id
int64 959M
2.55B
| title
stringlengths 3
133
| body
stringlengths 1
65.5k
β | description
stringlengths 5
65.6k
| state
stringclasses 2
values | created_at
stringlengths 20
20
| updated_at
stringlengths 20
20
| closed_at
stringlengths 20
20
β | user
stringclasses 174
values |
|---|---|---|---|---|---|---|---|---|
1,981,190,891 | Add missing runners to job runner factory | null | Add missing runners to job runner factory: | closed | 2023-11-07T12:01:16Z | 2023-11-07T12:10:33Z | 2023-11-07T12:10:22Z | lhoestq |
1,981,048,062 | create /healthcheck and /metrics endpoints in workers | fixes #2067
- [x] Publish /metrics and /healthcheck endpoints in workers
- [x] Check worker's health with the new endpoint
- [x] Fetch worker's Prometheus metrics
- [x] Instrument parts of the worker's code | create /healthcheck and /metrics endpoints in workers: fixes #2067
- [x] Publish /metrics and /healthcheck endpoints in workers
- [x] Check worker's health with the new endpoint
- [x] Fetch worker's Prometheus metrics
- [x] Instrument parts of the worker's code | closed | 2023-11-07T10:37:40Z | 2023-11-08T18:22:55Z | 2023-11-08T18:22:54Z | severo |
1,980,331,649 | try to remove assets and cached-assets | Second try to delete all dependencies to assets and cached assets.
In https://github.com/huggingface/datasets-server/pull/2040, reverse-proxy failed (Maybe because of a wrong nginx configuration file, I didn't find any log error).
| try to remove assets and cached-assets: Second try to delete all dependencies to assets and cached assets.
In https://github.com/huggingface/datasets-server/pull/2040, reverse-proxy failed (Maybe because of a wrong nginx configuration file, I didn't find any log error).
| closed | 2023-11-07T00:44:07Z | 2023-11-07T12:14:49Z | 2023-11-07T11:29:26Z | AndreaFrancis |
1,979,752,070 | Instrument the workers to profile jobs duration | The workers do not send metrics to Prometheus, which makes it hard to understand where the time is spent. | Instrument the workers to profile jobs duration: The workers do not send metrics to Prometheus, which makes it hard to understand where the time is spent. | closed | 2023-11-06T18:04:49Z | 2023-11-08T18:22:55Z | 2023-11-08T18:22:55Z | severo |
1,979,667,830 | Revert delete assets | null | Revert delete assets: | closed | 2023-11-06T17:25:08Z | 2023-11-06T17:26:51Z | 2023-11-06T17:26:49Z | AndreaFrancis |
1,979,660,702 | Revert "remove assets and cached assets (#2040)" | This reverts commit 7d1213c05598e2cb4aabbb829b3df7e4a9cbca6f. | Revert "remove assets and cached assets (#2040)": This reverts commit 7d1213c05598e2cb4aabbb829b3df7e4a9cbca6f. | closed | 2023-11-06T17:20:48Z | 2023-11-06T17:23:03Z | 2023-11-06T17:22:59Z | AndreaFrancis |
1,979,610,600 | UnexpectedApiError on viewer | https://huggingface.co/datasets/RepoFusion/Stack-Repo/viewer/bm25_contexts
> Error code: UnexpectedApiError
| UnexpectedApiError on viewer: https://huggingface.co/datasets/RepoFusion/Stack-Repo/viewer/bm25_contexts
> Error code: UnexpectedApiError
| closed | 2023-11-06T16:55:07Z | 2023-11-06T17:54:04Z | 2023-11-06T17:02:28Z | severo |
1,979,357,282 | fix reverse proxy | After deploying https://github.com/huggingface/datasets-server/pull/2040 in staging, reverse-proxy component failed (I could not see the specific error but suspect it is because of an alias to a no longer existent location /storage).
| fix reverse proxy: After deploying https://github.com/huggingface/datasets-server/pull/2040 in staging, reverse-proxy component failed (I could not see the specific error but suspect it is because of an alias to a no longer existent location /storage).
| closed | 2023-11-06T14:52:19Z | 2023-11-06T16:59:24Z | 2023-11-06T16:59:23Z | AndreaFrancis |
1,979,344,129 | Fix cache directory for partial datasets for `split-descriptive-statistics` runner | fixes https://github.com/huggingface/datasets-server/issues/1831
fix glob pattern: preserve 'partial' in a split dir name | Fix cache directory for partial datasets for `split-descriptive-statistics` runner: fixes https://github.com/huggingface/datasets-server/issues/1831
fix glob pattern: preserve 'partial' in a split dir name | closed | 2023-11-06T14:46:20Z | 2023-11-06T17:21:41Z | 2023-11-06T17:21:40Z | polinaeterna |
1,978,834,505 | Support "maxDays: -1" to disable deleting old cache entries | See https://github.com/huggingface/datasets-server/pull/2060#pullrequestreview-1714775842 for reference. | Support "maxDays: -1" to disable deleting old cache entries: See https://github.com/huggingface/datasets-server/pull/2060#pullrequestreview-1714775842 for reference. | closed | 2023-11-06T10:47:56Z | 2024-02-06T15:55:02Z | 2024-02-06T15:55:02Z | severo |
1,978,804,114 | invalidate the datasets cache after 3 years instead of 3 months | Quick fix to avoid deleting old cache entries (in particular, to be sure to keep the old dataset cache entries for datasets with a dataset script, now that we don't support scripts anymore) | invalidate the datasets cache after 3 years instead of 3 months: Quick fix to avoid deleting old cache entries (in particular, to be sure to keep the old dataset cache entries for datasets with a dataset script, now that we don't support scripts anymore) | closed | 2023-11-06T10:31:48Z | 2023-11-06T10:48:05Z | 2023-11-06T10:46:32Z | severo |
1,978,729,245 | UnexpectedError on /rows for splits with images | https://huggingface.co/datasets/diegomiranda/teste-new-repo-1/discussions/1
> The dataset viewer is not working.
>
> Error details:
>
> ```
> Error code: UnexpectedError
> ```
>
> Hi,
>
> I'm encountering an issue with loading images in the dataset. I'm working with five datasets: three of them include both images and CSV files, while the remaining two contain only CSV files. When I test the datasets with only CSV files, everything works fine. However, I'm facing challenges when working with the datasets that include images. I would greatly appreciate any assistance or guidance on resolving this matter. | UnexpectedError on /rows for splits with images: https://huggingface.co/datasets/diegomiranda/teste-new-repo-1/discussions/1
> The dataset viewer is not working.
>
> Error details:
>
> ```
> Error code: UnexpectedError
> ```
>
> Hi,
>
> I'm encountering an issue with loading images in the dataset. I'm working with five datasets: three of them include both images and CSV files, while the remaining two contain only CSV files. When I test the datasets with only CSV files, everything works fine. However, I'm facing challenges when working with the datasets that include images. I would greatly appreciate any assistance or guidance on resolving this matter. | closed | 2023-11-06T09:51:54Z | 2024-07-30T16:08:11Z | 2024-07-30T16:08:11Z | severo |
1,978,612,223 | Stop worker loop executor on SIGTERM | Fix #1999. | Stop worker loop executor on SIGTERM: Fix #1999. | closed | 2023-11-06T08:53:42Z | 2023-11-07T10:29:23Z | 2023-11-07T10:29:23Z | albertvillanova |
1,976,978,765 | The "Recreate dataset" admin action first fails, then works | First click on the button:
<img width="1526" alt="Capture dβeΜcran 2023-11-03 aΜ 22 55 31" src="https://github.com/huggingface/datasets-server/assets/1676121/ebbcb349-ca39-4a6e-9aee-f244ef0ebbf6">
Second click (a few seconds later):
<img width="1526" alt="Capture dβeΜcran 2023-11-03 aΜ 22 55 38" src="https://github.com/huggingface/datasets-server/assets/1676121/96ecc3d3-e5a8-4d95-a20f-2c239589ed7d">
| The "Recreate dataset" admin action first fails, then works: First click on the button:
<img width="1526" alt="Capture dβeΜcran 2023-11-03 aΜ 22 55 31" src="https://github.com/huggingface/datasets-server/assets/1676121/ebbcb349-ca39-4a6e-9aee-f244ef0ebbf6">
Second click (a few seconds later):
<img width="1526" alt="Capture dβeΜcran 2023-11-03 aΜ 22 55 38" src="https://github.com/huggingface/datasets-server/assets/1676121/96ecc3d3-e5a8-4d95-a20f-2c239589ed7d">
| closed | 2023-11-03T21:57:55Z | 2023-11-07T20:22:43Z | 2023-11-07T19:09:50Z | severo |
1,976,962,566 | feat: πΈ upgrade nginx image to 1.25.3 | fixes #1961.
RELEASE NOTES are here: https://nginx.org/en/CHANGES.
The change we're interested in is:
> *) Change: improved detection of misbehaving clients when using
HTTP/2. | feat: πΈ upgrade nginx image to 1.25.3: fixes #1961.
RELEASE NOTES are here: https://nginx.org/en/CHANGES.
The change we're interested in is:
> *) Change: improved detection of misbehaving clients when using
HTTP/2. | closed | 2023-11-03T21:37:53Z | 2023-11-04T09:30:39Z | 2023-11-04T09:30:38Z | severo |
1,976,953,097 | fix: π hide the error to avoid disclosing a token | see internal discussion in Slack:
https://huggingface.slack.com/archives/C04L6P8KNQ5/p1692803953736769 | fix: π hide the error to avoid disclosing a token: see internal discussion in Slack:
https://huggingface.slack.com/archives/C04L6P8KNQ5/p1692803953736769 | closed | 2023-11-03T21:27:33Z | 2023-11-06T17:17:57Z | 2023-11-06T17:17:56Z | severo |
1,976,941,021 | Use ruff in vscode | Can you test if it works for you? (edit: I'm merging, but still: please give me feedback if it breaks something)
For me, after some struggle, it works well. A nice improvement is that it will sort the imports.
https://github.com/huggingface/datasets-server/assets/1676121/ba698454-b6df-42d6-b402-676b6d030e8c
| Use ruff in vscode: Can you test if it works for you? (edit: I'm merging, but still: please give me feedback if it breaks something)
For me, after some struggle, it works well. A nice improvement is that it will sort the imports.
https://github.com/huggingface/datasets-server/assets/1676121/ba698454-b6df-42d6-b402-676b6d030e8c
| closed | 2023-11-03T21:15:07Z | 2023-11-03T21:17:01Z | 2023-11-03T21:16:31Z | severo |
1,976,560,588 | Delete duckdb files when a split is deleted | See https://huggingface.co/datasets/HuggingFaceH4/ultrachat_200k/tree/refs%2Fconvert%2Fparquet/default/train
The `train` split existed, but has been deleted. Now, the datasets has `train_sft` and `train_gen`. But the duckdb index file still exists for `train` in `refs/convert/parquet`. Note that the parquet files have been deleted. | Delete duckdb files when a split is deleted: See https://huggingface.co/datasets/HuggingFaceH4/ultrachat_200k/tree/refs%2Fconvert%2Fparquet/default/train
The `train` split existed, but has been deleted. Now, the datasets has `train_sft` and `train_gen`. But the duckdb index file still exists for `train` in `refs/convert/parquet`. Note that the parquet files have been deleted. | closed | 2023-11-03T16:33:43Z | 2024-01-19T14:40:11Z | 2024-01-19T14:12:47Z | severo |
1,976,529,844 | Support JWT on cookies | private conversion: https://huggingface.slack.com/archives/D030YA5BW91/p1696507761676679
When the users goes to https://huggingface.co/datasets/emrgnt-cmplxty/sciphi-textbooks-are-all-you-need, moonlanding can put a cookie on `datasets-server.huggingface.co` with name `hf_jwt_[sha256(/datasets/emrgnt-cmplxty/sciphi-textbooks-are-all-you-need)]` and with the JWT as the value
This cookie would be read on datasets server when accessing a gated dataset
Doing so would simplify a lot the code on the Hub (moonlanding) by removing the need to refresh the JWT (remove an endpoint), and avoid the logic in the frontend code that refreshes the JWT. It would be a security improvement too, because the Hub's frontend code (javascript) would no more have access to the JWT (the browser directly adds the cookie to the HTTP request) | Support JWT on cookies: private conversion: https://huggingface.slack.com/archives/D030YA5BW91/p1696507761676679
When the users goes to https://huggingface.co/datasets/emrgnt-cmplxty/sciphi-textbooks-are-all-you-need, moonlanding can put a cookie on `datasets-server.huggingface.co` with name `hf_jwt_[sha256(/datasets/emrgnt-cmplxty/sciphi-textbooks-are-all-you-need)]` and with the JWT as the value
This cookie would be read on datasets server when accessing a gated dataset
Doing so would simplify a lot the code on the Hub (moonlanding) by removing the need to refresh the JWT (remove an endpoint), and avoid the logic in the frontend code that refreshes the JWT. It would be a security improvement too, because the Hub's frontend code (javascript) would no more have access to the JWT (the browser directly adds the cookie to the HTTP request) | closed | 2023-11-03T16:17:20Z | 2024-03-13T09:49:15Z | 2024-03-13T09:49:14Z | severo |
1,976,239,331 | Update index.mdx | Add a sentence to highlight conversion to Parquet | Update index.mdx: Add a sentence to highlight conversion to Parquet | closed | 2023-11-03T13:44:58Z | 2023-11-03T13:51:41Z | 2023-11-03T13:51:10Z | severo |
1,976,215,817 | Should we support video datasets? | Like https://huggingface.co/datasets/commaai/commavq
There was a previous intent in datasets: https://github.com/huggingface/datasets/pull/5339 | Should we support video datasets?: Like https://huggingface.co/datasets/commaai/commavq
There was a previous intent in datasets: https://github.com/huggingface/datasets/pull/5339 | closed | 2023-11-03T13:33:00Z | 2023-12-11T15:04:08Z | 2023-12-11T15:04:08Z | severo |
1,975,982,656 | Retry jobs that finish with `ClientConnection` error? | Maybe here: https://github.com/huggingface/datasets-server/blob/f311a9212aaa91dd0373e5c2d4f5da9b6bdabcb5/chart/env/prod.yaml#L209
Internal conversation on Slack: https://huggingface.slack.com/archives/C0311GZ7R6K/p1698224875005729
Anyway: I'm wondering if we can have the error now that the dataset scripts are disabled by default. | Retry jobs that finish with `ClientConnection` error?: Maybe here: https://github.com/huggingface/datasets-server/blob/f311a9212aaa91dd0373e5c2d4f5da9b6bdabcb5/chart/env/prod.yaml#L209
Internal conversation on Slack: https://huggingface.slack.com/archives/C0311GZ7R6K/p1698224875005729
Anyway: I'm wondering if we can have the error now that the dataset scripts are disabled by default. | closed | 2023-11-03T11:28:19Z | 2024-02-06T17:29:45Z | 2024-02-06T17:29:45Z | severo |
1,975,936,480 | Improve selection of the next job to priorize datasets with few configs/splits | Internal Slack conversation: https://huggingface.slack.com/archives/C04L6P8KNQ5/p1698323643019099?thread_ts=1698323528.797149&cid=C04L6P8KNQ5
> `opus_euconst` is squatting the pods, and the small datasets are waiting for too long before being processed.
> maybe switch it to "low" prior ?
> It could be a good solution for these big datasets, indeed! If number of configs or splits is over some threshold -> change the priority to low.
> I like it, because it would not change our bottleneck query, which is to find the next job to process.
| Improve selection of the next job to priorize datasets with few configs/splits: Internal Slack conversation: https://huggingface.slack.com/archives/C04L6P8KNQ5/p1698323643019099?thread_ts=1698323528.797149&cid=C04L6P8KNQ5
> `opus_euconst` is squatting the pods, and the small datasets are waiting for too long before being processed.
> maybe switch it to "low" prior ?
> It could be a good solution for these big datasets, indeed! If number of configs or splits is over some threshold -> change the priority to low.
> I like it, because it would not change our bottleneck query, which is to find the next job to process.
| closed | 2023-11-03T10:59:19Z | 2024-02-06T17:04:07Z | 2024-02-06T17:04:06Z | severo |
1,975,908,066 | descriptive-statistics job is sometimes stuck | Conversation on Slack (internal) https://huggingface.slack.com/archives/C04L6P8KNQ5/p1698325340414399?thread_ts=1698323528.797149&cid=C04L6P8KNQ5
> Note that stats jobs get stuck very often for some reason, I think Polina was investigating
> e.g. this job has been stuck for 30min at the duckdb data ingestion step (only 22MB of data)
```
INFO: 2023-10-26 12:44:15,764 - root - [split-descriptive-statistics] compute JobManager(job_id=653a5e8e2a3bd4b73b210849 dataset=onuralp/open-otter job_info={'job_id': '653a5e8e2a3bd4b73b210849', 'type': 'split-descriptive-statistics', 'params': {'dataset': 'onuralp/open-otter', 'revision': '17db84f97d83d3d782869fa3767b812ba4f7e407', 'config': 'default', 'split': 'train'}, 'priority': <Priority.NORMAL: 'normal'>, 'difficulty': 70}
INFO: 2023-10-26 12:44:15,771 - root - Compute descriptive statistics for dataset='onuralp/open-otter', config='default', split='train'
INFO: 2023-10-26 12:44:15,774 - root - Downloading remote parquet files to a local directory /storage/stats-cache/89781749453102-split-descriptive-statistics-onuralp-open-otter-776f2760.
