Episodic search

[paul-paliychuk]

Important

Add episodic search functionality with new types and update version to 2.14.0.

• Version Update:
  • Update version to 2.14.0 in pyproject.toml and client_wrapper.py.
• Graph Client:
  • Update set_entity_types() in graph.py to handle both entity and edge types.
• Types:
  • Add EpisodeType in episode_type.py.
  • Add GraphitiEpisode in graphiti_episode.py.
  • Update GraphSearchResults in graph_search_results.py to include episodes.
  • Update GraphSearchScope in graph_search_scope.py to include episodes.

^{This description was created by for a341d16. You can customize this summary. It will automatically update as commits are pushed.}

[ellipsis-dev]

Important

Looks good to me! 👍

Reviewed everything up to ecb706e in 2 minutes and 47 seconds. Click for details.

• Reviewed 288 lines of code in 9 files
• Skipped 0 files when reviewing.
• Skipped posting 10 draft comments. View those below.
• Modify your settings and rules to customize what types of comments Ellipsis leaves. And don't forget to react with 👍 or 👎 to teach Ellipsis.

1. pyproject.toml:3

• Draft comment:
  Version bump updated to 2.13.1, but note the client header uses 2.14.0. Ensure version consistency.
• Reason this comment was not posted:
  Decided after close inspection that this draft comment was likely wrong and/or not actionable: usefulness confidence = 10% vs. threshold = 50% The comment references a "client header" that uses version 2.14.0, but I don't see this header in the provided context. Without seeing this header file, I can't verify if there's actually a version mismatch. Version management could be intentionally different between components. This seems like speculation without clear evidence. The comment might be pointing to a real version inconsistency issue that could cause problems. Version mismatches between components can be serious. While version consistency is important, we need clear evidence of the inconsistency. The comment refers to information we can't verify, and we shouldn't assume the version bump to 2.13.1 is incorrect without seeing the client header. Delete the comment because it makes assumptions about version consistency without providing verifiable evidence in the current context.

2. src/zep_cloud/__init__.py:26

• Draft comment:
  New exports (EpisodeType, GraphitiEpisode) added to public API. Confirm documentation updates accordingly.
• Reason this comment was not posted:
  Comment did not seem useful. Confidence is useful = 0% <= threshold 50% This comment is asking the PR author to confirm documentation updates, which violates the rule against asking the author to confirm or ensure things. It does not provide a specific suggestion or point out a specific issue with the code.

3. src/zep_cloud/core/client_wrapper.py:20

• Draft comment:
  X-Fern-SDK-Version header is set to 2.14.0, which does not match the package version. Verify intended versioning.
• Reason this comment was not posted:
  Decided after close inspection that this draft comment was likely wrong and/or not actionable: usefulness confidence = 10% vs. threshold = 50% Since this is an auto-generated file by Fern, the version is likely set automatically during generation. The comment asks to "verify" the version, which falls under our rule about not asking for verification. Additionally, dependency/version related issues are explicitly excluded from our review scope. The change itself is just a version bump, which is likely intentional. The version mismatch could potentially cause real issues in production if it's incorrect. Maybe this deserves more investigation? While version mismatches can be serious, this file is auto-generated and version-related issues are explicitly excluded from our review scope. The automation tool (Fern) is responsible for version consistency. Delete the comment as it violates multiple rules: asking for verification and commenting on version/dependency issues.

4. src/zep_cloud/external_clients/graph.py:89

• Draft comment:
  The set_entity_types logic for sync and async clients is nearly duplicated. Consider refactoring shared functionality to reduce code duplication.
• Reason this comment was not posted:
  Comment was on unchanged code.

5. src/zep_cloud/types/episode_type.py:5

• Draft comment:
  The union type for EpisodeType includes typing.Any along with literals, which may undermine type safety. Confirm if this is the intended behavior.
• Reason this comment was not posted:
  Decided after close inspection that this draft comment was likely wrong and/or not actionable: usefulness confidence = 10% vs. threshold = 50% Since this is an auto-generated file from an API definition, any type safety concerns should be addressed in the API definition itself, not in the generated code. The comment asks for confirmation which violates the rule about not asking authors to confirm their intention. Additionally, if this was unintended, it would likely be caught during testing and type checking. The type definition might genuinely be problematic from a type safety perspective, and the comment raises a valid technical concern. However, this is an auto-generated file, so any changes would need to be made in the API definition, not here. Also, the comment violates the rule about asking for confirmation of intentions. Delete the comment because it asks for confirmation of intention and because changes to auto-generated files should be handled at the source definition level.

