lei.chen/myxce
1
0
Fork 1
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

feat(buglog): one Markdown record per bug, checked against a record standard

Browse source 
The buglog was one 27K-line JSON file whose fields covered only symptom,
root cause and fix, as free text: 2% of entries named the code version a
bug was found on, 12% the build profile, 36% a way to reproduce, and 58% of
the file was OpenWolf hook auto-detections ("Missing guard clause",
"Function not marked async" on .md files).

- .claude/skills/buglog/bug-record.md: the record standard. Eight
  sections (Summary as the title, Description, Environment, Reproduction;
  Root cause, Fix, Verification, References once fixed), each with the
  content it must contain -- e.g. Environment needs the OS, the libc the
  failing binary loads (from its ELF interpreter: glibc 2.42 in the nix
  shell vs 2.40 on the host), toolchain, build profile and commit.
- .wolf/bugs/<id>.md: 500 records migrated from buglog.json; every field
  verified to round-trip. They carry legacy: true and an explicit "not
  recorded" where the old log had no such field; component derived from
  file paths, severity from keywords (else unknown), fix_commits from
  hashes in status_note/fix that git resolves (91 of 463 fixed).
- .wolf/buglog-archive.json: the 710 hook auto-detections, verbatim.
- .wolf/buglog.json is now an index generated from the records, keeping
  the fields OpenWolf's hooks and `openwolf bug search` read.
- scripts/buglog.py rewritten: new (template + Environment collected from
  the machine and --binary), check, status (refuses a status the record
  doesn't support; fixed needs Root cause, Fix, Verification and
  --commit hashes git resolves), note, recur, set, show, list, triage,
  env, index. 29 unittest cases in scripts/tests/test_buglog.py.
- scripts/git-hooks/pre-commit runs `check --staged`: incomplete staged
  records, a legacy record hand-changed to fixed without the fix-time
  sections, a new record claiming legacy: true, or a stale index refuse
  the commit.
- scripts/gitea-sync.py reads the records: issue body = the record's
  sections plus linked fix commits (origin remote, GITEA_COMMIT_REPO to
  override), and a body hash so an issue's text is refreshed when the
  record changes.
- bug-1866 rewritten to the full standard as the worked example.

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
This commit is contained in:
Rongsong Shen 2026-09-28 03:40:11 +00:00
parent 27b7a84980
commit d26ff8870d
Signed by untrusted user: rshen
GPG key ID: 80533536B0EB651C
512 changed files with 40724 additions and 17843 deletions
Download patch file Download diff file Expand all files Collapse all files

340
scripts/tests/test_buglog.py Normal file
View file

@ -0,0 +1,340 @@
"""Tests for scripts/buglog.py (one Markdown record per bug).
Run: python3 -m unittest discover -s scripts/tests -v
"""
import json
import os
import subprocess
import sys
import tempfile
import unittest
from pathlib import Path
sys.path.insert(0, str(Path(__file__).resolve().parent.parent))
import buglog # noqa: E402
FULL_ENV = """\
- OS: Ubuntu 24.04 LTS; kernel Linux 6.18.33 x86_64
- libc: glibc 2.42 (from the binary's interpreter)
- Toolchain: gcc 15.2.0, meson 1.9.1
- Build: asan (xce.build)
- Code: scheduler-framework @ {sha}
- Deployment: single host, xcecd_virtual"""
FULL_BODY = """\
# {id}: xcecd_virtual leaks worker strings on every finished job
## Description
**Observed:**
```
Direct leak of 12 byte(s) in 12 object(s)
```
**Expected:** no leak at exit
**Impact:** grows with job count
**Found by:** LeakSanitizer after a job burst
## Environment
{env}
## Reproduction
1. `/xce-build worker`
2. run `finished_job_retention_test`
- Frequency: deterministic
- Evidence: the unit test
## Root cause
{root}
## Fix
{fix}
## Verification
{verif}
## References
- Related: none
"""
def git(root, *args):
