You are viewing a plain text version of this content. The canonical link for it is here.
Posted to github@beam.apache.org by "claudevdm (via GitHub)" <gi...@apache.org> on 2026/06/02 12:30:40 UTC
[PR] Jsonpick ver [beam]
claudevdm opened a new pull request, #38769:
URL: https://github.com/apache/beam/pull/38769
Bump jsonpickle upper bound.
Version 4.0.0 + Uses `decode(safe=True)` by default which avoids eval during decoding, which is a breaking change.
https://github.com/jsonpickle/jsonpickle/blob/c3fa14e6c4a1099f7d1c8b0898b6bbf8aade5143/CHANGES.rst#v400.
------------------------
Thank you for your contribution! Follow this checklist to help us incorporate your contribution quickly and easily:
- [ ] Mention the appropriate issue in your description (for example: `addresses #123`), if applicable. This will automatically add a link to the pull request in the issue. If you would like the issue to automatically close on merging the pull request, comment `fixes #<ISSUE NUMBER>` instead.
- [ ] Update `CHANGES.md` with noteworthy changes.
- [ ] If this contribution is large, please file an Apache [Individual Contributor License Agreement](https://www.apache.org/licenses/icla.pdf).
See the [Contributor Guide](https://beam.apache.org/contribute) for more tips on [how to make review process smoother](https://github.com/apache/beam/blob/master/CONTRIBUTING.md#make-the-reviewers-job-easier).
To check the build health, please visit [https://github.com/apache/beam/blob/master/.test-infra/BUILD_STATUS.md](https://github.com/apache/beam/blob/master/.test-infra/BUILD_STATUS.md)
GitHub Actions Tests Status (on master branch)
------------------------------------------------------------------------------------------------
[](https://github.com/apache/beam/actions?query=workflow%3A%22Build+python+source+distribution+and+wheels%22+branch%3Amaster+event%3Aschedule)
[](https://github.com/apache/beam/actions?query=workflow%3A%22Python+Tests%22+branch%3Amaster+event%3Aschedule)
[](https://github.com/apache/beam/actions?query=workflow%3A%22Java+Tests%22+branch%3Amaster+event%3Aschedule)
[](https://github.com/apache/beam/actions?query=workflow%3A%22Go+tests%22+branch%3Amaster+event%3Aschedule)
See [CI.md](https://github.com/apache/beam/blob/master/CI.md) for more information about GitHub Actions CI or the [workflows README](https://github.com/apache/beam/blob/master/.github/workflows/README.md) to see a list of phrases to trigger workflows.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscribe@beam.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] Bump jsonpickle upper bound. [beam]
Posted by "github-actions[bot] (via GitHub)" <gi...@apache.org>.
github-actions[bot] commented on PR #38769:
URL: https://github.com/apache/beam/pull/38769#issuecomment-4603642195
Assigning reviewers:
R: @shunping for label python.
Note: If you would like to opt out of this review, comment `assign to next reviewer`.
Available commands:
- `stop reviewer notifications` - opt out of the automated review tooling
- `remind me after tests pass` - tag the comment author after tests pass
- `waiting on author` - shift the attention set back to the author (any comment or push by the author will return the attention set to the reviewers)
The PR bot will only process comments in the main thread (not review comments).
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscribe@beam.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] Bump jsonpickle upper bound. [beam]
Posted by "damccorm (via GitHub)" <gi...@apache.org>.
damccorm commented on code in PR #38769:
URL: https://github.com/apache/beam/pull/38769#discussion_r3341627633
##########
sdks/python/apache_beam/ml/transforms/base.py:
##########
@@ -591,7 +592,18 @@ def save_attributes(
def load_attributes(artifact_location):
with FileSystems.open(os.path.join(artifact_location, _ATTRIBUTE_FILE_NAME),
'rb') as f:
- return jsonpickle.decode(f.read())
+ # load_attributes runs eagerly during MLTransform.expand() at pipeline
+ # construction time, so the pipeline's options are available via the
+ # construction-time context.
