You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This update introduces minor modifications by adding trailing commas to enhance code readability and maintain consistency across various files. These changes do not affect the functionality of the code.
Refactored multiple modules across the codebase to improve readability, maintainability, and consistency by adding trailing commas, reformatting multiline arguments, and enhancing logic.
Enhanced recall options and MMR logic in session models, queries, and workflows to handle edge cases and improve functionality.
Updated and added test cases to align with new formatting standards and improve test coverage for various modules.
Improved documentation for session recall options and hybrid search parameters.
Updated JSON schemas for agent and task creation, removing outdated references and improving definitions.
Updated dependencies across multiple pyproject.toml files for flexibility and consistency.
Adjusted configuration files, including Ruff and environment variable settings, for better maintainability.
Changes walkthrough 📝
Relevant files
Enhancement
35 files
sync.py
Refactor sync logic and improve argument formatting.
cli/src/julep_cli/sync.py
Added trailing commas to function calls for consistency.
Improved readability by reformatting multiline function arguments.
Enhanced error messages with better formatting.
Adjusted logic for handling YAML changes in watch mode.
The PR consists of minor changes, primarily adding trailing commas to function calls and list definitions across multiple files for improved code readability and consistency. No significant impact on functionality.
Entelligence.ai can learn from your feedback. Simply add 👍 / 👎 emojis to teach it your preferences. More shortcuts below
Emoji Descriptions:
⚠️Potential Issue - May require further investigation.
🔒 Security Vulnerability - Fix to ensure system safety.
💻 Code Improvement - Suggestions to enhance code quality.
🔨 Refactor Suggestion - Recommendations for restructuring code.
ℹ️ Others - General comments and information.
Interact with the Bot:
Send a message or request using the format: @bot + *your message*
Example: @bot Can you suggest improvements for this code?
Help the Bot learn by providing feedback on its responses. @bot + *feedback*
Example: @bot Do not comment on `save_auth` function !
The gather_messages function was significantly refactored with new logic for handling recall options and MMR. Verify that the new implementation properly handles edge cases and maintains existing functionality.
# If recall is disabled, return earlyifnotrecall:
returnpast_messages, []
# Get recall configrecall_options=chat_context.session.recall_options# Get messages to search fromsearch_messages= [
msgformsgin (past_messages+new_raw_messages)[-(recall_options.num_search_messages) :]
ifisinstance(msg["content"], str) andmsg["role"] in ["user", "assistant"]
]
ifnotsearch_messages:
returnpast_messages, []
# Build search text and get embedding if neededembed_text="\n\n".join(
f"{msg.get('name') ormsg['role']}: {msg['content']}"formsginsearch_messages
).strip()
query_embedding=Noneifrecall_options.mode!="text":
[query_embedding, *_] =awaitlitellm.aembedding(
inputs=embed_text[-(recall_options.max_query_length) :],
embed_instruction="Represent the query for retrieving supporting documents: ",
)
# Get query text from last messagequery_text=search_messages[-1]["content"].strip()[: recall_options.max_query_length]
# Get owners to search docs fromactive_agent_id=chat_context.get_active_agent().idowners= [("user", user.id) foruserinchat_context.users]
owners.append(("agent", active_agent_id))
# Build search params based on modesearch_params=Noneifrecall_options.mode=="vector":
search_params=VectorDocSearchRequest(
lang=recall_options.lang,
limit=recall_options.limit,
metadata_filter=recall_options.metadata_filter,
confidence=recall_options.confidence,
mmr_strength=recall_options.mmr_strength,
vector=query_embedding,
)
