mirror of
https://github.com/headroomlabs-ai/headroom.git
synced 2026-11-22 13:29:19 +08:00
fix(learn): harden analyzer JSON parsing against malformed LLM output (#2471)
## Description
Two spots in the `learn` analyzer crash on JSON a model can
realistically emit. Both parse fully model-controlled output.
**1. `_strip_fenced_json` returns a non-object instead of raising.** It
is documented to return a dict and to raise `json.JSONDecodeError` when
no candidate is an object, but its final fallback was `result: dict =
json.loads(text); return result`. `json.loads` raises for invalid JSON,
but for valid JSON that is not an object (a bare array, string, or
number) it returns a non-dict. The `-> dict`-annotated value is then a
list/str, and all three callers do `.get(...)` on it, raising
`AttributeError` instead of the clean `JSONDecodeError` they already
handle.
**2. `_parse_llm_response` crashes on null rule lists/fields.** It read
`raw.get("context_file_rules", [])` / `raw.get("memory_file_rules", [])`
and `rule.get("section", "").strip()` / `rule.get("content",
"").strip()`. `.get(key, default)` only falls back when the key is
absent, so a present-but-null `"context_file_rules": null` made `for
rule in None` raise `TypeError`, and a null `section`/`content` made
`None.strip()` raise `AttributeError`.
## Fix
- `_strip_fenced_json`: reject a non-dict fallback result explicitly
(`raise json.JSONDecodeError(...)`), restoring the documented contract.
The lenient first-brace..last-brace slice still recovers a single object
a model wrapped in an array; only genuinely object-less JSON raises.
- `_parse_llm_response`: coerce the rule lists and the
`section`/`content` fields with `or` so a null is treated like an absent
value. Valid rules parse unchanged.
## Type of Change
- [x] Bug fix (non-breaking change that fixes an issue)
- [ ] New feature (non-breaking change that adds functionality)
- [ ] Breaking change (fix or feature that would cause existing
functionality to change)
- [ ] Documentation update
- [ ] Performance improvement
- [ ] Code refactoring (no functional changes)
## Changes Made
- `headroom/learn/analyzer.py`: raise `JSONDecodeError` when
`_strip_fenced_json`'s fallback parse yields a non-dict; `or`-coerce
null rule lists and null `section`/`content` in `_parse_llm_response`.
- `tests/test_learn/test_analyzer.py`: regressions for non-object JSON
raising (and array-wrapped-object still extracted), and for null rule
lists / null fields being tolerated with valid rules still parsed.
## Testing
- [x] Unit tests pass (`pytest`)
- [x] Linting passes (`ruff check .`)
- [x] Type checking passes (`mypy headroom`)
- [x] New tests added for new functionality
- [ ] Manual testing performed
### Test Output
```text
$ python -m pytest tests/test_learn/test_analyzer.py::TestStripFencedJson tests/test_learn/test_analyzer.py::TestLLMResponseParser -q
# all pass; each new test fails with the corresponding fix reverted
# (JSONDecodeError contract / 'NoneType' object is not iterable / has no attribute 'strip')
$ uvx ruff@0.15.17 check headroom/learn/analyzer.py tests/test_learn/test_analyzer.py
All checks passed!
$ uvx mypy@1.20.2 --ignore-missing-imports headroom/learn/analyzer.py
Success: no issues found in 1 source file
```
Note: one unrelated digest-builder test
(`TestDigestBuilder::test_includes_project_info`) fails in my local
Windows venv because it hardcodes a `/tmp/...` path; it fails
identically on a clean `main` and is not touched by this change.
## Real Behavior Proof
- Environment: Windows 11, Python 3.12, project venv (`uv sync --extra
proxy`), `uvx ruff@0.15.17` / `uvx mypy@1.20.2`, pytest in the venv.
- Exact command / steps: called the real `_strip_fenced_json` with
object-less JSON (`[1,2,3]`, `"str"`, `42`, ...) and valid/extractable
objects, and the real `_parse_llm_response` with `{"context_file_rules":
null}`, a rule with `section: null`, and a valid rule; then reverted
each fix and re-ran.
- Observed result: with the fixes the object-less inputs raise
`JSONDecodeError`, the null shapes yield an empty recommendation list,
and valid inputs parse as before; with each fix reverted the respective
input raises (`AttributeError: 'list' object has no attribute 'get'`,
`TypeError: 'NoneType' object is not iterable`, `AttributeError:
'NoneType' object has no attribute 'strip'`). Ran against the actual
module.
- Not tested: a live `headroom learn` run where the backing model
returns these shapes.