+ pipeline_options = get_pipeline_options()
+ safe = True
+ if (pipeline_options is not None and
+ pipeline_options.is_compat_version_prior_to("2.75.0")):
+ # Keep the pre-2.75.0 jsonpickle behavior (safe=False permits
+ # eval-based decoding) for backwards compatibility with already-staged
+ # artifacts.
+ safe = False
Review Comment:
Is there actually a downside to always setting safe=False? IIUC, this is protecting against remote code execution exploits that allow escalating privileges. But if you are able to compromise this artifact file, I think you already functionally have code execution privileges since it allows you to define (semi) arbitrary transforms.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscribe@beam.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] Bump jsonpickle upper bound. [beam]
Posted by "claudevdm (via GitHub)" <gi...@apache.org>.
claudevdm commented on code in PR #38769:
URL: https://github.com/apache/beam/pull/38769#discussion_r3341858513
##########
sdks/python/apache_beam/ml/transforms/base.py:
##########
@@ -591,7 +592,18 @@ def save_attributes(
def load_attributes(artifact_location):
with FileSystems.open(os.path.join(artifact_location, _ATTRIBUTE_FILE_NAME),
'rb') as f:
- return jsonpickle.decode(f.read())
+ # load_attributes runs eagerly during MLTransform.expand() at pipeline
+ # construction time, so the pipeline's options are available via the
+ # construction-time context.
+ pipeline_options = get_pipeline_options()
+ safe = True
+ if (pipeline_options is not None and
+ pipeline_options.is_compat_version_prior_to("2.75.0")):
+ # Keep the pre-2.75.0 jsonpickle behavior (safe=False permits
+ # eval-based decoding) for backwards compatibility with already-staged
+ # artifacts.
+ safe = False
Review Comment:
I also think it is reasonable to just skip compat flag completely. I doubt it will break anyone, and if it does they can file a bug and we can add a compat flag in an upcoming beam version?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscribe@beam.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] Bump jsonpickle upper bound. [beam]
Posted by "damccorm (via GitHub)" <gi...@apache.org>.
damccorm commented on code in PR #38769:
URL: https://github.com/apache/beam/pull/38769#discussion_r3342226784
##########
sdks/python/apache_beam/ml/transforms/base.py:
##########
@@ -591,7 +592,18 @@ def save_attributes(
def load_attributes(artifact_location):
with FileSystems.open(os.path.join(artifact_location, _ATTRIBUTE_FILE_NAME),
'rb') as f:
- return jsonpickle.decode(f.read())
+ # load_attributes runs eagerly during MLTransform.expand() at pipeline
+ # construction time, so the pipeline's options are available via the
+ # construction-time context.
+ pipeline_options = get_pipeline_options()
+ safe = True
+ if (pipeline_options is not None and
+ pipeline_options.is_compat_version_prior_to("2.75.0")):
+ # Keep the pre-2.75.0 jsonpickle behavior (safe=False permits
+ # eval-based decoding) for backwards compatibility with already-staged
+ # artifacts.
+ safe = False
Review Comment:
> Is the benefit of always setting to False simplicity? I don't have a strong opinion here.
Yeah - IMO this is either a vulnerability which we should unconditionally patch, or it is not and we can retain a single behavior.
> I also think it is reasonable to just skip compat flag completely. I doubt it will break anyone, and if it does they can file a bug and we can add a compat flag in an upcoming beam version?
I agree with this. I actually also realized that the compat flag doesn't work here anyways - these artifacts persist across (usually batch) pipeline runs, so it isn't coming from an update, its coming from a subsequent pipeline run.
My slight preference would be `safe=False`, but I'm good with either approach as long as we skip the compat flag piece and will defer to you here.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscribe@beam.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] Bump jsonpickle upper bound. [beam]
Posted by "github-actions[bot] (via GitHub)" <gi...@apache.org>.
github-actions[bot] commented on PR #38769:
URL: https://github.com/apache/beam/pull/38769#issuecomment-4670728527
Reminder, please take a look at this pr: @shunping
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscribe@beam.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] Jsonpick ver [beam]
Posted by "gemini-code-assist[bot] (via GitHub)" <gi...@apache.org>.
gemini-code-assist[bot] commented on code in PR #38769:
URL: https://github.com/apache/beam/pull/38769#discussion_r3341187524
##########
sdks/python/apache_beam/ml/transforms/base_test.py:
##########
@@ -841,6 +843,46 @@ def test_save_and_load_run_inference(self):
self.assertListEqual(
get_keys(model_handler), get_keys(loaded_model_handler))
+ @parameterized.expand([
+ # Pipelines pinned to a version older than 2.75.0 keep the pre-2.75.0
+ # jsonpickle behavior (safe=False, which permits eval-based decoding).
