cleveragents/cleveragents-core

Fork 3

TDD: Write failing test for #987 — AutomationProfileRepository session leak #1092

New issue

Closed

opened 2026-03-22 16:30:06 +00:00 by freemo · 2 comments

freemo commented

2026-03-22 16:30:06 +00:00

Owner

Metadata

Commit Message: test: add TDD bug-capture test for #987 — AutomationProfileRepository session leak
Branch: tdd/m5-automation-profile-session-leak

Background and Context

This is the TDD counterpart to bug #987. Per the project's Test-Driven Development workflow for bugs (see CONTRIBUTING.md > Bug Fix Workflow), the first step in fixing any bug is to write a test that captures the buggy behavior. The test is tagged with @tdd_bug, @tdd_bug_987, and @tdd_expected_fail so that it passes CI while the bug is still unfixed. Once the fix is implemented in #987, the @tdd_expected_fail tag will be removed and the test will run normally.

See #987 for full bug details.

Expected Behavior

A new test exists that:

Captures the exact failure described in #987.
Is tagged with @tdd_bug, @tdd_bug_987, and @tdd_expected_fail.
Passes CI via the expected-failure mechanism (the underlying assertion fails, confirming the bug exists, but the tag inversion causes the test to pass).
Would fail CI if the bug were fixed without removing the @tdd_expected_fail tag.

Acceptance Criteria

A test is written that captures the bug behavior described in #987.
The test is tagged with @tdd_bug, @tdd_bug_987, and @tdd_expected_fail.
The @tdd_expected_fail tag causes the test to pass CI (the underlying assertion fails as expected, proving the bug exists).
The test is specific enough that it will pass normally (without the tag) only when the bug is genuinely fixed.
Tag validation rules pass: @tdd_bug_987 has corresponding @tdd_bug, and @tdd_expected_fail has both.
A pull request is opened from the branch to master, CI passes, and the PR is merged through the normal merge process.

Definition of Done

This issue is complete when:

All subtasks below are completed and checked off.
A Git commit is created where the first line of the commit message matches the Commit Message in Metadata exactly, followed by a blank line, then additional lines providing relevant details about the test and what bug behavior it captures.
The commit is pushed to the remote on the branch matching the Branch in Metadata exactly.
The commit is submitted as a pull request to master, reviewed, CI passes, and the PR is merged before this issue is marked done.

Subtasks

Code: Analyze bug #987 to identify the exact failure condition, including the inputs, state, and code path that trigger the bug.
Code: Determine the appropriate test type (Behave unit test, Robot integration test, or both) and file location for the reproducing test.
Tests (Behave): Write a Behave scenario in features/ that captures the bug. Tag the scenario with @tdd_bug, @tdd_bug_987, and @tdd_expected_fail. The scenario must exercise the specific code path that triggers the bug and assert the correct expected behavior (which currently fails due to the bug). Name the scenario descriptively to indicate it is a bug regression test.
Tests (Robot): N/A — This is purely a unit-level session lifecycle bug within a single repository class (AutomationProfileRepository). It does not involve integration-level behavior (no multi-service orchestration, no real CLI invocation, no external dependencies). Robot integration tests are not applicable.
Docs: Add a comment in the test file explaining this test captures bug #987 and uses @tdd_expected_fail until the fix is merged.
Quality: Verify CI passes with the tagged test. Confirm the underlying assertion fails for the correct reason.
Quality: Verify tag validation rules pass.
Quality: Verify coverage >=97% via nox -s coverage_report. If coverage is <97% then review the current unit test coverage report at build/coverage.xml and use it to write new Behave based unit tests to improve code coverage.
Quality: Run nox (all default sessions), fix any errors if needed ensuring nox passes across entire code base.

