feat: add codebase genome diff script for structural change detection

Introduces genome_diff.py — a tool for detecting structural changes between
two git refs: file-level changes, function/class signature modifications,
and dependency import changes.

Addresses #132.

scripts/genome_diff.py

@@ -0,0 +1,288 @@
+#!/usr/bin/env python3
+"""
+Codebase Genome Diff — Detect structural changes between two versions.
+
+Compares two git refs (commits, branches, tags) and produces a human-readable
+report of structural changes:
+  • Added/removed/renamed files
+  • Changed functions/classes (signature modifications)
+  • New dependencies (imports, requirements, etc.)
+
+Usage:
+    python3 scripts/genome_diff.py --ref1 <commit1> --ref2 <commit2>
+    python3 scripts/genome_diff.py --ref1 main --ref2 feature-branch
+    python3 scripts/genome_diff.py --ref1 v1.0 --ref2 v2.0 --output report.txt
+"""
+
+import argparse
+import json
+import os
+import re
+import subprocess
+import sys
+from dataclasses import dataclass, field
+from typing import List, Dict, Any, Optional
+
+SCRIPT_DIR = os.path.dirname(os.path.abspath(__file__))
+sys.path.insert(0, SCRIPT_DIR)
+from diff_analyzer import DiffAnalyzer, ChangeCategory
+
+
+@dataclass
+class FunctionChange:
+    file: str
+    name: str
+    kind: str  # 'function' or 'class'
+    change_type: str  # 'added' or 'removed' (simplified)
+    old_line: Optional[int] = None
+    new_line: Optional[int] = None
+
+
+@dataclass
+class DependencyChange:
+    file: str
+    module: str
+    change_type: str  # 'added' or 'removed' or 'modified'
+    line: int = 0
+
+
+@dataclass
+class GenomeDiffReport:
+    ref1: str
+    ref2: str
+    file_changes: List[Dict[str, Any]] = field(default_factory=list)
+    function_changes: List[FunctionChange] = field(default_factory=list)
+    dependency_changes: List[DependencyChange] = field(default_factory=list)
+    total_files_changed: int = 0
+    total_functions_changed: int = 0
+    total_dependencies_changed: int = 0
+
+    def to_dict(self) -> Dict[str, Any]:
+        return {
+            "ref1": self.ref1,
+            "ref2": self.ref2,
+            "summary": {
+                "files": self.total_files_changed,
+                "functions": self.total_functions_changed,
+                "dependencies": self.total_dependencies_changed,
+            },
+            "file_changes": self.file_changes,
+            "function_changes": [fc.__dict__ for fc in self.function_changes],
+            "dependency_changes": [dc.__dict__ for dc in self.dependency_changes],
+        }
+
+    def human_report(self) -> str:
+        lines = []
+        lines.append(f"Codebase Genome Diff: {self.ref1} → {self.ref2}")
+        lines.append("=" * 60)
+        lines.append(f"  Files changed: {self.total_files_changed}")
+        lines.append(f"  Functions changed: {self.total_functions_changed}")
+        lines.append(f"  Dependencies changed: {self.total_dependencies_changed}")
+        lines.append("")
+
+        for fc in self.file_changes:
+            kind = []
+            if fc.get('is_new'):
+                kind.append("NEW")
+            if fc.get('is_deleted'):
+                kind.append("DELETED")
+            if fc.get('is_renamed'):
+                kind.append("RENAMED")
+            if fc.get('is_binary'):
+                kind.append("BINARY")
+            kind_str = f" [{', '.join(kind)}]" if kind else ""
+            lines.append(f"  {fc['path']}{kind_str}  (+{fc['added_lines']}/-{fc['deleted_lines']})")
+        lines.append("")
+
+        for fc in self.function_changes:
+            op = {'added': '+', 'removed': '-', 'modified': '~'}.get(fc.change_type, '?')
+            lines.append(f"  [{op}] {fc.file}: {fc.kind} '{fc.name}'")
+        lines.append("")
+
+        for dc in self.dependency_changes:
+            op = '+' if dc.change_type == 'added' else '-'
+            lines.append(f"  [{op}] {dc.file}: {dc.module}")
+        lines.append("")
+
+        return "\n".join(lines)
+
+
+def run_git_diff(ref1: str, ref2: str) -> str:
+    result = subprocess.run(
+        ['git', 'diff', '--unified=0', f'{ref1}...{ref2}'],
+        capture_output=True, text=True, cwd=SCRIPT_DIR
+    )
+    if result.returncode not in (0, 1):
+        print(f"git diff failed: {result.stderr}", file=sys.stderr)
+        sys.exit(1)
+    return result.stdout
+
+
+def extract_function_changes(diff_text: str) -> List[FunctionChange]:
+    changes: List[FunctionChange] = []
+    pattern = re.compile(r'^([+\-])\s*(def|class)\s+(\w+)', re.MULTILINE)
+    hunk_header_re = re.compile(r'^@@\s+-(\d+)(?:,(\d+))?\s+\+(\d+)(?:,(\d+))?\s+@@')
+    current_old_line: Optional[int] = None
+    current_new_line: Optional[int] = None
+
+    for line in diff_text.split('\n'):
+        hdr = hunk_header_re.match(line)
+        if hdr:
+            current_old_line = int(hdr.group(1))
+            current_new_line = int(hdr.group(3))
+            continue
+        m = pattern.match(line)
+        if m:
+            op = m.group(1)
+            kind = m.group(2)
+            name = m.group(3)
+            change_type = "added" if op == '+' else "removed"
+            line_num = current_new_line if change_type == "added" else current_old_line
+            changes.append(FunctionChange(
+                file="<unknown>",
+                name=name,
+                kind=kind,
+                change_type=change_type,
+                new_line=line_num if change_type == "added" else None,
+                old_line=line_num if change_type == "removed" else None,
+            ))
+            # Advance line counters heuristically
