found -- Looking for stdint.h -- Looking for stdint.h - found -- Looking for strings.h -- Looking for strings.h - found -- Looking for sys/mman.h -- Looking for sys/mman.h - found -- Looking for sys/stat.h -- Looking for sys/stat.h - found -- Looking for sys/types.h -- Looking for sys/types.h - found -- Looking for regex.h -- Looking for regex.h - found -- Looking for inttypes.h -- Looking for inttypes.h - found -- Performing Test COMPILER_HAS_DEPRECATED_ATTR -- Performing Test COMPILER_HAS_DEPRECATED_ATTR - Success -- Install prefix: /mnt/data/GIT/6.x/project/bundles/mxe/build.win64/usr/x86_64-w64-mingw32.shared -- ------------------------------------------------------------------ -- CMake Generator: Unix Makefiles -- Compiler info: GNU (/mnt/data/GIT/6.x/project/bundles/mxe/build.win64/usr/bin/x86_64-w64-mingw32.shared-g++) ; version: 5.5.0 -- --- Compiler flags --- -- General: -Wall -Wcast-align -Wpointer-arith -Wformat-security -Wmissing-format-attribute -Woverloaded-virtual -W -std=gnu++98 -- Extra: -- Debug: -g3 -gstrict-dwarf -O0 -- Release: -O3 -DNDEBUG -- RelWithDebInfo: -O2 -g -DNDEBUG -- MinSizeRel: -Os -DNDEBUG -- --- Linker flags --- -- General: -- Debug: -- Release: -- RelWithDebInfo: -- MinSizeRel: -- -- Compiler Options -- Warnings as errors: NO -- Use extra compiler warning flags: NO -- -- Compiler info: GNU (/mnt/data/GIT/6.x/project/bundles/mxe/build.win64/usr/bin/x86_64-w64-mingw32.shared-g++) ; version: 5.5.0 -- ------------------------------------------------------------------ -- Building shared library: YES -- Building PNG support: YES -- XMP metadata support: YES -- Native language support: YES -- Conversion of Windows XP tags: YES -- Nikon lens database: YES -- Building video support: NO -- Building webready support: NO -- Dynamic runtime override: NO -- Unicode paths (wstring): YES -- Building exiv2 command: NO -- Building samples: NO -- Building PO files: NO -- Building unit tests: NO -- Building doc: NO -- Building with coverage flags: NO -- Using ccache: NO -- ------------------------------------------------------------------ -- Configuring done -- Generating done -- Build files have been written to: /mnt/data/GIT/exiv2/build [gilles@localhost exiv2 (master)]$ [gilles@localhost exiv2 (master)]$ cd build [gilles@localhost build (master)]$ make Scanning dependencies of target xmp [ 1%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/ExpatAdapter.cpp.obj [ 2%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/MD5.cpp.obj [ 3%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/ParseRDF.cpp.obj [ 4%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/UnicodeConversions.cpp.obj [ 6%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/WXMPIterator.cpp.obj [ 7%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/WXMPMeta.cpp.obj [ 8%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/WXMPUtils.cpp.obj [ 9%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/XML_Node.cpp.obj [ 10%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/XMPCore_Impl.cpp.obj [ 12%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/XMPIterator.cpp.obj [ 13%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/XMPMeta-GetSet.cpp.obj [ 14%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/XMPMeta-Parse.cpp.obj [ 15%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/XMPMeta-Serialize.cpp.obj [ 16%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/XMPMeta.cpp.obj [ 18%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/XMPUtils-FileInfo.cpp.obj [ 19%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/XMPUtils.cpp.obj [ 20%] Linking CXX static library ../lib/libxmp.a [ 20%] Built target xmp Scanning dependencies of target exiv2lib_int [ 21%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/canonmn_int.cpp.obj [ 22%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/casiomn_int.cpp.obj [ 24%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/cr2header_int.cpp.obj [ 25%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/crwimage_int.cpp.obj [ 26%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/fujimn_int.cpp.obj [ 27%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/makernote_int.cpp.obj [ 28%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/minoltamn_int.cpp.obj [ 30%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/nikonmn_int.cpp.obj [ 31%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/olympusmn_int.cpp.obj [ 32%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/orfimage_int.cpp.obj [ 33%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/panasonicmn_int.cpp.obj [ 34%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/pentaxmn_int.cpp.obj [ 36%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/rw2image_int.cpp.obj [ 37%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/samsungmn_int.cpp.obj [ 38%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/sigmamn_int.cpp.obj [ 39%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/sonymn_int.cpp.obj [ 40%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/tags_int.cpp.obj [ 42%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/tiffcomposite_int.cpp.obj [ 43%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/tiffimage_int.cpp.obj [ 44%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/tiffvisitor_int.cpp.obj [ 45%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/helper_functions.cpp.obj [ 46%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/pngchunk_int.cpp.obj [ 46%] Built target exiv2lib_int Scanning dependencies of target exiv2lib [ 48%] Building CXX object src/CMakeFiles/exiv2lib.dir/asfvideo.cpp.obj [ 49%] Building CXX object src/CMakeFiles/exiv2lib.dir/basicio.cpp.obj [ 50%] Building CXX object src/CMakeFiles/exiv2lib.dir/bigtiffimage.cpp.obj [ 51%] Building CXX object src/CMakeFiles/exiv2lib.dir/bmpimage.cpp.obj [ 53%] Building CXX object src/CMakeFiles/exiv2lib.dir/convert.cpp.obj /mnt/data/GIT/exiv2/src/convert.cpp:1511:10: warning: 'bool {anonymous}::convertStringCharsetWindows(std::__cxx11::string&, const char*, const char*)' defined but not used [-Wunused-function] bool convertStringCharsetWindows(std::string& str, const char* from, const char* to) ^ [ 54%] Building CXX object src/CMakeFiles/exiv2lib.dir/cr2image.cpp.obj [ 55%] Building CXX object src/CMakeFiles/exiv2lib.dir/crwimage.cpp.obj [ 56%] Building CXX object src/CMakeFiles/exiv2lib.dir/datasets.cpp.obj [ 57%] Building CXX object src/CMakeFiles/exiv2lib.dir/easyaccess.cpp.obj [ 59%] Building CXX object src/CMakeFiles/exiv2lib.dir/epsimage.cpp.obj [ 60%] Building CXX object src/CMakeFiles/exiv2lib.dir/error.cpp.obj [ 61%] Building CXX object src/CMakeFiles/exiv2lib.dir/exif.cpp.obj [ 62%] Building CXX object src/CMakeFiles/exiv2lib.dir/futils.cpp.obj [ 63%] Building CXX object src/CMakeFiles/exiv2lib.dir/gifimage.cpp.obj [ 65%] Building CXX object src/CMakeFiles/exiv2lib.dir/http.cpp.obj [ 66%] Building CXX object src/CMakeFiles/exiv2lib.dir/image.cpp.obj /mnt/data/GIT/exiv2/src/image.cpp:861:77: warning: unused parameter 'useCurl' [-Wunused-parameter] BasicIo::AutoPtr ImageFactory::createIo(const std::wstring& wpath, bool useCurl) ^ [ 67%] Building CXX object src/CMakeFiles/exiv2lib.dir/ini.cpp.obj [ 68%] Building CXX object src/CMakeFiles/exiv2lib.dir/iptc.cpp.obj [ 69%] Building CXX object src/CMakeFiles/exiv2lib.dir/jp2image.cpp.obj [ 71%] Building CXX object src/CMakeFiles/exiv2lib.dir/jpgimage.cpp.obj [ 72%] Building CXX object src/CMakeFiles/exiv2lib.dir/matroskavideo.cpp.obj [ 73%] Building CXX object src/CMakeFiles/exiv2lib.dir/metadatum.cpp.obj [ 74%] Building CXX object src/CMakeFiles/exiv2lib.dir/mrwimage.cpp.obj [ 75%] Building CXX object src/CMakeFiles/exiv2lib.dir/orfimage.cpp.obj [ 77%] Building CXX object src/CMakeFiles/exiv2lib.dir/pgfimage.cpp.obj [ 78%] Building CXX object src/CMakeFiles/exiv2lib.dir/preview.cpp.obj [ 79%] Building CXX object src/CMakeFiles/exiv2lib.dir/properties.cpp.obj [ 80%] Building CXX object src/CMakeFiles/exiv2lib.dir/psdimage.cpp.obj [ 81%] Building CXX object src/CMakeFiles/exiv2lib.dir/quicktimevideo.cpp.obj [ 83%] Building CXX object src/CMakeFiles/exiv2lib.dir/rafimage.cpp.obj [ 84%] Building CXX object src/CMakeFiles/exiv2lib.dir/riffvideo.cpp.obj [ 85%] Building CXX object src/CMakeFiles/exiv2lib.dir/rw2image.cpp.obj [ 86%] Building CXX object src/CMakeFiles/exiv2lib.dir/tags.cpp.obj [ 87%] Building CXX object src/CMakeFiles/exiv2lib.dir/tgaimage.cpp.obj [ 89%] Building CXX object src/CMakeFiles/exiv2lib.dir/tiffimage.cpp.obj [ 90%] Building CXX object src/CMakeFiles/exiv2lib.dir/types.cpp.obj [ 91%] Building CXX object src/CMakeFiles/exiv2lib.dir/utilsvideo.cpp.obj [ 92%] Building CXX object src/CMakeFiles/exiv2lib.dir/value.cpp.obj [ 93%] Building CXX object src/CMakeFiles/exiv2lib.dir/version.cpp.obj [ 95%] Building CXX object src/CMakeFiles/exiv2lib.dir/webpimage.cpp.obj [ 96%] Building CXX object src/CMakeFiles/exiv2lib.dir/xmp.cpp.obj [ 97%] Building CXX object src/CMakeFiles/exiv2lib.dir/xmpsidecar.cpp.obj [ 98%] Building CXX object src/CMakeFiles/exiv2lib.dir/pngimage.cpp.obj [100%] Linking CXX shared library ../bin/libexiv2.dll [100%] Built target exiv2lib [gilles@localhost build (master)]$ ``` My patch applied is very simple : ``` patch diff --git a/include/exiv2/config.h b/include/exiv2/config.h index 41082d83..7b28f15c 100644 --- a/include/exiv2/config.h +++ b/include/exiv2/config.h @@ -67,10 +67,10 @@ typedef int pid_t; # ifndef __MINGW__ # define __MINGW__ 1 # endif - +/* #ifdef EXV_UNICODE_PATH #error EXV_UNICODE_PATH is not supported for MinGW builds -#endif +#endif*/ #endif #ifndef __CYGWIN__ ``` Gilles Caulier

