Add tests for max_tokens propagation in LLM methods

[saiganesh47]

This test verifies that the max_tokens parameter is correctly propagated to the generate_typed method for different extraction functions.

Before you submit: make sure you followed the issue workflow in CONTRIBUTING.md — comment on the issue and wait for assignment before opening a PR, to avoid duplicate work.

Description

Type of Change

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• Documentation update
• Performance improvement
• Code refactoring

Related Issues

Closes #
Fixes #

Changes Made

Testing

• Tested locally
• Added tests for new functionality
• Package builds successfully (python -m build)

Test Commands

# Build the package
pip install build
python -m build

# Optional: Run your own tests
pytest tests/

# Optional: Format code
black semantica/
isort semantica/

Documentation

• Updated relevant documentation
• Added code examples if applicable
• Updated API reference if adding new APIs
• Updated cookbook if adding new examples
• No documentation changes needed

Breaking Changes

Breaking Changes: [Yes/No]

Checklist

• My code follows the project's style guidelines
• I have performed a self-review of my code
• I have commented my code, particularly in hard-to-understand areas
• My changes generate no new warnings
• Package builds successfully

Additional Notes

[qodo-code-review]

ⓘ Qodo reviews are paused because the subscription is no longer active. Ask your workspace admin to reactivate the subscription to resume reviews. Manage billing

[qodo-free-for-open-source-projects]

PR Summary by Qodo

Add unit tests ensuring max_tokens is forwarded to LLM generate_typed()

🧪 Tests 🕐 10-20 Minutes

AI Description

• Add unit tests covering max_tokens propagation across entity, relation, and triplet extraction
 helpers.
• Mock provider creation to assert generate_typed() receives the configured max_tokens value.

Diagram

graph TD
  T["TestMaxTokensPropagation"] --> M["extract_*_llm"] --> P["create_provider()"] --> L["LLM provider"] --> G["generate_typed()"]

Loading

High-Level Assessment

The following are alternative approaches to this PR:

1. Parametrize with pytest

• ➕ Less repetition: one parametrized test can cover all extraction functions
• ➕ Cleaner assertion reporting per parameter set
• ➖ Requires pytest adoption/usage patterns if the repo primarily uses unittest
• ➖ May need minor test harness changes (fixtures, markers)

2. Assert full call signature for generate_typed()

• ➕ Catches accidental changes to other forwarded kwargs beyond max_tokens
• ➖ More brittle tests (fails on unrelated internal refactors)
• ➖ Harder to maintain across provider implementations

Recommendation: Current approach (unittest + shared helper method + patched create_provider) is a good balance of readability and stability for verifying max_tokens forwarding. If the project standardizes on pytest, consider converting this to a single parametrized test to reduce duplication, but it’s not required for correctness.

Files changed (1) +93 / -0

Tests (1) +93 / -0

optimize reproduce_issue_176.pyAdd regression tests for max_tokens propagation into generate_typed() +93/-0

Add regression tests for max_tokens propagation into generate_typed()

• Introduces a unittest.TestCase that patches create_provider to return a mocked LLM provider. Adds three tests (entities, relations, triplets) that call the public extraction helpers with max_tokens and assert generate_typed() receives the same value.

tests/optimize reproduce_issue_176.py

[qodo-free-for-open-source-projects]

Code Review by Qodo

🐞 Bugs (3) 📘 Rule violations (0) 📜 Skill insights (0)

1. Test not collected 🐞 Bug ☼ Reliability

Description

The added file tests/optimize reproduce_issue_176.py is unlikely to be collected by pytest because
it doesn’t match default test_*.py / *_test.py patterns and includes a space in the filename. As
a result, CI likely won’t execute these tests, so the PR won’t actually add coverage.

Code

tests/optimize reproduce_issue_176.py[R1-4]

+import unittest
+from unittest.mock import MagicMock, patch
+
+from semantica.semantic_extract.methods import (

Evidence

Pytest is configured with testpaths = ["tests"] but no python_files override, so only default
patterns are collected; the added file’s name does not match those patterns and contains a space.

pyproject.toml[274-276]
tests/optimize reproduce_issue_176.py[1-4]

Agent prompt

The issue below was found during a code review. Follow the provided context and guidance below and implement a solution

## Issue description
The new test module name (`optimize reproduce_issue_176.py`) won’t be discovered by pytest’s default collection rules and the space in the filename makes it awkward/brittle to run from shells/CI.