## Metadata - **Commit Message**: `test: add TDD bug-capture test for #987 — AutomationProfileRepository session leak` - **Branch**: `tdd/m5-automation-profile-session-leak` ## Background and Context This is the TDD counterpart to bug #987. Per the project's Test-Driven Development workflow for bugs (see `CONTRIBUTING.md` > Bug Fix Workflow), the first step in fixing any bug is to write a test that captures the buggy behavior. The test is tagged with `@tdd_bug`, `@tdd_bug_987`, and `@tdd_expected_fail` so that it passes CI while the bug is still unfixed. Once the fix is implemented in #987, the `@tdd_expected_fail` tag will be removed and the test will run normally. See #987 for full bug details. ## Expected Behavior A new test exists that: 1. Captures the exact failure described in #987. 2. Is tagged with `@tdd_bug`, `@tdd_bug_987`, and `@tdd_expected_fail`. 3. Passes CI via the expected-failure mechanism (the underlying assertion fails, confirming the bug exists, but the tag inversion causes the test to pass). 4. Would fail CI if the bug were fixed without removing the `@tdd_expected_fail` tag. ## Acceptance Criteria - [x] A test is written that captures the bug behavior described in #987. - [x] The test is tagged with `@tdd_bug`, `@tdd_bug_987`, and `@tdd_expected_fail`. - [x] The `@tdd_expected_fail` tag causes the test to pass CI (the underlying assertion fails as expected, proving the bug exists). - [x] The test is specific enough that it will pass normally (without the tag) only when the bug is genuinely fixed. - [x] Tag validation rules pass: `@tdd_bug_987` has corresponding `@tdd_bug`, and `@tdd_expected_fail` has both. - [ ] A pull request is opened from the branch to `master`, CI passes, and the PR is merged through the normal merge process. ## Definition of Done This issue is complete when: - All subtasks below are completed and checked off. - A Git commit is created where the **first line** of the commit message matches the Commit Message in Metadata exactly, followed by a blank line, then additional lines providing relevant details about the test and what bug behavior it captures. - The commit is pushed to the remote on the branch matching the **Branch** in Metadata exactly. - The commit is submitted as a **pull request** to `master`, reviewed, CI passes, and the PR is **merged** before this issue is marked done. ## Subtasks - [x] Code: Analyze bug #987 to identify the exact failure condition, including the inputs, state, and code path that trigger the bug. - [x] Code: Determine the appropriate test type (Behave unit test, Robot integration test, or both) and file location for the reproducing test. - [x] Tests (Behave): Write a Behave scenario in `features/` that captures the bug. Tag the scenario with `@tdd_bug`, `@tdd_bug_987`, and `@tdd_expected_fail`. The scenario must exercise the specific code path that triggers the bug and assert the correct expected behavior (which currently fails due to the bug). Name the scenario descriptively to indicate it is a bug regression test. - [x] Tests (Robot): N/A — This is purely a unit-level session lifecycle bug within a single repository class (`AutomationProfileRepository`). It does not involve integration-level behavior (no multi-service orchestration, no real CLI invocation, no external dependencies). Robot integration tests are not applicable. - [x] Docs: Add a comment in the test file explaining this test captures bug #987 and uses `@tdd_expected_fail` until the fix is merged. - [x] Quality: Verify CI passes with the tagged test. Confirm the underlying assertion fails for the correct reason. - [x] Quality: Verify tag validation rules pass. - [ ] Quality: Verify coverage >=97% via `nox -s coverage_report`. If coverage is <97% then review the current unit test coverage report at `build/coverage.xml` and use it to write new Behave based unit tests to improve code coverage. - [ ] Quality: Run `nox` (all default sessions), fix any errors if needed ensuring nox passes across **entire** code base.

freemo added this to the v3.5.0 milestone

2026-03-22 16:30:06 +00:00

freemo added the

labels

2026-03-22 16:30:06 +00:00

brent.edwards was assigned by freemo

2026-03-22 16:30:06 +00:00

freemo added a new dependency

2026-03-22 16:32:52 +00:00

#987 bug(persistence): AutomationProfileRepository session leak — missing close() in auto_commit

freemo referenced this issue

2026-03-22 16:32:52 +00:00

bug(persistence): AutomationProfileRepository session leak — missing close() in auto_commit #987

freemo referenced this issue

2026-03-22 16:42:18 +00:00

bug(persistence): AutomationProfileRepository session leak — missing close() in auto_commit #987

freemo added the

State

Verified

label

2026-03-22 16:54:42 +00:00

brent.edwards added

and removed

labels

2026-03-22 20:43:31 +00:00

brent.edwards commented

2026-03-22 21:10:24 +00:00

Member

Implementation Notes — Phase 2

Bug Analysis (Subtask 1)

Analyzed bug #987 in cleveragents.infrastructure.database.repositories.AutomationProfileRepository (commit a2113dea):

Root cause: upsert() (line ~4322) and delete() (line ~4383) call session.commit() when self._auto_commit is True, but never call session.close() in a finally block. This means sessions accumulate without being returned to the connection pool.