return subprocess.run(["git", "-C", str(root), *args], check=True,
capture_output=True, text=True).stdout.strip()
class BuglogTestCase(unittest.TestCase):
def setUp(self):
self.tmp = tempfile.TemporaryDirectory()
self.root = Path(self.tmp.name)
(self.root / ".wolf" / "bugs").mkdir(parents=True)
git(self.root, "init", "-q")
git(self.root, "-c", "user.email=t@t", "-c", "user.name=t",
"commit", "-q", "--allow-empty", "-m", "init")
self.sha = git(self.root, "rev-parse", "--short=8", "HEAD")
buglog.configure(self.root)
def tearDown(self):
self.tmp.cleanup()
def write_record(self, bug_id="bug-100", status="open", legacy=False,
root="TODO", fix="TODO", verif="TODO", commits=None,
env=None, extra_meta=None):
meta = {"id": bug_id, "status": status, "status_updated": "2026-09-27",
"status_note": "" if status == "fixed" else "why",
"component": "worker", "severity": "leak",
"timestamp": "2026-09-25T15:42:02Z"}
if legacy:
meta["legacy"] = True
if commits is not None:
meta["fix_commits"] = commits
meta.update(extra_meta or {})
body = FULL_BODY.format(id=bug_id, env=env if env is not None
else FULL_ENV.format(sha=self.sha),
root=root, fix=fix, verif=verif)
text = buglog.render_meta(meta) + "\n" + body
path = self.root / ".wolf" / "bugs" / f"{bug_id}.md"
path.write_text(text, encoding="utf-8")
return path
class ParseRender(BuglogTestCase):
def test_round_trip_keeps_every_field(self):
path = self.write_record(extra_meta={"tags": ["worker", "asan"],
"gitea_issue": 285})
rec = buglog.load("bug-100")
self.assertEqual(rec.meta["tags"], ["worker", "asan"])
self.assertEqual(rec.meta["gitea_issue"], 285)
self.assertEqual(rec.summary,
"xcecd_virtual leaks worker strings on every finished job")
self.assertIn("Direct leak", rec.sections["Description"])
self.assertEqual(buglog.render(rec), path.read_text(encoding="utf-8"))
def test_section_header_inside_code_fence_is_not_a_section(self):
self.write_record(root="See:\n```\n## Fix\nnot a header\n```")
rec = buglog.load("bug-100")
self.assertIn("not a header", rec.sections["Root cause"])
self.assertEqual(rec.sections["Fix"], "TODO")
def test_escaped_section_header_round_trips(self):
text = "a\n## Fix\nb"
rec = buglog.Record({"id": "bug-7", "status": "open"}, "s",
{"Description": buglog.escape_text(text)})
again = buglog.parse(buglog.render(rec))
self.assertEqual(buglog.unescape_text(again.sections["Description"]), text)
self.assertNotIn("Fix", again.sections)
def test_set_meta_leaves_body_bytes_unchanged(self):
path = self.write_record()
body_before = path.read_text(encoding="utf-8").split("\n---\n", 1)[1]
buglog.set_meta("bug-100", {"gitea_issue": 9, "status_note": "x: y"})
text = path.read_text(encoding="utf-8")
self.assertEqual(text.split("\n---\n", 1)[1], body_before)
self.assertEqual(buglog.load("bug-100").meta["status_note"], "x: y")
class Check(BuglogTestCase):
def problems(self, bug_id="bug-100", target=None):
return buglog.check(buglog.load(bug_id), target_status=target)
def test_complete_open_record_passes(self):
self.write_record()
self.assertEqual(self.problems(), [])
def test_environment_needs_every_item(self):
env = FULL_ENV.format(sha=self.sha).replace(
"- libc: glibc 2.42 (from the binary's interpreter)\n", "")
self.write_record(env=env)
self.assertTrue(any("libc" in p for p in self.problems()))
def test_libc_needs_a_version(self):
env = FULL_ENV.format(sha=self.sha).replace("glibc 2.42", "glibc")
self.write_record(env=env)
self.assertTrue(any("libc" in p for p in self.problems()))
def test_code_needs_a_commit(self):
env = FULL_ENV.format(sha=self.sha).replace(self.sha, "latest")
self.write_record(env=env)
self.assertTrue(any("Code" in p for p in self.problems()))
def test_todo_counts_as_missing(self):
env = FULL_ENV.format(sha=self.sha).replace(
"single host, xcecd_virtual", "TODO: topology")
self.write_record(env=env)
self.assertTrue(any("Deployment" in p for p in self.problems()))