+ param(update_compatibility_version='2.74.0', expected_safe=False),
+ # The breaking-change version itself and newer decode securely.
+ param(update_compatibility_version='2.75.0', expected_safe=True),
+ # Pipelines that do not set the option (the common case) decode securely.
+ param(update_compatibility_version=None, expected_safe=True),
+ ])
+ def test_load_attributes_safe_flag_follows_compat_version(
+ self, update_compatibility_version, expected_safe):
+ data = [{'x': 'Hello world'}, {'x': 'Apache Beam'}]
+ with beam.Pipeline() as p:
+ _ = (
+ p
+ | beam.Create(data)
+ | base.MLTransform(
+ write_artifact_location=self.artifact_location).with_transform(
+ FakeEmbeddingsManager(columns=['x'])))
+
+ # FakeEmbeddingsManager reverses the values of the embedded columns.
+ expected_data = [{'x': d['x'][::-1]} for d in data]
+
+ options = PipelineOptions(
+ update_compatibility_version=update_compatibility_version)
+ with mock.patch.object(base.jsonpickle,
+ 'decode',
+ wraps=base.jsonpickle.decode) as mock_decode:
+ with beam.Pipeline(options=options) as p:
+ result = (
+ p
+ | beam.Create(data)
+ | base.MLTransform(read_artifact_location=self.artifact_location))
+ assert_that(result, equal_to(expected_data))
+
+ safe_flags = [
+ call.kwargs.get('safe') for call in mock_decode.call_args_list
+ ]
+ self.assertEqual(safe_flags, [expected_safe])
Review Comment:
To prevent test failures on environments running `jsonpickle < 4.0.0`, we should conditionally assert the `safe` flag only if the parameter is supported by the installed version of `jsonpickle`.
```python
import inspect
if 'safe' in inspect.signature(base.jsonpickle.decode).parameters:
safe_flags = [
call.kwargs.get('safe') for call in mock_decode.call_args_list
]
self.assertEqual(safe_flags, [expected_safe])
else:
self.assertTrue(mock_decode.called)
```
##########
sdks/python/apache_beam/ml/transforms/base.py:
##########
@@ -591,7 +592,18 @@ def save_attributes(
def load_attributes(artifact_location):
with FileSystems.open(os.path.join(artifact_location, _ATTRIBUTE_FILE_NAME),
'rb') as f:
- return jsonpickle.decode(f.read())
+ # load_attributes runs eagerly during MLTransform.expand() at pipeline
+ # construction time, so the pipeline's options are available via the
+ # construction-time context.
+ pipeline_options = get_pipeline_options()
+ safe = True
+ if (pipeline_options is not None and
+ pipeline_options.is_compat_version_prior_to("2.75.0")):
+ # Keep the pre-2.75.0 jsonpickle behavior (safe=False permits
+ # eval-based decoding) for backwards compatibility with already-staged
+ # artifacts.
+ safe = False
+ return jsonpickle.decode(f.read(), safe=safe)
Review Comment:
Since `jsonpickle` versions `< 4.0.0` are still supported (as specified in `setup.py` with `jsonpickle>=3.0.0,<5.0.0`), calling `jsonpickle.decode` with the `safe` keyword argument will raise a `TypeError` on environments running older versions of `jsonpickle`. We should conditionally pass the `safe` argument only if it is supported by the installed version of `jsonpickle`.
