fix: correct file status and metrics to use latest status per step

- Fix _compute_status_from_logs to track latest status per unique step
- Fix _compute_step_summary to count only latest status per step
- Add comprehensive tests for both fixes
- Resolves issue where completed files showed as "Processing"
- Resolves issue where metrics showed incorrect counts

Co-authored-by: christianlouis <361235+christianlouis@users.noreply.github.com>
This commit is contained in:
copilot-swe-agent[bot]
2026-02-11 18:33:41 +00:00
parent 1e5cbc4f70
commit 4fb696e1eb
3 changed files with 257 additions and 14 deletions
+12 -5
View File
@@ -76,13 +76,20 @@ def _compute_status_from_logs(logs: List[ProcessingLog]) -> Dict:
if not logs:
return {"status": "pending", "last_step": None, "has_errors": False, "total_steps": 0}
# Check for failures
has_errors = any(log.status == "failure" for log in logs)
# Get the latest status for each unique step
# Since logs are ordered by timestamp desc, the first occurrence is the latest
latest_by_step = {}
for log in logs:
if log.step_name not in latest_by_step:
latest_by_step[log.step_name] = log
# Check if any in progress
in_progress = any(log.status == "in_progress" for log in logs)
# Check for failures in latest statuses
has_errors = any(log.status == "failure" for log in latest_by_step.values())
# Get the latest log
# Check if any step is currently in progress (based on latest status per step)
in_progress = any(log.status == "in_progress" for log in latest_by_step.values())
# Get the overall latest log
latest_log = logs[0]
# Determine overall status
+14 -9
View File
@@ -347,8 +347,8 @@ def _compute_step_summary(logs):
main_counts = {"queued": 0, "in_progress": 0, "success": 0, "failure": 0}
upload_counts = {"queued": 0, "in_progress": 0, "success": 0, "failure": 0}
# Track which steps we've seen
main_steps_seen = set()
# Track latest status for each step (logs are ordered by timestamp desc)
main_steps_seen = {}
upload_tasks_seen = {}
for log in logs:
@@ -363,15 +363,20 @@ def _compute_step_summary(logs):
is_upload = any(step_name.startswith(prefix) for prefix in upload_prefixes)
if is_upload:
# Track latest status for each unique upload task
upload_tasks_seen[step_name] = status
# Track latest status for each unique upload task (first seen is latest)
if step_name not in upload_tasks_seen:
upload_tasks_seen[step_name] = status
elif step_name in main_steps:
# Track latest status for main steps
main_steps_seen.add(step_name)
if status in main_counts:
main_counts[status] += 1
# Track latest status for main steps (first seen is latest)
if step_name not in main_steps_seen:
main_steps_seen[step_name] = status
# Count upload task statuses
# Count main step statuses from latest status per step
for task_status in main_steps_seen.values():
if task_status in main_counts:
main_counts[task_status] += 1
# Count upload task statuses from latest status per task
for task_status in upload_tasks_seen.values():
if task_status in upload_counts:
upload_counts[task_status] += 1