Exiv2 and MinGW : The library compile fine with Unicode paths support

https://api.github.com/repos/Exiv2/exiv2/issues/468/comments

https://github.com/Exiv2/exiv2/issues/468

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : Exiv2 and MinGW : The library compile fine with Unicode paths support Issue date: --- start body --- ``` [gilles@localhost exiv2 (master)]$ ./bootstrap.mxe /mnt/data/GIT/6.x/project/bundles/mxe/build.win64 MXE target : 64 bits shared Installing to /mnt/data/GIT/6.x/project/bundles/mxe/build.win64 for target x86_64-w64-mingw32.shared with build mode RelWithDebInfo and configure options == Using MXE wrapper: /mnt/data/GIT/6.x/project/bundles/mxe/build.win64/usr/bin/x86_64-w64-mingw32.shared-cmake - cmake version 3.10.0 - warnings for unused CMAKE_POLICY_DEFAULT variables can be ignored == Using MXE toolchain: /mnt/data/GIT/6.x/project/bundles/mxe/build.win64/usr/x86_64-w64-mingw32.shared/share/cmake/mxe-conf.cmake == Using MXE runresult: /mnt/data/GIT/6.x/project/bundles/mxe/build.win64/usr/share/cmake/modules/TryRunResults.cmake loading initial cache file /mnt/data/GIT/6.x/project/bundles/mxe/build.win64/usr/share/cmake/modules/TryRunResults.cmake -- The CXX compiler identification is GNU 5.5.0 -- The C compiler identification is GNU 5.5.0 -- Check for working CXX compiler: /mnt/data/GIT/6.x/project/bundles/mxe/build.win64/usr/bin/x86_64-w64-mingw32.shared-g++ -- Check for working CXX compiler: /mnt/data/GIT/6.x/project/bundles/mxe/build.win64/usr/bin/x86_64-w64-mingw32.shared-g++ -- works -- Detecting CXX compiler ABI info -- Detecting CXX compiler ABI info - done -- Detecting CXX compile features -- Detecting CXX compile features - done -- Check for working C compiler: /mnt/data/GIT/6.x/project/bundles/mxe/build.win64/usr/bin/x86_64-w64-mingw32.shared-gcc -- Check for working C compiler: /mnt/data/GIT/6.x/project/bundles/mxe/build.win64/usr/bin/x86_64-w64-mingw32.shared-gcc -- works -- Detecting C compiler ABI info -- Detecting C compiler ABI info - done -- Detecting C compile features -- Detecting C compile features - done -- Looking for pthread.h -- Looking for pthread.h - found -- Looking for pthread_create -- Looking for pthread_create - not found -- Looking for pthread_create in pthreads -- Looking for pthread_create in pthreads - not found -- Looking for pthread_create in pthread -- Looking for pthread_create in pthread - found -- Found Threads: TRUE -- Found Regex: /mnt/data/GIT/6.x/project/bundles/mxe/build.win64/usr/x86_64-w64-mingw32.shared/include -- Found ZLIB: /mnt/data/GIT/6.x/project/bundles/mxe/build.win64/usr/x86_64-w64-mingw32.shared/lib/libz.dll.a (found version "1.2.11") -- Found EXPAT: /mnt/data/GIT/6.x/project/bundles/mxe/build.win64/usr/x86_64-w64-mingw32.shared/lib/libexpat.dll.a (found version "2.2.6") -- Found Intl: /mnt/data/GIT/6.x/project/bundles/mxe/build.win64/usr/x86_64-w64-mingw32.shared/include -- Found Iconv: /mnt/data/GIT/6.x/project/bundles/mxe/build.win64/usr/x86_64-w64-mingw32.shared/lib/libiconv.dll.a -- ICONV_INCLUDE_DIR : /mnt/data/GIT/6.x/project/bundles/mxe/build.win64/usr/x86_64-w64-mingw32.shared/include -- ICONV_LIBRARIES : /mnt/data/GIT/6.x/project/bundles/mxe/build.win64/usr/x86_64-w64-mingw32.shared/lib/libiconv.dll.a -- Looking for gmtime_r -- Looking for gmtime_r - not found -- Looking for mmap -- Looking for mmap - not found -- Looking for munmap -- Looking for munmap - not found -- Looking for strerror_r -- Looking for strerror_r - not found -- Looking for libintl.h -- Looking for libintl.h - found -- Looking for unistd.h -- Looking for unistd.h - found -- Looking for memory.h -- Looking for memory.h - found -- Looking for process.h -- Looking for process.h - found -- Looking for stdbool.h -- Looking for stdbool.h - found -- Looking for stdint.h -- Looking for stdint.h - found -- Looking for strings.h -- Looking for strings.h - found -- Looking for sys/mman.h -- Looking for sys/mman.h - found -- Looking for sys/stat.h -- Looking for sys/stat.h - found -- Looking for sys/types.h -- Looking for sys/types.h - found -- Looking for regex.h -- Looking for regex.h - found -- Looking for inttypes.h -- Looking for inttypes.h - found -- Performing Test COMPILER_HAS_DEPRECATED_ATTR -- Performing Test COMPILER_HAS_DEPRECATED_ATTR - Success -- Install prefix: /mnt/data/GIT/6.x/project/bundles/mxe/build.win64/usr/x86_64-w64-mingw32.shared -- ------------------------------------------------------------------ -- CMake Generator: Unix Makefiles -- Compiler info: GNU (/mnt/data/GIT/6.x/project/bundles/mxe/build.win64/usr/bin/x86_64-w64-mingw32.shared-g++) ; version: 5.5.0 -- --- Compiler flags --- -- General: -Wall -Wcast-align -Wpointer-arith -Wformat-security -Wmissing-format-attribute -Woverloaded-virtual -W -std=gnu++98 -- Extra: -- Debug: -g3 -gstrict-dwarf -O0 -- Release: -O3 -DNDEBUG -- RelWithDebInfo: -O2 -g -DNDEBUG -- MinSizeRel: -Os -DNDEBUG -- --- Linker flags --- -- General: -- Debug: -- Release: -- RelWithDebInfo: -- MinSizeRel: -- -- Compiler Options -- Warnings as errors: NO -- Use extra compiler warning flags: NO -- -- Compiler info: GNU (/mnt/data/GIT/6.x/project/bundles/mxe/build.win64/usr/bin/x86_64-w64-mingw32.shared-g++) ; version: 5.5.0 -- ------------------------------------------------------------------ -- Building shared library: YES -- Building PNG support: YES -- XMP metadata support: YES -- Native language support: YES -- Conversion of Windows XP tags: YES -- Nikon lens database: YES -- Building video support: NO -- Building webready support: NO -- Dynamic runtime override: NO -- Unicode paths (wstring): YES -- Building exiv2 command: NO -- Building samples: NO -- Building PO files: NO -- Building unit tests: NO -- Building doc: NO -- Building with coverage flags: NO -- Using ccache: NO -- ------------------------------------------------------------------ -- Configuring done -- Generating done -- Build files have been written to: /mnt/data/GIT/exiv2/build [gilles@localhost exiv2 (master)]$ [gilles@localhost exiv2 (master)]$ cd build [gilles@localhost build (master)]$ make Scanning dependencies of target xmp [ 1%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/ExpatAdapter.cpp.obj [ 2%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/MD5.cpp.obj [ 3%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/ParseRDF.cpp.obj [ 4%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/UnicodeConversions.cpp.obj [ 6%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/WXMPIterator.cpp.obj [ 7%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/WXMPMeta.cpp.obj [ 8%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/WXMPUtils.cpp.obj [ 9%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/XML_Node.cpp.obj [ 10%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/XMPCore_Impl.cpp.obj [ 12%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/XMPIterator.cpp.obj [ 13%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/XMPMeta-GetSet.cpp.obj [ 14%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/XMPMeta-Parse.cpp.obj [ 15%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/XMPMeta-Serialize.cpp.obj [ 16%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/XMPMeta.cpp.obj [ 18%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/XMPUtils-FileInfo.cpp.obj [ 19%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/XMPUtils.cpp.obj [ 20%] Linking CXX static library ../lib/libxmp.a [ 20%] Built target xmp Scanning dependencies of target exiv2lib_int [ 21%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/canonmn_int.cpp.obj [ 22%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/casiomn_int.cpp.obj [ 24%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/cr2header_int.cpp.obj [ 25%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/crwimage_int.cpp.obj [ 26%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/fujimn_int.cpp.obj [ 27%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/makernote_int.cpp.obj [ 28%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/minoltamn_int.cpp.obj [ 30%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/nikonmn_int.cpp.obj [ 31%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/olympusmn_int.cpp.obj [ 32%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/orfimage_int.cpp.obj [ 33%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/panasonicmn_int.cpp.obj [ 34%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/pentaxmn_int.cpp.obj [ 36%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/rw2image_int.cpp.obj [ 37%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/samsungmn_int.cpp.obj [ 38%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/sigmamn_int.cpp.obj [ 39%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/sonymn_int.cpp.obj [ 40%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/tags_int.cpp.obj [ 42%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/tiffcomposite_int.cpp.obj [ 43%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/tiffimage_int.cpp.obj [ 44%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/tiffvisitor_int.cpp.obj [ 45%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/helper_functions.cpp.obj [ 46%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/pngchunk_int.cpp.obj [ 46%] Built target exiv2lib_int Scanning dependencies of target exiv2lib [ 48%] Building CXX object src/CMakeFiles/exiv2lib.dir/asfvideo.cpp.obj [ 49%] Building CXX object src/CMakeFiles/exiv2lib.dir/basicio.cpp.obj [ 50%] Building CXX object src/CMakeFiles/exiv2lib.dir/bigtiffimage.cpp.obj [ 51%] Building CXX object src/CMakeFiles/exiv2lib.dir/bmpimage.cpp.obj [ 53%] Building CXX object src/CMakeFiles/exiv2lib.dir/convert.cpp.obj /mnt/data/GIT/exiv2/src/convert.cpp:1511:10: warning: 'bool {anonymous}::convertStringCharsetWindows(std::__cxx11::string&, const char*, const char*)' defined but not used [-Wunused-function] bool convertStringCharsetWindows(std::string& str, const char* from, const char* to) ^ [ 54%] Building CXX object src/CMakeFiles/exiv2lib.dir/cr2image.cpp.obj [ 55%] Building CXX object src/CMakeFiles/exiv2lib.dir/crwimage.cpp.obj [ 56%] Building CXX object src/CMakeFiles/exiv2lib.dir/datasets.cpp.obj [ 57%] Building CXX object src/CMakeFiles/exiv2lib.dir/easyaccess.cpp.obj [ 59%] Building CXX object src/CMakeFiles/exiv2lib.dir/epsimage.cpp.obj [ 60%] Building CXX object src/CMakeFiles/exiv2lib.dir/error.cpp.obj [ 61%] Building CXX object src/CMakeFiles/exiv2lib.dir/exif.cpp.obj [ 62%] Building CXX object src/CMakeFiles/exiv2lib.dir/futils.cpp.obj [ 63%] Building CXX object src/CMakeFiles/exiv2lib.dir/gifimage.cpp.obj [ 65%] Building CXX object src/CMakeFiles/exiv2lib.dir/http.cpp.obj [ 66%] Building CXX object src/CMakeFiles/exiv2lib.dir/image.cpp.obj /mnt/data/GIT/exiv2/src/image.cpp:861:77: warning: unused parameter 'useCurl' [-Wunused-parameter] BasicIo::AutoPtr ImageFactory::createIo(const std::wstring& wpath, bool useCurl) ^ [ 67%] Building CXX object src/CMakeFiles/exiv2lib.dir/ini.cpp.obj [ 68%] Building CXX object src/CMakeFiles/exiv2lib.dir/iptc.cpp.obj [ 69%] Building CXX object src/CMakeFiles/exiv2lib.dir/jp2image.cpp.obj [ 71%] Building CXX object src/CMakeFiles/exiv2lib.dir/jpgimage.cpp.obj [ 72%] Building CXX object src/CMakeFiles/exiv2lib.dir/matroskavideo.cpp.obj [ 73%] Building CXX object src/CMakeFiles/exiv2lib.dir/metadatum.cpp.obj [ 74%] Building CXX object src/CMakeFiles/exiv2lib.dir/mrwimage.cpp.obj [ 75%] Building CXX object src/CMakeFiles/exiv2lib.dir/orfimage.cpp.obj [ 77%] Building CXX object src/CMakeFiles/exiv2lib.dir/pgfimage.cpp.obj [ 78%] Building CXX object src/CMakeFiles/exiv2lib.dir/preview.cpp.obj [ 79%] Building CXX object src/CMakeFiles/exiv2lib.dir/properties.cpp.obj [ 80%] Building CXX object src/CMakeFiles/exiv2lib.dir/psdimage.cpp.obj [ 81%] Building CXX object src/CMakeFiles/exiv2lib.dir/quicktimevideo.cpp.obj [ 83%] Building CXX object src/CMakeFiles/exiv2lib.dir/rafimage.cpp.obj [ 84%] Building CXX object src/CMakeFiles/exiv2lib.dir/riffvideo.cpp.obj [ 85%] Building CXX object src/CMakeFiles/exiv2lib.dir/rw2image.cpp.obj [ 86%] Building CXX object src/CMakeFiles/exiv2lib.dir/tags.cpp.obj [ 87%] Building CXX object src/CMakeFiles/exiv2lib.dir/tgaimage.cpp.obj [ 89%] Building CXX object src/CMakeFiles/exiv2lib.dir/tiffimage.cpp.obj [ 90%] Building CXX object src/CMakeFiles/exiv2lib.dir/types.cpp.obj [ 91%] Building CXX object src/CMakeFiles/exiv2lib.dir/utilsvideo.cpp.obj [ 92%] Building CXX object src/CMakeFiles/exiv2lib.dir/value.cpp.obj [ 93%] Building CXX object src/CMakeFiles/exiv2lib.dir/version.cpp.obj [ 95%] Building CXX object src/CMakeFiles/exiv2lib.dir/webpimage.cpp.obj [ 96%] Building CXX object src/CMakeFiles/exiv2lib.dir/xmp.cpp.obj [ 97%] Building CXX object src/CMakeFiles/exiv2lib.dir/xmpsidecar.cpp.obj [ 98%] Building CXX object src/CMakeFiles/exiv2lib.dir/pngimage.cpp.obj [100%] Linking CXX shared library ../bin/libexiv2.dll [100%] Built target exiv2lib [gilles@localhost build (master)]$ ``` My patch applied is very simple : ``` patch diff --git a/include/exiv2/config.h b/include/exiv2/config.h index 41082d83..7b28f15c 100644 --- a/include/exiv2/config.h +++ b/include/exiv2/config.h @@ -67,10 +67,10 @@ typedef int pid_t; # ifndef __MINGW__ # define __MINGW__ 1 # endif - +/* #ifdef EXV_UNICODE_PATH #error EXV_UNICODE_PATH is not supported for MinGW builds -#endif +#endif*/ #endif #ifndef __CYGWIN__ ``` Gilles Caulier --- end body --- comments : --- start comments --- --- end comments ---