elifrecall_options.mode=="hybrid":
search_params=HybridDocSearchRequest(
lang=recall_options.lang,
limit=recall_options.limit,
metadata_filter=recall_options.metadata_filter,
confidence=recall_options.confidence,
mmr_strength=recall_options.mmr_strength,
alpha=recall_options.alpha,
text=query_text,
vector=query_embedding,
)
elifrecall_options.mode=="text":
search_params=TextOnlyDocSearchRequest(
lang=recall_options.lang,
limit=recall_options.limit,
metadata_filter=recall_options.metadata_filter,
text=query_text,
)
else:
# Invalid mode, return earlyreturnpast_messages, []
# Execute searchsearch_fn, params=get_search_fn_and_params(search_params)
doc_references: list[DocReference] =awaitsearch_fn(
developer_id=developer.id,
owners=owners,
connection_pool=connection_pool,
**params,
)
# Apply MMR if enabled and applicableif (
recall_options.mmr_strength>0andlen(doc_references) >recall_options.limitandrecall_options.mode!="text"
):
# Filter docs with embeddingsdocs_with_embeddings= [
docfordocindoc_referencesifdoc.snippet.embeddingisnotNone
]
iflen(docs_with_embeddings) >=2:
# Apply MMRindices=maximal_marginal_relevance(
np.asarray(query_embedding),
[doc.snippet.embeddingfordocindocs_with_embeddings],
k=min(recall_options.limit, len(docs_with_embeddings)),
lambda_mult=1-recall_options.mmr_strength,
)
doc_references= [
docfori, docinenumerate(docs_with_embeddings) ifiinset(indices)
]
Major changes to session recall options types and defaults. Verify that these changes are backwards compatible and properly validated.
recall_options: VectorDocSearch|TextOnlyDocSearch|HybridDocSearch|None=None""" Recall options for the session """metadata: dict[str, Any] |None=NoneclassHybridDocSearch(BaseDocSearch):
model_config=ConfigDict(
populate_by_name=True,
)
mode: Literal["hybrid"] ="hybrid"""" The mode to use for the search. """confidence: Annotated[float, Field(ge=-1.0, le=1.0)] =0""" The confidence cutoff level """alpha: Annotated[float, Field(ge=0.0, le=1.0)] =0.5""" The weight to apply to BM25 vs Vector search results. 0 => pure BM25; 1 => pure vector; """mmr_strength: Annotated[float, Field(ge=0.0, lt=1.0)] =0.5""" MMR Strength (mmr_strength = 1 - mmr_lambda) """classHybridDocSearchUpdate(BaseDocSearchUpdate):
model_config=ConfigDict(
populate_by_name=True,
)
mode: Literal["hybrid"] ="hybrid"""" The mode to use for the search. """confidence: Annotated[float, Field(ge=-1.0, le=1.0)] =0""" The confidence cutoff level """alpha: Annotated[float, Field(ge=0.0, le=1.0)] =0.5""" The weight to apply to BM25 vs Vector search results. 0 => pure BM25; 1 => pure vector; """
⚠️ This PR is too big for Ellipsis, but support for larger PRs is coming soon. If you want us to prioritize this feature, let us know at [email protected]
Add error handling for empty or invalid query_embedding when using vector or hybrid search modes. Currently, if litellm.aembedding() fails, the code will crash without graceful fallback.
if recall_options.mode != "text":
- [query_embedding, *_] = await litellm.aembedding(- inputs=embed_text[-(recall_options.max_query_length) :],- embed_instruction="Represent the query for retrieving supporting documents: ",- )+ try:+ [query_embedding, *_] = await litellm.aembedding(+ inputs=embed_text[-(recall_options.max_query_length) :],+ embed_instruction="Represent the query for retrieving supporting documents: ",+ )+ except Exception as e:+ logger.error(f"Failed to generate embedding: {e}")+ return past_messages, []
Apply this suggestion
Suggestion importance[1-10]: 8
__
Why: The suggestion adds critical error handling for the embedding generation process, which could fail due to API issues or invalid inputs. Without this, the application would crash instead of gracefully handling the error.
Medium
Review hybrid search weight change
The default alpha value for hybrid search mode was changed from 0.75 to 0.5 without clear justification. This could significantly impact search behavior. Consider documenting the rationale for this change or reverting if unintentional.
