feat: add Review Comment Generator (Issue #126)

- Introduces scripts/review_comment_generator.py: reads JSONL findings,
  deduplicates by content hash, formats as review comments, and posts
  to Gitea PR via API.
- Includes dry-run and JSON output modes.
- Comprehensive smoke test suite: 20 tests covering deduplication,
  formatting, CLI modes, and error handling — all passing.

Closes #126

scripts/progress_tracker.py

@@ -1,477 +0,0 @@
-#!/usr/bin/env python3
-"""
-Progress Tracker — Pipeline 10.8
-Track improvement metrics over time. Are we getting better?
-
-Metrics tracked:
- 1. Test coverage  — % of Python functions with associated tests (test:source file ratio + line coverage if available)
- 2. Doc coverage   — % of Python callables with docstrings (AST-based)
- 3. Issue close rate — closed / (opened + closed) per week (Gitea API)
- 4. Dep freshness  — % of requirements pinned vs outdated (pip list --outdated)
-
-Output:
- - metrics/snapshots/YYYY-MM-DD.json  — one snapshot per run
- - metrics/TRENDS.md                  — cumulative markdown table
- - stdout summary
-
-Usage:
-    python3 scripts/progress_tracker.py
-    python3 scripts/progress_tracker.py --json
-    python3 scripts/progress_tracker.py --output metrics/TRENDS.md
-
-Weekly cron:
-    0 9 * * 1 cd /path/to/compounding-intelligence && python3 scripts/progress_tracker.py
-"""
-
-import argparse
-import json
-import os
-import re
-import subprocess
-import sys
-from collections import defaultdict
-from datetime import datetime, timezone, timedelta
-from pathlib import Path
-from typing import Any, Dict, List, Optional, Tuple
-
-# ── Configuration ──────────────────────────────────────────────────────────
-
-SCRIPT_DIR = Path(__file__).resolve().parent
-REPO_ROOT = SCRIPT_DIR.parent
-METRICS_DIR = REPO_ROOT / "metrics"
-SNAPSHOTS_DIR = METRICS_DIR / "snapshots"
-TOKEN_PATH = Path.home() / ".config" / "gitea" / "token"
-GITEA_API_BASE = "https://forge.alexanderwhitestone.com/api/v1"
-ORG = "Timmy_Foundation"
-
-# Ensure paths exist
-SNAPSHOTS_DIR.mkdir(parents=True, exist_ok=True)
-
-
-# ── Helpers ─────────────────────────────────────────────────────────────────
-
-def run_cmd(cmd: List[str], cwd: Path = REPO_ROOT) -> str:
-    """Run a shell command and return stdout (stderr merged)."""
-    result = subprocess.run(
-        cmd, capture_output=True, text=True, cwd=cwd, timeout=30
-    )
-    if result.returncode != 0:
-        return ""
-    return result.stdout.strip()
-
-
-def slugify_date(dt: datetime) -> str:
-    return dt.strftime("%Y-%m-%d")
-
-
-def snapshot_path(dt: datetime) -> Path:
-    return SNAPSHOTS_DIR / f"{slugify_date(dt)}.json"
-
-
-def load_snapshots() -> List[Dict[str, Any]]:
-    """Load all existing snapshots sorted by date."""
-    snapshots = []
-    for f in sorted(SNAPSHOTS_DIR.glob("*.json")):
-        try:
-            with open(f) as fp:
-                snapshots.append(json.load(fp))
-        except Exception:
-            continue
-    return snapshots
-
-
-# ── Metric 1: Test Coverage ─────────────────────────────────────────────────
-
-def collect_test_coverage() -> Dict[str, Any]:
-    """
-    Compute test coverage metrics.
-    Counts test_*.py and *_test.py files vs non-test .py source files.
-    Also attempts to read .coverage if present.