def test_fixed_needs_root_cause_fix_verification_and_commit(self):
self.write_record(status="fixed")
probs = " | ".join(self.problems())
for word in ("Root cause", "Fix", "Verification", "fix_commits"):
self.assertIn(word, probs)
def test_fixed_with_everything_passes(self):
self.write_record(status="fixed", root="remove_worker skips five strings",
fix="free them", commits=[self.sha],
verif="- Test: test_x: RED before, GREEN after\n"
"- Suites: worker 12/12, asan")
self.assertEqual(self.problems(), [])
def test_unknown_commit_is_rejected(self):
self.write_record(status="fixed", root="r", fix="f", commits=["deadbeef"],
verif="- Test: t: RED before, GREEN after\n- Suites: s")
self.assertTrue(any("deadbeef" in p for p in self.problems()))
def test_open_status_needs_note(self):
self.write_record(extra_meta={"status_note": ""})
self.assertTrue(any("status_note" in p for p in self.problems()))
def test_legacy_record_is_exempt_until_it_changes_status(self):
self.write_record(legacy=True, env=buglog.NOT_RECORDED)
self.assertEqual(self.problems(), [])
# marking it fixed now demands the fix-time sections
probs = " | ".join(self.problems(target="fixed"))
self.assertIn("Verification", probs)
self.assertIn("fix_commits", probs)
def test_migrated_placeholder_is_not_verification(self):
self.write_record(legacy=True, root="r", fix="f", commits=[self.sha],
verif=buglog.NOT_RECORDED)
self.assertTrue(any("Verification" in p
for p in self.problems(target="fixed")))
class Commands(BuglogTestCase):
def run_cli(self, *argv):
try:
buglog.main(list(argv))
return 0
except SystemExit as e:
return e.code
def test_status_refuses_incomplete_fixed_and_leaves_file_alone(self):
path = self.write_record()
before = path.read_text(encoding="utf-8")
rc = self.run_cli("status", "bug-100", "fixed", "--commit", self.sha)
self.assertNotEqual(rc, 0)
self.assertEqual(path.read_text(encoding="utf-8"), before)
def test_status_fixed_records_commits(self):
self.write_record(root="r", fix="f",
verif="- Test: t: RED before, GREEN after\n- Suites: s")
rc = self.run_cli("status", "bug-100", "fixed", "--commit", self.sha)
self.assertIn(rc, (0, None))
meta = buglog.load("bug-100").meta
self.assertEqual(meta["status"], "fixed")
self.assertEqual(meta["fix_commits"], [self.sha])
def test_new_creates_record_with_environment_filled(self):
self.run_cli("new", "--summary", "scheduler hangs on empty queue",
"--component", "scheduler", "--severity", "hang")
rec = buglog.load("bug-001")
self.assertEqual(rec.meta["status"], "open")
env = rec.sections["Environment"]
self.assertRegex(env, r"- OS: .*Linux")
self.assertRegex(env, r"- Code: .*" + self.sha)
# the author still has to write the description
self.assertTrue(buglog.check(rec))
def test_next_id_counts_archive(self):
self.write_record(bug_id="bug-100")
(self.root / ".wolf" / "buglog-archive.json").write_text(
json.dumps({"bugs": [{"id": "bug-900"}]}), encoding="utf-8")
self.assertEqual(buglog.next_id(), "bug-901")
def test_index_keeps_openwolf_fields(self):
self.write_record(extra_meta={"files": ["src/worker/a.c", "src/worker/b.c"],
"tags": ["worker"]}, root="cause text")
buglog.build_index()
idx = json.loads((self.root / ".wolf" / "buglog.json").read_text())
b = idx["bugs"][0]
self.assertEqual(b["id"], "bug-100")
self.assertTrue(b["error_message"].startswith("xcecd_virtual leaks"))
self.assertIn("Direct leak", b["error_message"])
self.assertEqual(b["root_cause"], "cause text")
self.assertEqual(b["file"], "src/worker/a.c, src/worker/b.c")
self.assertEqual(b["tags"], ["worker"])
self.assertEqual(b["status"], "open")
def test_write_commands_refresh_the_index(self):
self.write_record()
self.run_cli("note", "bug-100", "seen again under load")
idx = json.loads((self.root / ".wolf" / "buglog.json").read_text())
self.assertIn("seen again under load", idx["bugs"][0]["error_message"])
class Staged(BuglogTestCase):
"""`check --staged`, which the pre-commit hook runs."""