Run CMake with Visual Studio generator (tested with Visual Studio 2013 and Visual Studio 2017). Try running tests: ``` >ctest -C Release Test project E:/dev/packages/exiv2/0.27-1/aligntech/stable/build/b76c459f2beaee629c93c8b89349fd8a04f10ba5 No tests were found!!! >cmake --build . --config Release --target RUN_TESTS Microsoft (R) Build Engine version 12.0.31101.0 [Microsoft .NET Framework, version 4.0.30319.42000] Copyright (C) Microsoft Corporation. All rights reserved. MSBUILD : error MSB1009: Project file does not exist. Switch: RUN_TESTS.vcxproj ``` Tested with this revision: https://github.com/Exiv2/exiv2/commit/11e66c6c9eceeddd2263c3591af6317cbd05c1b6

CMake tests (ctest) do not work with Visual Studio (on Windows)

https://api.github.com/repos/Exiv2/exiv2/issues/467/comments

https://github.com/Exiv2/exiv2/issues/467

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : CMake tests (ctest) do not work with Visual Studio (on Windows) Issue date: --- start body --- Run CMake with Visual Studio generator (tested with Visual Studio 2013 and Visual Studio 2017). Try running tests: ``` >ctest -C Release Test project E:/dev/packages/exiv2/0.27-1/aligntech/stable/build/b76c459f2beaee629c93c8b89349fd8a04f10ba5 No tests were found!!! >cmake --build . --config Release --target RUN_TESTS Microsoft (R) Build Engine version 12.0.31101.0 [Microsoft .NET Framework, version 4.0.30319.42000] Copyright (C) Microsoft Corporation. All rights reserved. MSBUILD : error MSB1009: Project file does not exist. Switch: RUN_TESTS.vcxproj ``` Tested with this revision: https://github.com/Exiv2/exiv2/commit/11e66c6c9eceeddd2263c3591af6317cbd05c1b6 --- end body --- comments : --- start comments --- --- end comments ---

As the whole project is based on Cmake, no need to use PKG_Config with cmake to configure. PKG_config is and older Unix based tool, and With Mingw the introduce complexity and problem. No need to re-invent the wheel. Cmake can find expected dependency files as well. The patch given in this file fix the Regex lib detection and use a Find.cmake script to do the job in place. Now, the compilation under MinGw become more portable and generic (and work as expected...) [regex-cmake-detect.patch.txt](https://github.com/Exiv2/exiv2/files/2460876/regex-cmake-detect.patch.txt) Gilles Caulier

Exiv2 0.27 : Regex dependency : do not use PKG_CONFIG with Mingw but cmake script instead

https://api.github.com/repos/Exiv2/exiv2/issues/466/comments

https://github.com/Exiv2/exiv2/issues/466

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : Exiv2 0.27 : Regex dependency : do not use PKG_CONFIG with Mingw but cmake script instead Issue date: --- start body --- As the whole project is based on Cmake, no need to use PKG_Config with cmake to configure. PKG_config is and older Unix based tool, and With Mingw the introduce complexity and problem. No need to re-invent the wheel. Cmake can find expected dependency files as well. The patch given in this file fix the Regex lib detection and use a Find.cmake script to do the job in place. Now, the compilation under MinGw become more portable and generic (and work as expected...) [regex-cmake-detect.patch.txt](https://github.com/Exiv2/exiv2/files/2460876/regex-cmake-detect.patch.txt) Gilles Caulier --- end body --- comments : --- start comments --- --- end comments ---