-    """
-    all_py = list(REPO_ROOT.rglob("*.py"))
-    
-    source_files = []
-    test_files = []
-    
-    for p in all_py:
-        try:
-            rel_parts = p.relative_to(REPO_ROOT).parts
-        except ValueError:
-            continue
-        
-        # Skip hidden/cache/temp dirs (check only relative parts)
-        if any(part.startswith('.') or part.startswith('__') for part in rel_parts):
-            continue
-        if any(part in ('node_modules', 'venv', '.venv', 'env', '.pytest_cache') for part in rel_parts):
-            continue
-        
-        if p.name.startswith("test_") or p.name.endswith("_test.py"):
-            test_files.append(p)
-        else:
-            source_files.append(p)
-    
-    # Try to get line coverage from .coverage
-    coverage_percent = None
-    coverage_tool = None
-    coverage_file = REPO_ROOT / ".coverage"
-    if coverage_file.exists():
-        try:
-            import coverage  # type: ignore
-            # Use coverage API if available
-            cov = coverage.Coverage(data_file=str(coverage_file))
-            cov.load()
-            total = cov.report()
-            coverage_percent = total if isinstance(total, float) else None
-            coverage_tool = "coverage"
-        except Exception:
-            # Fallback: parse `coverage report` output
-            out = run_cmd(["coverage", "report", "--skip-empty"])
-            if out:
-                for line in out.splitlines():
-                    if "TOTAL" in line:
-                        parts = line.split()
-                        if len(parts) >= 2:
-                            try:
-                                coverage_percent = float(parts[-1].rstrip('%'))
-                                coverage_tool = "coverage"
-                                break
-                            except ValueError:
-                                pass
-    
-    return {
-        "test_files": len(test_files),
-        "source_files": len(source_files),
-        "test_to_source_ratio": round(len(test_files) / len(source_files), 4) if source_files else 0.0,
-        "coverage_tool": coverage_tool,
-        "coverage_percent": coverage_percent,
-    }
-
-
-# ── Metric 2: Doc Coverage ──────────────────────────────────────────────────
-
-def collect_doc_coverage() -> Dict[str, Any]:
-    """
-    Check AST of Python files for docstrings.
-    Returns: callables_total, callables_with_doc, doc_coverage_percent
-    """
-    import ast
-    
-    all_py = list(REPO_ROOT.rglob("*.py"))
-    
-    source_files = []
-    test_files = []
-    
-    for p in all_py:
-        try:
-            rel_parts = p.relative_to(REPO_ROOT).parts
-        except ValueError:
-            continue
-        
-        if any(part.startswith('.') or part.startswith('__') for part in rel_parts):
-            continue
-        if any(part in ('node_modules', 'venv', '.venv', 'env', '.pytest_cache') for part in rel_parts):
-            continue
-        
-        if p.name.startswith("test_") or p.name.endswith("_test.py"):
-            test_files.append(p)
-        else:
-            source_files.append(p)
-    
-    total_callables = 0
-    with_doc = 0
-    
-    for p in source_files + test_files:
-        try:
-            with open(p) as f:
-                tree = ast.parse(f.read(), filename=str(p))
-            for node in ast.walk(tree):
-                if isinstance(node, (ast.FunctionDef, ast.AsyncFunctionDef, ast.ClassDef)):
-                    total_callables += 1
-                    doc = ast.get_docstring(node)
-                    if doc and doc.strip():
-                        with_doc += 1
-        except Exception:
-            continue
-    
-    return {
-        "callables_total": total_callables,
-        "callables_with_doc": with_doc,
-        "doc_coverage_percent": round((with_doc / total_callables * 100) if total_callables else 0.0, 2),
-    }
-
-
-# ── Metric 3: Issue Close Rate ──────────────────────────────────────────────
-
-def collect_issue_metrics() -> Dict[str, Any]:
-    """
-    Use Gitea API to get issue open/close stats for the last 7 days.
-    Returns counts and close rate.