def commit(self, *paths):
git(self.root, "add", *paths)
git(self.root, "-c", "user.email=t@t", "-c", "user.name=t",
"commit", "-q", "-m", "x")
def check_staged(self):
try:
buglog.main(["check", "--staged"])
return 0
except SystemExit as e:
return e.code or 0
def test_complete_record_with_fresh_index_passes(self):
self.write_record()
buglog.build_index()
git(self.root, "add", ".wolf")
self.assertEqual(self.check_staged(), 0)
def test_incomplete_record_is_refused(self):
self.write_record(env="- OS: TODO")
buglog.build_index()
git(self.root, "add", ".wolf")
self.assertNotEqual(self.check_staged(), 0)
def test_stale_index_is_refused(self):
self.write_record()
git(self.root, "add", ".wolf")
self.assertNotEqual(self.check_staged(), 0)
def test_hand_edited_index_alone_is_refused(self):
self.write_record()
buglog.build_index()
self.commit(".wolf")
idx = self.root / ".wolf" / "buglog.json"
idx.write_text(idx.read_text().replace("leaks", "LEAKS"))
git(self.root, "add", ".wolf/buglog.json")
self.assertNotEqual(self.check_staged(), 0)
def test_legacy_record_marked_fixed_by_hand_needs_fix_sections(self):
self.write_record(legacy=True, env=buglog.NOT_RECORDED)
buglog.build_index()
self.commit(".wolf")
buglog.set_meta("bug-100", {"status": "fixed", "status_note": ""})
git(self.root, "add", ".wolf")
self.assertNotEqual(self.check_staged(), 0)
def test_migrated_record_matching_the_old_index_passes(self):
# the migration commit: the record is new, but HEAD's buglog.json
# already had this bug with the same status
(self.root / ".wolf" / "buglog.json").write_text(
json.dumps({"bugs": [{"id": "bug-100", "status": "fixed"}]}))
self.commit(".wolf")
self.write_record(legacy=True, status="fixed", env=buglog.NOT_RECORDED)
buglog.build_index()
git(self.root, "add", ".wolf")
self.assertEqual(self.check_staged(), 0)
def test_new_record_cannot_claim_legacy(self):
self.write_record(legacy=True, status="fixed", env=buglog.NOT_RECORDED)
buglog.build_index()
git(self.root, "add", ".wolf")
self.assertNotEqual(self.check_staged(), 0)
def test_untouched_legacy_record_passes(self):
self.write_record(legacy=True, env=buglog.NOT_RECORDED)
buglog.build_index()
self.commit(".wolf")
buglog.set_meta("bug-100", {"gitea_issue": 5})
git(self.root, "add", ".wolf")
self.assertEqual(self.check_staged(), 0)
if __name__ == "__main__":
unittest.main()