+            if op == '-':
+                if current_old_line is not None:
+                    current_old_line += 1
+            elif op == '+':
+                if current_new_line is not None:
+                    current_new_line += 1
+        elif line.startswith(' '):
+            if current_old_line is not None:
+                current_old_line += 1
+            if current_new_line is not None:
+                current_new_line += 1
+        # lines starting with other prefixes (like \\ No newline) ignored
+    return changes
+
+
+def extract_dependency_changes(diff_text: str, analyzer: DiffAnalyzer) -> List[DependencyChange]:
+    changes: List[DependencyChange] = []
+    import_pattern = re.compile(
+        r'^([+\-])\s*(?:import\s+([\w\.]+)|from\s+([\w\.]+)\s+import)',
+        re.MULTILINE
+    )
+    file_diffs = analyzer._split_files(diff_text)
+    for file_diff in file_diffs:
+        file_match = re.search(r'^diff --git a/.*? b/(.*?)$', file_diff, re.MULTILINE)
+        if not file_match:
+            continue
+        filepath = file_match.group(1)
+
+        # Scan each line for import changes
+        for line in file_diff.split('\n'):
+            m = import_pattern.match(line)
+            if m:
+                change_type = "added" if m.group(1) == '+' else "removed"
+                module = m.group(2) or m.group(3)
+                changes.append(DependencyChange(
+                    file=filepath,
+                    module=module,
+                    change_type=change_type,
+                    line=0
+                ))
+
+        # Detect if this file is a dependency manifest
+        req_file_pattern = re.compile(
+            r'^[\+\-].*?(requirements(.*?)\.txt|pyproject\.toml|setup\.py|Pipfile)'
+        )
+        if any(req_file_pattern.match(line) for line in file_diff.split('\n')):
+            if not any(c.file == filepath and c.module == "<file>" for c in changes):
+                changes.append(DependencyChange(
+                    file=filepath,
+                    module="<file>",
+                    change_type="modified",
+                    line=0
+                ))
+    return changes
+
+
+def correlate_function_changes_with_files(diff_text: str, functions: List[FunctionChange]) -> List[FunctionChange]:
+    result: List[FunctionChange] = []
+    # Split diff into per-file sections
+    file_sections: List[tuple[str, str]] = []
+    current_file: Optional[str] = None
+    current_lines: List[str] = []
+    for line in diff_text.split('\n'):
+        if line.startswith('diff --git'):
+            if current_file is not None:
+                file_sections.append((current_file, '\n'.join(current_lines)))
+            m = re.match(r'^diff --git a/.*? b/(.*?)$', line)
+            current_file = m.group(1) if m else "unknown"
+            current_lines = [line]
+        else:
+            current_lines.append(line)
+    if current_file is not None:
+        file_sections.append((current_file, '\n'.join(current_lines)))
+
+    pattern = re.compile(r'^([+\-])\s*(def|class)\s+(\w+)', re.MULTILINE)
+    for filepath, section in file_sections:
+        for m in pattern.finditer(section):
+            op = m.group(1)
+            kind = m.group(2)
+            name = m.group(3)
+            change_type = "added" if op == '+' else "removed"
+            result.append(FunctionChange(
+                file=filepath,
+                name=name,
+                kind=kind,
+                change_type=change_type
+            ))
+    return result
+
+
+def main():
+    parser = argparse.ArgumentParser(description="Codebase Genome Diff — structural changes between versions")
+    parser.add_argument("--ref1", required=True, help="First git ref (commit, branch, tag)")
+    parser.add_argument("--ref2", required=True, help="Second git ref")
+    parser.add_argument("--output", help="Write report to file")
+    parser.add_argument("--json", action="store_true", help="Output JSON instead of human report")
+    args = parser.parse_args()
+
+    try:
+        diff_text = run_git_diff(args.ref1, args.ref2)
+    except Exception as e:
+        print(f"Error: {e}", file=sys.stderr)
+        sys.exit(1)
+
+    if not diff_text.strip():
+        print(f"No differences between {args.ref1} and {args.ref2}.")
+        sys.exit(0)
+
+    analyzer = DiffAnalyzer()
+    summary = analyzer.analyze(diff_text)
+
+    file_changes = [fc.to_dict() for fc in summary.files]
+    func_changes = extract_function_changes(diff_text)
+    func_changes = correlate_function_changes_with_files(diff_text, func_changes)
+    dep_changes = extract_dependency_changes(diff_text, analyzer)
+
+    report = GenomeDiffReport(
+        ref1=args.ref1,
+        ref2=args.ref2,
+        file_changes=file_changes,
+        function_changes=func_changes,
+        dependency_changes=dep_changes,
+        total_files_changed=len(file_changes),
+        total_functions_changed=len(func_changes),
+        total_dependencies_changed=len(dep_changes),
+    )
+
+    output = json.dumps(report.to_dict(), indent=2) if args.json else report.human_report()
+
+    if args.output:
+        with open(args.output, 'w') as f:
+            f.write(output + '\n')
+        print(f"Report written to {args.output}")
+    else:
+        print(output)
+
+
+if __name__ == '__main__':
+    main()

scripts/review_comment_generator.py

@@ -1,185 +0,0 @@
-#!/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

@@ -1,5 +0,0 @@
-{"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

@@ -1,234 +0,0 @@
-#!/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"])