Downloading 0000.parquet: 100%|ββββββββββ| 22.3M/22.3M [00:01<00:00, 17.0MB/s]
INFO: 2023-10-26 12:47:00,397 - root - Original number of threads=2
INFO: 2023-10-26 12:47:00,398 - root - Current number of threads=8
INFO: 2023-10-26 12:47:00,398 - root - Original max_memory='105.7GB'
INFO: 2023-10-26 12:47:00,399 - root - Current max_memory='28.0GB'
INFO: 2023-10-26 12:47:00,399 - root - Loading data into in-memory table.
INFO: 2023-10-26 12:47:00,399 - root -
CREATE OR REPLACE TABLE data AS SELECT "input","output","instruction","data_source" FROM read_parquet('/storage/stats-cache/89781749453102-split-descriptive-statistics-onuralp-open-otter-776f2760/default/train/*.parquet');
```
> (running this duckdb command in a shell is instantaneous !!) (edited)
> (and the subsequent duckdb-index job took a few seconds)
| descriptive-statistics job is sometimes stuck: Conversation on Slack (internal) https://huggingface.slack.com/archives/C04L6P8KNQ5/p1698325340414399?thread_ts=1698323528.797149&cid=C04L6P8KNQ5
> Note that stats jobs get stuck very often for some reason, I think Polina was investigating
> e.g. this job has been stuck for 30min at the duckdb data ingestion step (only 22MB of data)
```
INFO: 2023-10-26 12:44:15,764 - root - [split-descriptive-statistics] compute JobManager(job_id=653a5e8e2a3bd4b73b210849 dataset=onuralp/open-otter job_info={'job_id': '653a5e8e2a3bd4b73b210849', 'type': 'split-descriptive-statistics', 'params': {'dataset': 'onuralp/open-otter', 'revision': '17db84f97d83d3d782869fa3767b812ba4f7e407', 'config': 'default', 'split': 'train'}, 'priority': <Priority.NORMAL: 'normal'>, 'difficulty': 70}
INFO: 2023-10-26 12:44:15,771 - root - Compute descriptive statistics for dataset='onuralp/open-otter', config='default', split='train'
INFO: 2023-10-26 12:44:15,774 - root - Downloading remote parquet files to a local directory /storage/stats-cache/89781749453102-split-descriptive-statistics-onuralp-open-otter-776f2760.
Downloading 0000.parquet: 100%|ββββββββββ| 22.3M/22.3M [00:01<00:00, 17.0MB/s]
INFO: 2023-10-26 12:47:00,397 - root - Original number of threads=2
INFO: 2023-10-26 12:47:00,398 - root - Current number of threads=8
INFO: 2023-10-26 12:47:00,398 - root - Original max_memory='105.7GB'
INFO: 2023-10-26 12:47:00,399 - root - Current max_memory='28.0GB'
INFO: 2023-10-26 12:47:00,399 - root - Loading data into in-memory table.
INFO: 2023-10-26 12:47:00,399 - root -
CREATE OR REPLACE TABLE data AS SELECT "input","output","instruction","data_source" FROM read_parquet('/storage/stats-cache/89781749453102-split-descriptive-statistics-onuralp-open-otter-776f2760/default/train/*.parquet');
```
> (running this duckdb command in a shell is instantaneous !!) (edited)
> (and the subsequent duckdb-index job took a few seconds)
| closed | 2023-11-03T10:44:55Z | 2024-07-30T17:07:22Z | 2024-07-30T17:07:22Z | severo |
1,975,787,190 | Fix Hub datasets after updating datasets to 2.14.6 | After deploying PR:
- #2007
We need to fix/refresh Hub datasets:
- [x] Raise `DefunctDatasetError` from datasets with taken down data
- [x] Refresh viewer of datasets with `metadata.csv`
- [x] 27 datasets, e.g. https://huggingface.co/datasets/lukarape/public_small_papers/discussions/1
- [x] Refresh viewer of datasets with 'data' word twice
- [x] 270 datasets, e.g. https://huggingface.co/datasets/piuba-bigdata/articles_and_comments/discussions/1 | Fix Hub datasets after updating datasets to 2.14.6: After deploying PR:
- #2007
We need to fix/refresh Hub datasets:
- [x] Raise `DefunctDatasetError` from datasets with taken down data
- [x] Refresh viewer of datasets with `metadata.csv`
- [x] 27 datasets, e.g. https://huggingface.co/datasets/lukarape/public_small_papers/discussions/1
- [x] Refresh viewer of datasets with 'data' word twice
- [x] 270 datasets, e.g. https://huggingface.co/datasets/piuba-bigdata/articles_and_comments/discussions/1 | closed | 2023-11-03T09:36:10Z | 2024-02-06T18:27:29Z | 2024-02-06T18:27:29Z | albertvillanova |
1,974,386,677 | Audio asset with pydub conversion fails to write on S3 | See https://huggingface.co/datasets/zeio/auto-pale/discussions/1 (ogg audio files)
```
File \"/src/libs/libcommon/src/libcommon/viewer_utils/asset.py\", line 111, in create_audio_file
segment.export(f, format=suffix[1:])
File \"/src/services/worker/.venv/lib/python3.9/site-packages/pydub/audio_segment.py\", line 868, in export
out_f.seek(0)
File \"/src/services/worker/.venv/lib/python3.9/site-packages/fsspec/spec.py\", line 1737, in seek
aise OSError(ESPIPE, \"Seek only available in read mode\")
OSError: [Errno 29] Seek only available in read mode
```
we might have to write to a BytesIO buffer before writing the audio actual audio asset file, since fsspec doesn't allow seek for wb files somehow. | Audio asset with pydub conversion fails to write on S3: See https://huggingface.co/datasets/zeio/auto-pale/discussions/1 (ogg audio files)
```
File \"/src/libs/libcommon/src/libcommon/viewer_utils/asset.py\", line 111, in create_audio_file
segment.export(f, format=suffix[1:])
File \"/src/services/worker/.venv/lib/python3.9/site-packages/pydub/audio_segment.py\", line 868, in export
out_f.seek(0)
File \"/src/services/worker/.venv/lib/python3.9/site-packages/fsspec/spec.py\", line 1737, in seek
aise OSError(ESPIPE, \"Seek only available in read mode\")
OSError: [Errno 29] Seek only available in read mode
```
we might have to write to a BytesIO buffer before writing the audio actual audio asset file, since fsspec doesn't allow seek for wb files somehow. | closed | 2023-11-02T14:30:10Z | 2023-11-16T12:45:39Z | 2023-11-16T12:45:39Z | lhoestq |
1,974,288,216 | Support ruff in vscode | See https://github.com/huggingface/datasets-server/pull/2031#issuecomment-1790455688 and https://github.com/huggingface/datasets-server/pull/2031#issuecomment-1790718911 | Support ruff in vscode: See https://github.com/huggingface/datasets-server/pull/2031#issuecomment-1790455688 and https://github.com/huggingface/datasets-server/pull/2031#issuecomment-1790718911 | closed | 2023-11-02T13:41:26Z | 2023-11-10T15:07:41Z | 2023-11-10T15:07:40Z | severo |
1,973,807,306 | Explain single quote string SQL syntax in filter docs | Explain single quote string SQL syntax in filter docs.
Close #2038. | Explain single quote string SQL syntax in filter docs: Explain single quote string SQL syntax in filter docs.
Close #2038. | closed | 2023-11-02T09:07:45Z | 2023-11-02T13:53:02Z | 2023-11-02T13:52:33Z | albertvillanova |
1,973,757,952 | Use hf-doc-builder version from GitHub to build docs locally | This PR aligns the `hf-doc-builder` version used to build docs locally with the one used by our CI: now both use the version in the main branch in the GitHub repository: https://github.com/huggingface/doc-builder
Note that before this PR, local docs building:
- used an old version of `hf-doc-builder` (0.3.0) released on Apr 29, 2022: https://pypi.org/project/hf-doc-builder/0.3.0/
- that version had a bug (it didn't install required package `black`), that we were avoiding by requiring the "quality" extra
Now:
- the above mentioned bug has been fixed in main branch; see:
- https://github.com/huggingface/doc-builder/pull/434
- we no longer need to install the "quality" extra, so no unnecessary packages are installed: `flake8`, `isort`, `mccabe`, `pycodestyle`, `pyflakes`,... | Use hf-doc-builder version from GitHub to build docs locally: This PR aligns the `hf-doc-builder` version used to build docs locally with the one used by our CI: now both use the version in the main branch in the GitHub repository: https://github.com/huggingface/doc-builder
Note that before this PR, local docs building:
- used an old version of `hf-doc-builder` (0.3.0) released on Apr 29, 2022: https://pypi.org/project/hf-doc-builder/0.3.0/
- that version had a bug (it didn't install required package `black`), that we were avoiding by requiring the "quality" extra
Now:
- the above mentioned bug has been fixed in main branch; see:
- https://github.com/huggingface/doc-builder/pull/434
- we no longer need to install the "quality" extra, so no unnecessary packages are installed: `flake8`, `isort`, `mccabe`, `pycodestyle`, `pyflakes`,... | closed | 2023-11-02T08:35:46Z | 2023-11-02T12:50:09Z | 2023-11-02T12:49:36Z | albertvillanova |
1,970,839,681 | Add information about filter feature to the Hub's docs on the Dataset Viewer | Maybe I didn't find it but I think it's missing, at least [here](https://huggingface.co/docs/hub/datasets-viewer) | Add information about filter feature to the Hub's docs on the Dataset Viewer: Maybe I didn't find it but I think it's missing, at least [here](https://huggingface.co/docs/hub/datasets-viewer) | closed | 2023-10-31T16:24:18Z | 2024-02-06T15:45:25Z | 2024-02-06T15:45:25Z | polinaeterna |
1,970,662,057 | remove assets and cached assets | Since /storage/assets and /storage/cached-assets are no more needed for staging and prod environments, I am removing all chart/k8s related artifacts:
- persistance claim
- containers
- volume mounts
- pvc | remove assets and cached assets: Since /storage/assets and /storage/cached-assets are no more needed for staging and prod environments, I am removing all chart/k8s related artifacts:
- persistance claim
- containers
- volume mounts
- pvc | closed | 2023-10-31T14:58:06Z | 2023-11-06T14:13:24Z | 2023-11-06T14:13:23Z | AndreaFrancis |
1,970,262,386 | Revert "Revert assets fsspec (#2036)" | This reverts commit 5716638a683d83652da672f0f59bf68935d0a93e. | Revert "Revert assets fsspec (#2036)": This reverts commit 5716638a683d83652da672f0f59bf68935d0a93e. | closed | 2023-10-31T11:36:27Z | 2023-10-31T12:13:58Z | 2023-10-31T12:13:57Z | AndreaFrancis |
1,969,315,934 | How to pass single quote in /filter endpoint "where" parameter? | See `https://huggingface.co/datasets/albertvillanova/lm_en_dummy2/viewer/default/train?f[meta][value]='{'file': 'file_4.txt'}'`
From `https://datasets-server.huggingface.co/filter?dataset=albertvillanova/lm_en_dummy2&config=default&split=train&where=meta='{'file': 'file_4.txt'}'`, we get:
```
{"error":"Parameter 'where' is invalid"}
```
We want to search he value `{'file': 'file_4.txt'}` in the column `meta`
| How to pass single quote in /filter endpoint "where" parameter?: See `https://huggingface.co/datasets/albertvillanova/lm_en_dummy2/viewer/default/train?f[meta][value]='{'file': 'file_4.txt'}'`
From `https://datasets-server.huggingface.co/filter?dataset=albertvillanova/lm_en_dummy2&config=default&split=train&where=meta='{'file': 'file_4.txt'}'`, we get:
```
{"error":"Parameter 'where' is invalid"}
```
We want to search he value `{'file': 'file_4.txt'}` in the column `meta`
| closed | 2023-10-30T22:21:24Z | 2023-11-02T17:22:54Z | 2023-11-02T13:52:34Z | severo |
1,969,160,211 | feat: πΈ allow dataset script on togethercomputer/RedPajama-Data | null | feat: πΈ allow dataset script on togethercomputer/RedPajama-Data: | closed | 2023-10-30T20:29:46Z | 2023-10-30T20:36:23Z | 2023-10-30T20:36:22Z | severo |
1,969,087,295 | Revert assets fsspec | Once deployed in staging env, it looks like URL path resolve is not working for assets and cached-assets, will revert related PRs.
For example image https://datasets-server.us.dev.moon.huggingface.tech/assets/asoria/bluey/--/2479b12a83d3ac64cecf3ef[β¦]3b2fbf88fb1/--/default/train/0/image/image.png](https://datasets-server.us.dev.moon.huggingface.tech/assets/asoria/bluey/--/2479b12a83d3ac64cecf3ef1c33cc3b2fbf88fb1/--/default/train/0/image/image.png cant be found. | Revert assets fsspec: Once deployed in staging env, it looks like URL path resolve is not working for assets and cached-assets, will revert related PRs.
For example image https://datasets-server.us.dev.moon.huggingface.tech/assets/asoria/bluey/--/2479b12a83d3ac64cecf3ef[β¦]3b2fbf88fb1/--/default/train/0/image/image.png](https://datasets-server.us.dev.moon.huggingface.tech/assets/asoria/bluey/--/2479b12a83d3ac64cecf3ef1c33cc3b2fbf88fb1/--/default/train/0/image/image.png cant be found. | closed | 2023-10-30T19:40:34Z | 2023-10-30T20:09:42Z | 2023-10-30T20:09:41Z | AndreaFrancis |
1,969,025,760 | Update pip to 23.3.1 to fix vulnerability | Update `pip` to 23.3.1 to fix vulnerability.
See: https://github.com/huggingface/datasets-server/actions/runs/6697364484/job/18197081786?pr=2031
```
Found 1 known vulnerability in 1 package
Name Version ID Fix Versions
---- ------- ------------------- ------------
pip 23.1.2 GHSA-mq26-g339-26xf 23.3
``` | Update pip to 23.3.1 to fix vulnerability: Update `pip` to 23.3.1 to fix vulnerability.
See: https://github.com/huggingface/datasets-server/actions/runs/6697364484/job/18197081786?pr=2031
```
Found 1 known vulnerability in 1 package
Name Version ID Fix Versions
---- ------- ------------------- ------------
pip 23.1.2 GHSA-mq26-g339-26xf 23.3
``` | closed | 2023-10-30T18:59:39Z | 2023-10-30T19:45:04Z | 2023-10-30T19:45:03Z | albertvillanova |
1,968,937,161 | Install `ps` and `htop` in workers to monitor processes inside pods | also added them to other services in case we ever need them | Install `ps` and `htop` in workers to monitor processes inside pods: also added them to other services in case we ever need them | closed | 2023-10-30T18:01:15Z | 2023-10-31T15:33:13Z | 2023-10-31T15:33:12Z | polinaeterna |
1,968,658,253 | Add S3 configs to admin | null | Add S3 configs to admin: | closed | 2023-10-30T15:33:27Z | 2023-10-30T15:34:20Z | 2023-10-30T15:34:18Z | AndreaFrancis |
1,968,607,482 | Fix assets volumes | null | Fix assets volumes: | closed | 2023-10-30T15:10:35Z | 2023-10-30T15:14:39Z | 2023-10-30T15:14:38Z | AndreaFrancis |
1,968,553,996 | Use ruff for CI quality | Use `ruff` for CI quality: 10-100x faster than existing linters (like Flake8) and formatters (like Black).
Note this is most useful when developing: `make style` now is much faster and additionally runs the linter (before `flake8` was only called by `make quality`). | Use ruff for CI quality: Use `ruff` for CI quality: 10-100x faster than existing linters (like Flake8) and formatters (like Black).
Note this is most useful when developing: `make style` now is much faster and additionally runs the linter (before `flake8` was only called by `make quality`). | closed | 2023-10-30T14:47:14Z | 2023-11-02T14:01:49Z | 2023-11-02T14:01:48Z | albertvillanova |
1,967,941,143 | Support async file I/O to get duckdb index file | Support asynchronous file I/O to get duckdb index file by replacing `pathlib.Path` with `anyio.Path`. | Support async file I/O to get duckdb index file: Support asynchronous file I/O to get duckdb index file by replacing `pathlib.Path` with `anyio.Path`. | closed | 2023-10-30T09:50:48Z | 2023-10-30T12:34:09Z | 2023-10-30T12:34:08Z | albertvillanova |
1,967,882,566 | Update pandas dependency to 2.0.3 | Update `pandas` dependency to latest patch release 2.0.3 in 2.0 series: https://github.com/pandas-dev/pandas/releases/tag/v2.0.3
> This is a patch release in the 2.0.x series and includes some regression and bug fixes. We recommend that all users upgrade to this version.
Note that most components had `pandas` 2.0.2.
Also note that we have respected the constrain `<2.1`, until we update `duckdb` as well. More details in:
- #1914 | Update pandas dependency to 2.0.3: Update `pandas` dependency to latest patch release 2.0.3 in 2.0 series: https://github.com/pandas-dev/pandas/releases/tag/v2.0.3
> This is a patch release in the 2.0.x series and includes some regression and bug fixes. We recommend that all users upgrade to this version.
Note that most components had `pandas` 2.0.2.
Also note that we have respected the constrain `<2.1`, until we update `duckdb` as well. More details in:
- #1914 | closed | 2023-10-30T09:19:08Z | 2023-10-30T10:02:19Z | 2023-10-30T10:02:18Z | albertvillanova |
1,966,126,224 | adding tests for recreate dataset | Adding test for recreate dataset in admin service.
Also found a bug for cancel_jobs, it was returning always 0 after the update statement.
| adding tests for recreate dataset: Adding test for recreate dataset in admin service.
