The dataset viewer is not available for this split.
Error code: StreamingRowsError
Exception: ValueError
Message: Expected object or value
Traceback: Traceback (most recent call last):
File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/json/json.py", line 290, in _generate_tables
pa_table = paj.read_json(
io.BytesIO(batch), read_options=paj.ReadOptions(block_size=block_size)
)
File "pyarrow/_json.pyx", line 364, in pyarrow._json.read_json
File "pyarrow/error.pxi", line 155, in pyarrow.lib.pyarrow_internal_check_status
File "pyarrow/error.pxi", line 92, in pyarrow.lib.check_status
raise convert_status(status)
pyarrow.lib.ArrowInvalid: JSON parse error: Column() changed from object to string in row 0
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "/src/services/worker/src/worker/utils.py", line 147, in get_rows_or_raise
return get_rows(
dataset=dataset,
...<4 lines>...
column_names=column_names,
)
File "/src/libs/libcommon/src/libcommon/utils.py", line 272, in decorator
return func(*args, **kwargs)
File "/src/services/worker/src/worker/utils.py", line 127, in get_rows
rows_plus_one = list(itertools.islice(safe_iter(ds, dataset=dataset), rows_max_number + 1))
File "/src/services/worker/src/worker/utils.py", line 483, in safe_iter
yield from ds.decode(False) if ds.features else ds
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2951, in __iter__
for key, example in ex_iterable:
^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2461, in __iter__
for key, pa_table in self._iter_arrow():
~~~~~~~~~~~~~~~~^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2486, in _iter_arrow
for key, pa_table in self.ex_iterable._iter_arrow():
~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 547, in _iter_arrow
for key, pa_table in iterator:
^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 430, in _iter_arrow
for key, pa_table in self.generate_tables_fn(**gen_kwags):
~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/json/json.py", line 304, in _generate_tables
batch = json_encode_fields_in_json_lines(original_batch, json_field_paths)
File "/usr/local/lib/python3.14/site-packages/datasets/utils/json.py", line 111, in json_encode_fields_in_json_lines
examples = [ujson_loads(line) for line in original_batch.splitlines()]
~~~~~~~~~~~^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/utils/json.py", line 20, in ujson_loads
return pd.io.json.ujson_loads(*args, **kwargs)
~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
ValueError: Expected object or valueNeed help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.
Bendyline knowledge catalogs
Searchable .gezk archives for Gezel, with pinned sources, license evidence, checksums, and retrieval verification. Downloads use immutable Hugging Face revisions; GitHub Releases provide mirrors and verification reports.
Published Microsoft technology catalogs
Version 2026.10.16 contains 73,625 indexed documents across 13 catalogs. These are preserved source snapshots, with complete source revisions, attribution, and license evidence; they do not claim to track the live Microsoft Learn site.
| Catalog | Documents | Size | Downloads |
|---|---|---|---|
| .NET and C# | 20,170 | 349.9 MiB | Download .gezk · GitHub and checksums |
| ASP.NET Core | 12,339 | 155.8 MiB | Download .gezk · GitHub and checksums |
| Azure AI, Foundry, and Machine Learning | 2,843 | 185.1 MiB | Download .gezk · GitHub and checksums |
| Azure applications and compute | 4,307 | 205.0 MiB | Download .gezk · GitHub and checksums |
| Azure data, storage, and IoT | 3,976 | 182.8 MiB | Download .gezk · GitHub and checksums |
| Azure management and hybrid infrastructure | 2,996 | 134.6 MiB | Download .gezk · GitHub and checksums |
| Azure networking and security | 2,311 | 112.2 MiB | Download .gezk · GitHub and checksums |
| Entity Framework | 1,157 | 26.5 MiB | Download .gezk · GitHub and checksums |
| Microsoft 365 and Copilot | 1,246 | 41.3 MiB | Download .gezk · GitHub and checksums |
| PowerShell | 3,079 | 142.3 MiB | Download .gezk · GitHub and checksums |
| SQL Server and Azure SQL | 13,848 | 375.0 MiB | Download .gezk · GitHub and checksums |
| Windows app development | 2,791 | 93.9 MiB | Download .gezk · GitHub and checksums |
| Windows Server | 2,562 | 77.0 MiB | Download .gezk · GitHub and checksums |
Microsoft documentation in these catalogs is licensed under CC BY 4.0, and included code under MIT, as specified by each source repository. See LICENSE, the archive's LICENSES directory, and the preservation scope. Archive integrity and configured full-text and semantic retrieval checks passed; both public mirrors were downloaded and hash-verified.
Published how-to library
| Catalog | Version | Documents | Size | Downloads |
|---|---|---|---|---|
| How-to books | 2026.10.8 | 69,635 | 2027.2 MiB | Download .gezk · GitHub and checksums |
The catalog contains every English book on Project Gutenberg's "How To ..." bookshelf that Project Gutenberg marks public domain in the USA. That is 2,212 books on cooking, gardening, animals, crafts, building, outdoor skills, household management and more, split into 69,509 chapter sections that read in book order. It also contains 126 original how-to guides in 18 topics; each cites the book sections it draws on and adds modern safety notes. Book text carries the Public Domain Mark 1.0. That status comes from Project Gutenberg's metadata for the United States, so check local law elsewhere. Project Gutenberg's license and trademark text are removed, and this catalog is not affiliated with or endorsed by Project Gutenberg. The guides are licensed CC BY-SA 4.0. The books are historical works: some of their advice is obsolete or unsafe and is not current professional guidance. Archive integrity checks, and all configured full-text and semantic retrieval checks, passed; downloads from both hosts were verified. See LICENSE, the archive's LICENSES directory, and the catalog documentation.