## Issue Context
`pyproject.toml` configures `testpaths` but does not override `python_files`, so pytest uses its default module patterns.

## Fix Focus Areas
- pyproject.toml[274-276]
- tests/optimize reproduce_issue_176.py[1-93]

## Suggested fix
- Rename/move the file to a pytest-discoverable name, e.g. `tests/semantic_extract/test_max_tokens_propagation.py` (or `tests/test_max_tokens_propagation.py`).
- Avoid spaces in filenames.
- (Optional) If you truly want nonstandard names, explicitly set `[tool.pytest.ini_options].python_files` in `pyproject.toml`, but renaming is preferred.

ⓘ Copy this prompt and use it to remediate the issue with your preferred AI generation tools

---

2. Relations max_tokens dropped 🐞 Bug ≡ Correctness

Description

The new test asserts extract_relations_llm() forwards max_tokens into llm.generate_typed(),
but extract_relations_llm() currently builds call_kwargs without max_tokens before calling
generate_typed. Once the test is made discoverable, it will fail unless production code is updated
(or the assertion is changed).

Code

tests/optimize reproduce_issue_176.py[R45-48]

+        self.assertEqual(
+            call_kwargs.get("max_tokens"),
+            self.MAX_TOKENS,
+            "max_tokens was not propagated correctly",

Evidence

The test passes max_tokens and asserts it appears in the generate_typed kwargs, but the
production implementation for relations explicitly constructs call_kwargs without max_tokens and
passes only that dict into generate_typed.

tests/optimize reproduce_issue_176.py[33-49]
semantica/semantic_extract/methods.py[1904-1916]

Agent prompt

The issue below was found during a code review. Follow the provided context and guidance below and implement a solution

## Issue description
`extract_relations_llm()` does not propagate `max_tokens` to `llm.generate_typed()`, unlike `extract_entities_llm()` / `extract_triplets_llm()`. The newly added test expects `max_tokens` to be present in the `generate_typed` call kwargs.

## Issue Context
In `extract_relations_llm()`, the code constructs a limited `call_kwargs` dict (temperature/verbose/max_retries) and calls `llm.generate_typed(..., **call_kwargs)`, which drops `max_tokens`.

## Fix Focus Areas
- semantica/semantic_extract/methods.py[1904-1916]
- tests/optimize reproduce_issue_176.py[33-49]

## Suggested fix
- Include `max_tokens` (and any other intended generation limit knobs) in `call_kwargs` when present in `kwargs`, e.g.:
 - `if "max_tokens" in kwargs: call_kwargs["max_tokens"] = kwargs["max_tokens"]`
- Ensure the updated behavior is consistent with `extract_entities_llm()` / `extract_triplets_llm()` so the test can pass once properly collected.

ⓘ Copy this prompt and use it to remediate the issue with your preferred AI generation tools

---

3. Duplicate issue-176 test script 🐞 Bug ⚙ Maintainability

Description

The PR adds another issue-176 max_tokens propagation test script, but the repository already
contains tests/reproduce_issue_176.py covering the same three methods. Keeping multiple
near-duplicates makes it unclear which one is canonical and increases the chance of inconsistent
future edits.

Code

tests/optimize reproduce_issue_176.py[R12-18]

+class TestMaxTokensPropagation(unittest.TestCase):
+    MAX_TOKENS = 128000
+
+    def setUp(self):
+        self.mock_llm = MagicMock()
+        self.mock_llm.is_available.return_value = True
+

Evidence

The newly added file defines the same TestMaxTokensPropagation suite already present in
tests/reproduce_issue_176.py (relations/entities/triplets propagation checks).

tests/optimize reproduce_issue_176.py[12-90]
tests/reproduce_issue_176.py[1-98]

Agent prompt

The issue below was found during a code review. Follow the provided context and guidance below and implement a solution

## Issue description
There are now two scripts testing the same max_tokens propagation scenario (`tests/reproduce_issue_176.py` and the newly added `tests/optimize reproduce_issue_176.py`). This duplication is confusing and makes it easy for one to drift.

## Issue Context
The new file appears to be a cleaner refactor of the existing reproduce script.

## Fix Focus Areas
- tests/optimize reproduce_issue_176.py[1-93]
- tests/reproduce_issue_176.py[1-100]

## Suggested fix
- Choose a single canonical test module (preferably pytest-discoverable, e.g. `tests/semantic_extract/test_max_tokens_propagation.py`).
- Move/rename the better version there.
- Delete the redundant reproduce script(s) or keep *one* reproduction script outside `tests/` if it’s meant for manual runs only.