Also found a bug for cancel_jobs, it was returning always 0 after the update statement.
| closed | 2023-10-27T20:04:15Z | 2023-10-30T11:30:21Z | 2023-10-30T11:30:20Z | AndreaFrancis |
1,965,849,698 | fix duckdb circular import | When deploying to staging (and also locally), the following error appeared for /search service:
| fix duckdb circular import: When deploying to staging (and also locally), the following error appeared for /search service:
| closed | 2023-10-27T16:35:20Z | 2023-10-27T17:45:23Z | 2023-10-27T16:59:55Z | AndreaFrancis |
1,965,756,311 | Make create_response async and run _transform_row map async in separate thread | Make `create_response` (and inner functions) asynchronous and run `_transform_row` map asynchronously in separate thread. | Make create_response async and run _transform_row map async in separate thread: Make `create_response` (and inner functions) asynchronous and run `_transform_row` map asynchronously in separate thread. | closed | 2023-10-27T15:28:21Z | 2023-10-31T16:03:25Z | 2023-10-31T16:03:23Z | albertvillanova |
1,965,475,341 | 2023 reduce number comments | fixes #2023
note: `__PARQUET_CONVERTER_USER__` will be shown as `parquet-converter` in prod, without the bold formatting (due to the `__` characters).
<img width="597" alt="Capture dβeΜcran 2023-10-27 aΜ 14 53 26" src="https://github.com/huggingface/datasets-server/assets/1676121/df164999-709c-4afe-b825-c547ce7d4ca6">
---
<img width="767" alt="Capture dβeΜcran 2023-10-27 aΜ 15 01 00" src="https://github.com/huggingface/datasets-server/assets/1676121/08c3e0fb-a8b1-4950-ae88-b21bf0797d60">
| 2023 reduce number comments: fixes #2023
note: `__PARQUET_CONVERTER_USER__` will be shown as `parquet-converter` in prod, without the bold formatting (due to the `__` characters).
<img width="597" alt="Capture dβeΜcran 2023-10-27 aΜ 14 53 26" src="https://github.com/huggingface/datasets-server/assets/1676121/df164999-709c-4afe-b825-c547ce7d4ca6">
---
<img width="767" alt="Capture dβeΜcran 2023-10-27 aΜ 15 01 00" src="https://github.com/huggingface/datasets-server/assets/1676121/08c3e0fb-a8b1-4950-ae88-b21bf0797d60">
| closed | 2023-10-27T12:56:59Z | 2023-10-30T20:57:12Z | 2023-10-30T20:57:11Z | severo |
1,965,382,173 | Reduce the amount of comments after parquet conversion | See https://huggingface.co/datasets/open-source-metrics/datasets-dependents/discussions/1#6537d488b33b913282d076de
> Would it make sense to just update the original message to add the different revisions that have been converted?
> I agree that editing original might be better. Translating this same situation to GitHub, I would not be happy with a bot that pings for every single commit. For example, the doc-builder bot does not write a comment for every commit in a PR, it just does a single comment and link will work. I don't think most users will want to get a new notification for every single commit. We can also not silence notifications at a repo level, so as a user i would be inclined to silence the whole org.
Let's just create a discussion that points to the refs/convert/parquet page, and ignore if a discussion already exists. | Reduce the amount of comments after parquet conversion: See https://huggingface.co/datasets/open-source-metrics/datasets-dependents/discussions/1#6537d488b33b913282d076de
> Would it make sense to just update the original message to add the different revisions that have been converted?
> I agree that editing original might be better. Translating this same situation to GitHub, I would not be happy with a bot that pings for every single commit. For example, the doc-builder bot does not write a comment for every commit in a PR, it just does a single comment and link will work. I don't think most users will want to get a new notification for every single commit. We can also not silence notifications at a repo level, so as a user i would be inclined to silence the whole org.
Let's just create a discussion that points to the refs/convert/parquet page, and ignore if a discussion already exists. | closed | 2023-10-27T12:03:42Z | 2023-10-30T20:57:12Z | 2023-10-30T20:57:12Z | severo |
1,965,321,728 | fix: π do not set the JWT public key URL by default | it's only useful for the Hugging Face production datasets-server, since any other instance would not receive requests from the Hugging Face Hub.
Fixes #2019 | fix: π do not set the JWT public key URL by default: it's only useful for the Hugging Face production datasets-server, since any other instance would not receive requests from the Hugging Face Hub.
Fixes #2019 | closed | 2023-10-27T11:27:22Z | 2023-10-27T11:28:06Z | 2023-10-27T11:28:05Z | severo |
1,965,181,589 | Remove duckdb as libapi dependency | Remove `duckdb` from libapi by moving the function `duckdb_connect` to the "search" service.
In the future, if this function is required by other components, I would suggest to create a new duckdb-specific lib, so that we decouple API and DuckDB functionalities.
Additionally update all components that depend on `libapi` and remove their dependency on `duckdb` if they do not need it. These are:
- admin
- api
- rows
- sse-api | Remove duckdb as libapi dependency: Remove `duckdb` from libapi by moving the function `duckdb_connect` to the "search" service.
In the future, if this function is required by other components, I would suggest to create a new duckdb-specific lib, so that we decouple API and DuckDB functionalities.
Additionally update all components that depend on `libapi` and remove their dependency on `duckdb` if they do not need it. These are:
- admin
- api
- rows
- sse-api | closed | 2023-10-27T10:01:00Z | 2023-10-27T11:19:16Z | 2023-10-27T11:19:15Z | albertvillanova |
1,965,046,181 | Update duckdb minor version to 0.8.1 in admin_ui | Update `duckdb` dependency minor version to 0.8.1 in admin_ui, which is a patch version that includes only bug fixes: https://github.com/duckdb/duckdb/releases/tag/v0.8.1 | Update duckdb minor version to 0.8.1 in admin_ui: Update `duckdb` dependency minor version to 0.8.1 in admin_ui, which is a patch version that includes only bug fixes: https://github.com/duckdb/duckdb/releases/tag/v0.8.1 | closed | 2023-10-27T08:39:57Z | 2023-10-27T09:20:52Z | 2023-10-27T09:20:51Z | albertvillanova |
1,964,976,711 | Some jwt error after `make start` the newest dataset-server | My system is Windows10 and use the newest version of Docker Desktop, which the wsl engine was go on to execute docker together with WSL2. My WSL2 system is Ubuntu-22.04, and I cloned the newest version of datasets-server in it. Then I ran `make start` to build the server and successfully execute and can be shown in Docker Desktop dashboard.
When i saw the log about them, I found the worker docker: api, rows and search docker occred the same error as follows, which make the proxy docker receive 502 when i requests them:
``` bash
2023-10-27 15:11:50 INFO: Stopping parent process [1]
2023-10-27 15:11:52 INFO: Uvicorn running on http://0.0.0.0:8180 (Press CTRL+C to quit)
2023-10-27 15:11:52 INFO: Started parent process [1]
2023-10-27 15:11:53 Process SpawnProcess-2:
2023-10-27 15:11:53 Traceback (most recent call last):
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpcore/_exceptions.py", line 10, in map_exceptions
2023-10-27 15:11:53 yield
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpcore/_backends/sync.py", line 174, in start_tls
2023-10-27 15:11:53 raise exc
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpcore/_backends/sync.py", line 169, in start_tls
2023-10-27 15:11:53 sock = ssl_context.wrap_socket(
2023-10-27 15:11:53 File "/usr/local/lib/python3.9/ssl.py", line 501, in wrap_socket
2023-10-27 15:11:53 return self.sslsocket_class._create(
2023-10-27 15:11:53 File "/usr/local/lib/python3.9/ssl.py", line 1041, in _create
2023-10-27 15:11:53 self.do_handshake()
2023-10-27 15:11:53 File "/usr/local/lib/python3.9/ssl.py", line 1310, in do_handshake
2023-10-27 15:11:53 self._sslobj.do_handshake()
2023-10-27 15:11:53 socket.timeout: _ssl.c:1112: The handshake operation timed out
2023-10-27 15:11:53
2023-10-27 15:11:53 The above exception was the direct cause of the following exception:
2023-10-27 15:11:53
2023-10-27 15:11:53 Traceback (most recent call last):
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpx/_transports/default.py", line 66, in map_httpcore_exceptions
2023-10-27 15:11:53 yield
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpx/_transports/default.py", line 228, in handle_request
2023-10-27 15:11:53 resp = self._pool.handle_request(req)
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpcore/_sync/connection_pool.py", line 262, in handle_request
2023-10-27 15:11:53 raise exc
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpcore/_sync/connection_pool.py", line 245, in handle_request
2023-10-27 15:11:53 response = connection.handle_request(request)
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpcore/_sync/connection.py", line 99, in handle_request
2023-10-27 15:11:53 raise exc
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpcore/_sync/connection.py", line 76, in handle_request
2023-10-27 15:11:53 stream = self._connect(request)
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpcore/_sync/connection.py", line 156, in _connect
2023-10-27 15:11:53 stream = stream.start_tls(**kwargs)
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpcore/_backends/sync.py", line 174, in start_tls
2023-10-27 15:11:53 raise exc
2023-10-27 15:11:53 File "/usr/local/lib/python3.9/contextlib.py", line 137, in __exit__
2023-10-27 15:11:53 self.gen.throw(typ, value, traceback)
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpcore/_exceptions.py", line 14, in map_exceptions
2023-10-27 15:11:53 raise to_exc(exc) from exc
2023-10-27 15:11:53 httpcore.ConnectTimeout: _ssl.c:1112: The handshake operation timed out
2023-10-27 15:11:53
2023-10-27 15:11:53 The above exception was the direct cause of the following exception:
2023-10-27 15:11:53
2023-10-27 15:11:53 Traceback (most recent call last):
2023-10-27 15:11:53 File "/src/libs/libapi/src/libapi/jwt_token.py", line 177, in fetch_jwt_public_key_json
2023-10-27 15:11:53 response = httpx.get(url, timeout=hf_timeout_seconds)
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpx/_api.py", line 189, in get
2023-10-27 15:11:53 return request(
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpx/_api.py", line 100, in request
2023-10-27 15:11:53 return client.request(
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpx/_client.py", line 814, in request
2023-10-27 15:11:53 return self.send(request, auth=auth, follow_redirects=follow_redirects)
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpx/_client.py", line 901, in send
2023-10-27 15:11:53 response = self._send_handling_auth(
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpx/_client.py", line 929, in _send_handling_auth
2023-10-27 15:11:53 response = self._send_handling_redirects(
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpx/_client.py", line 966, in _send_handling_redirects
2023-10-27 15:11:53 response = self._send_single_request(request)
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpx/_client.py", line 1002, in _send_single_request
2023-10-27 15:11:53 response = transport.handle_request(request)
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpx/_transports/default.py", line 228, in handle_request
2023-10-27 15:11:53 resp = self._pool.handle_request(req)
2023-10-27 15:11:53 File "/usr/local/lib/python3.9/contextlib.py", line 137, in __exit__
2023-10-27 15:11:53 self.gen.throw(typ, value, traceback)
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpx/_transports/default.py", line 83, in map_httpcore_exceptions
2023-10-27 15:11:53 raise mapped_exc(message) from exc
2023-10-27 15:11:53 httpx.ConnectTimeout: _ssl.c:1112: The handshake operation timed out
2023-10-27 15:11:53
2023-10-27 15:11:53 The above exception was the direct cause of the following exception:
2023-10-27 15:11:53
2023-10-27 15:11:53 Traceback (most recent call last):
2023-10-27 15:11:53 File "/src/libs/libapi/src/libapi/jwt_token.py", line 220, in get_jwt_public_keys
2023-10-27 15:11:53 payload = fetch_jwt_public_key_json(
2023-10-27 15:11:53 File "/src/libs/libapi/src/libapi/jwt_token.py", line 181, in fetch_jwt_public_key_json
2023-10-27 15:11:53 raise RuntimeError(f"Failed to fetch the JWT public key from {url}. ") from err
2023-10-27 15:11:53 RuntimeError: Failed to fetch the JWT public key from https://huggingface.co/api/keys/jwt.
2023-10-27 15:11:53
2023-10-27 15:11:53 The above exception was the direct cause of the following exception:
2023-10-27 15:11:53
2023-10-27 15:11:53 Traceback (most recent call last):
2023-10-27 15:11:53 File "/usr/local/lib/python3.9/multiprocessing/process.py", line 315, in _bootstrap
2023-10-27 15:11:53 self.run()
2023-10-27 15:11:53 File "/usr/local/lib/python3.9/multiprocessing/process.py", line 108, in run
2023-10-27 15:11:53 self._target(*self._args, **self._kwargs)
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/uvicorn/_subprocess.py", line 76, in subprocess_started
2023-10-27 15:11:53 target(sockets=sockets)
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/uvicorn/server.py", line 60, in run
2023-10-27 15:11:53 return asyncio.run(self.serve(sockets=sockets))
2023-10-27 15:11:53 File "/usr/local/lib/python3.9/asyncio/runners.py", line 44, in run
2023-10-27 15:11:53 return loop.run_until_complete(main)
2023-10-27 15:11:53 File "/usr/local/lib/python3.9/asyncio/base_events.py", line 647, in run_until_complete
2023-10-27 15:11:53 return future.result()
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/uvicorn/server.py", line 67, in serve
2023-10-27 15:11:53 config.load()
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/uvicorn/config.py", line 483, in load
2023-10-27 15:11:53 self.loaded_app = self.loaded_app()
2023-10-27 15:11:53 File "/src/services/api/src/api/app.py", line 28, in create_app
2023-10-27 15:11:53 return create_app_with_config(app_config=app_config, endpoint_config=endpoint_config)
2023-10-27 15:11:53 File "/src/services/api/src/api/app.py", line 37, in create_app_with_config
2023-10-27 15:11:53 hf_jwt_public_keys = get_jwt_public_keys(
2023-10-27 15:11:53 File "/src/libs/libapi/src/libapi/jwt_token.py", line 232, in get_jwt_public_keys
2023-10-27 15:11:53 raise JWTKeysError("Failed to create the JWT public keys.") from err
2023-10-27 15:11:53 libapi.exceptions.JWTKeysError: Failed to create the JWT public keys.
2023-10-27 15:11:53 Process SpawnProcess-1:
2023-10-27 15:11:53 Traceback (most recent call last):
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpcore/_exceptions.py", line 10, in map_exceptions
2023-10-27 15:11:53 yield
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpcore/_backends/sync.py", line 174, in start_tls
2023-10-27 15:11:53 raise exc
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpcore/_backends/sync.py", line 169, in start_tls
2023-10-27 15:11:53 sock = ssl_context.wrap_socket(
2023-10-27 15:11:53 File "/usr/local/lib/python3.9/ssl.py", line 501, in wrap_socket
2023-10-27 15:11:53 return self.sslsocket_class._create(
2023-10-27 15:11:53 File "/usr/local/lib/python3.9/ssl.py", line 1041, in _create
2023-10-27 15:11:53 self.do_handshake()
2023-10-27 15:11:53 File "/usr/local/lib/python3.9/ssl.py", line 1310, in do_handshake
2023-10-27 15:11:53 self._sslobj.do_handshake()
2023-10-27 15:11:53 socket.timeout: _ssl.c:1112: The handshake operation timed out
2023-10-27 15:11:53
2023-10-27 15:11:53 The above exception was the direct cause of the following exception:
2023-10-27 15:11:53
2023-10-27 15:11:53 Traceback (most recent call last):
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpx/_transports/default.py", line 66, in map_httpcore_exceptions
2023-10-27 15:11:53 yield
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpx/_transports/default.py", line 228, in handle_request
2023-10-27 15:11:53 resp = self._pool.handle_request(req)
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpcore/_sync/connection_pool.py", line 262, in handle_request
2023-10-27 15:11:53 raise exc
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpcore/_sync/connection_pool.py", line 245, in handle_request
2023-10-27 15:11:53 response = connection.handle_request(request)
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpcore/_sync/connection.py", line 99, in handle_request
2023-10-27 15:11:53 raise exc
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpcore/_sync/connection.py", line 76, in handle_request
2023-10-27 15:11:53 stream = self._connect(request)
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpcore/_sync/connection.py", line 156, in _connect
2023-10-27 15:11:53 stream = stream.start_tls(**kwargs)
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpcore/_backends/sync.py", line 174, in start_tls
2023-10-27 15:11:53 raise exc
2023-10-27 15:11:53 File "/usr/local/lib/python3.9/contextlib.py", line 137, in __exit__
2023-10-27 15:11:53 self.gen.throw(typ, value, traceback)
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpcore/_exceptions.py", line 14, in map_exceptions
2023-10-27 15:11:53 raise to_exc(exc) from exc
2023-10-27 15:11:53 httpcore.ConnectTimeout: _ssl.c:1112: The handshake operation timed out
2023-10-27 15:11:53
2023-10-27 15:11:53 The above exception was the direct cause of the following exception:
2023-10-27 15:11:53
2023-10-27 15:11:53 Traceback (most recent call last):
2023-10-27 15:11:53 File "/src/libs/libapi/src/libapi/jwt_token.py", line 177, in fetch_jwt_public_key_json
2023-10-27 15:11:53 response = httpx.get(url, timeout=hf_timeout_seconds)
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpx/_api.py", line 189, in get
2023-10-27 15:11:53 return request(
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpx/_api.py", line 100, in request
2023-10-27 15:11:53 return client.request(
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpx/_client.py", line 814, in request
2023-10-27 15:11:53 return self.send(request, auth=auth, follow_redirects=follow_redirects)
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpx/_client.py", line 901, in send
2023-10-27 15:11:53 response = self._send_handling_auth(
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpx/_client.py", line 929, in _send_handling_auth
2023-10-27 15:11:53 response = self._send_handling_redirects(
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpx/_client.py", line 966, in _send_handling_redirects
2023-10-27 15:11:53 response = self._send_single_request(request)
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpx/_client.py", line 1002, in _send_single_request
2023-10-27 15:11:53 response = transport.handle_request(request)
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpx/_transports/default.py", line 228, in handle_request
2023-10-27 15:11:53 resp = self._pool.handle_request(req)
2023-10-27 15:11:53 File "/usr/local/lib/python3.9/contextlib.py", line 137, in __exit__
2023-10-27 15:11:53 self.gen.throw(typ, value, traceback)
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpx/_transports/default.py", line 83, in map_httpcore_exceptions
2023-10-27 15:11:53 raise mapped_exc(message) from exc
2023-10-27 15:11:53 httpx.ConnectTimeout: _ssl.c:1112: The handshake operation timed out
2023-10-27 15:11:53
2023-10-27 15:11:53 The above exception was the direct cause of the following exception:
2023-10-27 15:11:53
2023-10-27 15:11:53 Traceback (most recent call last):
2023-10-27 15:11:53 File "/src/libs/libapi/src/libapi/jwt_token.py", line 220, in get_jwt_public_keys
2023-10-27 15:11:53 payload = fetch_jwt_public_key_json(
2023-10-27 15:11:53 File "/src/libs/libapi/src/libapi/jwt_token.py", line 181, in fetch_jwt_public_key_json
2023-10-27 15:11:53 raise RuntimeError(f"Failed to fetch the JWT public key from {url}. ") from err
2023-10-27 15:11:53 RuntimeError: Failed to fetch the JWT public key from https://huggingface.co/api/keys/jwt.