```suggestion
pipeline_options = get_pipeline_options()
safe = True
if (pipeline_options is not None and
pipeline_options.is_compat_version_prior_to("2.75.0")):
# Keep the pre-2.75.0 jsonpickle behavior (safe=False permits
# eval-based decoding) for backwards compatibility with already-staged
# artifacts.
safe = False
import inspect
kwargs = {}
if 'safe' in inspect.signature(jsonpickle.decode).parameters:
kwargs['safe'] = safe
return jsonpickle.decode(f.read(), **kwargs)
```
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscribe@beam.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] Bump jsonpickle upper bound. [beam]
Posted by "github-actions[bot] (via GitHub)" <gi...@apache.org>.
github-actions[bot] commented on PR #38769:
URL: https://github.com/apache/beam/pull/38769#issuecomment-4719631946
Assigning new set of reviewers because Pr has gone too long without review. If you would like to opt out of this review, comment `assign to next reviewer`:
R: @damccorm for label python.
Available commands:
- `stop reviewer notifications` - opt out of the automated review tooling
- `remind me after tests pass` - tag the comment author after tests pass
- `waiting on author` - shift the attention set back to the author (any comment or push by the author will return the attention set to the reviewers)
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscribe@beam.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] Bump jsonpickle upper bound. [beam]
Posted by "claudevdm (via GitHub)" <gi...@apache.org>.
claudevdm merged PR #38769:
URL: https://github.com/apache/beam/pull/38769
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscribe@beam.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] Bump jsonpickle upper bound. [beam]
Posted by "claudevdm (via GitHub)" <gi...@apache.org>.
claudevdm commented on code in PR #38769:
URL: https://github.com/apache/beam/pull/38769#discussion_r3341850410
##########
sdks/python/apache_beam/ml/transforms/base.py:
##########
@@ -591,7 +592,18 @@ def save_attributes(
def load_attributes(artifact_location):
with FileSystems.open(os.path.join(artifact_location, _ATTRIBUTE_FILE_NAME),
'rb') as f:
- return jsonpickle.decode(f.read())
+ # load_attributes runs eagerly during MLTransform.expand() at pipeline
+ # construction time, so the pipeline's options are available via the
+ # construction-time context.
+ pipeline_options = get_pipeline_options()
+ safe = True
+ if (pipeline_options is not None and
+ pipeline_options.is_compat_version_prior_to("2.75.0")):
+ # Keep the pre-2.75.0 jsonpickle behavior (safe=False permits
+ # eval-based decoding) for backwards compatibility with already-staged
+ # artifacts.
+ safe = False
Review Comment:
The default in v 4+ is safe=True, so I think it is best if we align with the default behavior in the underlying library?
Is the benefit of always setting to False simplicity? I don't have a strong opinion here.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscribe@beam.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] Bump jsonpickle upper bound. [beam]
Posted by "claudevdm (via GitHub)" <gi...@apache.org>.
claudevdm commented on code in PR #38769:
URL: https://github.com/apache/beam/pull/38769#discussion_r3341277715
##########
sdks/python/apache_beam/ml/transforms/base_test.py:
##########
@@ -841,6 +843,46 @@ def test_save_and_load_run_inference(self):
self.assertListEqual(
get_keys(model_handler), get_keys(loaded_model_handler))
+ @parameterized.expand([
+ # Pipelines pinned to a version older than 2.75.0 keep the pre-2.75.0
+ # jsonpickle behavior (safe=False, which permits eval-based decoding).
+ param(update_compatibility_version='2.74.0', expected_safe=False),
+ # The breaking-change version itself and newer decode securely.
+ param(update_compatibility_version='2.75.0', expected_safe=True),
+ # Pipelines that do not set the option (the common case) decode securely.
+ param(update_compatibility_version=None, expected_safe=True),
+ ])
+ def test_load_attributes_safe_flag_follows_compat_version(
+ self, update_compatibility_version, expected_safe):
+ data = [{'x': 'Hello world'}, {'x': 'Apache Beam'}]
+ with beam.Pipeline() as p:
+ _ = (
+ p
+ | beam.Create(data)
+ | base.MLTransform(
+ write_artifact_location=self.artifact_location).with_transform(
+ FakeEmbeddingsManager(columns=['x'])))
+
+ # FakeEmbeddingsManager reverses the values of the embedded columns.