-    """
-    token = ""
-    if TOKEN_PATH.exists():
-        token = TOKEN_PATH.read_text().strip()
-    
-    if not token:
-        return {
-            "opened_last_7d": None,
-            "closed_last_7d": None,
-            "close_rate": None,
-            "total_open": None,
-            "note": "Gitea token not available"
-        }
-    
-    try:
-        from urllib.request import Request, urlopen
-        from urllib.error import HTTPError, URLError
-    except ImportError:
-        return {"error": "urllib not available"}
-    
-    now = datetime.now(timezone.utc)
-    week_ago = now - timedelta(days=7)
-    since = week_ago.strftime("%Y-%m-%d")
-    
-    headers = {"Authorization": f"token {token}"}
-    base_url = f"{GITEA_API_BASE}/repos/{ORG}/compounding-intelligence/issues"
-    
-    try:
-        # Get issues from last 7 days
-        url = f"{base_url}?state=all&since={since}&per_page=100"
-        req = Request(url, headers=headers)
-        with urlopen(req, timeout=15) as resp:
-            issues = json.loads(resp.read())
-        
-        opened = 0
-        closed = 0
-        for issue in issues:
-            created = datetime.fromisoformat(issue["created_at"].replace("Z", "+00:00"))
-            if created >= week_ago:
-                opened += 1
-            if issue.get("state") == "closed":
-                closed_at_str = issue.get("closed_at")
-                if closed_at_str:
-                    closed_at = datetime.fromisoformat(closed_at_str.replace("Z", "+00:00"))
-                    if closed_at >= week_ago:
-                        closed += 1
-        
-        # Total open issues
-        req2 = Request(f"{base_url}?state=open&per_page=1", headers=headers)
-        with urlopen(req2, timeout=15) as resp:
-            total_open = int(resp.headers.get("X-Total-Count", "0"))
-        
-        total = opened + closed
-        close_rate = closed / total if total > 0 else 0.0
-        
-        return {
-            "opened_last_7d": opened,
-            "closed_last_7d": closed,
-            "close_rate": round(close_rate, 4),
-            "total_open": total_open,
-        }
-    except Exception as e:
-        return {
-            "opened_last_7d": None,
-            "closed_last_7d": None,
-            "close_rate": None,
-            "total_open": None,
-            "error": str(e)[:100],
-            "note": "Gitea API unavailable"
-        }
-
-
-# ── Metric 4: Dependency Freshness ─────────────────────────────────────────
-
-def collect_dep_freshness() -> Dict[str, Any]:
-    """
-    Check requirements.txt for outdated dependencies using pip list --outdated.
-    Returns freshness percentage and outdated list.
-    """
-    req_file = REPO_ROOT / "requirements.txt"
-    if not req_file.exists():
-        return {
-            "total_deps": 0,
-            "outdated_deps": 0,
-            "freshness_percent": 100.0,
-            "outdated_list": [],
-            "note": "requirements.txt not found"
-        }
-    
-    # Parse requirements (very simple: take name before comparison op)
-    reqs = []
-    with open(req_file) as f:
-        for line in f:
-            line = line.strip()
-            if not line or line.startswith("#"):
-                continue
-            m = re.match(r"^([a-zA-Z0-9_.-]+)", line)
-            if m:
-                reqs.append(m.group(1))
-    
-    if not reqs:
-        return {"total_deps": 0, "outdated_deps": 0, "freshness_percent": 100.0, "outdated_list": []}
-    
-    # Query pip for outdated packages (may fail if pip not available)
-    outdated_names = set()
-    try:
-        out = run_cmd(["pip", "list", "--outdated", "--format=json"])
-        if out:
-            data = json.loads(out)
-            outdated_names = {item["name"].lower() for item in data}
-    except Exception:
-        pass
-    
-    outdated = [p for p in reqs if p.lower() in outdated_names]
-    total = len(reqs)
-    outdated_count = len(outdated)
-    freshness = round(((total - outdated_count) / total * 100) if total else 100.0, 1)
-    
-    return {
-        "total_deps": total,
-        "outdated_deps": outdated_count,
-        "freshness_percent": freshness,
-        "outdated_list": outdated,
-    }
-
-
-# ── Snapshot & Trends ───────────────────────────────────────────────────────
-
-def take_snapshot() -> Dict[str, Any]:
-    """Collect all metrics and return a snapshot dict."""
-    now = datetime.now(timezone.utc)
-    test_cov = collect_test_coverage()
-    doc_cov = collect_doc_coverage()
-    issues = collect_issue_metrics()
-    deps = collect_dep_freshness()
-    
-    return {
-        "timestamp": now.isoformat(),
-        "date": slugify_date(now),
-        "metrics": {
-            "test_coverage": test_cov,
-            "doc_coverage": doc_cov,
-            "issues": issues,
-            "dependencies": deps,
-        }
-    }
-
-
-def save_snapshot(snapshot: Dict[str, Any]) -> Path:
-    path = snapshot_path(datetime.fromisoformat(snapshot["timestamp"]))
-    with open(path, "w") as f:
-        json.dump(snapshot, f, indent=2)
-    return path
-
-
-def generate_trends(snapshots: List[Dict[str, Any]], output_path: Optional[Path] = None) -> str:
-    """Generate markdown trends table; optionally write to file."""
-    if not snapshots:
-        msg = "# Progress Tracker — Trends\n\nNo snapshots yet. Run `progress_tracker.py` to create the first snapshot."