With new API 0.27, the ExifTags::groupList() disappears. In digiKam we use this list to build a tags view categorized by group name. Code is here for example : https://cgit.kde.org/digikam.git/tree/core/libs/dmetadata/metaengine_exif.cpp#n1077 https://cgit.kde.org/digikam.git/tree/core/libs/dmetadata/metaengine_exif.cpp#n1138 Best

Exiv 0.27 : broken API with ExifTags::groupList()

https://api.github.com/repos/Exiv2/exiv2/issues/465/comments

https://github.com/Exiv2/exiv2/issues/465

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : Exiv 0.27 : broken API with ExifTags::groupList() Issue date: --- start body --- With new API 0.27, the ExifTags::groupList() disappears. In digiKam we use this list to build a tags view categorized by group name. Code is here for example : https://cgit.kde.org/digikam.git/tree/core/libs/dmetadata/metaengine_exif.cpp#n1077 https://cgit.kde.org/digikam.git/tree/core/libs/dmetadata/metaengine_exif.cpp#n1138 Best --- end body --- comments : --- start comments --- --- end comments ---

A first try to use Exiv2 from trunk with digiKam give this compilation result : ``` [gilles@localhost exiv2 (master)]$ ./bootstrap-exiv2.sh -- The CXX compiler identification is GNU 5.5.0 -- The C compiler identification is GNU 5.5.0 -- Check for working CXX compiler: /bin/c++ -- Check for working CXX compiler: /bin/c++ -- works -- Detecting CXX compiler ABI info -- Detecting CXX compiler ABI info - done -- Detecting CXX compile features -- Detecting CXX compile features - done -- Check for working C compiler: /bin/cc -- Check for working C compiler: /bin/cc -- works -- Detecting C compiler ABI info -- Detecting C compiler ABI info - done -- Detecting C compile features -- Detecting C compile features - done -- Looking for pthread.h -- Looking for pthread.h - found -- Looking for pthread_create -- Looking for pthread_create - not found -- Looking for pthread_create in pthreads -- Looking for pthread_create in pthreads - not found -- Looking for pthread_create in pthread -- Looking for pthread_create in pthread - found -- Found Threads: TRUE -- Found ZLIB: /usr/lib64/libz.so (found version "1.2.11") -- Found EXPAT: /usr/lib64/libexpat.so (found version "2.2.1") -- Found Intl: /usr/include -- Performing Test Iconv_IS_BUILT_IN -- Performing Test Iconv_IS_BUILT_IN - Success -- Found Iconv: /usr/lib64/libc.so -- ICONV_INCLUDE_DIR : -- ICONV_LIBRARIES : /usr/lib64/libc.so -- Looking for gmtime_r -- Looking for gmtime_r - found -- Looking for mmap -- Looking for mmap - found -- Looking for munmap -- Looking for munmap - found -- Looking for strerror_r -- Looking for strerror_r - found -- Looking for libintl.h -- Looking for libintl.h - found -- Looking for unistd.h -- Looking for unistd.h - found -- Looking for memory.h -- Looking for memory.h - found -- Looking for process.h -- Looking for process.h - not found -- Looking for stdbool.h -- Looking for stdbool.h - found -- Looking for stdint.h -- Looking for stdint.h - found -- Looking for strings.h -- Looking for strings.h - found -- Looking for sys/mman.h -- Looking for sys/mman.h - found -- Looking for sys/stat.h -- Looking for sys/stat.h - found -- Looking for sys/types.h -- Looking for sys/types.h - found -- Looking for regex.h -- Looking for regex.h - found -- Looking for inttypes.h -- Looking for inttypes.h - found -- Performing Test COMPILER_HAS_HIDDEN_VISIBILITY -- Performing Test COMPILER_HAS_HIDDEN_VISIBILITY - Success -- Performing Test COMPILER_HAS_HIDDEN_INLINE_VISIBILITY -- Performing Test COMPILER_HAS_HIDDEN_INLINE_VISIBILITY - Success -- Performing Test COMPILER_HAS_DEPRECATED_ATTR -- Performing Test COMPILER_HAS_DEPRECATED_ATTR - Success -- Install prefix: /usr -- ------------------------------------------------------------------ -- CMake Generator: Unix Makefiles -- Compiler info: GNU (/bin/c++) ; version: 5.5.0 -- --- Compiler flags --- -- General: -Wall -Wcast-align -Wpointer-arith -Wformat-security -Wmissing-format-attribute -Woverloaded-virtual -W -std=gnu++98 -- Extra: -- Debug: -g3 -gstrict-dwarf -O0 -- Release: -O3 -DNDEBUG -- RelWithDebInfo: -O2 -g -DNDEBUG -- MinSizeRel: -Os -DNDEBUG -- --- Linker flags --- -- General: -- Debug: -- Release: -- RelWithDebInfo: -- MinSizeRel: -- -- Compiler Options -- Warnings as errors: NO -- Use extra compiler warning flags: NO -- -- Compiler info: GNU (/bin/c++) ; version: 5.5.0 -- ------------------------------------------------------------------ -- Building shared library: YES -- Building PNG support: YES -- XMP metadata support: YES -- Native language support: YES -- Conversion of Windows XP tags: YES -- Nikon lens database: YES -- Building video support: NO -- Building webready support: NO -- Building exiv2 command: YES -- Building samples: YES -- Building PO files: NO -- Building unit tests: NO -- Building doc: NO -- Building with coverage flags: NO -- Using ccache: NO -- ------------------------------------------------------------------ -- Configuring done -- Generating done -- Build files have been written to: /mnt/data/GIT/exiv2/build [gilles@localhost exiv2 (master)]$ [gilles@localhost build (master)]$ make Scanning dependencies of target xmp [ 1%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/ExpatAdapter.cpp.o [ 2%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/MD5.cpp.o [ 2%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/ParseRDF.cpp.o [ 3%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/UnicodeConversions.cpp.o [ 3%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/WXMPIterator.cpp.o [ 4%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/WXMPMeta.cpp.o [ 5%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/WXMPUtils.cpp.o [ 5%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/XML_Node.cpp.o [ 6%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/XMPCore_Impl.cpp.o [ 6%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/XMPIterator.cpp.o [ 7%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/XMPMeta-GetSet.cpp.o [ 8%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/XMPMeta-Parse.cpp.o [ 8%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/XMPMeta-Serialize.cpp.o [ 9%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/XMPMeta.cpp.o [ 9%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/XMPUtils-FileInfo.cpp.o [ 10%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/XMPUtils.cpp.o [ 11%] Linking CXX static library ../lib/libxmp.a [ 11%] Built target xmp Scanning dependencies of target exiv2lib_int [ 12%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/canonmn_int.cpp.o [ 13%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/casiomn_int.cpp.o [ 13%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/cr2header_int.cpp.o [ 14%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/crwimage_int.cpp.o [ 14%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/fujimn_int.cpp.o [ 15%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/makernote_int.cpp.o [ 16%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/minoltamn_int.cpp.o [ 16%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/nikonmn_int.cpp.o [ 17%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/olympusmn_int.cpp.o [ 17%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/orfimage_int.cpp.o [ 18%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/panasonicmn_int.cpp.o [ 19%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/pentaxmn_int.cpp.o [ 19%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/rw2image_int.cpp.o [ 20%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/samsungmn_int.cpp.o [ 20%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/sigmamn_int.cpp.o [ 21%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/sonymn_int.cpp.o [ 22%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/tags_int.cpp.o [ 22%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/tiffcomposite_int.cpp.o [ 23%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/tiffimage_int.cpp.o [ 23%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/tiffvisitor_int.cpp.o [ 24%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/helper_functions.cpp.o [ 25%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/pngchunk_int.cpp.o [ 25%] Built target exiv2lib_int Scanning dependencies of target exiv2lib [ 26%] Building CXX object src/CMakeFiles/exiv2lib.dir/asfvideo.cpp.o [ 26%] Building CXX object src/CMakeFiles/exiv2lib.dir/basicio.cpp.o In file included from /mnt/data/GIT/exiv2/src/basicio.cpp:31:0: /mnt/data/GIT/exiv2/include/exiv2/http.hpp:23:23: erreur fatale : curl/curl.h : Aucun fichier ou dossier de ce type compilation terminée. src/CMakeFiles/exiv2lib.dir/build.make:86 : la recette pour la cible « src/CMakeFiles/exiv2lib.dir/basicio.cpp.o » a échouée make[2]: *** [src/CMakeFiles/exiv2lib.dir/basicio.cpp.o] Erreur 1 CMakeFiles/Makefile2:283 : la recette pour la cible « src/CMakeFiles/exiv2lib.dir/all » a échouée make[1]: *** [src/CMakeFiles/exiv2lib.dir/all] Erreur 2 Makefile:149 : la recette pour la cible « all » a échouée make: *** [all] Erreur 2 ``` libcurl-devel package is not installed. So the error appear on my computer. The expected result must be a complete compilation in this case. Best