2023-10-27 15:11:53
2023-10-27 15:11:53 The above exception was the direct cause of the following exception:
2023-10-27 15:11:53
2023-10-27 15:11:53 Traceback (most recent call last):
2023-10-27 15:11:53 File "/usr/local/lib/python3.9/multiprocessing/process.py", line 315, in _bootstrap
2023-10-27 15:11:53 self.run()
2023-10-27 15:11:53 File "/usr/local/lib/python3.9/multiprocessing/process.py", line 108, in run
2023-10-27 15:11:53 self._target(*self._args, **self._kwargs)
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/uvicorn/_subprocess.py", line 76, in subprocess_started
2023-10-27 15:11:53 target(sockets=sockets)
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/uvicorn/server.py", line 60, in run
2023-10-27 15:11:53 return asyncio.run(self.serve(sockets=sockets))
2023-10-27 15:11:53 File "/usr/local/lib/python3.9/asyncio/runners.py", line 44, in run
2023-10-27 15:11:53 return loop.run_until_complete(main)
2023-10-27 15:11:53 File "/usr/local/lib/python3.9/asyncio/base_events.py", line 647, in run_until_complete
2023-10-27 15:11:53 return future.result()
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/uvicorn/server.py", line 67, in serve
2023-10-27 15:11:53 config.load()
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/uvicorn/config.py", line 483, in load
2023-10-27 15:11:53 self.loaded_app = self.loaded_app()
2023-10-27 15:11:53 File "/src/services/api/src/api/app.py", line 28, in create_app
2023-10-27 15:11:53 return create_app_with_config(app_config=app_config, endpoint_config=endpoint_config)
2023-10-27 15:11:53 File "/src/services/api/src/api/app.py", line 37, in create_app_with_config
2023-10-27 15:11:53 hf_jwt_public_keys = get_jwt_public_keys(
2023-10-27 15:11:53 File "/src/libs/libapi/src/libapi/jwt_token.py", line 232, in get_jwt_public_keys
2023-10-27 15:11:53 raise JWTKeysError("Failed to create the JWT public keys.") from err
2023-10-27 15:11:53 libapi.exceptions.JWTKeysError: Failed to create the JWT public keys.
```
At first I thought the main error is network and i use `ping huggingface.co` and `curl https://huggingface.co/api/keys/jwt` in these dockers, but I had 200 request later, which means the network in dockers are not the main reason cause the prolems.
Can anyone give a solutions or mothod to solve this bug, thanks ! | Some jwt error after `make start` the newest dataset-server: My system is Windows10 and use the newest version of Docker Desktop, which the wsl engine was go on to execute docker together with WSL2. My WSL2 system is Ubuntu-22.04, and I cloned the newest version of datasets-server in it. Then I ran `make start` to build the server and successfully execute and can be shown in Docker Desktop dashboard.
When i saw the log about them, I found the worker docker: api, rows and search docker occred the same error as follows, which make the proxy docker receive 502 when i requests them:
``` bash
2023-10-27 15:11:50 INFO: Stopping parent process [1]
2023-10-27 15:11:52 INFO: Uvicorn running on http://0.0.0.0:8180 (Press CTRL+C to quit)
2023-10-27 15:11:52 INFO: Started parent process [1]
2023-10-27 15:11:53 Process SpawnProcess-2:
2023-10-27 15:11:53 Traceback (most recent call last):
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpcore/_exceptions.py", line 10, in map_exceptions
2023-10-27 15:11:53 yield
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpcore/_backends/sync.py", line 174, in start_tls
2023-10-27 15:11:53 raise exc
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpcore/_backends/sync.py", line 169, in start_tls
2023-10-27 15:11:53 sock = ssl_context.wrap_socket(
2023-10-27 15:11:53 File "/usr/local/lib/python3.9/ssl.py", line 501, in wrap_socket
2023-10-27 15:11:53 return self.sslsocket_class._create(
2023-10-27 15:11:53 File "/usr/local/lib/python3.9/ssl.py", line 1041, in _create
2023-10-27 15:11:53 self.do_handshake()
2023-10-27 15:11:53 File "/usr/local/lib/python3.9/ssl.py", line 1310, in do_handshake
2023-10-27 15:11:53 self._sslobj.do_handshake()
2023-10-27 15:11:53 socket.timeout: _ssl.c:1112: The handshake operation timed out
2023-10-27 15:11:53
2023-10-27 15:11:53 The above exception was the direct cause of the following exception:
2023-10-27 15:11:53
2023-10-27 15:11:53 Traceback (most recent call last):
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpx/_transports/default.py", line 66, in map_httpcore_exceptions
2023-10-27 15:11:53 yield
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpx/_transports/default.py", line 228, in handle_request
2023-10-27 15:11:53 resp = self._pool.handle_request(req)
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpcore/_sync/connection_pool.py", line 262, in handle_request
2023-10-27 15:11:53 raise exc
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpcore/_sync/connection_pool.py", line 245, in handle_request
2023-10-27 15:11:53 response = connection.handle_request(request)
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpcore/_sync/connection.py", line 99, in handle_request
2023-10-27 15:11:53 raise exc
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpcore/_sync/connection.py", line 76, in handle_request
2023-10-27 15:11:53 stream = self._connect(request)
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpcore/_sync/connection.py", line 156, in _connect
2023-10-27 15:11:53 stream = stream.start_tls(**kwargs)
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpcore/_backends/sync.py", line 174, in start_tls
2023-10-27 15:11:53 raise exc
2023-10-27 15:11:53 File "/usr/local/lib/python3.9/contextlib.py", line 137, in __exit__
2023-10-27 15:11:53 self.gen.throw(typ, value, traceback)
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpcore/_exceptions.py", line 14, in map_exceptions
2023-10-27 15:11:53 raise to_exc(exc) from exc
2023-10-27 15:11:53 httpcore.ConnectTimeout: _ssl.c:1112: The handshake operation timed out
2023-10-27 15:11:53
2023-10-27 15:11:53 The above exception was the direct cause of the following exception:
2023-10-27 15:11:53
2023-10-27 15:11:53 Traceback (most recent call last):
2023-10-27 15:11:53 File "/src/libs/libapi/src/libapi/jwt_token.py", line 177, in fetch_jwt_public_key_json
2023-10-27 15:11:53 response = httpx.get(url, timeout=hf_timeout_seconds)
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpx/_api.py", line 189, in get
2023-10-27 15:11:53 return request(
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpx/_api.py", line 100, in request
2023-10-27 15:11:53 return client.request(
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpx/_client.py", line 814, in request
2023-10-27 15:11:53 return self.send(request, auth=auth, follow_redirects=follow_redirects)
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpx/_client.py", line 901, in send
2023-10-27 15:11:53 response = self._send_handling_auth(
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpx/_client.py", line 929, in _send_handling_auth
2023-10-27 15:11:53 response = self._send_handling_redirects(
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpx/_client.py", line 966, in _send_handling_redirects
2023-10-27 15:11:53 response = self._send_single_request(request)
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpx/_client.py", line 1002, in _send_single_request
2023-10-27 15:11:53 response = transport.handle_request(request)
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpx/_transports/default.py", line 228, in handle_request
2023-10-27 15:11:53 resp = self._pool.handle_request(req)
2023-10-27 15:11:53 File "/usr/local/lib/python3.9/contextlib.py", line 137, in __exit__
2023-10-27 15:11:53 self.gen.throw(typ, value, traceback)
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpx/_transports/default.py", line 83, in map_httpcore_exceptions
2023-10-27 15:11:53 raise mapped_exc(message) from exc
2023-10-27 15:11:53 httpx.ConnectTimeout: _ssl.c:1112: The handshake operation timed out
2023-10-27 15:11:53
2023-10-27 15:11:53 The above exception was the direct cause of the following exception:
2023-10-27 15:11:53
2023-10-27 15:11:53 Traceback (most recent call last):
2023-10-27 15:11:53 File "/src/libs/libapi/src/libapi/jwt_token.py", line 220, in get_jwt_public_keys
2023-10-27 15:11:53 payload = fetch_jwt_public_key_json(
2023-10-27 15:11:53 File "/src/libs/libapi/src/libapi/jwt_token.py", line 181, in fetch_jwt_public_key_json
2023-10-27 15:11:53 raise RuntimeError(f"Failed to fetch the JWT public key from {url}. ") from err
2023-10-27 15:11:53 RuntimeError: Failed to fetch the JWT public key from https://huggingface.co/api/keys/jwt.
2023-10-27 15:11:53
2023-10-27 15:11:53 The above exception was the direct cause of the following exception:
2023-10-27 15:11:53
2023-10-27 15:11:53 Traceback (most recent call last):
2023-10-27 15:11:53 File "/usr/local/lib/python3.9/multiprocessing/process.py", line 315, in _bootstrap
2023-10-27 15:11:53 self.run()
2023-10-27 15:11:53 File "/usr/local/lib/python3.9/multiprocessing/process.py", line 108, in run
2023-10-27 15:11:53 self._target(*self._args, **self._kwargs)
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/uvicorn/_subprocess.py", line 76, in subprocess_started
2023-10-27 15:11:53 target(sockets=sockets)
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/uvicorn/server.py", line 60, in run
2023-10-27 15:11:53 return asyncio.run(self.serve(sockets=sockets))
2023-10-27 15:11:53 File "/usr/local/lib/python3.9/asyncio/runners.py", line 44, in run
2023-10-27 15:11:53 return loop.run_until_complete(main)
2023-10-27 15:11:53 File "/usr/local/lib/python3.9/asyncio/base_events.py", line 647, in run_until_complete
2023-10-27 15:11:53 return future.result()
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/uvicorn/server.py", line 67, in serve
2023-10-27 15:11:53 config.load()
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/uvicorn/config.py", line 483, in load
2023-10-27 15:11:53 self.loaded_app = self.loaded_app()
2023-10-27 15:11:53 File "/src/services/api/src/api/app.py", line 28, in create_app
2023-10-27 15:11:53 return create_app_with_config(app_config=app_config, endpoint_config=endpoint_config)
2023-10-27 15:11:53 File "/src/services/api/src/api/app.py", line 37, in create_app_with_config
2023-10-27 15:11:53 hf_jwt_public_keys = get_jwt_public_keys(
2023-10-27 15:11:53 File "/src/libs/libapi/src/libapi/jwt_token.py", line 232, in get_jwt_public_keys
2023-10-27 15:11:53 raise JWTKeysError("Failed to create the JWT public keys.") from err
2023-10-27 15:11:53 libapi.exceptions.JWTKeysError: Failed to create the JWT public keys.
2023-10-27 15:11:53 Process SpawnProcess-1:
2023-10-27 15:11:53 Traceback (most recent call last):
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpcore/_exceptions.py", line 10, in map_exceptions
2023-10-27 15:11:53 yield
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpcore/_backends/sync.py", line 174, in start_tls
2023-10-27 15:11:53 raise exc
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpcore/_backends/sync.py", line 169, in start_tls
2023-10-27 15:11:53 sock = ssl_context.wrap_socket(
2023-10-27 15:11:53 File "/usr/local/lib/python3.9/ssl.py", line 501, in wrap_socket
2023-10-27 15:11:53 return self.sslsocket_class._create(
2023-10-27 15:11:53 File "/usr/local/lib/python3.9/ssl.py", line 1041, in _create
2023-10-27 15:11:53 self.do_handshake()
2023-10-27 15:11:53 File "/usr/local/lib/python3.9/ssl.py", line 1310, in do_handshake
2023-10-27 15:11:53 self._sslobj.do_handshake()
2023-10-27 15:11:53 socket.timeout: _ssl.c:1112: The handshake operation timed out
2023-10-27 15:11:53
2023-10-27 15:11:53 The above exception was the direct cause of the following exception:
2023-10-27 15:11:53
2023-10-27 15:11:53 Traceback (most recent call last):
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpx/_transports/default.py", line 66, in map_httpcore_exceptions
2023-10-27 15:11:53 yield
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpx/_transports/default.py", line 228, in handle_request
2023-10-27 15:11:53 resp = self._pool.handle_request(req)
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpcore/_sync/connection_pool.py", line 262, in handle_request
2023-10-27 15:11:53 raise exc
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpcore/_sync/connection_pool.py", line 245, in handle_request
2023-10-27 15:11:53 response = connection.handle_request(request)
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpcore/_sync/connection.py", line 99, in handle_request
2023-10-27 15:11:53 raise exc
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpcore/_sync/connection.py", line 76, in handle_request
2023-10-27 15:11:53 stream = self._connect(request)
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpcore/_sync/connection.py", line 156, in _connect
2023-10-27 15:11:53 stream = stream.start_tls(**kwargs)
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpcore/_backends/sync.py", line 174, in start_tls
2023-10-27 15:11:53 raise exc
2023-10-27 15:11:53 File "/usr/local/lib/python3.9/contextlib.py", line 137, in __exit__
2023-10-27 15:11:53 self.gen.throw(typ, value, traceback)
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpcore/_exceptions.py", line 14, in map_exceptions
2023-10-27 15:11:53 raise to_exc(exc) from exc
2023-10-27 15:11:53 httpcore.ConnectTimeout: _ssl.c:1112: The handshake operation timed out
2023-10-27 15:11:53
2023-10-27 15:11:53 The above exception was the direct cause of the following exception:
2023-10-27 15:11:53
2023-10-27 15:11:53 Traceback (most recent call last):
2023-10-27 15:11:53 File "/src/libs/libapi/src/libapi/jwt_token.py", line 177, in fetch_jwt_public_key_json
2023-10-27 15:11:53 response = httpx.get(url, timeout=hf_timeout_seconds)
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpx/_api.py", line 189, in get
2023-10-27 15:11:53 return request(
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpx/_api.py", line 100, in request
2023-10-27 15:11:53 return client.request(
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpx/_client.py", line 814, in request
2023-10-27 15:11:53 return self.send(request, auth=auth, follow_redirects=follow_redirects)
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpx/_client.py", line 901, in send
2023-10-27 15:11:53 response = self._send_handling_auth(
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpx/_client.py", line 929, in _send_handling_auth
2023-10-27 15:11:53 response = self._send_handling_redirects(
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpx/_client.py", line 966, in _send_handling_redirects
2023-10-27 15:11:53 response = self._send_single_request(request)
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpx/_client.py", line 1002, in _send_single_request
2023-10-27 15:11:53 response = transport.handle_request(request)
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpx/_transports/default.py", line 228, in handle_request
2023-10-27 15:11:53 resp = self._pool.handle_request(req)
2023-10-27 15:11:53 File "/usr/local/lib/python3.9/contextlib.py", line 137, in __exit__
2023-10-27 15:11:53 self.gen.throw(typ, value, traceback)
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/httpx/_transports/default.py", line 83, in map_httpcore_exceptions
2023-10-27 15:11:53 raise mapped_exc(message) from exc
2023-10-27 15:11:53 httpx.ConnectTimeout: _ssl.c:1112: The handshake operation timed out
2023-10-27 15:11:53
2023-10-27 15:11:53 The above exception was the direct cause of the following exception:
2023-10-27 15:11:53
2023-10-27 15:11:53 Traceback (most recent call last):
2023-10-27 15:11:53 File "/src/libs/libapi/src/libapi/jwt_token.py", line 220, in get_jwt_public_keys
2023-10-27 15:11:53 payload = fetch_jwt_public_key_json(
2023-10-27 15:11:53 File "/src/libs/libapi/src/libapi/jwt_token.py", line 181, in fetch_jwt_public_key_json
2023-10-27 15:11:53 raise RuntimeError(f"Failed to fetch the JWT public key from {url}. ") from err
2023-10-27 15:11:53 RuntimeError: Failed to fetch the JWT public key from https://huggingface.co/api/keys/jwt.