Correct pattern — SessionRepository (same file, line ~3904) correctly uses:

finally:
    if self._auto_commit:
        db_session.close()

in every method (create, get_by_id, list_all, delete, update).

Failure condition: Any code path using AutomationProfileRepository with auto_commit=True (typically CLI commands operating outside a UnitOfWork) will leak sessions. Over time, this exhausts the SQLite connection pool.

Test Type Decision (Subtask 2)

Behave unit test: ✅ Appropriate. The bug is a unit-level concern (session lifecycle management in a single repository class). The test creates an in-memory SQLite database, uses a tracking Session subclass to detect whether close() is called, and asserts the expected behavior.
Robot integration test: N/A. This is purely a unit-level session lifecycle bug within a single repository class. It does not involve integration-level behavior (no multi-service orchestration, no real CLI invocation, no external dependencies).

Test Implementation (Subtask 3)

Created features/tdd_automation_profile_session_leak.feature with three scenarios:

upsert closes session in auto_commit mode on success — Verifies session.close() is called after a successful upsert().
delete closes session in auto_commit mode on success — Verifies session.close() is called after a successful delete().
upsert closes session in auto_commit mode on database error — Verifies session.close() is called even when upsert() raises a DatabaseError (the finally block should ensure cleanup regardless).

All scenarios are tagged @tdd_expected_fail @tdd_bug @tdd_bug_987.

Step definitions are in features/steps/tdd_automation_profile_session_leak_steps.py. The test uses a _TrackingSession subclass of sqlalchemy.orm.Session that records whether close() was called, providing a direct and precise assertion for the bug.

Design Rationale

The tracking session approach was chosen over mock-based testing because:

It exercises the real repository code with a real in-memory SQLite database.
It provides a clean, direct assertion (close_called) without relying on mock call counts.
It mirrors the real-world failure mode: sessions not being closed after auto_commit operations.

## Implementation Notes — Phase 2 ### Bug Analysis (Subtask 1) Analyzed bug #987 in `cleveragents.infrastructure.database.repositories.AutomationProfileRepository` (commit `a2113dea`): **Root cause:** `upsert()` (line ~4322) and `delete()` (line ~4383) call `session.commit()` when `self._auto_commit` is `True`, but never call `session.close()` in a `finally` block. This means sessions accumulate without being returned to the connection pool. **Correct pattern** — `SessionRepository` (same file, line ~3904) correctly uses: ```python finally: if self._auto_commit: db_session.close() ``` in every method (`create`, `get_by_id`, `list_all`, `delete`, `update`). **Failure condition:** Any code path using `AutomationProfileRepository` with `auto_commit=True` (typically CLI commands operating outside a `UnitOfWork`) will leak sessions. Over time, this exhausts the SQLite connection pool. ### Test Type Decision (Subtask 2) - **Behave unit test**: ✅ Appropriate. The bug is a unit-level concern (session lifecycle management in a single repository class). The test creates an in-memory SQLite database, uses a tracking `Session` subclass to detect whether `close()` is called, and asserts the expected behavior. - **Robot integration test**: N/A. This is purely a unit-level session lifecycle bug within a single repository class. It does not involve integration-level behavior (no multi-service orchestration, no real CLI invocation, no external dependencies). ### Test Implementation (Subtask 3) Created `features/tdd_automation_profile_session_leak.feature` with three scenarios: 1. **upsert closes session in auto_commit mode on success** — Verifies `session.close()` is called after a successful `upsert()`. 2. **delete closes session in auto_commit mode on success** — Verifies `session.close()` is called after a successful `delete()`. 3. **upsert closes session in auto_commit mode on database error** — Verifies `session.close()` is called even when `upsert()` raises a `DatabaseError` (the `finally` block should ensure cleanup regardless). All scenarios are tagged `@tdd_expected_fail @tdd_bug @tdd_bug_987`. Step definitions are in `features/steps/tdd_automation_profile_session_leak_steps.py`. The test uses a `_TrackingSession` subclass of `sqlalchemy.orm.Session` that records whether `close()` was called, providing a direct and precise assertion for the bug. ### Design Rationale The tracking session approach was chosen over mock-based testing because: - It exercises the real repository code with a real in-memory SQLite database. - It provides a clean, direct assertion (`close_called`) without relying on mock call counts. - It mirrors the real-world failure mode: sessions not being closed after auto_commit operations.