libcurl dependency required even if cmake configuration option is turned OFF

https://api.github.com/repos/Exiv2/exiv2/issues/464/comments

https://github.com/Exiv2/exiv2/issues/464

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : libcurl dependency required even if cmake configuration option is turned OFF Issue date: --- start body --- A first try to use Exiv2 from trunk with digiKam give this compilation result : ``` [gilles@localhost exiv2 (master)]$ ./bootstrap-exiv2.sh -- The CXX compiler identification is GNU 5.5.0 -- The C compiler identification is GNU 5.5.0 -- Check for working CXX compiler: /bin/c++ -- Check for working CXX compiler: /bin/c++ -- works -- Detecting CXX compiler ABI info -- Detecting CXX compiler ABI info - done -- Detecting CXX compile features -- Detecting CXX compile features - done -- Check for working C compiler: /bin/cc -- Check for working C compiler: /bin/cc -- works -- Detecting C compiler ABI info -- Detecting C compiler ABI info - done -- Detecting C compile features -- Detecting C compile features - done -- Looking for pthread.h -- Looking for pthread.h - found -- Looking for pthread_create -- Looking for pthread_create - not found -- Looking for pthread_create in pthreads -- Looking for pthread_create in pthreads - not found -- Looking for pthread_create in pthread -- Looking for pthread_create in pthread - found -- Found Threads: TRUE -- Found ZLIB: /usr/lib64/libz.so (found version "1.2.11") -- Found EXPAT: /usr/lib64/libexpat.so (found version "2.2.1") -- Found Intl: /usr/include -- Performing Test Iconv_IS_BUILT_IN -- Performing Test Iconv_IS_BUILT_IN - Success -- Found Iconv: /usr/lib64/libc.so -- ICONV_INCLUDE_DIR : -- ICONV_LIBRARIES : /usr/lib64/libc.so -- Looking for gmtime_r -- Looking for gmtime_r - found -- Looking for mmap -- Looking for mmap - found -- Looking for munmap -- Looking for munmap - found -- Looking for strerror_r -- Looking for strerror_r - found -- Looking for libintl.h -- Looking for libintl.h - found -- Looking for unistd.h -- Looking for unistd.h - found -- Looking for memory.h -- Looking for memory.h - found -- Looking for process.h -- Looking for process.h - not found -- Looking for stdbool.h -- Looking for stdbool.h - found -- Looking for stdint.h -- Looking for stdint.h - found -- Looking for strings.h -- Looking for strings.h - found -- Looking for sys/mman.h -- Looking for sys/mman.h - found -- Looking for sys/stat.h -- Looking for sys/stat.h - found -- Looking for sys/types.h -- Looking for sys/types.h - found -- Looking for regex.h -- Looking for regex.h - found -- Looking for inttypes.h -- Looking for inttypes.h - found -- Performing Test COMPILER_HAS_HIDDEN_VISIBILITY -- Performing Test COMPILER_HAS_HIDDEN_VISIBILITY - Success -- Performing Test COMPILER_HAS_HIDDEN_INLINE_VISIBILITY -- Performing Test COMPILER_HAS_HIDDEN_INLINE_VISIBILITY - Success -- Performing Test COMPILER_HAS_DEPRECATED_ATTR -- Performing Test COMPILER_HAS_DEPRECATED_ATTR - Success -- Install prefix: /usr -- ------------------------------------------------------------------ -- CMake Generator: Unix Makefiles -- Compiler info: GNU (/bin/c++) ; version: 5.5.0 -- --- Compiler flags --- -- General: -Wall -Wcast-align -Wpointer-arith -Wformat-security -Wmissing-format-attribute -Woverloaded-virtual -W -std=gnu++98 -- Extra: -- Debug: -g3 -gstrict-dwarf -O0 -- Release: -O3 -DNDEBUG -- RelWithDebInfo: -O2 -g -DNDEBUG -- MinSizeRel: -Os -DNDEBUG -- --- Linker flags --- -- General: -- Debug: -- Release: -- RelWithDebInfo: -- MinSizeRel: -- -- Compiler Options -- Warnings as errors: NO -- Use extra compiler warning flags: NO -- -- Compiler info: GNU (/bin/c++) ; version: 5.5.0 -- ------------------------------------------------------------------ -- Building shared library: YES -- Building PNG support: YES -- XMP metadata support: YES -- Native language support: YES -- Conversion of Windows XP tags: YES -- Nikon lens database: YES -- Building video support: NO -- Building webready support: NO -- Building exiv2 command: YES -- Building samples: YES -- Building PO files: NO -- Building unit tests: NO -- Building doc: NO -- Building with coverage flags: NO -- Using ccache: NO -- ------------------------------------------------------------------ -- Configuring done -- Generating done -- Build files have been written to: /mnt/data/GIT/exiv2/build [gilles@localhost exiv2 (master)]$ [gilles@localhost build (master)]$ make Scanning dependencies of target xmp [ 1%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/ExpatAdapter.cpp.o [ 2%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/MD5.cpp.o [ 2%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/ParseRDF.cpp.o [ 3%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/UnicodeConversions.cpp.o [ 3%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/WXMPIterator.cpp.o [ 4%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/WXMPMeta.cpp.o [ 5%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/WXMPUtils.cpp.o [ 5%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/XML_Node.cpp.o [ 6%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/XMPCore_Impl.cpp.o [ 6%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/XMPIterator.cpp.o [ 7%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/XMPMeta-GetSet.cpp.o [ 8%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/XMPMeta-Parse.cpp.o [ 8%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/XMPMeta-Serialize.cpp.o [ 9%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/XMPMeta.cpp.o [ 9%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/XMPUtils-FileInfo.cpp.o [ 10%] Building CXX object xmpsdk/CMakeFiles/xmp.dir/src/XMPUtils.cpp.o [ 11%] Linking CXX static library ../lib/libxmp.a [ 11%] Built target xmp Scanning dependencies of target exiv2lib_int [ 12%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/canonmn_int.cpp.o [ 13%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/casiomn_int.cpp.o [ 13%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/cr2header_int.cpp.o [ 14%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/crwimage_int.cpp.o [ 14%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/fujimn_int.cpp.o [ 15%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/makernote_int.cpp.o [ 16%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/minoltamn_int.cpp.o [ 16%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/nikonmn_int.cpp.o [ 17%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/olympusmn_int.cpp.o [ 17%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/orfimage_int.cpp.o [ 18%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/panasonicmn_int.cpp.o [ 19%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/pentaxmn_int.cpp.o [ 19%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/rw2image_int.cpp.o [ 20%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/samsungmn_int.cpp.o [ 20%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/sigmamn_int.cpp.o [ 21%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/sonymn_int.cpp.o [ 22%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/tags_int.cpp.o [ 22%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/tiffcomposite_int.cpp.o [ 23%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/tiffimage_int.cpp.o [ 23%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/tiffvisitor_int.cpp.o [ 24%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/helper_functions.cpp.o [ 25%] Building CXX object src/CMakeFiles/exiv2lib_int.dir/pngchunk_int.cpp.o [ 25%] Built target exiv2lib_int Scanning dependencies of target exiv2lib [ 26%] Building CXX object src/CMakeFiles/exiv2lib.dir/asfvideo.cpp.o [ 26%] Building CXX object src/CMakeFiles/exiv2lib.dir/basicio.cpp.o In file included from /mnt/data/GIT/exiv2/src/basicio.cpp:31:0: /mnt/data/GIT/exiv2/include/exiv2/http.hpp:23:23: erreur fatale : curl/curl.h : Aucun fichier ou dossier de ce type compilation terminée. src/CMakeFiles/exiv2lib.dir/build.make:86 : la recette pour la cible « src/CMakeFiles/exiv2lib.dir/basicio.cpp.o » a échouée make[2]: *** [src/CMakeFiles/exiv2lib.dir/basicio.cpp.o] Erreur 1 CMakeFiles/Makefile2:283 : la recette pour la cible « src/CMakeFiles/exiv2lib.dir/all » a échouée make[1]: *** [src/CMakeFiles/exiv2lib.dir/all] Erreur 2 Makefile:149 : la recette pour la cible « all » a échouée make: *** [all] Erreur 2 ``` libcurl-devel package is not installed. So the error appear on my computer. The expected result must be a complete compilation in this case. Best --- end body --- comments : --- start comments --- --- end comments ---

Using the API, I'm unable to set the following Exif metadata with tiff files: - Exif.Image.XResolution - Exif.Image.YResolution - Exif.Image.ResolutionUnit I can read them but not write. Any changes I make or add are ignored and untouched for those fields. I've tried it with png, jpg, and jp2 files and it works there. Just with tiff files. Tested on Windows 7. Compiled with VS14 MSVC and MSYS2 (MinGW/GCC)

Unable to set Exif resolution metadata in tiff files

https://api.github.com/repos/Exiv2/exiv2/issues/463/comments

https://github.com/Exiv2/exiv2/issues/463

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : Unable to set Exif resolution metadata in tiff files Issue date: --- start body --- Using the API, I'm unable to set the following Exif metadata with tiff files: - Exif.Image.XResolution - Exif.Image.YResolution - Exif.Image.ResolutionUnit I can read them but not write. Any changes I make or add are ignored and untouched for those fields. I've tried it with png, jpg, and jp2 files and it works there. Just with tiff files. Tested on Windows 7. Compiled with VS14 MSVC and MSYS2 (MinGW/GCC) --- end body --- comments : --- start comments --- --- end comments ---

