Fixed a few stupid mistakes

2026-05-22 23:48:24 +02:00
parent a440931f7b
commit d9e7dce4e6
7 changed files with 379 additions and 51 deletions
--- a/src/gitea_codex_bot/main.py
+++ b/src/gitea_codex_bot/main.py
@@ -11,10 +11,12 @@ from fastapi import Depends, FastAPI, Header, HTTPException, Request, status
 from fastapi.exception_handlers import http_exception_handler
 from fastapi.responses import HTMLResponse
 from starlette.exceptions import HTTPException as StarletteHTTPException
+from sqlalchemy import select
 from sqlalchemy.orm import Session

 from gitea_codex_bot.config import Settings, get_settings
 from gitea_codex_bot.db import get_session
+from gitea_codex_bot.models import JobStatus, ReviewJob
 from gitea_codex_bot.services.commands import parse_command
 from gitea_codex_bot.services.gitea import GiteaClient
 from gitea_codex_bot.services.jobs import cooldown_remaining_seconds, enqueue_job, persist_webhook_event
@@ -147,6 +149,13 @@ def _load_repo_review_config_for_pr(gitea: GiteaClient, repo: str, pr_number: in
    return parse_repo_review_config_text(cfg_text, configured=True), head_sha


+def _resolve_pr_head_sha(gitea: GiteaClient, repo: str, pr_number: int, fallback: str) -> str:
+    try:
+        return gitea.get_pull_request(repo, pr_number).head_sha
+    except Exception:
+        return fallback
+
+
 def _render_landing_page() -> str:
    return """<!doctype html>
 <html lang="en">
@@ -164,6 +173,8 @@ def _render_landing_page() -> str:
        <p class="mt-4 text-base leading-7 text-slate-300">This endpoint powers automated pull request review workflows for Gitea. It validates signed webhook events, queues review jobs, and posts structured feedback back to pull requests.</p>
        <div class="mt-8 flex flex-wrap gap-3 text-sm">
          <button id="health-button" type="button" class="rounded-lg border border-slate-700 bg-slate-800/80 px-3 py-2 text-slate-200 transition hover:border-slate-500 hover:bg-slate-700">Health: <code>/healthz</code></button>
+          <button id="failure-button" type="button" class="rounded-lg border border-amber-500/40 bg-amber-500/10 px-3 py-2 text-amber-200 transition hover:border-amber-400 hover:bg-amber-500/20">Latest failure: <code>/healthz/latest-failure</code></button>
+          <button id="job-button" type="button" class="rounded-lg border border-cyan-500/40 bg-cyan-500/10 px-3 py-2 text-cyan-200 transition hover:border-cyan-400 hover:bg-cyan-500/20">Latest job: <code>/healthz/latest-job</code></button>
          <span class="rounded-lg border border-slate-700 bg-slate-800/80 px-3 py-2 text-slate-200">Webhook: <code>POST /webhook/gitea</code></span>
        </div>
      </section>
@@ -179,6 +190,8 @@ def _render_landing_page() -> str:
    </div>
    <script>
      const healthButton = document.getElementById("health-button");
+      const failureButton = document.getElementById("failure-button");
+      const jobButton = document.getElementById("job-button");
      const healthModal = document.getElementById("health-modal");
      const closeModal = document.getElementById("close-modal");
      const healthResult = document.getElementById("health-result");
@@ -196,6 +209,57 @@ def _render_landing_page() -> str:
        }
      }