2023-10-27 15:11:53
2023-10-27 15:11:53 The above exception was the direct cause of the following exception:
2023-10-27 15:11:53
2023-10-27 15:11:53 Traceback (most recent call last):
2023-10-27 15:11:53 File "/usr/local/lib/python3.9/multiprocessing/process.py", line 315, in _bootstrap
2023-10-27 15:11:53 self.run()
2023-10-27 15:11:53 File "/usr/local/lib/python3.9/multiprocessing/process.py", line 108, in run
2023-10-27 15:11:53 self._target(*self._args, **self._kwargs)
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/uvicorn/_subprocess.py", line 76, in subprocess_started
2023-10-27 15:11:53 target(sockets=sockets)
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/uvicorn/server.py", line 60, in run
2023-10-27 15:11:53 return asyncio.run(self.serve(sockets=sockets))
2023-10-27 15:11:53 File "/usr/local/lib/python3.9/asyncio/runners.py", line 44, in run
2023-10-27 15:11:53 return loop.run_until_complete(main)
2023-10-27 15:11:53 File "/usr/local/lib/python3.9/asyncio/base_events.py", line 647, in run_until_complete
2023-10-27 15:11:53 return future.result()
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/uvicorn/server.py", line 67, in serve
2023-10-27 15:11:53 config.load()
2023-10-27 15:11:53 File "/src/services/api/.venv/lib/python3.9/site-packages/uvicorn/config.py", line 483, in load
2023-10-27 15:11:53 self.loaded_app = self.loaded_app()
2023-10-27 15:11:53 File "/src/services/api/src/api/app.py", line 28, in create_app
2023-10-27 15:11:53 return create_app_with_config(app_config=app_config, endpoint_config=endpoint_config)
2023-10-27 15:11:53 File "/src/services/api/src/api/app.py", line 37, in create_app_with_config
2023-10-27 15:11:53 hf_jwt_public_keys = get_jwt_public_keys(
2023-10-27 15:11:53 File "/src/libs/libapi/src/libapi/jwt_token.py", line 232, in get_jwt_public_keys
2023-10-27 15:11:53 raise JWTKeysError("Failed to create the JWT public keys.") from err
2023-10-27 15:11:53 libapi.exceptions.JWTKeysError: Failed to create the JWT public keys.
```
At first I thought the main error is network and i use `ping huggingface.co` and `curl https://huggingface.co/api/keys/jwt` in these dockers, but I had 200 request later, which means the network in dockers are not the main reason cause the prolems.
Can anyone give a solutions or mothod to solve this bug, thanks ! | closed | 2023-10-27T07:57:38Z | 2023-10-27T11:28:07Z | 2023-10-27T11:28:06Z | vagitablebirdcode |
1,964,946,941 | Download duckdb index async in separate thread | Download `duckdb` index asynchronously in separate thread, and make `get_index_file_location_and_download_if_missing` asynchronous. | Download duckdb index async in separate thread: Download `duckdb` index asynchronously in separate thread, and make `get_index_file_location_and_download_if_missing` asynchronous. | closed | 2023-10-27T07:38:18Z | 2023-10-27T09:10:43Z | 2023-10-27T09:10:42Z | albertvillanova |
1,964,224,729 | Use get_request_parameter in endpoint | Refactor the "endpoint" route in the "api" service and use libapi `get_request_parameter`. | Use get_request_parameter in endpoint: Refactor the "endpoint" route in the "api" service and use libapi `get_request_parameter`. | closed | 2023-10-26T19:19:29Z | 2023-10-31T08:06:14Z | 2023-10-31T08:05:43Z | albertvillanova |
1,963,763,881 | add cron job to clean duckdb index cache | We have a cache folder in `/storage/duckdb-index`, this folder is used for cached downloads in `/search` and `/filter` while calling ` hf_hub_download` https://github.com/huggingface/datasets-server/blob/ef8a1cb34ba661b859962ac8d6c138374a0b0190/libs/libapi/src/libapi/duckdb.py#L78 (which needs a cache directory).
I noticed that we have old files in `/storage/duckdb-index/cache` which are using about 185G:
We can apply the same cleaning as we do for other folders so I added a new cron job.
I think another approach to avoid using `hf_hub_download` cache is to use `hf_transfer` directly, but I will investigate more about the advantages, meanwhile, the cron job will avoid increasing the storage usage.
| add cron job to clean duckdb index cache: We have a cache folder in `/storage/duckdb-index`, this folder is used for cached downloads in `/search` and `/filter` while calling ` hf_hub_download` https://github.com/huggingface/datasets-server/blob/ef8a1cb34ba661b859962ac8d6c138374a0b0190/libs/libapi/src/libapi/duckdb.py#L78 (which needs a cache directory).
I noticed that we have old files in `/storage/duckdb-index/cache` which are using about 185G:
We can apply the same cleaning as we do for other folders so I added a new cron job.
I think another approach to avoid using `hf_hub_download` cache is to use `hf_transfer` directly, but I will investigate more about the advantages, meanwhile, the cron job will avoid increasing the storage usage.
| closed | 2023-10-26T14:55:37Z | 2023-10-26T18:47:21Z | 2023-10-26T18:47:20Z | AndreaFrancis |
1,963,481,907 | increase number of replicas | Internal references:
API: https://huggingface.slack.com/archives/C04L6P8KNQ5/p1698224417526029
workers: https://huggingface.slack.com/archives/C04L6P8KNQ5/p1698323528797149 | increase number of replicas: Internal references:
API: https://huggingface.slack.com/archives/C04L6P8KNQ5/p1698224417526029
workers: https://huggingface.slack.com/archives/C04L6P8KNQ5/p1698323528797149 | closed | 2023-10-26T12:42:46Z | 2023-10-26T16:37:50Z | 2023-10-26T16:37:48Z | severo |
1,961,824,392 | Update werkzeug | fix
```
Name Version ID Fix Versions
-------- ------- ------------------- ------------
werkzeug 3.0.0 GHSA-hrfv-mqp8-q5rw 3.0.1
``` | Update werkzeug: fix
```
Name Version ID Fix Versions
-------- ------- ------------------- ------------
werkzeug 3.0.0 GHSA-hrfv-mqp8-q5rw 3.0.1
``` | closed | 2023-10-25T16:40:25Z | 2023-10-25T17:18:23Z | 2023-10-25T17:18:22Z | lhoestq |
1,961,471,039 | Factorize getting request parameters | Factorize getting request parameters for both required and non-required ones, by using the function `get_request_parameter` and merging `get_required_request_parameter` into it, with the addition of the parameters "required" and "default".
Use `get_request_parameter` to get the "cursor", "priority" and "all" parameters. | Factorize getting request parameters: Factorize getting request parameters for both required and non-required ones, by using the function `get_request_parameter` and merging `get_required_request_parameter` into it, with the addition of the parameters "required" and "default".
Use `get_request_parameter` to get the "cursor", "priority" and "all" parameters. | closed | 2023-10-25T13:45:54Z | 2023-10-26T17:35:22Z | 2023-10-26T17:35:21Z | albertvillanova |
1,961,360,943 | Update doc-build.yml - add missing token | Add missing token to doc build workflow | Update doc-build.yml - add missing token: Add missing token to doc build workflow | closed | 2023-10-25T12:55:28Z | 2023-10-25T12:57:43Z | 2023-10-25T12:57:42Z | mishig25 |
1,961,249,824 | Remove no_max_size_limit_datasets (no more full parquet conversion for Open-Orca/OpenOrca to fix the viewer) | The full Open-Orca/OpenOrca is too big to be processed correctly by most services, so let it work like all the other datasets (use the first 5GB) | Remove no_max_size_limit_datasets (no more full parquet conversion for Open-Orca/OpenOrca to fix the viewer): The full Open-Orca/OpenOrca is too big to be processed correctly by most services, so let it work like all the other datasets (use the first 5GB) | closed | 2023-10-25T12:05:53Z | 2023-10-25T12:28:49Z | 2023-10-25T12:28:48Z | lhoestq |
1,961,082,276 | Parquet stream-conversion fails to embed images/audio files from gated repos | e.g. for https://huggingface.co/datasets/SinKove/synthetic_mammography_csaw
it seems to be an issue with `datasets` not passing the token to `embed_table_storage` when generating a dataset
```json
{
"error": "An error occurred while generating the dataset",
"cause_exception": "DatasetGenerationError",
"cause_message": "An error occurred while generating the dataset",
"cause_traceback": [
"Traceback (most recent call last):\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/huggingface_hub/utils/_errors.py\", line 261, in hf_raise_for_status\n response.raise_for_status()\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/requests/models.py\", line 1021, in raise_for_status\n raise HTTPError(http_error_msg, response=self)\n",
"requests.exceptions.HTTPError: 401 Client Error: Unauthorized for url: https://huggingface.co/api/datasets/SinKove/synthetic_mammography_csaw/revision/490f9d72f59e2599b5d1eca5973b37a913270445\n",
"\nThe above exception was the direct cause of the following exception:\n\n",
"Traceback (most recent call last):\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/huggingface_hub/hf_file_system.py\", line 100, in _repo_and_revision_exist\n self._api.repo_info(repo_id, revision=revision, repo_type=repo_type)\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/huggingface_hub/utils/_validators.py\", line 118, in _inner_fn\n return fn(*args, **kwargs)\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/huggingface_hub/hf_api.py\", line 1868, in repo_info\n return method(\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/huggingface_hub/utils/_validators.py\", line 118, in _inner_fn\n return fn(*args, **kwargs)\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/huggingface_hub/hf_api.py\", line 1741, in dataset_info\n hf_raise_for_status(r)\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/huggingface_hub/utils/_errors.py\", line 277, in hf_raise_for_status\n raise GatedRepoError(message, response) from e\n",
"huggingface_hub.utils._errors.GatedRepoError: 401 Client Error. (Request ID: Root=1-6538ec77-55cac38800ae3763321e921e;1f6f0b56-19bb-4ea0-bfd6-0a1081822277)\n\nCannot access gated repo for url https://huggingface.co/api/datasets/SinKove/synthetic_mammography_csaw/revision/490f9d72f59e2599b5d1eca5973b37a913270445.\nRepo dataset SinKove/synthetic_mammography_csaw is gated. You must be authenticated to access it.\n",
"\nThe above exception was the direct cause of the following exception:\n\n",
"Traceback (most recent call last):\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/datasets/builder.py\", line 1703, in _prepare_split_single\n num_examples, num_bytes = writer.finalize()\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/datasets/arrow_writer.py\", line 586, in finalize\n self.write_examples_on_file()\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/datasets/arrow_writer.py\", line 448, in write_examples_on_file\n self.write_batch(batch_examples=batch_examples)\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/datasets/arrow_writer.py\", line 559, in write_batch\n self.write_table(pa_table, writer_batch_size)\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/datasets/arrow_writer.py\", line 574, in write_table\n pa_table = embed_table_storage(pa_table)\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/datasets/table.py\", line 2306, in embed_table_storage\n arrays = [\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/datasets/table.py\", line 2307, in <listcomp>\n embed_array_storage(table[name], feature) if require_storage_embed(feature) else table[name]\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/datasets/table.py\", line 1831, in wrapper\n return pa.chunked_array([func(chunk, *args, **kwargs) for chunk in array.chunks])\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/datasets/table.py\", line 1831, in <listcomp>\n return pa.chunked_array([func(chunk, *args, **kwargs) for chunk in array.chunks])\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/datasets/table.py\", line 2176, in embed_array_storage\n return feature.embed_storage(array)\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/datasets/features/image.py\", line 266, in embed_storage\n [\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/datasets/features/image.py\", line 267, in <listcomp>\n (path_to_bytes(x[\"path\"]) if x[\"bytes\"] is None else x[\"bytes\"]) if x is not None else None\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/datasets/utils/py_utils.py\", line 306, in wrapper\n return func(value) if value is not None else None\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/datasets/features/image.py\", line 261, in path_to_bytes\n with xopen(path, \"rb\") as f:\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/datasets/download/streaming_download_manager.py\", line 496, in xopen\n file_obj = fsspec.open(file, mode=mode, *args, **kwargs).open()\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/fsspec/core.py\", line 439, in open\n return open_files(\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/fsspec/core.py\", line 282, in open_files\n fs, fs_token, paths = get_fs_token_paths(\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/fsspec/core.py\", line 606, in get_fs_token_paths\n fs = filesystem(protocol, **inkwargs)\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/fsspec/registry.py\", line 261, in filesystem\n return cls(**storage_options)\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/fsspec/spec.py\", line 76, in __call__\n obj = super().__call__(*args, **kwargs)\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/fsspec/implementations/zip.py\", line 58, in __init__\n self.fo = fo.__enter__() # the whole instance is a context\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/fsspec/core.py\", line 102, in __enter__\n f = self.fs.open(self.path, mode=mode)\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/fsspec/spec.py\", line 1199, in open\n f = self._open(\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/huggingface_hub/hf_file_system.py\", line 200, in _open\n return HfFileSystemFile(self, path, mode=mode, revision=revision, **kwargs)\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/huggingface_hub/hf_file_system.py\", line 392, in __init__\n super().__init__(fs, path, **kwargs)\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/fsspec/spec.py\", line 1555, in __init__\n self.size = self.details[\"size\"]\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/fsspec/spec.py\", line 1568, in details\n self._details = self.fs.info(self.path)\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/huggingface_hub/hf_file_system.py\", line 351, in info\n resolved_path = self.resolve_path(path)\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/huggingface_hub/hf_file_system.py\", line 149, in resolve_path\n raise FileNotFoundError(path) from err\n",
"FileNotFoundError: SinKove/synthetic_mammography_csaw@490f9d72f59e2599b5d1eca5973b37a913270445/data/train.zip\n",
"\nThe above exception was the direct cause of the following exception:\n\n",
"Traceback (most recent call last):\n",
" File \"/src/services/worker/src/worker/job_manager.py\", line 168, in process\n job_result = self.job_runner.compute()\n",
" File \"/src/services/worker/src/worker/job_runners/config/parquet_and_info.py\", line 1348, in compute\n compute_config_parquet_and_info_response(\n",
" File \"/src/services/worker/src/worker/job_runners/config/parquet_and_info.py\", line 1255, in compute_config_parquet_and_info_response\n parquet_operations, partial = stream_convert_to_parquet(\n",
" File \"/src/services/worker/src/worker/job_runners/config/parquet_and_info.py\", line 852, in stream_convert_to_parquet\n builder._prepare_split(\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/datasets/builder.py\", line 1555, in _prepare_split\n for job_id, done, content in self._prepare_split_single(\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/datasets/builder.py\", line 1712, in _prepare_split_single\n raise DatasetGenerationError(\"An error occurred while generating the dataset\") from e\n",
"datasets.builder.DatasetGenerationError: An error occurred while generating the dataset\n"
]
}
``` | Parquet stream-conversion fails to embed images/audio files from gated repos: e.g. for https://huggingface.co/datasets/SinKove/synthetic_mammography_csaw
it seems to be an issue with `datasets` not passing the token to `embed_table_storage` when generating a dataset
```json
{
"error": "An error occurred while generating the dataset",
"cause_exception": "DatasetGenerationError",
"cause_message": "An error occurred while generating the dataset",
"cause_traceback": [
"Traceback (most recent call last):\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/huggingface_hub/utils/_errors.py\", line 261, in hf_raise_for_status\n response.raise_for_status()\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/requests/models.py\", line 1021, in raise_for_status\n raise HTTPError(http_error_msg, response=self)\n",
"requests.exceptions.HTTPError: 401 Client Error: Unauthorized for url: https://huggingface.co/api/datasets/SinKove/synthetic_mammography_csaw/revision/490f9d72f59e2599b5d1eca5973b37a913270445\n",
"\nThe above exception was the direct cause of the following exception:\n\n",
"Traceback (most recent call last):\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/huggingface_hub/hf_file_system.py\", line 100, in _repo_and_revision_exist\n self._api.repo_info(repo_id, revision=revision, repo_type=repo_type)\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/huggingface_hub/utils/_validators.py\", line 118, in _inner_fn\n return fn(*args, **kwargs)\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/huggingface_hub/hf_api.py\", line 1868, in repo_info\n return method(\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/huggingface_hub/utils/_validators.py\", line 118, in _inner_fn\n return fn(*args, **kwargs)\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/huggingface_hub/hf_api.py\", line 1741, in dataset_info\n hf_raise_for_status(r)\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/huggingface_hub/utils/_errors.py\", line 277, in hf_raise_for_status\n raise GatedRepoError(message, response) from e\n",
"huggingface_hub.utils._errors.GatedRepoError: 401 Client Error. (Request ID: Root=1-6538ec77-55cac38800ae3763321e921e;1f6f0b56-19bb-4ea0-bfd6-0a1081822277)\n\nCannot access gated repo for url https://huggingface.co/api/datasets/SinKove/synthetic_mammography_csaw/revision/490f9d72f59e2599b5d1eca5973b37a913270445.\nRepo dataset SinKove/synthetic_mammography_csaw is gated. You must be authenticated to access it.\n",
"\nThe above exception was the direct cause of the following exception:\n\n",
"Traceback (most recent call last):\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/datasets/builder.py\", line 1703, in _prepare_split_single\n num_examples, num_bytes = writer.finalize()\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/datasets/arrow_writer.py\", line 586, in finalize\n self.write_examples_on_file()\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/datasets/arrow_writer.py\", line 448, in write_examples_on_file\n self.write_batch(batch_examples=batch_examples)\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/datasets/arrow_writer.py\", line 559, in write_batch\n self.write_table(pa_table, writer_batch_size)\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/datasets/arrow_writer.py\", line 574, in write_table\n pa_table = embed_table_storage(pa_table)\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/datasets/table.py\", line 2306, in embed_table_storage\n arrays = [\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/datasets/table.py\", line 2307, in <listcomp>\n embed_array_storage(table[name], feature) if require_storage_embed(feature) else table[name]\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/datasets/table.py\", line 1831, in wrapper\n return pa.chunked_array([func(chunk, *args, **kwargs) for chunk in array.chunks])\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/datasets/table.py\", line 1831, in <listcomp>\n return pa.chunked_array([func(chunk, *args, **kwargs) for chunk in array.chunks])\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/datasets/table.py\", line 2176, in embed_array_storage\n return feature.embed_storage(array)\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/datasets/features/image.py\", line 266, in embed_storage\n [\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/datasets/features/image.py\", line 267, in <listcomp>\n (path_to_bytes(x[\"path\"]) if x[\"bytes\"] is None else x[\"bytes\"]) if x is not None else None\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/datasets/utils/py_utils.py\", line 306, in wrapper\n return func(value) if value is not None else None\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/datasets/features/image.py\", line 261, in path_to_bytes\n with xopen(path, \"rb\") as f:\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/datasets/download/streaming_download_manager.py\", line 496, in xopen\n file_obj = fsspec.open(file, mode=mode, *args, **kwargs).open()\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/fsspec/core.py\", line 439, in open\n return open_files(\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/fsspec/core.py\", line 282, in open_files\n fs, fs_token, paths = get_fs_token_paths(\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/fsspec/core.py\", line 606, in get_fs_token_paths\n fs = filesystem(protocol, **inkwargs)\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/fsspec/registry.py\", line 261, in filesystem\n return cls(**storage_options)\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/fsspec/spec.py\", line 76, in __call__\n obj = super().__call__(*args, **kwargs)\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/fsspec/implementations/zip.py\", line 58, in __init__\n self.fo = fo.__enter__() # the whole instance is a context\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/fsspec/core.py\", line 102, in __enter__\n f = self.fs.open(self.path, mode=mode)\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/fsspec/spec.py\", line 1199, in open\n f = self._open(\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/huggingface_hub/hf_file_system.py\", line 200, in _open\n return HfFileSystemFile(self, path, mode=mode, revision=revision, **kwargs)\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/huggingface_hub/hf_file_system.py\", line 392, in __init__\n super().__init__(fs, path, **kwargs)\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/fsspec/spec.py\", line 1555, in __init__\n self.size = self.details[\"size\"]\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/fsspec/spec.py\", line 1568, in details\n self._details = self.fs.info(self.path)\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/huggingface_hub/hf_file_system.py\", line 351, in info\n resolved_path = self.resolve_path(path)\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/huggingface_hub/hf_file_system.py\", line 149, in resolve_path\n raise FileNotFoundError(path) from err\n",
"FileNotFoundError: SinKove/synthetic_mammography_csaw@490f9d72f59e2599b5d1eca5973b37a913270445/data/train.zip\n",
"\nThe above exception was the direct cause of the following exception:\n\n",
"Traceback (most recent call last):\n",
" File \"/src/services/worker/src/worker/job_manager.py\", line 168, in process\n job_result = self.job_runner.compute()\n",
" File \"/src/services/worker/src/worker/job_runners/config/parquet_and_info.py\", line 1348, in compute\n compute_config_parquet_and_info_response(\n",
" File \"/src/services/worker/src/worker/job_runners/config/parquet_and_info.py\", line 1255, in compute_config_parquet_and_info_response\n parquet_operations, partial = stream_convert_to_parquet(\n",
" File \"/src/services/worker/src/worker/job_runners/config/parquet_and_info.py\", line 852, in stream_convert_to_parquet\n builder._prepare_split(\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/datasets/builder.py\", line 1555, in _prepare_split\n for job_id, done, content in self._prepare_split_single(\n",
" File \"/src/services/worker/.venv/lib/python3.9/site-packages/datasets/builder.py\", line 1712, in _prepare_split_single\n raise DatasetGenerationError(\"An error occurred while generating the dataset\") from e\n",
"datasets.builder.DatasetGenerationError: An error occurred while generating the dataset\n"
]
}
``` | open | 2023-10-25T10:38:47Z | 2023-11-03T11:23:49Z | null | lhoestq |
1,959,490,015 | Are URLs in rows response sanitized? | see https://github.com/huggingface/moon-landing/pull/7798#discussion_r1369813236 (internal)
> Is "src" validated / sanitized?