[To ensure code accuracy, apply this suggestion manually]
Suggestion importance[1-10]: 8
__
Why: Changing the default alpha value from 0.75 to 0.5 could significantly impact search behavior by giving equal weight to BM25 and vector search instead of favoring vector search. This change requires careful review.
Medium
Validate search messages content
Add validation for empty or whitespace-only search_messages content before generating embeddings to prevent potential issues with empty inputs.
embed_text = "\n\n".join(
f"{msg.get('name') or msg['role']}: {msg['content']}" for msg in search_messages
+ if msg['content'].strip()
).strip()
+if not embed_text:+ return past_messages, []
Apply this suggestion
Suggestion importance[1-10]: 7
__
Why: The suggestion adds important validation to prevent issues with empty or whitespace-only message content, which could cause problems in the embedding generation process. This improves robustness and prevents potential runtime errors.
Medium
General
Remove duplicate error handling code
Remove duplicate error handling code for watch and force_remote check. The same check and error message is repeated twice unnecessarily.
-# if watch is true and force_remote is true, we raise an error-if watch and force_remote:- error_console.print(- Text("Error: Cannot use both --watch and --force-remote", style="bold red"),- )- raise typer.Exit(1)-
# When watch mode is enabled, force a local sync and disable remote forcing.
if watch:
if force_remote:
error_console.print(
Text("Error: Cannot use both --watch and --force-remote", style="bold red"),
)
raise typer.Exit(1)
+ console.print(+ Text(+ "Watch mode enabled. Forcing local sync and disabling force_remote.",+ style="bold green",+ ),+ )+ force_local = True+ force_remote = False
[To ensure code accuracy, apply this suggestion manually]
Suggestion importance[1-10]: 7
__
Why: The suggestion correctly identifies and removes redundant error handling code that checks for the same condition twice. This improves code maintainability and reduces duplication.
Medium
Replace lambda with proper function
The lambda function definition uses a noqa comment to suppress E731 (lambda-assignment). Instead of suppressing the warning, refactor this into a proper function definition for better readability and maintainability.
[To ensure code accuracy, apply this suggestion manually]
Suggestion importance[1-10]: 5
__
Why: Converting the lambda to a proper function would improve code readability and maintainability, though the functional impact is minimal since both approaches work equivalently.
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
![entelligence-ai-pr-reviews[bot]](https://avatars.githubusercontent.com/in/932812?s=60&v=4){kind=small}
User description
EntelligenceAI PR Summary
This update introduces minor modifications by adding trailing commas to enhance code readability and maintain consistency across various files. These changes do not affect the functionality of the code.PR Type
Enhancement, Tests, Formatting, Documentation, Configuration changes, Dependencies
Description
Refactored multiple modules across the codebase to improve readability, maintainability, and consistency by adding trailing commas, reformatting multiline arguments, and enhancing logic.
Enhanced recall options and MMR logic in session models, queries, and workflows to handle edge cases and improve functionality.
Updated and added test cases to align with new formatting standards and improve test coverage for various modules.
Improved documentation for session recall options and hybrid search parameters.
Updated JSON schemas for agent and task creation, removing outdated references and improving definitions.
Updated dependencies across multiple
pyproject.tomlfiles for flexibility and consistency.Adjusted configuration files, including Ruff and environment variable settings, for better maintainability.