+      async function loadLatestFailure() {
+        healthResult.textContent = "Loading...";
+        try {
+          const response = await fetch("/healthz/latest-failure", { headers: { Accept: "application/json" } });
+          const payload = await response.json();
+          if (!payload.has_failed_job) {
+            healthResult.textContent = "No failed jobs found.";
+            return;
+          }
+          const failedAt = payload.failed_at ? payload.failed_at : "unknown";
+          const errorText = payload.error ? payload.error : "unknown";
+          healthResult.textContent =
+            "Latest failed job #" + payload.job_id +
+            " | " + payload.repo + "#" + payload.pr_number +
+            " | command=" + payload.command +
+            " | commit=" + payload.head_sha.slice(0, 7) +
+            " | failed_at=" + failedAt +
+            " | error=" + errorText;
+        } catch (_error) {
+          healthResult.textContent = "Could not load latest failure output.";
+        }
+      }
+
+      async function loadLatestJob() {
+        healthResult.textContent = "Loading...";
+        try {
+          const response = await fetch("/healthz/latest-job", { headers: { Accept: "application/json" } });
+          const payload = await response.json();
+          if (!payload.has_job) {
+            healthResult.textContent = "No jobs found yet.";
+            return;
+          }
+          const startedAt = payload.started_at ? payload.started_at : "not started";
+          const finishedAt = payload.finished_at ? payload.finished_at : "not finished";
+          const errorText = payload.error ? payload.error : "none";
+          const summary = payload.result_summary ? payload.result_summary : "none";
+          healthResult.textContent =
+            "Latest job #" + payload.job_id +
+            " | " + payload.repo + "#" + payload.pr_number +
+            " | command=" + payload.command +
+            " | status=" + payload.job_status +
+            " | commit=" + payload.head_sha.slice(0, 7) +
+            " | started_at=" + startedAt +
+            " | finished_at=" + finishedAt +
+            " | error=" + errorText +
+            " | summary=" + summary;
+        } catch (_error) {
+          healthResult.textContent = "Could not load latest job output.";
+        }
+      }
+
      function showModal() {
        healthModal.classList.remove("hidden");
        healthModal.classList.add("flex");
@@ -210,6 +274,14 @@ def _render_landing_page() -> str:
        showModal();
        await loadHealth();
      });
+      failureButton.addEventListener("click", async function () {
+        showModal();
+        await loadLatestFailure();
+      });
+      jobButton.addEventListener("click", async function () {
+        showModal();
+        await loadLatestJob();
+      });

      closeModal.addEventListener("click", hideModal);
      healthModal.addEventListener("click", function (event) {
@@ -264,6 +336,54 @@ def healthz(settings: Settings = Depends(get_settings)) -> dict[str, str]:
    return {"status": "ok"}


+@app.get("/healthz/latest-failure")
+def healthz_latest_failure(session: Session = Depends(get_session)) -> dict[str, Any]:
+    failed_job = session.execute(
+        select(ReviewJob).where(ReviewJob.status == JobStatus.failed).order_by(ReviewJob.created_at.desc(), ReviewJob.id.desc()).limit(1)
+    ).scalar_one_or_none()
+    if not failed_job:
+        return {"status": "ok", "has_failed_job": False}
+    return {
+        "status": "ok",
+        "has_failed_job": True,
+        "job_id": failed_job.id,
+        "repo": failed_job.repo,
+        "pr_number": failed_job.pr_number,
+        "command": failed_job.command,
+        "head_sha": failed_job.head_sha,
+        "error": failed_job.last_error or "",
+        "failed_at": failed_job.finished_at.isoformat() if failed_job.finished_at else None,
+    }
+
+
+@app.get("/healthz/latest-job")
+def healthz_latest_job(session: Session = Depends(get_session)) -> dict[str, Any]:
+    latest_job = session.execute(select(ReviewJob).order_by(ReviewJob.created_at.desc(), ReviewJob.id.desc()).limit(1)).scalar_one_or_none()
+    if not latest_job:
+        return {"status": "ok", "has_job": False}
+
+    result_summary = ""
+    if isinstance(latest_job.result_json, dict):
+        summary = latest_job.result_json.get("summary")
+        if isinstance(summary, str):
+            result_summary = summary
+    return {
+        "status": "ok",
+        "has_job": True,
+        "job_id": latest_job.id,
+        "repo": latest_job.repo,
+        "pr_number": latest_job.pr_number,
+        "command": latest_job.command,
+        "head_sha": latest_job.head_sha,
+        "job_status": latest_job.status.value if hasattr(latest_job.status, "value") else str(latest_job.status),
+        "error": latest_job.last_error or "",
+        "result_summary": result_summary,
+        "created_at": latest_job.created_at.isoformat() if latest_job.created_at else None,
+        "started_at": latest_job.started_at.isoformat() if latest_job.started_at else None,
+        "finished_at": latest_job.finished_at.isoformat() if latest_job.finished_at else None,
+    }
+
+
@app.post("/webhook/gitea")
 async def gitea_webhook(
    request: Request,
@@ -334,6 +454,7 @@ async def gitea_webhook(