A stack overflow exits in `CiffDirectory::readDirectory()` at `crwimage_int.cpp` due to a recursive function call causing the excessive stack consumption which leads to Denial of service. ###### **Affected version:** exiv2 0.27.0.0 (64 bit build) ###### **Command:** ./exiv2 -pi $POC ### **Debugging** ``` 0x7ffff6f0d35a call 0x7ffff6e86bc0 <__stack_chk_fail@plt> 0x7ffff6f0d35f nop 0x7ffff6f0d360 sub rsp, 0x8 → 0x7ffff6f0d364 call 0x7ffff6f06340 ↳ 0x7ffff6f06340 cmp BYTE PTR [rip+0x47b341], 0x0 # 0x7ffff7381688 0x7ffff6f06347 je 0x7ffff6f06358 0x7ffff6f06349 mov edi, DWORD PTR [rip+0x47b33d] # 0x7ffff738168c 0x7ffff6f0634f jmp 0x7ffff6e86e00 <pthread_getspecific@plt> 0x7ffff6f06354 nop DWORD PTR [rax+0x0] 0x7ffff6f06358 lea rdx, [rip+0x34986] # 0x7ffff6f3ace5 ``` ### ASAN Output ``` ASAN:SIGSEGV ================================================================= ==63879==ERROR: AddressSanitizer: stack-overflow on address 0x7ffd1c5e0ff8 (pc 0x7fce8fe99252 bp 0x000000000050 sp 0x7ffd1c5e1000 T0) #0 0x7fce8fe99251 (/usr/lib/x86_64-linux-gnu/libasan.so.2+0xb0251) #1 0x7fce8fe98d97 (/usr/lib/x86_64-linux-gnu/libasan.so.2+0xafd97) #2 0x7fce8fe0bf0f (/usr/lib/x86_64-linux-gnu/libasan.so.2+0x22f0f) #3 0x7fce8fe8255e in operator new(unsigned long) (/usr/lib/x86_64-linux-gnu/libasan.so.2+0x9955e) #4 0x7fce8f79c94c in Exiv2::Internal::CiffDirectory::readDirectory(unsigned char const*, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:292 #5 0x7fce8f79c733 in Exiv2::Internal::CiffDirectory::doRead(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:269 #6 0x7fce8f79c318 in Exiv2::Internal::CiffComponent::read(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:227 #7 0x7fce8f79ca54 in Exiv2::Internal::CiffDirectory::readDirectory(unsigned char const*, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:296 #8 0x7fce8f79c733 in Exiv2::Internal::CiffDirectory::doRead(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:269 #9 0x7fce8f79c318 in Exiv2::Internal::CiffComponent::read(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:227 #10 0x7fce8f79ca54 in Exiv2::Internal::CiffDirectory::readDirectory(unsigned char const*, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:296 #11 0x7fce8f79c733 in Exiv2::Internal::CiffDirectory::doRead(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:269 #12 0x7fce8f79c318 in Exiv2::Internal::CiffComponent::read(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:227 #13 0x7fce8f79ca54 in Exiv2::Internal::CiffDirectory::readDirectory(unsigned char const*, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:296 #14 0x7fce8f79c733 in Exiv2::Internal::CiffDirectory::doRead(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:269 #15 0x7fce8f79c318 in Exiv2::Internal::CiffComponent::read(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:227 #16 0x7fce8f79ca54 in Exiv2::Internal::CiffDirectory::readDirectory(unsigned char const*, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:296 #17 0x7fce8f79c733 in Exiv2::Internal::CiffDirectory::doRead(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:269 #18 0x7fce8f79c318 in Exiv2::Internal::CiffComponent::read(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:227 #19 0x7fce8f79ca54 in Exiv2::Internal::CiffDirectory::readDirectory(unsigned char const*, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:296 #20 0x7fce8f79c733 in Exiv2::Internal::CiffDirectory::doRead(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:269 #21 0x7fce8f79c318 in Exiv2::Internal::CiffComponent::read(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:227 #22 0x7fce8f79ca54 in Exiv2::Internal::CiffDirectory::readDirectory(unsigned char const*, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:296 #23 0x7fce8f79c733 in Exiv2::Internal::CiffDirectory::doRead(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:269 . . . . . . . . #68 0x7fce8f79c733 in Exiv2::Internal::CiffDirectory::doRead(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:269 #69 0x7fce8f79c318 in Exiv2::Internal::CiffComponent::read(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:227 #70 0x7fce8f79ca54 in Exiv2::Internal::CiffDirectory::readDirectory(unsigned char const*, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:296 #71 0x7fce8f79c733 in Exiv2::Internal::CiffDirectory::doRead(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:269 #72 0x7fce8f79c318 in Exiv2::Internal::CiffComponent::read(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:227 #73 0x7fce8f79ca54 in Exiv2::Internal::CiffDirectory::readDirectory(unsigned char const*, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:296 #74 0x7fce8f79c733 in Exiv2::Internal::CiffDirectory::doRead(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:269 #75 0x7fce8f79c318 in Exiv2::Internal::CiffComponent::read(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:227 #76 0x7fce8f79ca54 in Exiv2::Internal::CiffDirectory::readDirectory(unsigned char const*, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:296 #77 0x7fce8f79c733 in Exiv2::Internal::CiffDirectory::doRead(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:269 #78 0x7fce8f79c318 in Exiv2::Internal::CiffComponent::read(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:227 #79 0x7fce8f79ca54 in Exiv2::Internal::CiffDirectory::readDirectory(unsigned char const*, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:296 #80 0x7fce8f79c733 in Exiv2::Internal::CiffDirectory::doRead(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:269 #81 0x7fce8f79c318 in Exiv2::Internal::CiffComponent::read(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:227 #82 0x7fce8f79ca54 in Exiv2::Internal::CiffDirectory::readDirectory(unsigned char const*, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:296 #83 0x7fce8f79c733 in Exiv2::Internal::CiffDirectory::doRead(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:269 #84 0x7fce8f79c318 in Exiv2::Internal::CiffComponent::read(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:227 #85 0x7fce8f79ca54 in Exiv2::Internal::CiffDirectory::readDirectory(unsigned char const*, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:296 #86 0x7fce8f79c733 in Exiv2::Internal::CiffDirectory::doRead(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:269 #87 0x7fce8f79c318 in Exiv2::Internal::CiffComponent::read(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:227 #88 0x7fce8f79ca54 in Exiv2::Internal::CiffDirectory::readDirectory(unsigned char const*, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:296 #89 0x7fce8f79c733 in Exiv2::Internal::CiffDirectory::doRead(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:269 #90 0x7fce8f79c318 in Exiv2::Internal::CiffComponent::read(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:227 #91 0x7fce8f79ca54 in Exiv2::Internal::CiffDirectory::readDirectory(unsigned char const*, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:296 #92 0x7fce8f79c733 in Exiv2::Internal::CiffDirectory::doRead(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:269 #93 0x7fce8f79c318 in Exiv2::Internal::CiffComponent::read(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:227 #94 0x7fce8f79ca54 in Exiv2::Internal::CiffDirectory::readDirectory(unsigned char const*, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:296 #95 0x7fce8f79c733 in Exiv2::Internal::CiffDirectory::doRead(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:269 #96 0x7fce8f79c318 in Exiv2::Internal::CiffComponent::read(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:227 #97 0x7fce8f79ca54 in Exiv2::Internal::CiffDirectory::readDirectory(unsigned char const*, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:296 #98 0x7fce8f79c733 in Exiv2::Internal::CiffDirectory::doRead(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:269 #99 0x7fce8f79c318 in Exiv2::Internal::CiffComponent::read(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:227 #100 0x7fce8f79ca54 in Exiv2::Internal::CiffDirectory::readDirectory(unsigned char const*, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:296 #101 0x7fce8f79c733 in Exiv2::Internal::CiffDirectory::doRead(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:269 #102 0x7fce8f79c318 in Exiv2::Internal::CiffComponent::read(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:227 #103 0x7fce8f79ca54 in Exiv2::Internal::CiffDirectory::readDirectory(unsigned char const*, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:296 #104 0x7fce8f79c733 in Exiv2::Internal::CiffDirectory::doRead(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:269 . . . . . . . . . #240 0x7fce8f79c318 in Exiv2::Internal::CiffComponent::read(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:227 #241 0x7fce8f79ca54 in Exiv2::Internal::CiffDirectory::readDirectory(unsigned char const*, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:296 #242 0x7fce8f79c733 in Exiv2::Internal::CiffDirectory::doRead(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:269 #243 0x7fce8f79c318 in Exiv2::Internal::CiffComponent::read(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:227 #244 0x7fce8f79ca54 in Exiv2::Internal::CiffDirectory::readDirectory(unsigned char const*, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:296 #245 0x7fce8f79c733 in Exiv2::Internal::CiffDirectory::doRead(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:269 #246 0x7fce8f79c318 in Exiv2::Internal::CiffComponent::read(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:227 #247 0x7fce8f79ca54 in Exiv2::Internal::CiffDirectory::readDirectory(unsigned char const*, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:296 #248 0x7fce8f79c733 in Exiv2::Internal::CiffDirectory::doRead(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:269 #249 0x7fce8f79c318 in Exiv2::Internal::CiffComponent::read(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:227 #250 0x7fce8f79ca54 in Exiv2::Internal::CiffDirectory::readDirectory(unsigned char const*, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:296 #251 0x7fce8f79c733 in Exiv2::Internal::CiffDirectory::doRead(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:269 SUMMARY: AddressSanitizer: stack-overflow ??:0 ?? ``` [Reproducer file](https://github.com/SegfaultMasters/covering360/blob/master/Exiv2/readDirectory_stackoverflow_16)