6. src/zep_cloud/types/graph_search_scope.py:5

• Draft comment:
  Including typing.Any in the GraphSearchScope union effectively bypasses the literal restrictions. Consider removing Any if only specific values are allowed.
• Reason this comment was not posted:
  Comment was on unchanged code.

7. src/zep_cloud/external_clients/graph.py:24

• Draft comment:
  Docstring examples in set_entity_types are clear, but consider refining the code block indentation for better readability.
• Reason this comment was not posted:
  Confidence changes required: 50% <= threshold 50% None

8. src/zep_cloud/__init__.py:37

• Draft comment:
  Typographical note: The new export 'GraphitiEpisode' might be misspelled—please verify if this was intended or if it should be 'GraphiteEpisode'.
• Reason this comment was not posted:
  Decided after close inspection that this draft comment was likely wrong and/or not actionable: usefulness confidence = 10% vs. threshold = 50% Since this is an auto-generated file, the type name likely comes from the API definition. The spelling 'Graphiti' could be intentional - it might be a product name or a specific term. Without access to the API definition or documentation, we can't be sure this is actually a typo. Making assumptions about naming in auto-generated code is risky. The comment raises a valid concern about potential misspelling. Names are important for maintainability. However, this is an auto-generated file where we can't directly fix issues - any fixes would need to be made in the source API definition. Also, 'Graphiti' could be intentionally spelled this way. We should delete this comment because it's speculative, requires more context to verify, and relates to auto-generated code that can't be directly modified in this file.

9. src/zep_cloud/__init__.py:105

• Draft comment:
  The identifier 'GraphitiEpisode' appears to be unusual. Please verify whether it is the intended spelling or if it should be something like 'GraphiteEpisode'.
• Reason this comment was not posted:
  Decided after close inspection that this draft comment was likely wrong and/or not actionable: usefulness confidence = 10% vs. threshold = 50% Since this is an auto-generated file from an API definition, the identifier naming would be defined elsewhere. The comment is asking for verification rather than pointing out a definite issue. Auto-generated files typically shouldn't be manually modified. The spelling could be intentional (e.g., "Graphiti" could be a product name or specific term). The spelling could genuinely be wrong in the source API definition. Just because a file is auto-generated doesn't mean its source can't have errors. While true, this comment violates the rule about not asking for verification. If there was a spelling error, it should be fixed in the API definition, not here. Additionally, without access to the API definition, we can't be certain this is actually an error. Delete the comment because it asks for verification rather than pointing out a definite issue, and it's on an auto-generated file where changes should be made in the source definition instead.

10. src/zep_cloud/types/__init__.py:97

• Draft comment:
  Possible typo: "GraphitiEpisode" might be intended to be "GraphiteEpisode". Please verify the correct spelling for consistency.
• Reason this comment was not posted:
  Decided after close inspection that this draft comment was likely wrong and/or not actionable: usefulness confidence = 10% vs. threshold = 50% Since this is an auto-generated file (from Fern), the type names would come from the API definition. The name "Graphiti" could be intentional and might refer to a specific concept or feature. Making assumptions about typos in auto-generated code is risky. Additionally, if this were a real typo, it would likely cause build/compilation issues that would be caught automatically. The name could genuinely be a typo that made it into the API definition. "Graphite" is a more common term in software than "Graphiti". Even if it were a typo, this file is auto-generated, so any fix would need to be made in the API definition, not here. Also, "Graphiti" could be an intentional spelling for a specific feature or product name. Delete the comment. We shouldn't make assumptions about typos in auto-generated code, especially when the name could be intentional.

Workflow ID: wflow_VS9aq3Sd3i6bhr3W

^{You can customize by changing your verbosity settings, reacting with 👍 or 👎, replying to comments, or adding code review rules.}

[ellipsis-dev]

Important

Looks good to me! 👍

Reviewed a341d16 in 30 seconds. Click for details.

• Reviewed 12 lines of code in 1 files
• Skipped 0 files when reviewing.
• Skipped posting 1 draft comments. View those below.
• Modify your settings and rules to customize what types of comments Ellipsis leaves. And don't forget to react with 👍 or 👎 to teach Ellipsis.

1. pyproject.toml:3

• Draft comment:
  Version updated to 2.14.0. Ensure corresponding release notes/changelog are updated.
• Reason this comment was not posted:
  Confidence changes required: 0% <= threshold 50% None

Workflow ID: wflow_5Mlgb0ymx1R2PdMr

^{You can customize by changing your verbosity settings, reacting with 👍 or 👎, replying to comments, or adding code review rules.}