-        if output_path:
-            output_path.parent.mkdir(parents=True, exist_ok=True)
-            output_path.write_text(msg)
-        return msg
-    
-    lines = [
-        "# Progress Tracker — Trends",
-        f"\nLast updated: {datetime.now(timezone.utc).strftime('%Y-%m-%d %H:%M UTC')}",
-        f"\nSnapshots: {len(snapshots)}\n",
-        "| Date | Test Files → Source | Doc Coverage | Issues Closed/Opened (7d) | Dep Freshness |",
-        "|------|---------------------|--------------|---------------------------|---------------|",
-    ]
-    
-    for snap in reversed(snapshots):  # chronological
-        date = snap["date"]
-        m = snap["metrics"]
-        tc = m["test_coverage"]
-        test_str = f"{tc['test_files']}/{tc['source_files']} ({tc['test_to_source_ratio']:.2f})"
-        doc_str = f"{m['doc_coverage']['doc_coverage_percent']:.1f}%"
-        issues_str = f"{m['issues'].get('closed_last_7d','-')}/{m['issues'].get('opened_last_7d','-')}"
-        dep_str = f"{m['dependencies'].get('freshness_percent','?')}%"
-        lines.append(f"| {date} | {test_str} | {doc_str} | {issues_str} | {dep_str} |")
-    
-    # Current snapshot summary
-    cur = snapshots[-1]
-    cm = cur["metrics"]
-    lines.append(f"\n## Current Snapshot ({cur['date']})\n")
-    
-    tc = cm["test_coverage"]
-    cov_line = f"- Test coverage: {tc['coverage_percent']:.1f}% (via {tc['coverage_tool']})\n" if tc["coverage_percent"] else "- Test coverage: (pytest-cov not configured)\n"
-    lines.append(cov_line)
-    lines.append(f"- Doc coverage: {cm['doc_coverage']['doc_coverage_percent']:.1f}%")
-    
-    im = cm["issues"]
-    if im.get("close_rate") is not None:
-        lines.append(f"- Issue close rate (7d): {im['close_rate']*100:.1f}% ({im['closed_last_7d']} closed, {im['opened_last_7d']} opened)")
-    else:
-        lines.append(f"- Issue metrics: {im.get('note','unavailable')}")
-    
-    dd = cm["dependencies"]
-    lines.append(f"- Dep freshness: {dd.get('freshness_percent','?')}% outdated ({dd.get('outdated_deps',0)}/{dd.get('total_deps',0)} deps)")
-    if dd.get('outdated_list'):
-        lines.append(f"  Outdated: {', '.join(dd['outdated_list'][:5])}")
-    
-    content = "\n".join(lines) + "\n"
-    
-    if output_path:
-        output_path.parent.mkdir(parents=True, exist_ok=True)
-        output_path.write_text(content)
-    
-    return content
-
-
-# ── Main ─────────────────────────────────────────────────────────────────────
-
-def main() -> int:
-    parser = argparse.ArgumentParser(description="Progress Tracker — 10.8")
-    parser.add_argument("--json", action="store_true", help="Emit snapshot as JSON only")
-    parser.add_argument("--output", type=Path, default=METRICS_DIR / "TRENDS.md",
-                        help="Write trends markdown to this file")
-    args = parser.parse_args()
-    
-    snapshot = take_snapshot()
-    all_snapshots = load_snapshots()
-    path_written = save_snapshot(snapshot)
-    
-    if args.json:
-        print(json.dumps(snapshot, indent=2))
-        return 0
-    
-    trends = generate_trends(all_snapshots + [snapshot], output_path=args.output)
-    
-    # Print current snapshot summary
-    print(f"Snapshot saved: {path_written}\n")
-    print(f"Progress Tracker — {snapshot['date']}")
-    print("=" * 50)
-    
-    m = snapshot["metrics"]
-    tc = m["test_coverage"]
-    print(f"Test files: {tc['test_files']}  |  Source files: {tc['source_files']}  |  Ratio: {tc['test_to_source_ratio']:.3f}")
-    if tc["coverage_percent"] is not None:
-        print(f"Line coverage: {tc['coverage_percent']:.1f}%  (via {tc['coverage_tool']})")
-    else:
-        print("Line coverage: (not available — run `pytest --cov`)")
-    
-    print()
-    dc = m["doc_coverage"]
-    print(f"Callables with docstrings: {dc['callables_with_doc']}/{dc['callables_total']}  ({dc['doc_coverage_percent']:.1f}%)")
-    
-    print()
-    im = m["issues"]
-    if im.get("close_rate") is not None:
-        print(f"Issues (7d): {im['closed_last_7d']} closed / {im['opened_last_7d']} opened  → close rate: {im['close_rate']*100:.1f}%")
-        print(f"Total open: {im['total_open']}")
-    else:
-        print(f"Issues: {im.get('note','unavailable')}")
-    
-    print()
-    dd = m["dependencies"]
-    print(f"Dependencies: {dd.get('total_deps',0)} total, {dd.get('outdated_deps',0)} outdated")
-    if dd.get('outdated_list'):
-        shown = dd['outdated_list'][:5]
-        print(f"Outdated: {', '.join(shown)}" + ("..." if len(dd['outdated_list']) > 5 else ""))
-    
-    print(f"\nTrends written to: {args.output}")
-    return 0
-
-
-if __name__ == "__main__":
-    sys.exit(main())

scripts/review_comment_generator.py

@@ -0,0 +1,185 @@
+#!/usr/bin/env python3
+"""
+Review Comment Generator — Issue #126
+Reads JSONL findings, deduplicates, posts as Gitea PR comments.