Stack overflow due to excessive stack consumption (Recursive function)

https://api.github.com/repos/Exiv2/exiv2/issues/460/comments

https://github.com/Exiv2/exiv2/issues/460

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : Stack overflow due to excessive stack consumption (Recursive function) Issue date: --- start body --- A stack overflow exits in `CiffDirectory::readDirectory()` at `crwimage_int.cpp` due to a recursive function call causing the excessive stack consumption which leads to Denial of service. ###### **Affected version:** exiv2 0.27.0.0 (64 bit build) ###### **Command:** ./exiv2 -pi $POC ### **Debugging** ``` 0x7ffff6f0d35a call 0x7ffff6e86bc0 <__stack_chk_fail@plt> 0x7ffff6f0d35f nop 0x7ffff6f0d360 sub rsp, 0x8 → 0x7ffff6f0d364 call 0x7ffff6f06340 ↳ 0x7ffff6f06340 cmp BYTE PTR [rip+0x47b341], 0x0 # 0x7ffff7381688 0x7ffff6f06347 je 0x7ffff6f06358 0x7ffff6f06349 mov edi, DWORD PTR [rip+0x47b33d] # 0x7ffff738168c 0x7ffff6f0634f jmp 0x7ffff6e86e00 <pthread_getspecific@plt> 0x7ffff6f06354 nop DWORD PTR [rax+0x0] 0x7ffff6f06358 lea rdx, [rip+0x34986] # 0x7ffff6f3ace5 ``` ### ASAN Output ``` ASAN:SIGSEGV ================================================================= ==63879==ERROR: AddressSanitizer: stack-overflow on address 0x7ffd1c5e0ff8 (pc 0x7fce8fe99252 bp 0x000000000050 sp 0x7ffd1c5e1000 T0) #0 0x7fce8fe99251 (/usr/lib/x86_64-linux-gnu/libasan.so.2+0xb0251) #1 0x7fce8fe98d97 (/usr/lib/x86_64-linux-gnu/libasan.so.2+0xafd97) #2 0x7fce8fe0bf0f (/usr/lib/x86_64-linux-gnu/libasan.so.2+0x22f0f) #3 0x7fce8fe8255e in operator new(unsigned long) (/usr/lib/x86_64-linux-gnu/libasan.so.2+0x9955e) #4 0x7fce8f79c94c in Exiv2::Internal::CiffDirectory::readDirectory(unsigned char const*, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:292 #5 0x7fce8f79c733 in Exiv2::Internal::CiffDirectory::doRead(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:269 #6 0x7fce8f79c318 in Exiv2::Internal::CiffComponent::read(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:227 #7 0x7fce8f79ca54 in Exiv2::Internal::CiffDirectory::readDirectory(unsigned char const*, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:296 #8 0x7fce8f79c733 in Exiv2::Internal::CiffDirectory::doRead(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:269 #9 0x7fce8f79c318 in Exiv2::Internal::CiffComponent::read(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:227 #10 0x7fce8f79ca54 in Exiv2::Internal::CiffDirectory::readDirectory(unsigned char const*, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:296 #11 0x7fce8f79c733 in Exiv2::Internal::CiffDirectory::doRead(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:269 #12 0x7fce8f79c318 in Exiv2::Internal::CiffComponent::read(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:227 #13 0x7fce8f79ca54 in Exiv2::Internal::CiffDirectory::readDirectory(unsigned char const*, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:296 #14 0x7fce8f79c733 in Exiv2::Internal::CiffDirectory::doRead(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:269 #15 0x7fce8f79c318 in Exiv2::Internal::CiffComponent::read(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:227 #16 0x7fce8f79ca54 in Exiv2::Internal::CiffDirectory::readDirectory(unsigned char const*, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:296 #17 0x7fce8f79c733 in Exiv2::Internal::CiffDirectory::doRead(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:269 #18 0x7fce8f79c318 in Exiv2::Internal::CiffComponent::read(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:227 #19 0x7fce8f79ca54 in Exiv2::Internal::CiffDirectory::readDirectory(unsigned char const*, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:296 #20 0x7fce8f79c733 in Exiv2::Internal::CiffDirectory::doRead(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:269 #21 0x7fce8f79c318 in Exiv2::Internal::CiffComponent::read(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:227 #22 0x7fce8f79ca54 in Exiv2::Internal::CiffDirectory::readDirectory(unsigned char const*, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:296 #23 0x7fce8f79c733 in Exiv2::Internal::CiffDirectory::doRead(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:269 . . . . . . . . #68 0x7fce8f79c733 in Exiv2::Internal::CiffDirectory::doRead(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:269 #69 0x7fce8f79c318 in Exiv2::Internal::CiffComponent::read(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:227 #70 0x7fce8f79ca54 in Exiv2::Internal::CiffDirectory::readDirectory(unsigned char const*, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:296 #71 0x7fce8f79c733 in Exiv2::Internal::CiffDirectory::doRead(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:269 #72 0x7fce8f79c318 in Exiv2::Internal::CiffComponent::read(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:227 #73 0x7fce8f79ca54 in Exiv2::Internal::CiffDirectory::readDirectory(unsigned char const*, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:296 #74 0x7fce8f79c733 in Exiv2::Internal::CiffDirectory::doRead(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:269 #75 0x7fce8f79c318 in Exiv2::Internal::CiffComponent::read(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:227 #76 0x7fce8f79ca54 in Exiv2::Internal::CiffDirectory::readDirectory(unsigned char const*, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:296 #77 0x7fce8f79c733 in Exiv2::Internal::CiffDirectory::doRead(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:269 #78 0x7fce8f79c318 in Exiv2::Internal::CiffComponent::read(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:227 #79 0x7fce8f79ca54 in Exiv2::Internal::CiffDirectory::readDirectory(unsigned char const*, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:296 #80 0x7fce8f79c733 in Exiv2::Internal::CiffDirectory::doRead(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:269 #81 0x7fce8f79c318 in Exiv2::Internal::CiffComponent::read(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:227 #82 0x7fce8f79ca54 in Exiv2::Internal::CiffDirectory::readDirectory(unsigned char const*, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:296 #83 0x7fce8f79c733 in Exiv2::Internal::CiffDirectory::doRead(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:269 #84 0x7fce8f79c318 in Exiv2::Internal::CiffComponent::read(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:227 #85 0x7fce8f79ca54 in Exiv2::Internal::CiffDirectory::readDirectory(unsigned char const*, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:296 #86 0x7fce8f79c733 in Exiv2::Internal::CiffDirectory::doRead(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:269 #87 0x7fce8f79c318 in Exiv2::Internal::CiffComponent::read(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:227 #88 0x7fce8f79ca54 in Exiv2::Internal::CiffDirectory::readDirectory(unsigned char const*, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:296 #89 0x7fce8f79c733 in Exiv2::Internal::CiffDirectory::doRead(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:269 #90 0x7fce8f79c318 in Exiv2::Internal::CiffComponent::read(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:227 #91 0x7fce8f79ca54 in Exiv2::Internal::CiffDirectory::readDirectory(unsigned char const*, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:296 #92 0x7fce8f79c733 in Exiv2::Internal::CiffDirectory::doRead(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:269 #93 0x7fce8f79c318 in Exiv2::Internal::CiffComponent::read(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:227 #94 0x7fce8f79ca54 in Exiv2::Internal::CiffDirectory::readDirectory(unsigned char const*, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:296 #95 0x7fce8f79c733 in Exiv2::Internal::CiffDirectory::doRead(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:269 #96 0x7fce8f79c318 in Exiv2::Internal::CiffComponent::read(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:227 #97 0x7fce8f79ca54 in Exiv2::Internal::CiffDirectory::readDirectory(unsigned char const*, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:296 #98 0x7fce8f79c733 in Exiv2::Internal::CiffDirectory::doRead(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:269 #99 0x7fce8f79c318 in Exiv2::Internal::CiffComponent::read(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:227 #100 0x7fce8f79ca54 in Exiv2::Internal::CiffDirectory::readDirectory(unsigned char const*, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:296 #101 0x7fce8f79c733 in Exiv2::Internal::CiffDirectory::doRead(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:269 #102 0x7fce8f79c318 in Exiv2::Internal::CiffComponent::read(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:227 #103 0x7fce8f79ca54 in Exiv2::Internal::CiffDirectory::readDirectory(unsigned char const*, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:296 #104 0x7fce8f79c733 in Exiv2::Internal::CiffDirectory::doRead(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:269 . . . . . . . . . #240 0x7fce8f79c318 in Exiv2::Internal::CiffComponent::read(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:227 #241 0x7fce8f79ca54 in Exiv2::Internal::CiffDirectory::readDirectory(unsigned char const*, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:296 #242 0x7fce8f79c733 in Exiv2::Internal::CiffDirectory::doRead(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:269 #243 0x7fce8f79c318 in Exiv2::Internal::CiffComponent::read(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:227 #244 0x7fce8f79ca54 in Exiv2::Internal::CiffDirectory::readDirectory(unsigned char const*, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:296 #245 0x7fce8f79c733 in Exiv2::Internal::CiffDirectory::doRead(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:269 #246 0x7fce8f79c318 in Exiv2::Internal::CiffComponent::read(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:227 #247 0x7fce8f79ca54 in Exiv2::Internal::CiffDirectory::readDirectory(unsigned char const*, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:296 #248 0x7fce8f79c733 in Exiv2::Internal::CiffDirectory::doRead(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:269 #249 0x7fce8f79c318 in Exiv2::Internal::CiffComponent::read(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:227 #250 0x7fce8f79ca54 in Exiv2::Internal::CiffDirectory::readDirectory(unsigned char const*, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:296 #251 0x7fce8f79c733 in Exiv2::Internal::CiffDirectory::doRead(unsigned char const*, unsigned int, unsigned int, Exiv2::ByteOrder) /home/woot/Desktop/exiv2/src/crwimage_int.cpp:269 SUMMARY: AddressSanitizer: stack-overflow ??:0 ?? ``` [Reproducer file](https://github.com/SegfaultMasters/covering360/blob/master/Exiv2/readDirectory_stackoverflow_16) --- end body --- comments : --- start comments --- --- end comments ---

