Endpoint Changes

[shreyaspimpalgaonkar]

Important

Enhance chunk enrichment by allowing configuration overrides and handling empty strategies in ingestion_workflow.py and ingestion_service.py.

• Behavior:
  • In ingestion_workflow.py, override server chunk enrichment settings with ingestion config using update_settings_from_dict.
  • In ingestion_service.py, handle empty enrichment strategies in _get_enriched_chunk_text() and ensure UUID conversion for context_chunk_ids.
• Functions:
  • Update chunk_enrichment() in ingestion_service.py to log settings and handle chunk enrichment with updated settings.
  • Modify chunk_enrichment() to accept chunk_enrichment_settings as a parameter.
• Misc:
  • Add import for update_settings_from_dict in ingestion_workflow.py.
  • Change model in r2r.toml from openai/gpt-4o to azure/gpt-4o in multiple sections.

^{This description was created by for acf09ee. It will automatically update as commits are pushed.}

[vercel]

The latest updates on your projects. Learn more about Vercel for Git ↗︎

Name	Status	Preview	Comments	Updated (UTC)
yc_demo	✅ Ready (Inspect)	Visit Preview	💬 Add feedback	Oct 25, 2024 0:19am
yc-demo	✅ Ready (Inspect)	Visit Preview	💬 Add feedback	Oct 25, 2024 0:19am

1 Skipped Deployment

Name	Status	Preview	Comments	Updated (UTC)
recommendation_platform	⬜️ Ignored (Inspect)			Oct 25, 2024 0:19am

[ellipsis-dev]

❌ Changes requested. Reviewed everything up to 5b4a663 in 56 seconds

More details

• Looked at 108 lines of code in 2 files
• Skipped 0 files when reviewing.
• Skipped posting 2 drafted comments based on config settings.

1. py/core/main/orchestration/hatchet/ingestion_workflow.py:181

• Draft comment:
  Ensure chunk_enrichment_settings is not None before accessing enable_chunk_enrichment.
• Reason this comment was not posted:
  Comment did not seem useful.

2. py/core/main/services/ingestion_service.py:366

• Draft comment:
  The check if chunk_enrichment_settings.strategies == []: can be simplified to if not chunk_enrichment_settings.strategies:.
• Reason this comment was not posted:
  Confidence changes required: 50%
  The PR adds a check for empty enrichment strategies, which is a good addition. However, the check for chunk_enrichment_settings.strategies == [] is redundant since an empty list is falsy in Python.

Workflow ID: wflow_ctlRaasXxNCa3npW

---

Want Ellipsis to fix these issues? Tag @ellipsis-dev in a comment. You can customize Ellipsis with 👍 / 👎 feedback, review rules, user-specific overrides, quiet mode, and more.

[ellipsis-dev]

👍 Looks good to me! Incremental review on acf09ee in 38 seconds

More details

• Looked at 92 lines of code in 2 files
• Skipped 0 files when reviewing.
• Skipped posting 6 drafted comments based on config settings.

1. py/r2r.toml:61

• Draft comment:
  The comment 'disabled by default' is misleading since enable_chunk_enrichment is set to true. Consider updating the comment to reflect the current setting.
• Reason this comment was not posted:
  Confidence changes required: 50%
  The change from 'openai/gpt-4o' to 'azure/gpt-4o' in the model configuration is consistent across multiple sections, which is good for consistency. However, the comment 'disabled by default' is misleading since the setting is enabled.

2. py/r2r.toml:84

• Draft comment:
  The change from 'openai/gpt-4o' to 'azure/gpt-4o' is consistent across multiple sections, which is good for consistency. Ensure that this change is intentional and aligns with the overall system requirements.
• Reason this comment was not posted:
  Confidence changes required: 33%
  The change from 'openai/gpt-4o' to 'azure/gpt-4o' in the model configuration is consistent across multiple sections, which is good for consistency. However, the comment 'disabled by default' is misleading since the setting is enabled.

3. py/r2r.toml:90

• Draft comment:
  The change from 'openai/gpt-4o' to 'azure/gpt-4o' is consistent across multiple sections, which is good for consistency. Ensure that this change is intentional and aligns with the overall system requirements.
• Reason this comment was not posted:
  Confidence changes required: 33%
  The change from 'openai/gpt-4o' to 'azure/gpt-4o' in the model configuration is consistent across multiple sections, which is good for consistency. However, the comment 'disabled by default' is misleading since the setting is enabled.

4. py/r2r.toml:95

• Draft comment:
  The change from 'openai/gpt-4o' to 'azure/gpt-4o' is consistent across multiple sections, which is good for consistency. Ensure that this change is intentional and aligns with the overall system requirements.
• Reason this comment was not posted:
  Confidence changes required: 33%
  The change from 'openai/gpt-4o' to 'azure/gpt-4o' in the model configuration is consistent across multiple sections, which is good for consistency. However, the comment 'disabled by default' is misleading since the setting is enabled.

5. py/r2r.toml:102

• Draft comment:
  The change from 'openai/gpt-4o' to 'azure/gpt-4o' is consistent across multiple sections, which is good for consistency. Ensure that this change is intentional and aligns with the overall system requirements.
• Reason this comment was not posted:
  Confidence changes required: 33%
  The change from 'openai/gpt-4o' to 'azure/gpt-4o' in the model configuration is consistent across multiple sections, which is good for consistency. However, the comment 'disabled by default' is misleading since the setting is enabled.

6. py/core/providers/kg/postgres.py:517

• Draft comment:
  The condition if filter_query != "" and filter_ids: is used to determine whether to include filter_ids in the query execution. Ensure that filter_ids is always a list to avoid potential runtime errors.
• Reason this comment was not posted:
  Confidence changes required: 50%
  The change from 'openai/gpt-4o' to 'azure/gpt-4o' in the model configuration is consistent across multiple sections, which is good for consistency. However, the comment 'disabled by default' is misleading since the setting is enabled.

Workflow ID: wflow_wjhgUg8MOjQa5cyD

---

You can customize Ellipsis with 👍 / 👎 feedback, review rules, user-specific overrides, quiet mode, and more.