Skip to content

fix(swe-smith): parse pytest statuses that include ANSI color #599

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
Zhiyuan He (hzy46) merged 2 commits into from
Sep 17, 2026
Merged

fix(swe-smith): parse pytest statuses that include ANSI color #599

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
c247946
fix(swe-smith): parse pytest statuses that include ANSI color
xccElephant Sep 16, 2026
f175d05
fix(swe-smith): drop extra ANSI parser tests per review
xccElephant Sep 17, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
8 changes: 6 additions & 2 deletions examples/swe_smith/agents/smith_agent.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -31,6 +31,10 @@
_STATUS_RE = re.compile(
r"(?:^|\s)(PASSED|FAILED|ERROR|SKIPPED|XFAIL|XPASS)\s+(\S+)|(\S+)\s+(PASSED|FAILED|ERROR|SKIPPED|XFAIL|XPASS)(?:\s|$)"
)
# pytest --color=yes (or a repo that forces color) wraps PASSED in ANSI codes.
# Without stripping them, parse_test_statuses never sees a bare "PASSED" token
# and a fully passing suite is scored 0/N.
_ANSI_ESCAPE_RE = re.compile(r"\x1b\[[0-?]*[ -/]*[@-~]")

SUBMIT_MARKER = "COMPLETE_TASK_AND_SUBMIT_FINAL_OUTPUT"

Expand Down Expand Up @@ -571,7 +575,7 @@ def capture_patch() -> str:

def parse_test_statuses(test_output: str) -> dict[str, str]:
statuses: dict[str, str] = {}
for line in test_output.splitlines():
for line in _ANSI_ESCAPE_RE.sub("", test_output).splitlines():
m = _STATUS_RE.search(line)
if not m:
continue
Expand Down Expand Up @@ -603,7 +607,7 @@ def evaluate(eval_meta: dict[str, Any], timeout: int, f2p_only: bool = True) ->
# needs xdist, so probe first and fall back to serial `-p no:xdist` when absent.
xdist_flag = "-n4" if _run("python -c 'import xdist'", 30)[1] == 0 else "-p no:xdist"
output, rc = _run(
"python -m pytest -rA -p no:cacheprovider " + xdist_flag + " " + " ".join(map(_shq, nodes)),
"python -m pytest --color=no -rA -p no:cacheprovider " + xdist_flag + " " + " ".join(map(_shq, nodes)),
timeout,
)
# rc 124 == subprocess.TimeoutExpired (see _run); pytest itself never exits 124.
Expand Down