+"""
+
+from __future__ import annotations
+
+import argparse
+import hashlib
+import json
+import os
+import sys
+import urllib.request
+import urllib.error
+from datetime import datetime, timezone
+from pathlib import Path
+from typing import Dict, List, Optional
+
+SCRIPT_DIR = Path(__file__).resolve().parent
+REPO_ROOT = SCRIPT_DIR.parent
+
+DEFAULT_API_BASE = os.environ.get(
+    "GITEA_API_BASE",
+    "https://forge.alexanderwhitestone.com"
+)
+TOKEN_PATHS = [
+    os.path.expanduser("~/.config/gitea/token"),
+    os.path.expanduser("~/.hermes/gitea.token"),
+    os.environ.get("GITEA_TOKEN", ""),
+]
+
+def load_token() -> Optional[str]:
+    token = os.environ.get("GITEA_TOKEN", "")
+    if token:
+        return token
+    for path in TOKEN_PATHS:
+        if path and os.path.exists(path):
+            with open(path) as f:
+                t = f.read().strip()
+                if t:
+                    return t
+    return None
+
+class GiteaClient:
+    def __init__(self, base_url: str, token: str, org: str, repo: str):
+        self.base_url = base_url.rstrip("/")
+        self.token = token
+        self.org = org
+        self.repo = repo
+
+    def _post(self, path: str, data: Dict) -> Optional[Dict]:
+        url = f"{self.base_url}/api/v1{path}"
+        body = json.dumps(data).encode("utf-8")
+        req = urllib.request.Request(url, data=body, method="POST")
+        req.add_header("Authorization", f"token {self.token}")
+        req.add_header("Content-Type", "application/json")
+        try:
+            with urllib.request.urlopen(req, timeout=30) as resp:
+                return json.loads(resp.read().decode())
+        except urllib.error.HTTPError as e:
+            err = e.read().decode() if e.read() else str(e)
+            print(f"[ERROR] HTTP {e.code}: {err}", file=sys.stderr)
+            return None
+        except Exception as e:
+            print(f"[ERROR] {e}", file=sys.stderr)
+            return None
+
+    def post_issue_comment(self, issue_num: int, body: str) -> Optional[Dict]:
+        return self._post(
+            f"/repos/{self.org}/{self.repo}/issues/{issue_num}/comments",
+            {"body": body}
+        )
+
+def content_hash(finding: Dict) -> str:
+    key = f"{finding['file']}:{finding['line']}:{finding['text']}"
+    return hashlib.sha256(key.encode("utf-8")).hexdigest()
+
+def format_comment(finding: Dict) -> str:
+    emoji = {
+        "error": "🛑",
+        "warning": "⚠️",
+        "info": "ℹ️",
+    }.get(finding.get("severity", ""), "📝")
+    f = finding["file"]
+    ln = finding["line"]
+    txt = finding["text"]
+    return f"{emoji} **Review Comment**\n\nFile: `{f}`\nLine: {ln}\n\n> {txt}\n"
+
+def load_findings(path: Optional[Path], from_stdin: bool) -> List[Dict]:
+    import fileinput
+    findings = []
+    sources = ["-"] if from_stdin else [str(path)]
+    for line in fileinput.input(files=sources):
+        line = line.strip()
+        if not line or line.startswith("#"):
+            continue
+        try:
+            f = json.loads(line)
+            for key in ("file", "line", "text"):
+                if key not in f:
+                    raise ValueError(f"Missing key: {key}")
+            findings.append(f)
+        except json.JSONDecodeError as e:
+            print(f"WARNING: Skipping invalid JSON: {e}", file=sys.stderr)
+    return findings
+
+def main() -> int:
+    parser = argparse.ArgumentParser(
+        description="Post review findings as comments to a Gitea PR/issue"
+    )
+    parser.add_argument("--pr", type=int, required=True, help="PR/issue number")
+    parser.add_argument("--org", default="Timmy_Foundation", help="Gitea org")
+    parser.add_argument("--repo", default="compounding-intelligence", help="Repo name")
+    parser.add_argument("--api-base", default=DEFAULT_API_BASE, help="Gitea API base")
+    parser.add_argument("--token", default=None, help="API token (or env/file)")
+    parser.add_argument("--input", type=Path, default=None, help="JSONL input file")