Changes walkthrough 📝
35 files
sync.py
Refactor sync logic and improve argument formatting.cli/src/julep_cli/sync.py
gather_messages.py
Refactor gather_messages recall logic and MMR handling.agents-api/agents_api/queries/chat/gather_messages.py
Sessions.py
Enhance recall options in session models.agents-api/agents_api/autogen/Sessions.py
alphadefault value for hybrid mode.agents.py
Refactor agent CLI commands for readability.cli/src/julep_cli/agents.py
browserbase.py
Refactor Browserbase models for readability.integrations-service/integrations/models/browserbase.py
Sessions.py
Enhance recall options in session models.integrations-service/integrations/autogen/Sessions.py
alphadefault value for hybrid mode.__init__.py
Refactor task execution workflow for consistency.agents-api/agents_api/workflows/task_execution/init.py
tasks.py
Refactor task CLI commands for readability.cli/src/julep_cli/tasks.py
browserbase.py
Refactor Browserbase integration utilities.integrations-service/integrations/utils/integrations/browserbase.py
utils.py
Refactor CLI utilities for consistency.cli/src/julep_cli/utils.py
helpers.py
Refactor task execution helpers for readability.agents-api/agents_api/workflows/task_execution/helpers.py
search_docs.py
Refactor document search logic and MMR handling.agents-api/agents_api/routers/docs/search_docs.py
logs.py
Refactor CLI log display for consistency.cli/src/julep_cli/logs.py
importt.py
Refactor import CLI command for readability.cli/src/julep_cli/importt.py
tools.py
Refactor tool CLI commands for readability.cli/src/julep_cli/tools.py
db_exceptions.py
Refactor database exception handling.agents-api/agents_api/common/utils/db_exceptions.py
remote_browser.py
Refactor remote browser integration utilities.integrations-service/integrations/utils/integrations/remote_browser.py
litellm.py
Refactor LiteLLM client for readability.agents-api/agents_api/clients/litellm.py
tasks.py
Refactor task protocol for consistency.agents-api/agents_api/common/protocol/tasks.py
interceptors.py
Refactor interceptors for workflows and activities.agents-api/agents_api/common/interceptors.py
list_executions.py
Refactor execution listing query for consistency.agents-api/agents_api/queries/executions/list_executions.py
humanization_utils.py
Refactor humanization utilities for consistency.agents-api/agents_api/activities/humanization_utils.py
cloudinary.py
Refactor Cloudinary integration utilities.integrations-service/integrations/utils/integrations/cloudinary.py
codec.py
Refactor worker codec for consistency.agents-api/agents_api/worker/codec.py
chat.py
Refactor chat session router for consistency.agents-api/agents_api/routers/sessions/chat.py
cloudinary.py
Refactor Cloudinary models for readability.integrations-service/integrations/models/cloudinary.py
utils.py
Refactor activity utilities for consistency.agents-api/agents_api/activities/utils.py
temporal.py
Refactor Temporal client for consistency.agents-api/agents_api/clients/temporal.py
list_sessions.py
Enhanced session listing with transformation logic.agents-api/agents_api/queries/sessions/list_sessions.py
transformlambda to the@wrap_in_classdecorator.get_session.py
Enhanced session retrieval with transformation logic.agents-api/agents_api/queries/sessions/get_session.py
transformlambda to the@wrap_in_classdecorator.sessions.py
Enhanced type annotations in session protocol.agents-api/agents_api/common/protocol/sessions.py
dataparameter inmake_session.create_agent_request.json
Updated JSON schema for agent creation.schemas/create_agent_request.json
MessageandTools.VectorDocSearchandTextOnlyDocSearchreferences.create_task_request.json
Updated JSON schema for task creation.schemas/create_task_request.json
MessageandTools.VectorDocSearchandTextOnlyDocSearchreferences.openapi-1.0.0.yaml
Enhanced OpenAPI schema for session recall options.typespec/tsp-output/@typespec/openapi3/openapi-1.0.0.yaml
recall_optionsschema to useanyOf.models.tsp
Improved TypeSpec models for session recall options.typespec/sessions/models.tsp
modeproperties to use literal types.7 files
test_task_execution_workflow.py
Update task execution workflow tests for consistency.agents-api/tests/test_task_execution_workflow.py