    gitea = GiteaClient(settings)
    if parsed_command.name in {"review", "rerun"}:
+        head_sha = _resolve_pr_head_sha(gitea, repo, pr_number, head_sha)
        repo_cfg: RepoReviewConfig | None = None
        try:
            repo_cfg, resolved_head_sha = _load_repo_review_config_for_pr(gitea, repo, pr_number)
@@ -341,10 +462,7 @@ async def gitea_webhook(
        except Exception:
            repo_cfg = None
        if head_sha == "unknown":
-            try:
-                head_sha = gitea.get_pull_request(repo, pr_number).head_sha
-            except Exception:
-                pass
+            head_sha = _resolve_pr_head_sha(gitea, repo, pr_number, head_sha)
        if repo_cfg and not repo_cfg.enabled:
            gitea.post_issue_comment(repo, pr_number, format_disabled_ack())
            return {"accepted": True, "reason": "review disabled by repo config"}
--- a/src/gitea_codex_bot/services/gitea.py
+++ b/src/gitea_codex_bot/services/gitea.py
@@ -90,6 +90,16 @@ class GiteaClient:
        )
        return int(payload["id"])

+    def get_issue_comment(self, repo: str, comment_id: int) -> dict[str, Any]:
+        owner, name = self.split_repo(repo)
+        encoded_owner = quote(owner, safe="")
+        encoded_name = quote(name, safe="")
+        payload = self._request(
+            "GET",
+            f"/api/v1/repos/{encoded_owner}/{encoded_name}/issues/comments/{comment_id}",
+        )
+        return dict(payload)
+
    def list_issue_comments(self, repo: str, pr_number: int) -> list[dict[str, Any]]:
        owner, name = self.split_repo(repo)
        encoded_owner = quote(owner, safe="")
--- a/src/gitea_codex_bot/services/jobs.py
+++ b/src/gitea_codex_bot/services/jobs.py
@@ -1,5 +1,6 @@
 from __future__ import annotations

+import logging
 from datetime import datetime, timedelta, timezone

 from sqlalchemy import select
@@ -10,6 +11,8 @@ from gitea_codex_bot.models import JobStatus, ReviewJob, ReviewRun, RunStatus, W
 from gitea_codex_bot.services.security import payload_digest
 from gitea_codex_bot.types import ParsedCommand

+logger = logging.getLogger(__name__)
+

 def persist_webhook_event(
    session: Session,
@@ -79,6 +82,16 @@ def enqueue_job(
    session.add(job)
    session.commit()
    session.refresh(job)
+    logger.info(
+        "Job enqueued id=%s repo=%s pr=%s command=%s head_sha=%s trigger_comment_id=%s requested_by=%s",
+        job.id,
+        job.repo,
+        job.pr_number,
+        job.command,
+        job.head_sha,
+        job.trigger_comment_id,
+        job.requested_by,
+    )
    return job


@@ -95,6 +108,15 @@ def claim_next_job(session: Session) -> ReviewJob | None:
    session.add(run)
    session.commit()
    session.refresh(job)
+    logger.info(
+        "Job claimed id=%s repo=%s pr=%s command=%s head_sha=%s status=%s",
+        job.id,
+        job.repo,
+        job.pr_number,
+        job.command,
+        job.head_sha,
+        job.status.value if hasattr(job.status, "value") else job.status,
+    )
    return job


@@ -136,3 +158,13 @@ def finish_job(
        latest_run.error_message = error_message

    session.commit()
+    logger.info(
+        "Job finished id=%s repo=%s pr=%s status=%s run_status=%s skipped=%s error_present=%s",
+        job.id,
+        job.repo,
+        job.pr_number,
+        job.status.value if hasattr(job.status, "value") else job.status,
+        run_status.value if hasattr(run_status, "value") else run_status,
+        skipped,
+        bool(error_message),
+    )
--- a/src/gitea_codex_bot/workers/dispatcher.py
+++ b/src/gitea_codex_bot/workers/dispatcher.py
@@ -4,14 +4,13 @@ import asyncio
 import logging
 from typing import Any

-import httpx
 from sqlalchemy import select
 from sqlalchemy.orm import Session

 from gitea_codex_bot.config import Settings
 from gitea_codex_bot.db import get_session_factory
 from gitea_codex_bot.models import ReviewJob
-from gitea_codex_bot.services.comments import get_persistent_review_comment_id, upsert_persistent_review_comment_id
+from gitea_codex_bot.services.comments import upsert_persistent_review_comment_id
 from gitea_codex_bot.services.gitea import GiteaClient
 from gitea_codex_bot.services.jobs import claim_next_job, finish_job
 from gitea_codex_bot.services.review_format import format_disabled_ack, format_result_comment
@@ -75,6 +74,16 @@ def _handle_non_review_command(
    return False, False, None, None