> if not there is a potential XSS exploit here (you can inject javascript code in an image src)
> Are S3 object names sanitized? If no, it should be the case in dataset-server side | Are URLs in rows response sanitized?: see https://github.com/huggingface/moon-landing/pull/7798#discussion_r1369813236 (internal)
> Is "src" validated / sanitized?
> if not there is a potential XSS exploit here (you can inject javascript code in an image src)
> Are S3 object names sanitized? If no, it should be the case in dataset-server side | closed | 2023-10-24T15:10:29Z | 2023-11-21T15:39:13Z | 2023-11-21T15:39:12Z | severo |
1,959,370,917 | Fix trending datasets dashboard | <img width="794" alt="image" src="https://github.com/huggingface/datasets-server/assets/42851186/67a3310d-54ba-42d7-8f33-9e47d10179ab">
(already deployed on spaces) | Fix trending datasets dashboard: <img width="794" alt="image" src="https://github.com/huggingface/datasets-server/assets/42851186/67a3310d-54ba-42d7-8f33-9e47d10179ab">
(already deployed on spaces) | closed | 2023-10-24T14:19:01Z | 2023-10-26T10:21:53Z | 2023-10-24T16:09:09Z | lhoestq |
1,959,150,821 | Update datasets to 2.14.6 | Update `datasets` to 2.14.6.
Fix #2006. | Update datasets to 2.14.6: Update `datasets` to 2.14.6.
Fix #2006. | closed | 2023-10-24T12:30:03Z | 2023-10-25T06:02:01Z | 2023-10-25T06:02:00Z | albertvillanova |
1,959,132,059 | Update datasets to 2.14.6 | Update datasets to 2.14.6: https://github.com/huggingface/datasets/releases/tag/2.14.6
> Create DefunctDatasetError by @albertvillanova in huggingface/datasets#6286
- We will be able to raise the `DefunctDatasetError` from defunct datasets | Update datasets to 2.14.6: Update datasets to 2.14.6: https://github.com/huggingface/datasets/releases/tag/2.14.6
> Create DefunctDatasetError by @albertvillanova in huggingface/datasets#6286
- We will be able to raise the `DefunctDatasetError` from defunct datasets | closed | 2023-10-24T12:17:43Z | 2023-10-25T06:02:01Z | 2023-10-25T06:02:01Z | albertvillanova |
1,952,766,158 | Fix another test | from https://github.com/huggingface/datasets-server/pull/2003 | Fix another test: from https://github.com/huggingface/datasets-server/pull/2003 | closed | 2023-10-19T17:40:52Z | 2023-10-19T17:53:51Z | 2023-10-19T17:53:50Z | lhoestq |
1,952,749,167 | More disable_dataset_scripts_support | Satasets scripts were still loaded in some cases.
following https://github.com/huggingface/datasets-server/pull/2001 | More disable_dataset_scripts_support: Satasets scripts were still loaded in some cases.
following https://github.com/huggingface/datasets-server/pull/2001 | closed | 2023-10-19T17:28:31Z | 2023-10-19T17:32:23Z | 2023-10-19T17:32:22Z | lhoestq |
1,952,675,563 | Fix test | null | Fix test: | closed | 2023-10-19T16:42:48Z | 2023-10-19T16:44:35Z | 2023-10-19T16:44:34Z | lhoestq |
1,952,481,405 | Disable dataset scripts | The `config-parquet-and-info` step will now raise a `DatasetWithScriptNotSupportedError` for datasets with a script, except those in the allow list. This will prevent users from running arbitrary code using dataset scripts.
The error message is shown to the user on the website and it says
```python
raise DatasetWithScriptNotSupportedError(
"The dataset viewer doesn't support this dataset because it runs "
"arbitrary python code. Please open a discussion in the discussion tab "
"if you think this is an error and tag @lhoestq and @severo."
)
```
The allow list is hardcoded for now: `DATASET_SCRIPTS_ALLOW_LIST = ["canonical"]`
The keyword "canonical" means all the datasets without namespaces.
We can add other datasets to the allow list, and it supports `fnmatch`, for example to support all the datasets from `huggingface` we can add `huggingface/*` to the allow list.
cc @severo @XciD | Disable dataset scripts: The `config-parquet-and-info` step will now raise a `DatasetWithScriptNotSupportedError` for datasets with a script, except those in the allow list. This will prevent users from running arbitrary code using dataset scripts.
The error message is shown to the user on the website and it says
```python
raise DatasetWithScriptNotSupportedError(
"The dataset viewer doesn't support this dataset because it runs "
"arbitrary python code. Please open a discussion in the discussion tab "
"if you think this is an error and tag @lhoestq and @severo."
)
```
The allow list is hardcoded for now: `DATASET_SCRIPTS_ALLOW_LIST = ["canonical"]`
The keyword "canonical" means all the datasets without namespaces.
We can add other datasets to the allow list, and it supports `fnmatch`, for example to support all the datasets from `huggingface` we can add `huggingface/*` to the allow list.
cc @severo @XciD | closed | 2023-10-19T15:02:13Z | 2023-10-19T16:33:47Z | 2023-10-19T16:33:46Z | lhoestq |
1,952,166,518 | fix: remove useless token | This token is not used by your action.
Secret is removed from the repository. | fix: remove useless token: This token is not used by your action.
Secret is removed from the repository. | closed | 2023-10-19T12:42:15Z | 2023-10-19T12:42:37Z | 2023-10-19T12:42:36Z | rtrompier |
1,950,231,325 | Better handling of k8s SIGTERM when stopping workers | We've observed recently that worker pods can take more than 1h to get killed by kubernetes (see internal [slack thread](https://huggingface.slack.com/archives/C04L6P8KNQ5/p1697641144036899) for example).
According to the k8s docs, at one point it sends a TERM and then after a grace period a SIGKILL on all the running processes.
If this keeps happening we need to improve the way we handle those to kill the pods faster.
One idea is to catch SIGTERM in the worker executor (main python process of workers) that would stop the worker loop (its subprocess where the actual code runs). | Better handling of k8s SIGTERM when stopping workers: We've observed recently that worker pods can take more than 1h to get killed by kubernetes (see internal [slack thread](https://huggingface.slack.com/archives/C04L6P8KNQ5/p1697641144036899) for example).
According to the k8s docs, at one point it sends a TERM and then after a grace period a SIGKILL on all the running processes.
If this keeps happening we need to improve the way we handle those to kill the pods faster.
One idea is to catch SIGTERM in the worker executor (main python process of workers) that would stop the worker loop (its subprocess where the actual code runs). | closed | 2023-10-18T16:57:09Z | 2023-11-07T10:29:24Z | 2023-11-07T10:29:24Z | lhoestq |
1,949,834,163 | Store data in a local database file for stats instead of an in-memory connection | This is my almost last idea on how make things not freezing on big datasets. It's the only difference with `duckdb-index` worker which apparently works fine for 5Gb datasets: it uses local db file, stats worker didn't use it. | Store data in a local database file for stats instead of an in-memory connection: This is my almost last idea on how make things not freezing on big datasets. It's the only difference with `duckdb-index` worker which apparently works fine for 5Gb datasets: it uses local db file, stats worker didn't use it. | closed | 2023-10-18T14:00:23Z | 2023-10-18T14:11:33Z | 2023-10-18T14:11:32Z | polinaeterna |
1,948,988,533 | Fix quality in search by removing unused import | Fix quality issue in search by removing unused import.
This quality issue was introduced by:
- #1983 | Fix quality in search by removing unused import: Fix quality issue in search by removing unused import.
This quality issue was introduced by:
- #1983 | closed | 2023-10-18T06:52:47Z | 2023-10-18T07:07:44Z | 2023-10-18T07:07:42Z | albertvillanova |
1,948,970,916 | Update urllib3 from 1.26.17 to fix vulnerability | Update `urllib3` from 1.26.17 to fix vulnerability:
- to 2.0.7 in e2e
- to 1.26.18 in the rest of the components
This should fix 12 dependabot alerts.
Supersede and close #1994.
Supersede and close #1995. | Update urllib3 from 1.26.17 to fix vulnerability: Update `urllib3` from 1.26.17 to fix vulnerability:
- to 2.0.7 in e2e
- to 1.26.18 in the rest of the components
This should fix 12 dependabot alerts.
Supersede and close #1994.
Supersede and close #1995. | closed | 2023-10-18T06:40:30Z | 2023-10-18T15:39:11Z | 2023-10-18T15:39:10Z | albertvillanova |
1,948,568,266 | Bump urllib3 from 1.26.17 to 1.26.18 in /libs/libcommon | Bumps [urllib3](https://github.com/urllib3/urllib3) from 1.26.17 to 1.26.18.
<details>
<summary>Release notes</summary>
<p><em>Sourced from <a href="https://github.com/urllib3/urllib3/releases">urllib3's releases</a>.</em></p>
<blockquote>
<h2>1.26.18</h2>
<ul>
<li>Made body stripped from HTTP requests changing the request method to GET after HTTP 303 "See Other" redirect responses. (GHSA-g4mx-q9vg-27p4)</li>
</ul>
</blockquote>
</details>
<details>
<summary>Changelog</summary>
<p><em>Sourced from <a href="https://github.com/urllib3/urllib3/blob/main/CHANGES.rst">urllib3's changelog</a>.</em></p>
<blockquote>
<h1>1.26.18 (2023-10-17)</h1>
<ul>
<li>Made body stripped from HTTP requests changing the request method to GET after HTTP 303 "See Other" redirect responses.</li>
</ul>
</blockquote>
</details>
<details>
<summary>Commits</summary>
<ul>
<li><a href="https://github.com/urllib3/urllib3/commit/9c2c2307dd1d6af504e09aac0326d86ee3597a0b"><code>9c2c230</code></a> Release 1.26.18 (<a href="https://redirect.github.com/urllib3/urllib3/issues/3159">#3159</a>)</li>
<li><a href="https://github.com/urllib3/urllib3/commit/b594c5ceaca38e1ac215f916538fb128e3526a36"><code>b594c5c</code></a> Merge pull request from GHSA-g4mx-q9vg-27p4</li>
<li><a href="https://github.com/urllib3/urllib3/commit/944f0eb134485f41bc531be52de12ba5a37bca73"><code>944f0eb</code></a> [1.26] Use vendored six in urllib3.contrib.securetransport</li>
<li>See full diff in <a href="https://github.com/urllib3/urllib3/compare/1.26.17...1.26.18">compare view</a></li>
</ul>
</details>
<br />
[](https://docs.github.com/en/github/managing-security-vulnerabilities/about-dependabot-security-updates#about-compatibility-scores)
Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting `@dependabot rebase`.
[//]: # (dependabot-automerge-start)
[//]: # (dependabot-automerge-end)
---
<details>
<summary>Dependabot commands and options</summary>
<br />
You can trigger Dependabot actions by commenting on this PR:
- `@dependabot rebase` will rebase this PR
- `@dependabot recreate` will recreate this PR, overwriting any edits that have been made to it
- `@dependabot merge` will merge this PR after your CI passes on it
- `@dependabot squash and merge` will squash and merge this PR after your CI passes on it
- `@dependabot cancel merge` will cancel a previously requested merge and block automerging
- `@dependabot reopen` will reopen this PR if it is closed
- `@dependabot close` will close this PR and stop Dependabot recreating it. You can achieve the same result by closing it manually
- `@dependabot show <dependency name> ignore conditions` will show all of the ignore conditions of the specified dependency
- `@dependabot ignore this major version` will close this PR and stop Dependabot creating any more for this major version (unless you reopen the PR or upgrade to it yourself)
- `@dependabot ignore this minor version` will close this PR and stop Dependabot creating any more for this minor version (unless you reopen the PR or upgrade to it yourself)
- `@dependabot ignore this dependency` will close this PR and stop Dependabot creating any more for this dependency (unless you reopen the PR or upgrade to it yourself)
You can disable automated security fix PRs for this repo from the [Security Alerts page](https://github.com/huggingface/datasets-server/network/alerts).
</details> | Bump urllib3 from 1.26.17 to 1.26.18 in /libs/libcommon: Bumps [urllib3](https://github.com/urllib3/urllib3) from 1.26.17 to 1.26.18.
<details>
<summary>Release notes</summary>
<p><em>Sourced from <a href="https://github.com/urllib3/urllib3/releases">urllib3's releases</a>.</em></p>
<blockquote>
<h2>1.26.18</h2>
<ul>
<li>Made body stripped from HTTP requests changing the request method to GET after HTTP 303 "See Other" redirect responses. (GHSA-g4mx-q9vg-27p4)</li>
</ul>
</blockquote>
</details>
<details>
<summary>Changelog</summary>
<p><em>Sourced from <a href="https://github.com/urllib3/urllib3/blob/main/CHANGES.rst">urllib3's changelog</a>.</em></p>
<blockquote>
<h1>1.26.18 (2023-10-17)</h1>
<ul>
<li>Made body stripped from HTTP requests changing the request method to GET after HTTP 303 "See Other" redirect responses.</li>
</ul>
</blockquote>
</details>
<details>
<summary>Commits</summary>
<ul>
<li><a href="https://github.com/urllib3/urllib3/commit/9c2c2307dd1d6af504e09aac0326d86ee3597a0b"><code>9c2c230</code></a> Release 1.26.18 (<a href="https://redirect.github.com/urllib3/urllib3/issues/3159">#3159</a>)</li>
<li><a href="https://github.com/urllib3/urllib3/commit/b594c5ceaca38e1ac215f916538fb128e3526a36"><code>b594c5c</code></a> Merge pull request from GHSA-g4mx-q9vg-27p4</li>
<li><a href="https://github.com/urllib3/urllib3/commit/944f0eb134485f41bc531be52de12ba5a37bca73"><code>944f0eb</code></a> [1.26] Use vendored six in urllib3.contrib.securetransport</li>
<li>See full diff in <a href="https://github.com/urllib3/urllib3/compare/1.26.17...1.26.18">compare view</a></li>
</ul>
</details>
<br />
[](https://docs.github.com/en/github/managing-security-vulnerabilities/about-dependabot-security-updates#about-compatibility-scores)
Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting `@dependabot rebase`.