test_prepare_for_step.py
Improve readability in test_prepare_for_step.agents-api/tests/test_prepare_for_step.py
test_chat_routes.py
Add tests for recall options in chat routes.agents-api/tests/test_chat_routes.py
test_base_evaluate.py
Improve readability in test_base_evaluate.agents-api/tests/test_base_evaluate.py
test_workflow_helpers.py
Improve readability in test_workflow_helpers.agents-api/tests/test_workflow_helpers.py
utils.py
Refactor test utilities for consistency.agents-api/tests/utils.py
test_execution_workflow.py
Update execution workflow tests for consistency.agents-api/tests/test_execution_workflow.py
2 files
env.py
Refactor environment variable configurations.agents-api/agents_api/env.py
ruff.toml
Updated Ruff configuration for ignored rules.ruff.toml
UP046to ignored rules.62 files
chat.py
Improved formatting and consistency in CLI chat module.cli/src/julep_cli/chat.py
PanelandTextcalls for better readability.test_docs_queries.py
Improved formatting in test functions for documentation queries.agents-api/tests/test_docs_queries.py
entities.py
Improved formatting in entity prompt generation.agents-api/agents_api/rec_sum/entities.py
models.py
Improved formatting in protocol model functions.agents-api/agents_api/common/protocol/models.py
llama_parse.py
Improved formatting in LlamaParse integration utility.integrations-service/integrations/utils/integrations/llama_parse.py
delete_file.py
Improved formatting in file deletion router.agents-api/agents_api/routers/files/delete_file.py
execute_integration.py
Improved formatting in integration execution utility.integrations-service/integrations/utils/execute_integration.py
mmr.py
Improved formatting in MMR query logic.agents-api/agents_api/queries/docs/mmr.py
execute.py
Improved formatting in execution router.integrations-service/integrations/routers/execution/execute.py
delete_agent.py
Improved formatting in agent deletion router.agents-api/agents_api/routers/agents/delete_agent.py
transition.py
Improved formatting in task execution transition workflow.agents-api/agents_api/workflows/task_execution/transition.py
patch_tool.py
Improved formatting in tool patch query.agents-api/agents_api/queries/tools/patch_tool.py
create_or_update_agent.py
Improved formatting in agent creation or update query.agents-api/agents_api/queries/agents/create_or_update_agent.py
generate_changelog.py
Improved formatting in changelog generation script.scripts/generate_changelog.py
list_execution_transitions.py
Improved formatting in execution transitions query.agents-api/agents_api/queries/executions/list_execution_transitions.py
nlp.py
Improved formatting in NLP utility functions.agents-api/agents_api/common/nlp.py
get_user.py
Improved formatting in user retrieval query.agents-api/agents_api/queries/users/get_user.py
json.py
Improved formatting in JSON utility functions.agents-api/agents_api/common/utils/json.py
update_agent.py
Improved formatting in agent update query.agents-api/agents_api/queries/agents/update_agent.py
test_task_routes.py
Improved formatting in task routes test cases.agents-api/tests/test_task_routes.py
list_agents.py
Improved formatting in agent listing router.agents-api/agents_api/routers/agents/list_agents.py
patch_user.py
Improved formatting in user patch query.agents-api/agents_api/queries/users/patch_user.py
get_agent.py
Improved formatting in agent retrieval query.agents-api/agents_api/queries/agents/get_agent.py
create_or_update_user.py
Improved formatting in user creation or update query.agents-api/agents_api/queries/users/create_or_update_user.py
get_doc_search.py
Improved formatting in document search utility.agents-api/agents_api/common/utils/get_doc_search.py
update_user.py
Improved formatting in user update query.agents-api/agents_api/queries/users/update_user.py
transition_step.py
Improved formatting in task transition step activity.agents-api/agents_api/activities/task_steps/transition_step.py
pg.py
Improved formatting in PostgreSQL client utility.agents-api/agents_api/clients/pg.py
delete_entries.py
Improved formatting in entry deletion query.agents-api/agents_api/queries/entries/delete_entries.py
get_file.py
Improved formatting in file retrieval router.agents-api/agents_api/routers/files/get_file.py