+ expected_data = [{'x': d['x'][::-1]} for d in data]
+
+ options = PipelineOptions(
+ update_compatibility_version=update_compatibility_version)
+ with mock.patch.object(base.jsonpickle,
+ 'decode',
+ wraps=base.jsonpickle.decode) as mock_decode:
+ with beam.Pipeline(options=options) as p:
+ result = (
+ p
+ | beam.Create(data)
+ | base.MLTransform(read_artifact_location=self.artifact_location))
+ assert_that(result, equal_to(expected_data))
+
+ safe_flags = [
+ call.kwargs.get('safe') for call in mock_decode.call_args_list
+ ]
+ self.assertEqual(safe_flags, [expected_safe])
Review Comment:
Safe argument was added in 3.0.4. Bumped lower bound
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscribe@beam.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] Jsonpick ver [beam]
Posted by "gemini-code-assist[bot] (via GitHub)" <gi...@apache.org>.
gemini-code-assist[bot] commented on PR #38769:
URL: https://github.com/apache/beam/pull/38769#issuecomment-4602423470
## Summary of Changes
Hello, I'm Gemini Code Assist[^1]! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!
This pull request updates the jsonpickle dependency to allow versions up to 5.0.0. To accommodate the breaking change in jsonpickle 4.0.0, which introduced a more secure default decoding mode, the changes include logic to conditionally toggle the 'safe' decoding flag based on the pipeline's compatibility version. This ensures that existing artifacts remain readable while adopting secure defaults for newer pipelines.
### Highlights
* **Dependency Update**: Updated the jsonpickle dependency upper bound from 4.0.0 to 5.0.0 to support newer versions.
* **Security and Compatibility**: Implemented conditional logic in MLTransform to handle jsonpickle's breaking change in 4.0.0, which defaults to safe decoding (disabling eval). Pipelines using compatibility versions prior to 2.75.0 will retain the legacy behavior (safe=False) for backward compatibility.
* **Testing**: Added a new test case to verify that the jsonpickle 'safe' flag correctly respects the pipeline compatibility version settings.
<details>
<summary><b>New Features</b></summary>
🧠You can now enable **Memory** (public preview) to help **Gemini Code Assist** learn from your team's feedback. This makes future code reviews more consistent and personalized to your project's style. **Click [here](https://codeassist.google/code-review/login) to enable Memory in your admin console.**
</details>
<details>
<summary><b>Using Gemini Code Assist</b></summary>
<br>
The full guide for Gemini Code Assist can be found on our [documentation page](https://developers.google.com/gemini-code-assist/docs/review-github-code), here are some quick tips.
<b>Invoking Gemini</b>
You can request assistance from Gemini at any point by creating a comment using either `/gemini <command>` or `@gemini-code-assist <command>`. Below is a summary of the supported commands on the current page.
Feature | Command | Description
--- | --- | ---
Code Review | `/gemini review` | Performs a code review for the current pull request in its current state.
Pull Request Summary | `/gemini summary` | Provides a summary of the current pull request in its current state.
Comment | @gemini-code-assist | Responds in comments when explicitly tagged, both in pull request comments and review comments.
Help | `/gemini help` | Displays a list of available commands.
<b>Customization</b>
To customize the Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a `.gemini/` folder in the base of the repository. Detailed instructions can be found [here](https://developers.google.com/gemini-code-assist/docs/customize-gemini-behavior-github).
<b>Limitations & Feedback</b>
Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counterproductive. You can react with :thumbsup: and :thumbsdown: on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for GitHub and other Google products, sign up [here](https://google.qualtrics.com/jfe/form/SV_2cyuGuTWsEw84yG).
</details>
[^1]: Review the [Privacy Notices](https://policies.google.com/privacy), [Generative AI Prohibited Use Policy](https://policies.google.com/terms/generative-ai/use-policy), [Terms of Service](https://policies.google.com/terms), and learn how to configure Gemini Code Assist in GitHub [here](https://developers.google.com/gemini-code-assist/docs/customize-gemini-behavior-github). Gemini can make mistakes, so double check it and [use code with caution](https://support.google.com/legal/answer/13505487).
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscribe@beam.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org