[//]: # (dependabot-automerge-start)
[//]: # (dependabot-automerge-end)
---
<details>
<summary>Dependabot commands and options</summary>
<br />
You can trigger Dependabot actions by commenting on this PR:
- `@dependabot rebase` will rebase this PR
- `@dependabot recreate` will recreate this PR, overwriting any edits that have been made to it
- `@dependabot merge` will merge this PR after your CI passes on it
- `@dependabot squash and merge` will squash and merge this PR after your CI passes on it
- `@dependabot cancel merge` will cancel a previously requested merge and block automerging
- `@dependabot reopen` will reopen this PR if it is closed
- `@dependabot close` will close this PR and stop Dependabot recreating it. You can achieve the same result by closing it manually
- `@dependabot show <dependency name> ignore conditions` will show all of the ignore conditions of the specified dependency
- `@dependabot ignore this major version` will close this PR and stop Dependabot creating any more for this major version (unless you reopen the PR or upgrade to it yourself)
- `@dependabot ignore this minor version` will close this PR and stop Dependabot creating any more for this minor version (unless you reopen the PR or upgrade to it yourself)
- `@dependabot ignore this dependency` will close this PR and stop Dependabot creating any more for this dependency (unless you reopen the PR or upgrade to it yourself)
You can disable automated security fix PRs for this repo from the [Security Alerts page](https://github.com/huggingface/datasets-server/network/alerts).
</details> | closed | 2023-10-18T01:20:51Z | 2023-10-18T15:39:22Z | 2023-10-18T15:39:11Z | dependabot[bot] |
1,948,565,114 | Bump urllib3 from 1.26.17 to 1.26.18 in /e2e | Bumps [urllib3](https://github.com/urllib3/urllib3) from 1.26.17 to 1.26.18.
<details>
<summary>Release notes</summary>
<p><em>Sourced from <a href="https://github.com/urllib3/urllib3/releases">urllib3's releases</a>.</em></p>
<blockquote>
<h2>1.26.18</h2>
<ul>
<li>Made body stripped from HTTP requests changing the request method to GET after HTTP 303 "See Other" redirect responses. (GHSA-g4mx-q9vg-27p4)</li>
</ul>
</blockquote>
</details>
<details>
<summary>Changelog</summary>
<p><em>Sourced from <a href="https://github.com/urllib3/urllib3/blob/main/CHANGES.rst">urllib3's changelog</a>.</em></p>
<blockquote>
<h1>1.26.18 (2023-10-17)</h1>
<ul>
<li>Made body stripped from HTTP requests changing the request method to GET after HTTP 303 "See Other" redirect responses.</li>
</ul>
</blockquote>
</details>
<details>
<summary>Commits</summary>
<ul>
<li><a href="https://github.com/urllib3/urllib3/commit/9c2c2307dd1d6af504e09aac0326d86ee3597a0b"><code>9c2c230</code></a> Release 1.26.18 (<a href="https://redirect.github.com/urllib3/urllib3/issues/3159">#3159</a>)</li>
<li><a href="https://github.com/urllib3/urllib3/commit/b594c5ceaca38e1ac215f916538fb128e3526a36"><code>b594c5c</code></a> Merge pull request from GHSA-g4mx-q9vg-27p4</li>
<li><a href="https://github.com/urllib3/urllib3/commit/944f0eb134485f41bc531be52de12ba5a37bca73"><code>944f0eb</code></a> [1.26] Use vendored six in urllib3.contrib.securetransport</li>
<li>See full diff in <a href="https://github.com/urllib3/urllib3/compare/1.26.17...1.26.18">compare view</a></li>
</ul>
</details>
<br />
[](https://docs.github.com/en/github/managing-security-vulnerabilities/about-dependabot-security-updates#about-compatibility-scores)
Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting `@dependabot rebase`.
[//]: # (dependabot-automerge-start)
[//]: # (dependabot-automerge-end)
---
<details>
<summary>Dependabot commands and options</summary>
<br />
You can trigger Dependabot actions by commenting on this PR:
- `@dependabot rebase` will rebase this PR
- `@dependabot recreate` will recreate this PR, overwriting any edits that have been made to it
- `@dependabot merge` will merge this PR after your CI passes on it
- `@dependabot squash and merge` will squash and merge this PR after your CI passes on it
- `@dependabot cancel merge` will cancel a previously requested merge and block automerging
- `@dependabot reopen` will reopen this PR if it is closed
- `@dependabot close` will close this PR and stop Dependabot recreating it. You can achieve the same result by closing it manually
- `@dependabot show <dependency name> ignore conditions` will show all of the ignore conditions of the specified dependency
- `@dependabot ignore this major version` will close this PR and stop Dependabot creating any more for this major version (unless you reopen the PR or upgrade to it yourself)
- `@dependabot ignore this minor version` will close this PR and stop Dependabot creating any more for this minor version (unless you reopen the PR or upgrade to it yourself)
- `@dependabot ignore this dependency` will close this PR and stop Dependabot creating any more for this dependency (unless you reopen the PR or upgrade to it yourself)
You can disable automated security fix PRs for this repo from the [Security Alerts page](https://github.com/huggingface/datasets-server/network/alerts).
</details> | Bump urllib3 from 1.26.17 to 1.26.18 in /e2e: Bumps [urllib3](https://github.com/urllib3/urllib3) from 1.26.17 to 1.26.18.
<details>
<summary>Release notes</summary>
<p><em>Sourced from <a href="https://github.com/urllib3/urllib3/releases">urllib3's releases</a>.</em></p>
<blockquote>
<h2>1.26.18</h2>
<ul>
<li>Made body stripped from HTTP requests changing the request method to GET after HTTP 303 "See Other" redirect responses. (GHSA-g4mx-q9vg-27p4)</li>
</ul>
</blockquote>
</details>
<details>
<summary>Changelog</summary>
<p><em>Sourced from <a href="https://github.com/urllib3/urllib3/blob/main/CHANGES.rst">urllib3's changelog</a>.</em></p>
<blockquote>
<h1>1.26.18 (2023-10-17)</h1>
<ul>
<li>Made body stripped from HTTP requests changing the request method to GET after HTTP 303 "See Other" redirect responses.</li>
</ul>
</blockquote>
</details>
<details>
<summary>Commits</summary>
<ul>
<li><a href="https://github.com/urllib3/urllib3/commit/9c2c2307dd1d6af504e09aac0326d86ee3597a0b"><code>9c2c230</code></a> Release 1.26.18 (<a href="https://redirect.github.com/urllib3/urllib3/issues/3159">#3159</a>)</li>
<li><a href="https://github.com/urllib3/urllib3/commit/b594c5ceaca38e1ac215f916538fb128e3526a36"><code>b594c5c</code></a> Merge pull request from GHSA-g4mx-q9vg-27p4</li>
<li><a href="https://github.com/urllib3/urllib3/commit/944f0eb134485f41bc531be52de12ba5a37bca73"><code>944f0eb</code></a> [1.26] Use vendored six in urllib3.contrib.securetransport</li>
<li>See full diff in <a href="https://github.com/urllib3/urllib3/compare/1.26.17...1.26.18">compare view</a></li>
</ul>
</details>
<br />
[](https://docs.github.com/en/github/managing-security-vulnerabilities/about-dependabot-security-updates#about-compatibility-scores)
Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting `@dependabot rebase`.
[//]: # (dependabot-automerge-start)
[//]: # (dependabot-automerge-end)
---
<details>
<summary>Dependabot commands and options</summary>
<br />
You can trigger Dependabot actions by commenting on this PR:
- `@dependabot rebase` will rebase this PR
- `@dependabot recreate` will recreate this PR, overwriting any edits that have been made to it
- `@dependabot merge` will merge this PR after your CI passes on it
- `@dependabot squash and merge` will squash and merge this PR after your CI passes on it
- `@dependabot cancel merge` will cancel a previously requested merge and block automerging
- `@dependabot reopen` will reopen this PR if it is closed
- `@dependabot close` will close this PR and stop Dependabot recreating it. You can achieve the same result by closing it manually
- `@dependabot show <dependency name> ignore conditions` will show all of the ignore conditions of the specified dependency
- `@dependabot ignore this major version` will close this PR and stop Dependabot creating any more for this major version (unless you reopen the PR or upgrade to it yourself)
- `@dependabot ignore this minor version` will close this PR and stop Dependabot creating any more for this minor version (unless you reopen the PR or upgrade to it yourself)
- `@dependabot ignore this dependency` will close this PR and stop Dependabot creating any more for this dependency (unless you reopen the PR or upgrade to it yourself)
You can disable automated security fix PRs for this repo from the [Security Alerts page](https://github.com/huggingface/datasets-server/network/alerts).
</details> | closed | 2023-10-18T01:18:28Z | 2023-10-18T15:39:21Z | 2023-10-18T15:39:11Z | dependabot[bot] |
1,948,016,264 | making dataset_git_revision field mandatory | See https://github.com/huggingface/datasets-server/pull/1988#issuecomment-1766330413 for context
Depends on https://github.com/huggingface/datasets-server/pull/1996 | making dataset_git_revision field mandatory: See https://github.com/huggingface/datasets-server/pull/1988#issuecomment-1766330413 for context
Depends on https://github.com/huggingface/datasets-server/pull/1996 | closed | 2023-10-17T18:31:45Z | 2023-10-18T16:47:22Z | 2023-10-18T16:47:21Z | AndreaFrancis |
1,947,840,586 | Partial duckdb index | For big datasets, instead of using all the parquet files, use only the first ones until the limit of 5GB is reached.
It's used for both `search` and `filter`.
## Indexing
I added three fields in the duckdb-index job:
- partial (if it used a partial parquet export or if it didn't index the full parquet export)
- num_rows
- num_bytes
## New jobs
I also added two jobs and their responses can be used in the front-end instead of the `size` endpoint:
- config-duckdb-index-size
- dataset-duckdb-index-size
Both jobs return the same response format as the config-size and the dataset-size jobs, but with information about the size of the indexed data instead of the size of the parquet-exported data.
## Update of search and filter
I also added the `partial` field in the responses of `/search` and `/filter`, in case they use a partial index.
## Index file
The file name is `index.duckdb` if it indexes all the rows from the parquet export, otherwise it's named `partial-duckdb.index`. It is placed in the same folder as the parquet export in `refs/convert/parquet`.
Fix https://github.com/huggingface/datasets-server/issues/1742
TODO
- [x] tests
- [x] docs
- [x] openapi
- [x] update /search and /filter for partial indexes | Partial duckdb index: For big datasets, instead of using all the parquet files, use only the first ones until the limit of 5GB is reached.
It's used for both `search` and `filter`.
## Indexing
I added three fields in the duckdb-index job:
- partial (if it used a partial parquet export or if it didn't index the full parquet export)
- num_rows
- num_bytes
## New jobs
I also added two jobs and their responses can be used in the front-end instead of the `size` endpoint:
- config-duckdb-index-size
- dataset-duckdb-index-size
Both jobs return the same response format as the config-size and the dataset-size jobs, but with information about the size of the indexed data instead of the size of the parquet-exported data.
## Update of search and filter
I also added the `partial` field in the responses of `/search` and `/filter`, in case they use a partial index.
## Index file
The file name is `index.duckdb` if it indexes all the rows from the parquet export, otherwise it's named `partial-duckdb.index`. It is placed in the same folder as the parquet export in `refs/convert/parquet`.
Fix https://github.com/huggingface/datasets-server/issues/1742
TODO
- [x] tests
- [x] docs
- [x] openapi
- [x] update /search and /filter for partial indexes | closed | 2023-10-17T16:43:29Z | 2023-11-07T09:53:03Z | 2023-11-07T09:52:27Z | lhoestq |
1,947,739,879 | Limit max RAM that duckdb can use | Current value of `max_memory` in prod is 105.7Gb (despite the pod has only 34Gb but it's set automatically by duckdb as 80% of ram and ram of the node in total is 128Gb)
<img width="382" alt="image" src="https://github.com/huggingface/datasets-server/assets/16348744/b29d7829-43e3-431c-a453-546f80e722a5">
I want to try to lower it manually since we assume there might be some memory issues with duckdb causing the `split-descriptive-statistics` process to stuck at the step of loading data into in-memory db.
(also want to try to load data to local db file, this is the only difference with `duckdb-index` step) | Limit max RAM that duckdb can use: Current value of `max_memory` in prod is 105.7Gb (despite the pod has only 34Gb but it's set automatically by duckdb as 80% of ram and ram of the node in total is 128Gb)
<img width="382" alt="image" src="https://github.com/huggingface/datasets-server/assets/16348744/b29d7829-43e3-431c-a453-546f80e722a5">
I want to try to lower it manually since we assume there might be some memory issues with duckdb causing the `split-descriptive-statistics` process to stuck at the step of loading data into in-memory db.
(also want to try to load data to local db file, this is the only difference with `duckdb-index` step) | closed | 2023-10-17T15:47:35Z | 2023-10-18T13:50:53Z | 2023-10-17T16:22:15Z | polinaeterna |
1,947,439,157 | /filter returns Unexpected error for string columns | For example:
https://datasets-server.huggingface.co/filter?dataset=tapaco&config=af&split=train&where=language=%27af%27
```
{"error":"Unexpected error."}
```
| /filter returns Unexpected error for string columns: For example:
https://datasets-server.huggingface.co/filter?dataset=tapaco&config=af&split=train&where=language=%27af%27
```
{"error":"Unexpected error."}
```
| closed | 2023-10-17T13:28:57Z | 2023-10-19T14:45:10Z | 2023-10-19T14:45:10Z | severo |
1,947,101,711 | more search and filter pods | null | more search and filter pods: | closed | 2023-10-17T10:35:27Z | 2023-10-17T10:36:37Z | 2023-10-17T10:36:36Z | lhoestq |
1,945,813,806 | adding revision for assets creation path | Fix for https://github.com/huggingface/datasets-server/issues/1981
As suggested by @lhoestq in https://github.com/huggingface/datasets-server/issues/1981#issuecomment-1764559346 , adding dataset version (revision) as part of assets path:
- Removing old code to periodically clean cached-assets storage
- Add revision as part of the asset path {dataset}-{revision}-{config}-{split} if no revision, default is "main" same as in https://github.com/huggingface/datasets-server/blob/main/jobs/mongodb_migration/src/mongodb_migration/migrations/_20230516101500_queue_job_add_revision.py#L22
- For cached-assets, we dont need to delete previous version since they will be automatically removed by the bucket TTL policy (After 1 day of inactivity)
- We can keep overwrite=False in /search, /fiter and /rows since a new version will be created in another folder
- For previous assets I think there is no problem with backwards compatibility since we stored the file URL in the db
Open questions for upcoming PRs:
- Since all our cache records in [cachedResponsesBlue](https://github.com/huggingface/datasets-server/blob/main/libs/libcommon/src/libcommon/simple_cache.py#L129) collection have a revision value, should we change it to a required and non Optional same as in [Jobs](https://github.com/huggingface/datasets-server/blob/main/libs/libcommon/src/libcommon/queue.py#L178)? -> Done in https://github.com/huggingface/datasets-server/pull/1993
- For assets, if a new version is processed, we will be storing the previous files in S3, we need to find a way to clean them or, maybe in https://github.com/huggingface/datasets-server/issues/1823 it will make more sense since a "versioned" approach was proposed.
| adding revision for assets creation path: Fix for https://github.com/huggingface/datasets-server/issues/1981
As suggested by @lhoestq in https://github.com/huggingface/datasets-server/issues/1981#issuecomment-1764559346 , adding dataset version (revision) as part of assets path:
- Removing old code to periodically clean cached-assets storage
- Add revision as part of the asset path {dataset}-{revision}-{config}-{split} if no revision, default is "main" same as in https://github.com/huggingface/datasets-server/blob/main/jobs/mongodb_migration/src/mongodb_migration/migrations/_20230516101500_queue_job_add_revision.py#L22
- For cached-assets, we dont need to delete previous version since they will be automatically removed by the bucket TTL policy (After 1 day of inactivity)
- We can keep overwrite=False in /search, /fiter and /rows since a new version will be created in another folder
- For previous assets I think there is no problem with backwards compatibility since we stored the file URL in the db
Open questions for upcoming PRs:
- Since all our cache records in [cachedResponsesBlue](https://github.com/huggingface/datasets-server/blob/main/libs/libcommon/src/libcommon/simple_cache.py#L129) collection have a revision value, should we change it to a required and non Optional same as in [Jobs](https://github.com/huggingface/datasets-server/blob/main/libs/libcommon/src/libcommon/queue.py#L178)? -> Done in https://github.com/huggingface/datasets-server/pull/1993
- For assets, if a new version is processed, we will be storing the previous files in S3, we need to find a way to clean them or, maybe in https://github.com/huggingface/datasets-server/issues/1823 it will make more sense since a "versioned" approach was proposed.
| closed | 2023-10-16T18:11:08Z | 2023-10-19T15:23:40Z | 2023-10-19T15:23:04Z | AndreaFrancis |
1,945,237,423 | Add missing filter errors to openapi.json | Add missing filter errors to `openapi.json`. | Add missing filter errors to openapi.json: Add missing filter errors to `openapi.json`. | closed | 2023-10-16T13:35:03Z | 2023-10-20T07:29:18Z | 2023-10-20T07:28:43Z | albertvillanova |
1,945,181,398 | adding clean stats cache cron job | Same as we do now for `duckdb-index` and `hf-datasets-cache` (cleaning periodically obsolete/old files), we can sometimes keep files in stats because of job runner crash or zombie killer.