## Review Readiness
- [x] I have performed a self-review
- [x] This PR is ready for human review
## Checklist
- [x] My code follows the project's style guidelines
- [x] I have performed a self-review of my code
- [x] I have commented my code, particularly in hard-to-understand areas
- [ ] I have made corresponding changes to the documentation
- [x] My changes generate no new warnings
- [x] I have added tests that prove my fix is effective or that my
feature works
- [x] New and existing unit tests pass locally with my changes
- [ ] I have updated the CHANGELOG.md if applicable
<!-- devin-review-badge-begin -->
---
<a href="https://app.devin.ai/review/headroomlabs-ai/headroom/pull/2471"
target="_blank"><picture><source media="(prefers-color-scheme: dark)"
srcset="https://static.devin.ai/assets/gh-devin-review-dark.svg?v=4"><img
src="https://static.devin.ai/assets/gh-devin-review-light.svg?v=4"
alt="Devin Review"></picture></a>
<!-- devin-review-badge-end -->
### Maintainer current-main requalification
Exact head `27fa31eac507413415a00b54e29679bef08eabe9`, upstream main
`0ddaa56c21cdc01c4774391fdcc69e987ecc4e5c`.
Clean current upstream/main0ddaa56c integration; production
_strip_fenced_json/_parse_llm_response/_as_str ASTs are identical to
contributor published head. Main-relative scope remains two files. Small
maintainer test-only repair uses str(project.project_path) instead of a
POSIX-only literal in existing digest assertion; source and test fixture
data unchanged. Initial full analyzer run109pass/1fail and separate
clean-main failure independently establish inherited Windows path
expectation before repair. After repair399tests pass,11skip,1warning
(all110analyzer cases plus289incoming-main regressions). Two changed
files pass Rufflint/format,candidate offline lock and whitespace pass.
Complete native inputset/allrecordedhashes/artifactchecksum verified
before existing3982DLL reuse; no fresh build.
Independent paired parser probes: branch passes both truthy non-string
section and objectless JSON contract cases; clean main fails with
AttributeError and missing JSONDecodeError respectively. These are
controlled parser executions, not live LLM quality/provider behavior.
Outstanding Devin truthy non-list rule-container report also
independently reproduced on BOTH branch and clean main:
context_file_rules:42 and memory_file_rules:42 raise TypeError in both.
This is an inherited limitation, not a PR-introduced regression or
grounds to withhold this narrower fix. Do not claim arbitrary-container
safety. The later author suggestion to join list-valued rule content
changes intended normalization semantics; not selected or implemented by
maintainer. Existing _as_str drops non-string fields, preserving
contributor intent.
Read author replies/current discussions and unresolved thread. No new
unreviewed external findings at prepublication audit. Source/test diff
contains no workflow/setup/dependency/native/CLI paths; central
publication gate passed. Exact published-head Actions/rollup, current
main ancestry/mergeability and fresh discussion remain before final
approval. Maintainers handle CI; no PR merged.
Exact-head hosted checks are being handled by maintainers. No new
finding in the reviewed scope; formal approval awaits exact-head hosted
qualification; nothing merged.
---------
Co-authored-by: JerrettDavis <mxjerrett@gmail.com>
This commit is contained in:
co-authored by
JerrettDavis
parent
9eed58f84c
commit
12864fc8f0
@@ -661,7 +661,12 @@ def _strip_fenced_json(raw: str) -> dict:
|
||||
|
||||
# Nothing parsed as an object: re-raise the natural error on the raw text
|
||||
# so callers see a JSONDecodeError, preserving the documented contract.
|
||||
result: dict = json.loads(text)
|
||||
# json.loads raises for invalid JSON, but valid JSON that is not an object
|
||||
# (e.g. a bare array the model emitted) would otherwise be returned as a
|
||||
# non-dict and crash callers that do `.get(...)`, so reject it explicitly.
|
||||
result = json.loads(text)
|
||||
if not isinstance(result, dict):
|
||||
raise json.JSONDecodeError("Expected a JSON object", text, 0)
|
||||
return result
|
||||
|
||||
|
||||
@@ -1141,11 +1146,16 @@ def _parse_llm_response(raw: dict) -> list[Recommendation]:
|
||||
"""Convert LLM structured output into Recommendation objects."""
|
||||
recommendations: list[Recommendation] = []
|
||||
|
||||
for rule in raw.get("context_file_rules", []):
|
||||
# `.get(key, [])` / `.get(key, "")` only fall back when the key is absent;
|
||||
# the model controls this JSON, so a present-but-null `*_file_rules` list or
|
||||
# a non-string `section`/`content` would crash the iteration / `.strip()`.
|
||||
# `_as_str` coerces any non-string (null, number, list, object) to "" so a
|
||||
# malformed field is treated like an absent one instead of raising.
|
||||
for rule in raw.get("context_file_rules") or []:
|
||||
if not isinstance(rule, dict):
|
||||
continue
|
||||
section = rule.get("section", "").strip()
|
||||
content = rule.get("content", "").strip()
|
||||
section = _as_str(rule.get("section")).strip()
|
||||
content = _as_str(rule.get("content")).strip()
|
||||
if not section or not content:
|
||||
continue
|
||||
recommendations.append(
|
||||
@@ -1159,11 +1169,11 @@ def _parse_llm_response(raw: dict) -> list[Recommendation]:
|
||||
)
|
||||
)
|
||||
|
||||
for rule in raw.get("memory_file_rules", []):
|
||||
for rule in raw.get("memory_file_rules") or []:
|
||||
if not isinstance(rule, dict):
|
||||
continue
|
||||
section = rule.get("section", "").strip()
|
||||
content = rule.get("content", "").strip()
|
||||
section = _as_str(rule.get("section")).strip()
|
||||
content = _as_str(rule.get("content")).strip()
|
||||
if not section or not content:
|
||||
continue
|
||||
recommendations.append(
|
||||
@@ -1183,6 +1193,17 @@ def _parse_llm_response(raw: dict) -> list[Recommendation]:
|
||||
return recommendations
|
||||
|
||||
|
||||
def _as_str(val: object) -> str:
|
||||
"""Return ``val`` if it is a string, else ``""``.