trim.py
Improved formatting in message trimming logic.agents-api/agents_api/rec_sum/trim.py
list_agents.py
Improved formatting in agent listing query.agents-api/agents_api/queries/agents/list_agents.py
list_tasks.py
Improved formatting in task listing query.agents-api/agents_api/queries/tasks/list_tasks.py
list_docs.py
Improved formatting in document listing router.agents-api/agents_api/routers/docs/list_docs.py
delete_session.py
Improved formatting in session deletion router.agents-api/agents_api/routers/sessions/delete_session.py
delete_user.py
Improved formatting in user deletion router.agents-api/agents_api/routers/users/delete_user.py
test_agent_queries.py
Improved formatting in agent query test cases.agents-api/tests/test_agent_queries.py
exceptions.py
Improved formatting in exception definitions.agents-api/agents_api/exceptions.py
prompt_step.py
Improved formatting in task prompt step activity.agents-api/agents_api/activities/task_steps/prompt_step.py
get_session_history.py
Improved formatting in session history retrieval router.agents-api/agents_api/routers/sessions/get_session_history.py
test_get_doc_search.py
Improved formatting in document search test cases.agents-api/tests/test_get_doc_search.py
summarize.py
Improved formatting in message summarization logic.agents-api/agents_api/rec_sum/summarize.py
get_session.py
Improved formatting in session retrieval router.agents-api/agents_api/routers/sessions/get_session.py
exceptions.py
Improved formatting in session exception definitions.agents-api/agents_api/routers/sessions/exceptions.py
ffmpeg.py
Improved formatting in FFmpeg integration utility.integrations-service/integrations/utils/integrations/ffmpeg.py
test_activities_utils.py
Improved formatting in activity utility test cases.agents-api/tests/test_activities_utils.py
app.py
Improved formatting in application initialization.agents-api/agents_api/app.py
spider.py
Improved formatting in Spider integration utility.integrations-service/integrations/utils/integrations/spider.py
list_sessions.py
Improved formatting in session listing router.agents-api/agents_api/routers/sessions/list_sessions.py
wrapper.py
Improved formatting in CLI wrapper utility.cli/src/julep_cli/wrapper.py
list_users.py
Improved formatting in user listing router.agents-api/agents_api/routers/users/list_users.py
init.py
Improved formatting in CLI initialization logic.cli/src/julep_cli/init.py
get_integration_tool.py
Improved formatting in integration tool retrieval.integrations-service/integrations/routers/integrations/get_integration_tool.py
executions.py
Improved formatting in CLI execution logic.cli/src/julep_cli/executions.py
utils.py
Improved formatting in query utility functions.agents-api/agents_api/queries/utils.py
update_execution.py
Improved formatting in task execution update router.agents-api/agents_api/routers/tasks/update_execution.py
execute_system.py
Improved formatting in system execution activity.agents-api/agents_api/activities/execute_system.py
execute_api_call.py
Improved formatting in API call execution activity.agents-api/agents_api/activities/execute_api_call.py
execute_integration.py
Improved formatting in integration execution activity.agents-api/agents_api/activities/execute_integration.py
demo.py
Improved formatting in demo activity.agents-api/agents_api/activities/demo.py
render.py
Improved formatting in session rendering logic.agents-api/agents_api/routers/sessions/render.py
wikipedia.py
Improved formatting in Wikipedia integration model.integrations-service/integrations/models/wikipedia.py
3 files
pyproject.toml
Updated dependencies and cleaned configuration.agents-api/pyproject.toml
>=for flexibility.pyproject.toml
Updated CLI dependencies for flexibility.cli/pyproject.toml
>=for flexibility.pyproject.toml
Updated dependencies and formatting in integrations service.integrations-service/pyproject.toml
ruffdependency version.2 files
sessions.mdx
Improved documentation for session recall options.documentation/docs/concepts/sessions.mdx
recall_optionsis not set.docs.mdx
Updated documentation for hybrid search parameters.documentation/docs/concepts/docs.mdx
alphain hybrid search.