Adding a cron job to delete files/folders in /storage/stats-cache that are older than 3 hours ago.
| adding clean stats cache cron job: Same as we do now for `duckdb-index` and `hf-datasets-cache` (cleaning periodically obsolete/old files), we can sometimes keep files in stats because of job runner crash or zombie killer.
Adding a cron job to delete files/folders in /storage/stats-cache that are older than 3 hours ago.
| closed | 2023-10-16T13:06:51Z | 2023-10-16T16:51:26Z | 2023-10-16T16:51:25Z | AndreaFrancis |
1,945,108,580 | DuckDB parsing errors if a column name has quotes | Not super important but affects the `split-descriptive-statistics` and the `split-duckdb-index ` jobs
e.g. this dataset has a pretty long column name that has quotes and it raises this error
```
β INFO: 2023-10-16 12:27:43,231 - root - Compute descriptive statistics for dataset='lunaluan/chatbox3_history', config='default', split='train' β
β INFO: 2023-10-16 12:27:43,233 - root - Downloading remote parquet files to a local directory /storage/stats-cache/81004481536938-split-descriptive-statistics-lunaluan-chatbox3_hi-0d92723b. β
β Downloading 0000.parquet: 0%| | 0.00/6.75k [00:00<?, ?B/s]Downloading 0000.parquet: 100%|ββββββββββ| 6.75k/6.75k [00:00<00:00, 5.67MB/s] β
β INFO: 2023-10-16 12:27:43,912 - root - Loading data into in-memory table. β
β ERROR: 2023-10-16 12:27:44,068 - root - Parser Error: syntax error at or near "detail" β
β LINE 2: ... over his Professor's face. Mention "in detail description" how the professor ... β
β ^ β
β Traceback (most recent call last): β
β File "/src/services/worker/src/worker/job_manager.py", line 168, in process β
β job_result = self.job_runner.compute() β
β File "/src/services/worker/src/worker/job_runners/split/descriptive_statistics.py", line 591, in compute β
β compute_descriptive_statistics_response( β
β File "/src/services/worker/src/worker/job_runners/split/descriptive_statistics.py", line 485, in compute_descriptive_statistics_response β
β con.sql( β
β duckdb.ParserException: Parser Error: syntax error at or near "detail" β
β LINE 2: ... over his Professor's face. Mention "in detail description" how the professor ... β
β ^ β
``` | DuckDB parsing errors if a column name has quotes: Not super important but affects the `split-descriptive-statistics` and the `split-duckdb-index ` jobs
e.g. this dataset has a pretty long column name that has quotes and it raises this error
```
β INFO: 2023-10-16 12:27:43,231 - root - Compute descriptive statistics for dataset='lunaluan/chatbox3_history', config='default', split='train' β
β INFO: 2023-10-16 12:27:43,233 - root - Downloading remote parquet files to a local directory /storage/stats-cache/81004481536938-split-descriptive-statistics-lunaluan-chatbox3_hi-0d92723b. β
β Downloading 0000.parquet: 0%| | 0.00/6.75k [00:00<?, ?B/s]Downloading 0000.parquet: 100%|ββββββββββ| 6.75k/6.75k [00:00<00:00, 5.67MB/s] β
β INFO: 2023-10-16 12:27:43,912 - root - Loading data into in-memory table. β
β ERROR: 2023-10-16 12:27:44,068 - root - Parser Error: syntax error at or near "detail" β
β LINE 2: ... over his Professor's face. Mention "in detail description" how the professor ... β
β ^ β
β Traceback (most recent call last): β
β File "/src/services/worker/src/worker/job_manager.py", line 168, in process β
β job_result = self.job_runner.compute() β
β File "/src/services/worker/src/worker/job_runners/split/descriptive_statistics.py", line 591, in compute β
β compute_descriptive_statistics_response( β
β File "/src/services/worker/src/worker/job_runners/split/descriptive_statistics.py", line 485, in compute_descriptive_statistics_response β
β con.sql( β
β duckdb.ParserException: Parser Error: syntax error at or near "detail" β
β LINE 2: ... over his Professor's face. Mention "in detail description" how the professor ... β
β ^ β
``` | open | 2023-10-16T12:32:27Z | 2024-06-19T16:10:52Z | null | lhoestq |
1,944,692,347 | Run duckdb queries async in separate thread | Run `duckdb` queries asynchronously in a separate thread. | Run duckdb queries async in separate thread: Run `duckdb` queries asynchronously in a separate thread. | closed | 2023-10-16T08:44:19Z | 2023-10-20T08:00:00Z | 2023-10-20T07:59:59Z | albertvillanova |
1,944,606,683 | Move modification time update of duckdb index file to libapi | EDIT:
After the explanation by @AndreaFrancis, this PR moves the updating of the modification time of the duckdb index file from the /search endpoint to the `libapi` function `get_index_file_location_and_download_if_missing`.
This way, the modification time is also updated by the /filter endpoint.
~~Remove unnecessary `Path.touch` in /search.~~
~~I think this is not needed.~~
CC: @AndreaFrancis | Move modification time update of duckdb index file to libapi: EDIT:
After the explanation by @AndreaFrancis, this PR moves the updating of the modification time of the duckdb index file from the /search endpoint to the `libapi` function `get_index_file_location_and_download_if_missing`.
This way, the modification time is also updated by the /filter endpoint.
~~Remove unnecessary `Path.touch` in /search.~~
~~I think this is not needed.~~
CC: @AndreaFrancis | closed | 2023-10-16T07:56:24Z | 2023-10-16T15:48:45Z | 2023-10-16T15:48:44Z | albertvillanova |
1,942,211,081 | Remove unused httpfs duckdb extension | it's not used for files download anymore, we use `huggingface_hub` directly
**+ I added some logging to check what's going on with the stats computing in this PR to make it fast, to be removed!** | Remove unused httpfs duckdb extension: it's not used for files download anymore, we use `huggingface_hub` directly
**+ I added some logging to check what's going on with the stats computing in this PR to make it fast, to be removed!** | closed | 2023-10-13T15:57:09Z | 2023-10-17T12:13:33Z | 2023-10-17T12:13:32Z | polinaeterna |
1,942,060,762 | Overwrite image and audio cached assets when necessary | Currently we never overwrite image and audio cached assets, which can lead to outdated data in the viewer like here:
https://huggingface.co/datasets/ccmusic-database/instrument_timbre_eval/discussions/1
In particular we have `overwrite=False` here:
https://github.com/huggingface/datasets-server/blob/b84c8210ec023664c4780ab348b0232468afe116/libs/libapi/src/libapi/response.py#L36-L42
cc @AndreaFrancis @severo | Overwrite image and audio cached assets when necessary: Currently we never overwrite image and audio cached assets, which can lead to outdated data in the viewer like here:
https://huggingface.co/datasets/ccmusic-database/instrument_timbre_eval/discussions/1
In particular we have `overwrite=False` here:
https://github.com/huggingface/datasets-server/blob/b84c8210ec023664c4780ab348b0232468afe116/libs/libapi/src/libapi/response.py#L36-L42
cc @AndreaFrancis @severo | closed | 2023-10-13T14:21:51Z | 2023-10-19T17:49:42Z | 2023-10-19T17:49:42Z | lhoestq |
1,942,002,712 | increase ttl time for clean cron jobs | null | increase ttl time for clean cron jobs: | closed | 2023-10-13T13:47:54Z | 2023-10-13T13:50:33Z | 2023-10-13T13:50:32Z | AndreaFrancis |
1,941,961,874 | Remove requests as dependency of libapi | Remove `requests` as `libapi` dependency by using `httpx` instead. | Remove requests as dependency of libapi: Remove `requests` as `libapi` dependency by using `httpx` instead. | closed | 2023-10-13T13:24:09Z | 2023-10-19T06:35:21Z | 2023-10-19T06:35:19Z | albertvillanova |
1,941,617,329 | Make admin authentication asynchronous | Make admin service authentication asynchronous by using `httpx` instead of `requests`.
Related to:
- #1975 | Make admin authentication asynchronous: Make admin service authentication asynchronous by using `httpx` instead of `requests`.
Related to:
- #1975 | closed | 2023-10-13T09:45:14Z | 2023-10-13T12:37:24Z | 2023-10-13T12:37:23Z | albertvillanova |
1,941,379,218 | Remove requests as dependency of libcommon | Remove `requests` as dependency of `libcommon` because it is not used. | Remove requests as dependency of libcommon: Remove `requests` as dependency of `libcommon` because it is not used. | closed | 2023-10-13T07:13:43Z | 2023-10-13T08:09:56Z | 2023-10-13T08:09:55Z | albertvillanova |
1,940,736,477 | fsspec storage instead of boto3 | Related to comment https://github.com/huggingface/datasets-server/pull/1882#issuecomment-1740688443
- Use a storage agnostic tool (fsspec) to store images/audios for assets and cached assets.
- Adding an abstraction layer for "file" and "s3" storage.
- For docker containers, we will use "file" storage but for staging/prod we will use "s3".
- I removed all the dependencies of boto3 and moto (and added s3fs)
- I removed all the dependencies of `/storage/assets` and `/storage/cached-assets` (We don't need it anymore since images and audios are being written directly to the file system)
- I removed the disk metrics (They need to be added again but getting the values from S3, maybe can be handled in another PR)
- Will close https://github.com/huggingface/datasets-server/issues/1887 and https://github.com/huggingface/datasets-server/issues/1885
Pending issue:
- For reverse-proxy, I am not able to get the files stored in /assets and /cached-assets volume (My generated URL is http://localhost:8100/assets/asoria/image/--/666e73a/--/default/train/99/image/image.jpg but it only works when I get http://localhost:8100/assets/assets/asoria/image/--/666e73a/--/default/train/99/image/image.jpg) maybe it is a configuration with the nginx redirect configuration). | fsspec storage instead of boto3: Related to comment https://github.com/huggingface/datasets-server/pull/1882#issuecomment-1740688443
- Use a storage agnostic tool (fsspec) to store images/audios for assets and cached assets.
- Adding an abstraction layer for "file" and "s3" storage.
- For docker containers, we will use "file" storage but for staging/prod we will use "s3".
- I removed all the dependencies of boto3 and moto (and added s3fs)
- I removed all the dependencies of `/storage/assets` and `/storage/cached-assets` (We don't need it anymore since images and audios are being written directly to the file system)
- I removed the disk metrics (They need to be added again but getting the values from S3, maybe can be handled in another PR)
- Will close https://github.com/huggingface/datasets-server/issues/1887 and https://github.com/huggingface/datasets-server/issues/1885
Pending issue:
- For reverse-proxy, I am not able to get the files stored in /assets and /cached-assets volume (My generated URL is http://localhost:8100/assets/asoria/image/--/666e73a/--/default/train/99/image/image.jpg but it only works when I get http://localhost:8100/assets/assets/asoria/image/--/666e73a/--/default/train/99/image/image.jpg) maybe it is a configuration with the nginx redirect configuration). | closed | 2023-10-12T20:21:53Z | 2023-10-30T14:34:45Z | 2023-10-30T14:34:43Z | AndreaFrancis |
1,940,390,374 | Make authentication asynchronous | Make authentication asynchronous by using `httpx` instead of `requests`. | Make authentication asynchronous: Make authentication asynchronous by using `httpx` instead of `requests`. | closed | 2023-10-12T16:46:02Z | 2023-10-13T08:04:22Z | 2023-10-13T08:04:20Z | albertvillanova |
1,940,369,820 | Increase the number of CPUs from 2 to 8 for heavy workers | Currently loading data into table takes too much time but should be decreased proportionally to the number of cpus.
For example, on my local machine using 2 cpus loads 4Gb parquet files of [Open-Orca/OpenOrca](https://huggingface.co/datasets/Open-Orca/OpenOrca) for 20 sec and with 8 cpus for 5 sec.
Can we try this? | Increase the number of CPUs from 2 to 8 for heavy workers: Currently loading data into table takes too much time but should be decreased proportionally to the number of cpus.
For example, on my local machine using 2 cpus loads 4Gb parquet files of [Open-Orca/OpenOrca](https://huggingface.co/datasets/Open-Orca/OpenOrca) for 20 sec and with 8 cpus for 5 sec.
Can we try this? | closed | 2023-10-12T16:36:05Z | 2023-10-12T21:55:36Z | 2023-10-12T21:55:35Z | polinaeterna |
1,940,292,868 | fix cron job metadata name lenght | When deploying to staging, this error appeared:
`one or more objects failed to apply, reason: CronJob.batch "staging-datasets-server-job-clean-duckdb-index-downloads" is invalid: metadata.name: Invalid value: "staging-datasets-server-job-clean-duckdb-index-downloads": must be no more than 52 characters,CronJob.batch "staging-datasets-server-job-clean-duckdb-index-job-runner" is invalid: metadata.name: Invalid value: "staging-datasets-server-job-clean-duckdb-index-job-runner": must be no more than 52 characters`
Reducing the job name to see if it fix the error | fix cron job metadata name lenght: When deploying to staging, this error appeared:
`one or more objects failed to apply, reason: CronJob.batch "staging-datasets-server-job-clean-duckdb-index-downloads" is invalid: metadata.name: Invalid value: "staging-datasets-server-job-clean-duckdb-index-downloads": must be no more than 52 characters,CronJob.batch "staging-datasets-server-job-clean-duckdb-index-job-runner" is invalid: metadata.name: Invalid value: "staging-datasets-server-job-clean-duckdb-index-job-runner": must be no more than 52 characters`
Reducing the job name to see if it fix the error | closed | 2023-10-12T15:55:38Z | 2023-10-12T16:16:45Z | 2023-10-12T16:16:45Z | AndreaFrancis |
1,940,267,701 | Fix admin service in dev docker compose | It was missing libapi, a better timeout.
On my laptop the import of sounfile without libsndfile from datasets is causing an error so I also added it to the dev dockerfile. | Fix admin service in dev docker compose: It was missing libapi, a better timeout.
On my laptop the import of sounfile without libsndfile from datasets is causing an error so I also added it to the dev dockerfile. | closed | 2023-10-12T15:41:21Z | 2023-10-19T09:23:09Z | 2023-10-19T09:23:08Z | lhoestq |
1,940,255,563 | Add date type support in cache responses | Mongo only supports dates with time (not date alone) because of bson.
So the `date` python type from `date32` pyarrow type couldn't be saved in mongo.
I fixed it by saving the `date` the same way as `string` objects.
Fix https://github.com/huggingface/datasets-server/issues/86
e.g. in first-rows for [aborruso](https://huggingface.co/aborruso)[/pnrr](https://huggingface.co/datasets/aborruso/pnrr)
```json
{
"dataset": "__DUMMY_DATASETS_SERVER_USER__/pnrr",
"config": "default",
"split": "train",
"features": [
{
"feature_idx": 0,
"name": "Programma",
"type": {
"dtype": "string",
"_type": "Value"
}
},
...
{
"feature_idx": 50,
"name": "Data di Estrazione",
"type": {
"dtype": "date32",
"_type": "Value"
}
}
],
"rows": [
{
"row_idx": 0,
"row": {
"Programma": "PNRR",
...
"Data di Estrazione": "2023-06-13"
},
"truncated_cells": []
},
...
]
}
```
EDIT: also added support for time, timedelta and decimal
I also checked binary but it is already supported, though its format is not ideal (we get a list of integers from mongo) | Add date type support in cache responses: Mongo only supports dates with time (not date alone) because of bson.
So the `date` python type from `date32` pyarrow type couldn't be saved in mongo.
I fixed it by saving the `date` the same way as `string` objects.
Fix https://github.com/huggingface/datasets-server/issues/86
e.g. in first-rows for [aborruso](https://huggingface.co/aborruso)[/pnrr](https://huggingface.co/datasets/aborruso/pnrr)
```json
{
"dataset": "__DUMMY_DATASETS_SERVER_USER__/pnrr",
"config": "default",
"split": "train",
"features": [
{
"feature_idx": 0,
"name": "Programma",
"type": {
"dtype": "string",
"_type": "Value"
}
},
...
{
"feature_idx": 50,
"name": "Data di Estrazione",
"type": {
"dtype": "date32",
"_type": "Value"
}
}
],
"rows": [
{
"row_idx": 0,
"row": {
"Programma": "PNRR",
...
"Data di Estrazione": "2023-06-13"
},
"truncated_cells": []
},
...
]
}
```
EDIT: also added support for time, timedelta and decimal
I also checked binary but it is already supported, though its format is not ideal (we get a list of integers from mongo) | closed | 2023-10-12T15:34:20Z | 2023-10-17T12:56:02Z | 2023-10-16T15:40:43Z | lhoestq |
1,939,480,752 | Remove duplicate admin utils | Remove duplicate `admin.utils` and use `libapi.utils` instead. | Remove duplicate admin utils: Remove duplicate `admin.utils` and use `libapi.utils` instead. | closed | 2023-10-12T08:27:28Z | 2023-10-12T09:27:44Z | 2023-10-12T09:27:43Z | albertvillanova |
1,938,302,717 | minor logging improvement in clean hf dset cache | I noticed that the errors counter was incorrect (was counting subdirectories) | minor logging improvement in clean hf dset cache: I noticed that the errors counter was incorrect (was counting subdirectories) | closed | 2023-10-11T17:10:38Z | 2023-10-12T12:53:57Z | 2023-10-12T12:53:56Z | lhoestq |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.