Is the possible to wrap the internal `xmpsrc` XMP toolkit within an `Exiv2::internal` namespace or similar - since it appears to be only used internally and its interface never exported - to avoid potential translation unit/name clashes with other libraries that also have their own internal XMP toolkit? The use case/problem scope/justification below: ---- I have an `eog` (Eye of Gnome, image viewer) [plugin that uses Exiv2 to add XMP ratings](https://github.com/whatdoineed2do/eog-plugin-exif-rating-raw-viewer) to image files as I browse them. This plugin worked for me for many years and most recently on my Fedora26 systems. With the Fedora 28 release of `eog` I find that the plugin coredumps/`SEGV`s when calling `Exiv2::TiffImage::readMetaData()` and in `XMPData::RegisterNamespace()` where `sNamespaceURIToPrefixMap == NULL` ``` #0 0x00007efd47cd35b8 in XMPMeta::RegisterNamespace(char const*, char const*) () at /lib64/libexiv2.so.26 #1 0x00007efd47cb1537 in WXMPMeta_RegisterNamespace_1 () at /lib64/libexiv2.so.26 #2 0x00007efd47ca0ed8 in TXMPMeta<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >::RegisterNamespace(char const*, char const*) () at /lib64/libexiv2.so.26 #3 0x00007efd47c9cf76 in Exiv2::XmpParser::initialize(void (*)(void*, bool), void*) () at /lib64/libexiv2.so.26 #4 0x00007efd47c9f779 in Exiv2::XmpParser::decode(Exiv2::XmpData&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&) () at /lib64/libexiv2.so.26 #5 0x00007efd47c87939 in Exiv2::Internal::TiffDecoder::decodeXmp(Exiv2::Internal::TiffEntryBase const*) () at /lib64/libexiv2.so.26 #6 0x00007efd47c7091e in Exiv2::Internal::TiffDirectory::doAccept(Exiv2::Internal::TiffVisitor&) () at /lib64/libexiv2.so.26 #7 0x00007efd47c79653 in Exiv2::Internal::TiffParserWorker::decode(Exiv2::ExifData&, Exiv2::IptcData&, Exiv2::XmpData&, unsigned char const*, unsigned int, unsigned int, void (Exiv2::Internal::TiffDecoder::*(*)(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, unsigned int, Exiv2::Internal::IfdId))(Exiv2::Internal::TiffEntryBase const*), Exiv2::Internal::TiffHeaderBase*) () at /lib64/libexiv2.so.26 #8 0x00007efd47c7978b in Exiv2::TiffParser::decode(Exiv2::ExifData&, Exiv2::IptcData&, Exiv2::XmpData&, unsigned char const*, unsigned int) () at /lib64/libexiv2.so.26 #9 0x00007efd47c79851 in Exiv2::TiffImage::readMetadata() () at /lib64/libexiv2.so.26 ... ``` After some [investigation](https://github.com/whatdoineed2do/eog-plugin-exif-rating-raw-viewer/issues/3) I find that `eog` uses `libexempi` (which includes its own copy of the XMP toolkit) which when used in conjunction with `exiv2`’s XMPtool in the same process space causes problems - it appears both are trying to initialise their own versions of the their XMPtool kit but because of variable name clashes `exiv2` is left unitialised leading to `sNamespaceURIToPrefixMap == NULL` and the SEGV. I have a simple test case that demonstrates the problem: ``` #include <iostream> #include <stdexcept> #include <exiv2/exiv2.hpp> #include <exempi-2.0/exempi/xmp.h> int main(int argc, char* argv[]) { if (argc != 2) { std::cerr << "needs ONE image file" << std::endl; return -1; } try { xmp_init(); Exiv2::Image::AutoPtr img = Exiv2::ImageFactory::open(argv[1]); img->readMetadata(); } catch (const std::exception& ex) { std::cerr << ex.what() << std::endl; } return 0; } ``` On F28, `libexempi` -> `libexempi.so.3` and we can see the problem if `libexempi` is initialised/loaded first by the rt linker. ``` LD_PRELOAD=/usr/lib64/libexempi.so.3 ./a.out /tmp/foo.NEF ``` I don't think it matters what the file is as long as it has XMP data but my Nikon NEFs trigger the problem. Removing the `xmp_init()` call we we don't have the problem. Given that this appears to be a conflict with names for the 2 XMP toolkits, are you open to wrapping `Exiv` `xmpsrc` within an `Exiv2::internal` namespace? I think this would solve the problem and ensure there are no such conflicts with other instances of XMP toolkits in frameworks/programs?

add namespace to XMP toolkit - exiv2 and libexempi in same process coredumps

https://api.github.com/repos/Exiv2/exiv2/issues/459/comments

https://github.com/Exiv2/exiv2/issues/459

This is a GitHub Issue repo:exiv2 owner:Exiv2 Title : add namespace to XMP toolkit - exiv2 and libexempi in same process coredumps Issue date: --- start body --- Is the possible to wrap the internal `xmpsrc` XMP toolkit within an `Exiv2::internal` namespace or similar - since it appears to be only used internally and its interface never exported - to avoid potential translation unit/name clashes with other libraries that also have their own internal XMP toolkit? The use case/problem scope/justification below: ---- I have an `eog` (Eye of Gnome, image viewer) [plugin that uses Exiv2 to add XMP ratings](https://github.com/whatdoineed2do/eog-plugin-exif-rating-raw-viewer) to image files as I browse them. This plugin worked for me for many years and most recently on my Fedora26 systems. With the Fedora 28 release of `eog` I find that the plugin coredumps/`SEGV`s when calling `Exiv2::TiffImage::readMetaData()` and in `XMPData::RegisterNamespace()` where `sNamespaceURIToPrefixMap == NULL` ``` #0 0x00007efd47cd35b8 in XMPMeta::RegisterNamespace(char const*, char const*) () at /lib64/libexiv2.so.26 #1 0x00007efd47cb1537 in WXMPMeta_RegisterNamespace_1 () at /lib64/libexiv2.so.26 #2 0x00007efd47ca0ed8 in TXMPMeta<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >::RegisterNamespace(char const*, char const*) () at /lib64/libexiv2.so.26 #3 0x00007efd47c9cf76 in Exiv2::XmpParser::initialize(void (*)(void*, bool), void*) () at /lib64/libexiv2.so.26 #4 0x00007efd47c9f779 in Exiv2::XmpParser::decode(Exiv2::XmpData&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&) () at /lib64/libexiv2.so.26 #5 0x00007efd47c87939 in Exiv2::Internal::TiffDecoder::decodeXmp(Exiv2::Internal::TiffEntryBase const*) () at /lib64/libexiv2.so.26 #6 0x00007efd47c7091e in Exiv2::Internal::TiffDirectory::doAccept(Exiv2::Internal::TiffVisitor&) () at /lib64/libexiv2.so.26 #7 0x00007efd47c79653 in Exiv2::Internal::TiffParserWorker::decode(Exiv2::ExifData&, Exiv2::IptcData&, Exiv2::XmpData&, unsigned char const*, unsigned int, unsigned int, void (Exiv2::Internal::TiffDecoder::*(*)(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, unsigned int, Exiv2::Internal::IfdId))(Exiv2::Internal::TiffEntryBase const*), Exiv2::Internal::TiffHeaderBase*) () at /lib64/libexiv2.so.26 #8 0x00007efd47c7978b in Exiv2::TiffParser::decode(Exiv2::ExifData&, Exiv2::IptcData&, Exiv2::XmpData&, unsigned char const*, unsigned int) () at /lib64/libexiv2.so.26 #9 0x00007efd47c79851 in Exiv2::TiffImage::readMetadata() () at /lib64/libexiv2.so.26 ... ``` After some [investigation](https://github.com/whatdoineed2do/eog-plugin-exif-rating-raw-viewer/issues/3) I find that `eog` uses `libexempi` (which includes its own copy of the XMP toolkit) which when used in conjunction with `exiv2`’s XMPtool in the same process space causes problems - it appears both are trying to initialise their own versions of the their XMPtool kit but because of variable name clashes `exiv2` is left unitialised leading to `sNamespaceURIToPrefixMap == NULL` and the SEGV. I have a simple test case that demonstrates the problem: ``` #include <iostream> #include <stdexcept> #include <exiv2/exiv2.hpp> #include <exempi-2.0/exempi/xmp.h> int main(int argc, char* argv[]) { if (argc != 2) { std::cerr << "needs ONE image file" << std::endl; return -1; } try { xmp_init(); Exiv2::Image::AutoPtr img = Exiv2::ImageFactory::open(argv[1]); img->readMetadata(); } catch (const std::exception& ex) { std::cerr << ex.what() << std::endl; } return 0; } ``` On F28, `libexempi` -> `libexempi.so.3` and we can see the problem if `libexempi` is initialised/loaded first by the rt linker. ``` LD_PRELOAD=/usr/lib64/libexempi.so.3 ./a.out /tmp/foo.NEF ``` I don't think it matters what the file is as long as it has XMP data but my Nikon NEFs trigger the problem. Removing the `xmp_init()` call we we don't have the problem. Given that this appears to be a conflict with names for the 2 XMP toolkits, are you open to wrapping `Exiv` `xmpsrc` within an `Exiv2::internal` namespace? I think this would solve the problem and ensure there are no such conflicts with other instances of XMP toolkits in frameworks/programs? --- end body --- comments : --- start comments --- --- end comments ---