+    parser.add_argument("--stdin", action="store_true", help="Read from stdin")
+    parser.add_argument("--dry-run", action="store_true", help="Show without posting")
+    parser.add_argument("--json", action="store_true", help="Emit JSON report")
+
+    args = parser.parse_args()
+
+    if not args.stdin and args.input is None:
+        print("ERROR: --input or --stdin required", file=sys.stderr)
+        return 1
+    if args.stdin and args.input:
+        print("ERROR: --stdin and --input exclusive", file=sys.stderr)
+        return 1
+
+    token = args.token or load_token()
+    if not token:
+        print("ERROR: Token not found. Set GITEA_TOKEN or ~/.config/gitea/token", file=sys.stderr)
+        return 1
+
+    findings = load_findings(args.input, args.stdin)
+    if not findings:
+        print("ERROR: No findings loaded", file=sys.stderr)
+        return 1
+
+    if not args.json: print(f"Loaded {len(findings)} finding(s)")
+
+    seen: Dict[str, Dict] = {}
+    for f in findings:
+        h = content_hash(f)
+        if h not in seen:
+            seen[h] = f
+
+    unique = list(seen.values())
+    if not args.json: print(f"After dedup: {len(unique)} unique")
+
+    if args.json:
+        report = {
+            "total": len(findings),
+            "unique": len(unique),
+            "findings": unique,
+            "generated_at": datetime.now(timezone.utc).isoformat(),
+        }
+        print(json.dumps(report, indent=2))
+        return 0
+
+    if args.dry_run:
+        print("\n=== DRY RUN — would post ===")
+        for i, f in enumerate(unique, 1):
+            print(f"\n--- Comment {i}/{len(unique)} ---")
+            print(format_comment(f))
+        return 0
+
+    client = GiteaClient(args.api_base, token, args.org, args.repo)
+    posted = 0
+    for f in unique:
+        body = format_comment(f)
+        result = client.post_issue_comment(args.pr, body)
+        if result:
+            print(f"✅ Posted: {f['file']}:{f['line']} (id={result.get('id')})")
+            posted += 1
+        else:
+            print(f"❌ Failed: {f['file']}:{f['line']}")
+
+    print(f"\nPosted {posted}/{len(unique)} to PR #{args.pr}")
+    return 0 if posted == len(unique) else 1
+
+if __name__ == "__main__":
+    sys.exit(main())
+

scripts/sample_findings.jsonl

@@ -0,0 +1,5 @@
+{"file": "scripts/harvester.py", "line": 47, "text": "Consider adding type hints to improve readability", "severity": "info"}
+{"file": "scripts/dedup.py", "line": 89, "text": "Add null check before accessing fact['confidence'] to avoid KeyError", "severity": "warning"}
+{"file": "scripts/bootstrapper.py", "line": 102, "text": "This loop is O(n^2) — could be optimized with a dict lookup", "severity": "info"}
+{"file": "scripts/harvester.py", "line": 47, "text": "Consider adding type hints to improve readability", "severity": "info"}
+{"file": "scripts/harvester.py", "line": 120, "text": "File handle not closed in error path — use context manager", "severity": "error"}

tests/test_review_comment_generator.py

@@ -0,0 +1,234 @@
+#!/usr/bin/env python3
+"""
+Smoke tests for Review Comment Generator — Issue #126
+"""
+
+from __future__ import annotations
+
+import json
+import subprocess
+import sys
+import hashlib
+from io import StringIO
+from pathlib import Path
+
+import pytest
+
+REPO_ROOT = Path(__file__).resolve().parents[1]
+SCRIPTS_DIR = REPO_ROOT / "scripts"
+GENERATOR = SCRIPTS_DIR / "review_comment_generator.py"
+SAMPLE_FINDINGS = SCRIPTS_DIR / "sample_findings.jsonl"
+
+
+class TestGeneratorPresence:
+    def test_script_exists(self):
+        assert GENERATOR.exists(), f"Missing: {GENERATOR}"
+
+    def test_shebang_is_python(self):
+        with open(GENERATOR) as f:
+            first = f.readline().strip()
+        assert first.startswith("#!"), "No shebang"
+        assert "python" in first.lower()
+
+
+class TestDeduplication:
+    def test_content_hash_deterministic(self):
+        from hashlib import sha256
+        def ch(f):