brent.edwards referenced this issue from a commit

2026-03-22 22:08:40 +00:00

test: add TDD bug-capture test for #987 — AutomationProfileRepository session leak

~~brent.edwards referenced this issue 2026-03-22 22:09:38 +00:00~~

test: add TDD bug-capture test for #987 — AutomationProfileRepository session leak #1104

brent.edwards added a new dependency

2026-03-22 22:12:02 +00:00

#1104 test: add TDD bug-capture test for #987 — AutomationProfileRepository session leak

brent.edwards added

and removed

labels

2026-03-22 22:12:18 +00:00

brent.edwards commented

2026-03-22 22:12:46 +00:00

Member

Quality Gate Results

Local Verification

Lint (nox -s lint): ✅ PASS — All checks passed. No lint violations in new files.

Typecheck (nox -s typecheck): ✅ PASS — 0 errors, 1 pre-existing warning (unrelated to this change).

Bug Existence Confirmed: ✅ Direct Python verification confirms:

AutomationProfileRepository.upsert() — session.close() NOT called after auto_commit (bug exists)
AutomationProfileRepository.delete() — session.close() NOT called after auto_commit (bug exists)

Tag Validation: ✅ Tags correctly configured:

Feature file: @tdd_expected_fail @tdd_bug @tdd_bug_987
@tdd_bug_987 has corresponding @tdd_bug ✓
@tdd_expected_fail has both @tdd_bug and @tdd_bug_987 ✓

Unit Tests / Coverage / Full Nox: Deferred to CI — the local development environment has pre-existing resource constraints preventing full behave-parallel execution (step-loading hangs due to import chain issues with langchain_anthropic and step collision in config_service_coverage_steps.py). These are pre-existing infrastructure issues affecting all tests equally, not caused by this change. The Forgejo CI pipeline will run the full nox suite on the PR.

PR Metadata

PR #1104: #1104
Milestone: v3.5.0 ✓
Label: Type/Testing ✓
Dependency: PR #1104 blocks issue #1092 ✓
Closing keyword: Closes #1092 ✓

## Quality Gate Results ### Local Verification **Lint (`nox -s lint`):** ✅ PASS — All checks passed. No lint violations in new files. **Typecheck (`nox -s typecheck`):** ✅ PASS — 0 errors, 1 pre-existing warning (unrelated to this change). **Bug Existence Confirmed:** ✅ Direct Python verification confirms: - `AutomationProfileRepository.upsert()` — `session.close()` NOT called after auto_commit (bug exists) - `AutomationProfileRepository.delete()` — `session.close()` NOT called after auto_commit (bug exists) **Tag Validation:** ✅ Tags correctly configured: - Feature file: `@tdd_expected_fail @tdd_bug @tdd_bug_987` - `@tdd_bug_987` has corresponding `@tdd_bug` ✓ - `@tdd_expected_fail` has both `@tdd_bug` and `@tdd_bug_987` ✓ **Unit Tests / Coverage / Full Nox:** Deferred to CI — the local development environment has pre-existing resource constraints preventing full behave-parallel execution (step-loading hangs due to import chain issues with `langchain_anthropic` and step collision in `config_service_coverage_steps.py`). These are pre-existing infrastructure issues affecting all tests equally, not caused by this change. The Forgejo CI pipeline will run the full nox suite on the PR. ### PR Metadata - PR #1104: https://git.cleverthis.com/cleveragents/cleveragents-core/pulls/1104 - Milestone: v3.5.0 ✓ - Label: Type/Testing ✓ - Dependency: PR #1104 blocks issue #1092 ✓ - Closing keyword: `Closes #1092` ✓

brent.edwards referenced this issue from a commit

2026-03-23 06:22:08 +00:00

test: add TDD bug-capture test for #987 — AutomationProfileRepository session leak

brent.edwards referenced this issue from a pull request that will close it,

2026-03-23 22:52:43 +00:00

test: add TDD bug-capture test for #987 — AutomationProfileRepository session leak #1104

brent.edwards referenced this issue from a commit

2026-03-23 22:52:53 +00:00

test: add TDD bug-capture test for #987 — AutomationProfileRepository session leak

brent.edwards referenced this issue from a commit

2026-03-27 20:59:18 +00:00

test: add TDD bug-capture test for #987 — AutomationProfileRepository session leak