|
||||
|
||||
Fields like ``section``/``content`` come straight from model-controlled JSON,
|
||||
where a number/list/object is as possible as a string. Coercing a non-string
|
||||
to empty keeps ``.strip()`` from raising and lets the caller's emptiness check
|
||||
drop the malformed rule.
|
||||
"""
|
||||
return val if isinstance(val, str) else ""
|
||||
|
||||
|
||||
def _safe_int(val: object) -> int:
|
||||
"""Safely convert a value to int."""
|
||||
if isinstance(val, int):
|
||||
|
||||
@@ -70,7 +70,7 @@ class TestDigestBuilder:
|
||||
sessions = [SessionData(session_id="s1", tool_calls=[_tc()])]
|
||||
digest = _build_digest(project, sessions)
|
||||
assert "test-project" in digest
|
||||
assert "/tmp/test-project" in digest
|
||||
assert str(project.project_path) in digest
|
||||
|
||||
def test_includes_session_stats(self):
|
||||
sessions = [
|
||||
@@ -409,6 +409,40 @@ class TestLLMResponseParser:
|
||||
recs = _parse_llm_response(raw)
|
||||
assert recs == []
|
||||
|
||||
def test_handles_null_rule_lists_and_fields(self):
|
||||
# The model controls this JSON. A present-but-null rules list or a null
|
||||
# section/content must be treated like an absent value, not crash the
|
||||
# iteration / `.strip()`.
|
||||
assert _parse_llm_response({"context_file_rules": None, "memory_file_rules": None}) == []
|
||||
assert (
|
||||
_parse_llm_response({"context_file_rules": [{"section": None, "content": "x"}]}) == []
|
||||
)
|
||||
assert _parse_llm_response({"memory_file_rules": [{"section": "s", "content": None}]}) == []
|
||||
|
||||
# A truthy non-string section/content (number, list, object) is also
|
||||
# model-controlled and must not crash `.strip()`; it is dropped like null.
|
||||
assert _parse_llm_response({"context_file_rules": [{"section": 123, "content": "x"}]}) == []
|
||||
assert _parse_llm_response({"context_file_rules": [{"section": "s", "content": 123}]}) == []
|
||||
assert (
|
||||
_parse_llm_response({"memory_file_rules": [{"section": ["s"], "content": "x"}]}) == []
|
||||
)
|
||||
assert (
|
||||
_parse_llm_response({"memory_file_rules": [{"section": "s", "content": {"a": 1}}]})
|
||||
== []
|
||||
)
|
||||
|
||||
# A valid rule alongside the null shapes is still parsed.
|
||||
recs = _parse_llm_response(
|
||||
{
|
||||
"context_file_rules": [
|
||||
None,
|
||||
{"section": "Large Files", "content": "use offset", "evidence_count": 3},
|
||||
]
|
||||
}
|
||||
)
|
||||
assert [r.section for r in recs] == ["Large Files"]
|
||||
assert recs[0].evidence_count == 3
|
||||
|
||||
|
||||
# =============================================================================
|
||||
# Full Analyzer Integration Tests (mocked LLM)
|
||||
@@ -753,6 +787,19 @@ class TestStripFencedJson:
|
||||
with pytest.raises(json.JSONDecodeError):
|
||||
_strip_fenced_json("not json at all")
|
||||
|
||||
def test_non_object_json_raises(self):
|
||||
# A model may return valid JSON that is not an object (a bare array,
|
||||
# string, number, ...). The contract is to raise JSONDecodeError so
|
||||
# callers do not receive a non-dict and crash on `.get`.
|
||||
for raw in ("[1, 2, 3]", '["a", "b"]', '[{"a": 1}, {"b": 2}]', '"a string"', "42", "true"):
|
||||
with pytest.raises(json.JSONDecodeError):
|
||||
_strip_fenced_json(raw)
|
||||
|
||||
def test_array_wrapping_a_single_object_is_extracted(self):
|
||||
# The lenient first-{ .. last-} slice still recovers a single object a
|
||||
# model wrapped in an array; only genuinely object-less JSON raises.
|
||||
assert _strip_fenced_json('[{"key": "value"}]') == {"key": "value"}
|
||||
|
||||
def test_prose_preamble_before_fence(self):
|
||||
# Models sometimes add a preamble before the fence despite being told
|
||||
# to return JSON only (e.g. "Here it is:\n\n```json ...").
|
||||
|
||||
Reference in New Issue
Block a user