ⓘ Copy this prompt and use it to remediate the issue with your preferred AI generation tools

---

Tip of the day

💡 Did you know, you can enable the Remediation agent and Qodo fixes findings in a dedicated fix PR

More tips ↗ | Customize Qodo ↗ | Qodo docs ↗

[qodo-free-for-open-source-projects]

1. Test not collected 🐞 Bug ☼ Reliability

The added file tests/optimize reproduce_issue_176.py is unlikely to be collected by pytest because
it doesn’t match default test_*.py / *_test.py patterns and includes a space in the filename. As
a result, CI likely won’t execute these tests, so the PR won’t actually add coverage.

Agent Prompt

## Issue description
The new test module name (`optimize reproduce_issue_176.py`) won’t be discovered by pytest’s default collection rules and the space in the filename makes it awkward/brittle to run from shells/CI.

## Issue Context
`pyproject.toml` configures `testpaths` but does not override `python_files`, so pytest uses its default module patterns.

## Fix Focus Areas
- pyproject.toml[274-276]
- tests/optimize reproduce_issue_176.py[1-93]

## Suggested fix
- Rename/move the file to a pytest-discoverable name, e.g. `tests/semantic_extract/test_max_tokens_propagation.py` (or `tests/test_max_tokens_propagation.py`).
- Avoid spaces in filenames.
- (Optional) If you truly want nonstandard names, explicitly set `[tool.pytest.ini_options].python_files` in `pyproject.toml`, but renaming is preferred.

ⓘ Copy this prompt and use it to remediate the issue with your preferred AI generation tools

[qodo-free-for-open-source-projects]

2. Relations max_tokens dropped 🐞 Bug ≡ Correctness

The new test asserts extract_relations_llm() forwards max_tokens into llm.generate_typed(),
but extract_relations_llm() currently builds call_kwargs without max_tokens before calling
generate_typed. Once the test is made discoverable, it will fail unless production code is updated
(or the assertion is changed).

Agent Prompt

## Issue description
`extract_relations_llm()` does not propagate `max_tokens` to `llm.generate_typed()`, unlike `extract_entities_llm()` / `extract_triplets_llm()`. The newly added test expects `max_tokens` to be present in the `generate_typed` call kwargs.

## Issue Context
In `extract_relations_llm()`, the code constructs a limited `call_kwargs` dict (temperature/verbose/max_retries) and calls `llm.generate_typed(..., **call_kwargs)`, which drops `max_tokens`.

## Fix Focus Areas
- semantica/semantic_extract/methods.py[1904-1916]
- tests/optimize reproduce_issue_176.py[33-49]

## Suggested fix
- Include `max_tokens` (and any other intended generation limit knobs) in `call_kwargs` when present in `kwargs`, e.g.:
  - `if "max_tokens" in kwargs: call_kwargs["max_tokens"] = kwargs["max_tokens"]`
- Ensure the updated behavior is consistent with `extract_entities_llm()` / `extract_triplets_llm()` so the test can pass once properly collected.

ⓘ Copy this prompt and use it to remediate the issue with your preferred AI generation tools

[qodo-free-for-open-source-projects]

3. Duplicate issue-176 test script 🐞 Bug ⚙ Maintainability

The PR adds another issue-176 max_tokens propagation test script, but the repository already
contains tests/reproduce_issue_176.py covering the same three methods. Keeping multiple
near-duplicates makes it unclear which one is canonical and increases the chance of inconsistent
future edits.

Agent Prompt

## Issue description
There are now two scripts testing the same max_tokens propagation scenario (`tests/reproduce_issue_176.py` and the newly added `tests/optimize reproduce_issue_176.py`). This duplication is confusing and makes it easy for one to drift.

## Issue Context
The new file appears to be a cleaner refactor of the existing reproduce script.

## Fix Focus Areas
- tests/optimize reproduce_issue_176.py[1-93]
- tests/reproduce_issue_176.py[1-100]

## Suggested fix
- Choose a single canonical test module (preferably pytest-discoverable, e.g. `tests/semantic_extract/test_max_tokens_propagation.py`).
- Move/rename the better version there.
- Delete the redundant reproduce script(s) or keep *one* reproduction script outside `tests/` if it’s meant for manual runs only.

ⓘ Copy this prompt and use it to remediate the issue with your preferred AI generation tools

[Sameer6305]

@saiganesh47 can you please handle qodo findings, so we can proceed for the review.