Published caselaw collections
| Archive | Version | Case records | Size | Downloads |
|---|---|---|---|---|
| U.S. Reports, volumes 347–349 — CAP pilot | 2026.10.5 | 2,142 | 28.7 MiB | Download .gezk · GitHub and checksums |
| Alaska caselaw — CAP jurisdiction collection — 1887–2019 (part-0001) | 2026.10.1 | 10,749 | 993.2 MiB | Download .gezk · GitHub and checksums |
| Wyoming caselaw — Caselaw Access Project — 1870–2019 (part-0001) | 2026.10.3 | 10,931 | 1019.7 MiB | Download .gezk · GitHub and checksums |
| Nevada caselaw — CAP jurisdiction collection — 1865–2019 (part-0001) | 2026.10.1 | 13,707 | 730.8 MiB | Download .gezk · GitHub and checksums |
| Utah caselaw — Caselaw Access Project — 1861–1992 (part-0001) | 2026.10.1 | 15,254 | 972.9 MiB | Download .gezk · GitHub and checksums |
| Utah caselaw — Caselaw Access Project — 1993–2019 (part-0002) | 2026.10.1 | 7,650 | 695.7 MiB | Download .gezk · GitHub and checksums |
| U.S. Reports volumes 300–399 — Caselaw Access Project — 1910–1970 (part-0001) | 2026.10.2 | 61,926 | 985.4 MiB | Download .gezk · GitHub and checksums |
| Washington caselaw — Caselaw Access Project — 1854–1930 (part-0001) | 2026.10.1 | 20,139 | 931.6 MiB | Download .gezk · GitHub and checksums |
| Washington caselaw — Caselaw Access Project — 1931–1978 (part-0002) | 2026.10.1 | 18,778 | 967.7 MiB | Download .gezk · GitHub and checksums |
| Washington caselaw — Caselaw Access Project — 1979–1997 (part-0003) | 2026.10.1 | 28,210 | 921.9 MiB | Download .gezk · GitHub and checksums |
| Washington caselaw — Caselaw Access Project — 1998–2010 (part-0004) | 2026.10.1 | 24,936 | 878.5 MiB | Download .gezk · GitHub and checksums |
| Washington caselaw — Caselaw Access Project — 2011–2019 (part-0005) | 2026.10.1 | 14,114 | 538.0 MiB | Download .gezk · GitHub and checksums |
U.S. Reports 300–399 combines 100 source books into one archive near the 1 GiB target. It contains all 2,142 cases in the smaller 347–349 pilot. Choose the larger collection for broader coverage; installing both duplicates those cases. Both are bounded selections, not the complete U.S. Reports or federal collection. The original pilot remains available at its immutable release.
- Alaska caselaw — CAP jurisdiction collection: 10,749 records in 1 archive. Collection index with exact case membership and immutable downloads. Preserves CAP's Alaska jurisdiction assignment, including 21 U.S. District Court for the District of Alaska records printed in Federal Supplement, Third Series. Court names remain explicit; this is not exclusively state-court material.
- Wyoming caselaw — Caselaw Access Project: 10,931 records in 1 archive. Collection index with exact case membership and immutable downloads.
- Nevada caselaw — CAP jurisdiction collection: 13,707 records in 1 archive. Collection index with exact case membership and immutable downloads. Preserves CAP's Nevada jurisdiction assignment: 13,688 Supreme Court records, 18 Court of Appeals records, and one historical Eureka County District Court record. Coverage reflects this pinned historical archive, not every decision issued by those courts.
- Utah caselaw — Caselaw Access Project: 22,904 records in 2 archives. Collection index with exact case membership and immutable downloads. Preserves CAP's Utah jurisdiction assignment: 16,874 Supreme Court records, 6,029 Court of Appeals records under two source court labels, and one historical Salt Lake City Register and Receiver record. Coverage reflects the pinned historical archive, not every decision issued by those courts.
- U.S. Reports volumes 300–399 — Caselaw Access Project: 61,926 records in 1 archive. Collection index with exact case membership and immutable downloads. Every CAP case record in U.S. Reports volumes 300–399, preserving source jurisdiction, court labels, and decision dates, including four source dates earlier than the books' 1937–1970 period. This bounded historical reporter collection contains the earlier 347–349 pilot in full; installing both duplicates those records. It is not the complete U.S. Reports or federal collection.
- Washington caselaw — Caselaw Access Project: 106,177 records in 5 archives. Collection index with exact case membership and immutable downloads. Preserves CAP's Washington jurisdiction assignment: 41,045 Supreme Court records, 65,131 Court of Appeals records under four source court labels, and one historical Washington District Court record. Includes territorial-era records and upstream dates from 1854 through 2019. Coverage reflects the pinned historical archive, not every decision issued by those courts.
These are historical source collections. OCR errors and upstream metadata classifications are preserved; inclusion does not establish current legal authority. Headnotes are labeled separately from judicial opinions. Internal case links resolve locally within an archive; links to excluded cases point to CAP. Citation and semantic checks are smoke tests, not a broad legal retrieval evaluation. In U.S. Reports 300–399, the original broad right-to-counsel probe placed Gideon outside the top 20 chunks; the final smoke probe uses case-specific facts. The original failure remains documented.
Case data and metadata are designated CC0 1.0 Universal by the Caselaw Access Project. Credit to Harvard Law School Library is provided voluntarily. Each archive includes license texts, terms evidence, per-case source hashes, and provenance. The dataset contains catalogs with differing licenses; consult LICENSE and the relevant archive's LICENSES directory.
See the source repository, caselaw pipeline and scope, and provisional national publication list. Case bodies and generated Markdown are prepared outside Git; definitions and release tooling are versioned in the repository.
- Downloads last month
- 927