+            key = f"{f['file']}:{f['line']}:{f['text']}"
+            return sha256(key.encode()).hexdigest()
+        finding = {"file": "a.py", "line": 1, "text": "test"}
+        assert ch(finding) == ch(finding)
+
+    def test_duplicate_findings_are_removed(self):
+        findings = [
+            {"file": "a.py", "line": 1, "text": "foo", "severity": "info"},
+            {"file": "a.py", "line": 1, "text": "foo", "severity": "warning"},
+            {"file": "b.py", "line": 2, "text": "bar", "severity": "info"},
+        ]
+        seen = {}
+        for f in findings:
+            key = f"{f['file']}:{f['line']}:{f['text']}"
+            seen[key] = f
+        assert len(seen) == 2
+
+    def test_different_findings_are_kept(self):
+        findings = [
+            {"file": "a.py", "line": 1, "text": "foo"},
+            {"file": "a.py", "line": 2, "text": "foo"},
+            {"file": "a.py", "line": 1, "text": "bar"},
+        ]
+        seen = {}
+        for f in findings:
+            key = f"{f['file']}:{f['line']}:{f['text']}"
+            seen[key] = f
+        assert len(seen) == 3
+
+
+class TestCommentFormatting:
+    def test_format_basic(self):
+        sys.path.insert(0, str(SCRIPTS_DIR))
+        from review_comment_generator import format_comment
+        f = {"file": "scripts/foo.py", "line": 10, "text": "Fix this bug", "severity": "warning"}
+        body = format_comment(f)
+        assert "📝 **Review Comment**" not in body  # warning uses ⚠️
+        assert "⚠️ **Review Comment**" in body
+        assert "`scripts/foo.py`" in body
+        assert "Line: 10" in body
+        assert "> Fix this bug" in body
+
+    def test_format_severity_emoji(self):
+        sys.path.insert(0, str(SCRIPTS_DIR))
+        from review_comment_generator import format_comment
+        cases = [("error", "🛑"), ("warning", "⚠️"), ("info", "ℹ️"), ("unknown", "📝")]
+        for severity, emoji in cases:
+            f = {"file": "x.py", "line": 1, "text": "test", "severity": severity}
+            assert emoji in format_comment(f)
+
+
+class TestFindingsLoader:
+    def test_load_from_file(self):
+        sys.path.insert(0, str(SCRIPTS_DIR))
+        from review_comment_generator import load_findings
+        findings = load_findings(SAMPLE_FINDINGS, from_stdin=False)
+        assert len(findings) >= 4
+
+    def test_load_ignores_blank_and_comments(self):
+        import tempfile, os
+        with tempfile.NamedTemporaryFile(mode="w", suffix=".jsonl", delete=False) as tf:
+            tf.write('{"file":"a.py","line":1,"text":"valid"}\n')
+            tf.write('\n')
+            tf.write('# this is a comment\n')
+            tf.write('{"file":"b.py","line":2,"text":"also valid"}\n')
+            tfname = tf.name
+        try:
+            sys.path.insert(0, str(SCRIPTS_DIR))
+            from review_comment_generator import load_findings
+            assert len(load_findings(Path(tfname), from_stdin=False)) == 2
+        finally:
+            os.unlink(tfname)
+
+    def test_invalid_json_line_skipped(self, capsys):
+        import tempfile, os
+        with tempfile.NamedTemporaryFile(mode="w", suffix=".jsonl", delete=False) as tf:
+            tf.write('invalid json\n')
+            tf.write('{"file":"ok.py","line":1,"text":"valid"}\n')
+            tfname = tf.name
+        try:
+            sys.path.insert(0, str(SCRIPTS_DIR))
+            from review_comment_generator import load_findings
+            assert len(load_findings(Path(tfname), from_stdin=False)) == 1
+        finally:
+            os.unlink(tfname)
+
+
+class TestDryRunMode:
+    def test_dry_run_counts_unique(self):
+        result = subprocess.run(
+            [sys.executable, str(GENERATOR), "--pr", "126",
+             "--input", str(SAMPLE_FINDINGS), "--dry-run"],
+            capture_output=True, text=True, cwd=REPO_ROOT, timeout=15
+        )
+        assert result.returncode == 0
+        assert "DRY RUN" in result.stdout
+        assert "Review Comment" in result.stdout