+def _post_review_failure_comment(gitea: GiteaClient, job: ReviewJob, error_message: str) -> None:
+    message = (
+        "⚠️ Codex review run failed after queueing.\n\n"
+        f"- Commit: `{job.head_sha[:7]}`\n"
+        f"- Error: `{error_message[:500]}`\n\n"
+        "Please rerun `@codex rerun` after checking worker logs."
+    )
+    gitea.post_issue_comment(job.repo, job.pr_number, message)
+
+
 def process_one_job(settings: Settings) -> bool:
    session_factory = get_session_factory()
    with session_factory() as session:
@@ -84,11 +93,27 @@ def process_one_job(settings: Settings) -> bool:

    command = _command_from_job(job)
    gitea = GiteaClient(settings)
+    logger.info(
+        "Processing job id=%s repo=%s pr=%s command=%s args=%s head_sha=%s",
+        job.id,
+        job.repo,
+        job.pr_number,
+        command.name,
+        command.arguments,
+        job.head_sha,
+    )

    with session_factory() as session:
        db_job = session.execute(select(ReviewJob).where(ReviewJob.id == job.id)).scalar_one()
        handled, skipped, result, error = _handle_non_review_command(settings, session, gitea, db_job, command)
        if handled:
+            logger.info(
+                "Non-review command handled job id=%s command=%s skipped=%s error_present=%s",
+                db_job.id,
+                command.name,
+                skipped,
+                bool(error),
+            )
            finish_job(session, job_id=db_job.id, success=error is None, skipped=skipped, result=result, error_message=error)
            return True

@@ -108,6 +133,15 @@ def process_one_job(settings: Settings) -> bool:
                )
                return True
        result, repo_cfg = run_review_ephemeral(settings, repo=job.repo, pr_number=job.pr_number, command=command)
+        logger.info(
+            "Runner returned job id=%s repo=%s pr=%s repo_cfg_enabled=%s repo_cfg_configured=%s result_keys=%s",
+            job.id,
+            job.repo,
+            job.pr_number,
+            repo_cfg.enabled,
+            repo_cfg.configured,
+            sorted(result.keys()),
+        )
        if not repo_cfg.enabled:
            with session_factory() as session:
                gitea.post_issue_comment(job.repo, job.pr_number, format_disabled_ack())
@@ -122,22 +156,14 @@ def process_one_job(settings: Settings) -> bool:
                return True
        comment_body = format_result_comment(job.head_sha, result, repo_configured=repo_cfg.configured)
        with session_factory() as session:
-            comment_id = get_persistent_review_comment_id(session, job.repo, job.pr_number)
-            if comment_id:
-                try:
-                    gitea.edit_issue_comment(job.repo, comment_id, comment_body)
-                except httpx.HTTPStatusError as exc:
-                    if exc.response.status_code != 404:
-                        raise
-                    logger.warning(
-                        "Persistent review comment not found; posting a new one repo=%s pr=%s old_comment_id=%s",
-                        job.repo,
-                        job.pr_number,
-                        comment_id,
-                    )
-                    comment_id = gitea.post_issue_comment(job.repo, job.pr_number, comment_body)
-            else:
-                comment_id = gitea.post_issue_comment(job.repo, job.pr_number, comment_body)
+            comment_id = gitea.post_issue_comment(job.repo, job.pr_number, comment_body)
+            logger.info(
+                "Posted review comment job id=%s repo=%s pr=%s comment_id=%s",
+                job.id,
+                job.repo,
+                job.pr_number,
+                comment_id,
+            )
            upsert_persistent_review_comment_id(
                session,
                repo=job.repo,
@@ -145,11 +171,24 @@ def process_one_job(settings: Settings) -> bool:
                head_sha=job.head_sha,
                comment_id=comment_id,
            )
+            logger.info(
+                "Persistent comment mapping upserted job id=%s repo=%s pr=%s comment_id=%s head_sha=%s",
+                job.id,
+                job.repo,
+                job.pr_number,
+                comment_id,
+                job.head_sha,
+            )
            finish_job(session, job_id=job.id, success=True, skipped=False, result=result, error_message=None)
    except Exception as exc:
        logger.exception("Review job failed id=%s", job.id)
+        error_text = str(exc).strip() or exc.__class__.__name__
+        try:
+            _post_review_failure_comment(gitea, job, error_text)
+        except Exception:
+            logger.exception("Failed to post review failure comment id=%s", job.id)
        with session_factory() as session:
-            finish_job(session, job_id=job.id, success=False, skipped=False, result=None, error_message=str(exc))
+            finish_job(session, job_id=job.id, success=False, skipped=False, result=None, error_message=error_text)
    return True