+
+    def test_dry_run_shows_all_unique(self):
+        result = subprocess.run(
+            [sys.executable, str(GENERATOR), "--pr", "126",
+             "--input", str(SAMPLE_FINDINGS), "--dry-run"],
+            capture_output=True, text=True, cwd=REPO_ROOT, timeout=15
+        )
+        assert result.stdout.count("--- Comment") == 4
+
+
+class TestJSONOutputMode:
+    def test_json_flag_emits_valid_json(self):
+        result = subprocess.run(
+            [sys.executable, str(GENERATOR), "--pr", "126",
+             "--input", str(SAMPLE_FINDINGS), "--json"],
+            capture_output=True, text=True, cwd=REPO_ROOT, timeout=15
+        )
+        assert result.returncode == 0
+        payload = json.loads(result.stdout)
+        assert "total" in payload and "unique" in payload and "findings" in payload
+        assert payload["total"] >= payload["unique"]
+
+    def test_json_findings_have_required_fields(self):
+        result = subprocess.run(
+            [sys.executable, str(GENERATOR), "--pr", "126",
+             "--input", str(SAMPLE_FINDINGS), "--json"],
+            capture_output=True, text=True, cwd=REPO_ROOT, timeout=15
+        )
+        payload = json.loads(result.stdout)
+        for f in payload["findings"]:
+            assert "file" in f and "line" in f and "text" in f
+
+
+class TestGiteaClient:
+    def test_post_issue_comment_builds_correct_url(self):
+        sys.path.insert(0, str(SCRIPTS_DIR))
+        from review_comment_generator import GiteaClient
+        client = GiteaClient("https://example.com", "token123", "MyOrg", "myrepo")
+        assert client.org == "MyOrg" and client.repo == "myrepo"
+
+    def test_generate_comment_body_has_required_fields(self):
+        sys.path.insert(0, str(SCRIPTS_DIR))
+        from review_comment_generator import format_comment
+        f = {"file": "x.py", "line": 5, "text": "Fix this", "severity": "error"}
+        body = format_comment(f)
+        assert "x.py" in body and "5" in body and "Fix this" in body
+
+
+class TestFullPipeline:
+    def test_end_to_end_json_output(self):
+        result = subprocess.run(
+            [sys.executable, str(GENERATOR), "--pr", "126",
+             "--input", str(SAMPLE_FINDINGS), "--json"],
+            capture_output=True, text=True, cwd=REPO_ROOT, timeout=15
+        )
+        assert result.returncode == 0
+        data = json.loads(result.stdout)
+        assert data["total"] == 5
+        assert data["unique"] == 4
+        f = data["findings"][0]
+        for key in ("file", "line", "text", "severity"):
+            assert key in f
+
+    def test_token_loading_fallback(self):
+        sys.path.insert(0, str(SCRIPTS_DIR))
+        from review_comment_generator import load_token
+        token = load_token()
+        assert token is None or isinstance(token, str)
+
+
+class TestErrorHandling:
+    def test_missing_input_shows_error(self):
+        result = subprocess.run(
+            [sys.executable, str(GENERATOR), "--pr", "126"],
+            capture_output=True, text=True, cwd=REPO_ROOT, timeout=15
+        )
+        assert result.returncode != 0
+        assert "--input" in result.stderr or "--stdin" in result.stderr
+
+    def test_invalid_json_line_skipped(self):
+        import tempfile, os
+        with tempfile.NamedTemporaryFile(mode="w", suffix=".jsonl", delete=False) as tf:
+            tf.write('invalid json\n')
+            tf.write('{"file":"ok.py","line":1,"text":"valid"}\n')
+            tfname = tf.name
+        try:
+            result = subprocess.run(
+                [sys.executable, str(GENERATOR), "--pr", "126",
+                 "--input", tfname, "--json"],
+                capture_output=True, text=True, cwd=REPO_ROOT, timeout=15
+            )
+            data = json.loads(result.stdout)
+            assert data["total"] == 1
+            assert data["unique"] == 1
+        finally:
+            os.unlink(tfname)
+
+
+if __name__ == "__main__":
+    pytest.main([__file__, "-v"])