feat: add diff analyzer for PR change categorization (closes #118)

scripts/dependency_graph.py

@@ -1,249 +0,0 @@
-#!/usr/bin/env python3
-"""
-Cross-Repo Dependency Graph Builder
-
-Scans repos for import/require/reference patterns and builds a directed
-dependency graph. Detects circular dependencies. Outputs DOT and Mermaid.
-
-Usage:
-  python3 scripts/dependency_graph.py /path/to/repos/
-  python3 scripts/dependency_graph.py --repos repo1,repo2,repo3 --format mermaid
-  python3 scripts/dependency_graph.py --repos-dir /path/to/ --format dot --output deps.dot
-
-Patterns detected:
-  - Python: import X, from X import Y
-  - JavaScript: require("X"), import ... from "X"
-  - Go: import "X"
-  - Ansible: include_role, import_role
-  - Docker/Compose: image: X, depends_on
-  - Config references: repo-name in YAML/TOML/JSON
-"""
-
-import argparse
-import json
-import os
-import re
-import sys
-from collections import defaultdict
-from pathlib import Path
-
-
-# Known repo names for matching
-KNOWN_REPOS = [
-    "hermes-agent", "timmy-config", "timmy-home", "the-nexus", "the-door",
-    "the-beacon", "fleet-ops", "burn-fleet", "timmy-dispatch", "turboquant",
-    "compounding-intelligence", "the-playground", "second-son-of-timmy",
-    "ai-safety-review", "the-echo-pattern", "timmy-academy", "wolf",
-    "the-testament",
-]
-
-
-def normalize_repo_name(name: str) -> str:
-    """Normalize a repo name for comparison."""
-    return name.lower().replace("_", "-").replace(".git", "").strip()
-
-
-def scan_file_for_deps(filepath: str, content: str, own_repo: str) -> set:
-    """Scan a file's content for references to other repos."""
-    deps = set()
-    own_norm = normalize_repo_name(own_repo)
-
-    for repo in KNOWN_REPOS:
-        repo_norm = normalize_repo_name(repo)
-        if repo_norm == own_norm:
-            continue
-
-        # Direct name references
-        patterns = [
-            repo,  # exact name
-            repo.replace("-", "_"),  # underscore variant
-            repo.replace("-", ""),  # no separator
-            f"/{repo}/",  # path reference
-            f'"{repo}"',  # quoted
-            f"'{repo}'",  # single quoted
-            f"Timmy_Foundation/{repo}",  # full Gitea path
-            f"Timmy_Foundation.{repo}",  # Python module path
-        ]
-
-        for pattern in patterns:
-            if pattern in content:
-                deps.add(repo)
-                break
-
-    return deps
-
-
-def scan_repo(repo_path: str, repo_name: str = None) -> dict:
-    """Scan a repo directory for dependencies."""
-    path = Path(repo_path)
-    if not path.is_dir():
-        return {"error": f"Not a directory: {repo_path}"}
-
-    if not repo_name:
-        repo_name = path.name
-
-    deps = set()
-    files_scanned = 0
-    exts = {".py", ".js", ".ts", ".go", ".yaml", ".yml", ".toml", ".json",
-            ".md", ".sh", ".bash", ".Dockerfile", ".tf", ".hcl"}
-
-    for fpath in path.rglob("*"):
-        if not fpath.is_file():
-            continue
-        if fpath.suffix not in exts:
-            continue
-        # Skip common non-source dirs
-        parts = fpath.parts
-        if any(p in (".git", "node_modules", "__pycache__", ".venv", "venv",
-                      "vendor", "dist", "build", ".tox") for p in parts):
-            continue
-
-        try:
-            content = fpath.read_text(errors="ignore")
-        except:
-            continue
-
-        file_deps = scan_file_for_deps(str(fpath), content, repo_name)
-        deps.update(file_deps)
-        files_scanned += 1
-
-    return {
-        "repo": repo_name,
-        "dependencies": sorted(deps),
-        "files_scanned": files_scanned,
-    }
-
-
-def detect_cycles(graph: dict) -> list:
-    """Detect circular dependencies using DFS."""
-    cycles = []
-    visited = set()
-    rec_stack = set()
-
-    def dfs(node, path):
-        visited.add(node)
-        rec_stack.add(node)
-
-        for neighbor in graph.get(node, {}).get("dependencies", []):
-            if neighbor not in visited:
-                result = dfs(neighbor, path + [neighbor])
-                if result:
-                    return result
-            elif neighbor in rec_stack:
-                cycle_start = path.index(neighbor)
-                return path[cycle_start:] + [neighbor]
-
-        rec_stack.remove(node)
-        return None
-
-    for node in graph:
-        if node not in visited:
-            cycle = dfs(node, [node])
-            if cycle:
-                cycles.append(cycle)
-
-    return cycles
-
-
-def to_dot(graph: dict) -> str:
-    """Generate DOT format output."""
-    lines = ["digraph dependencies {"]
-    lines.append("  rankdir=LR;")
-    lines.append("  node [shape=box, style=filled, fillcolor="#1a1a2e", fontcolor="#e6edf3"];")
-    lines.append("  edge [color="#4a4a6a"];")
-    lines.append("")
-
-    for repo, data in sorted(graph.items()):
-        dep_count = len(data.get("dependencies", []))
-        fill = "#2d1b69" if dep_count > 2 else "#16213e"
-        lines.append(f'  "{repo}" [fillcolor="{fill}"];')
-        for dep in data.get("dependencies", []):
-            lines.append(f'  "{repo}" -> "{dep}";')
-
-    lines.append("}")
-    return "\n".join(lines)
-
-
-def to_mermaid(graph: dict) -> str:
-    """Generate Mermaid format output."""
-    lines = ["graph LR"]
-
-    for repo, data in sorted(graph.items()):
-        for dep in data.get("dependencies", []):
-            lines.append(f"    {repo.replace('-','_')} --> {dep.replace('-','_')}")
-
-    # Add node labels
-    lines.append("")
-    for repo in sorted(graph.keys()):
-        lines.append(f"    {repo.replace('-','_')}[{repo}]")
-
-    return "\n".join(lines)
-
-
-def main():
-    parser = argparse.ArgumentParser(description="Build cross-repo dependency graph")
-    parser.add_argument("repos_dir", nargs="?", help="Directory containing repos")
-    parser.add_argument("--repos", help="Comma-separated list of repo paths")
-    parser.add_argument("--format", choices=["dot", "mermaid", "json"], default="json")
-    parser.add_argument("--output", "-o", help="Output file (default: stdout)")
-    parser.add_argument("--cycles-only", action="store_true", help="Only report cycles")
-    args = parser.parse_args()
-
-    results = {}
-    repo_paths = []
-
-    if args.repos:
-        repo_paths = [p.strip() for p in args.repos.split(",")]
-    elif args.repos_dir:
-        base = Path(args.repos_dir)
-        repo_paths = [str(p) for p in base.iterdir() if p.is_dir() and not p.name.startswith(".")]
-    else:
-        parser.print_help()
-        sys.exit(1)
-
-    for rpath in repo_paths:
-        name = Path(rpath).name
-        print(f"Scanning {name}...", file=sys.stderr)
-        result = scan_repo(rpath, name)
-        if "error" not in result:
-            results[name] = result
-
-    # Detect cycles
-    cycles = detect_cycles(results)
-
-    if args.cycles_only:
-        if cycles:
-            print("CIRCULAR DEPENDENCIES DETECTED:")
-            for cycle in cycles:
-                print(f"  {' -> '.join(cycle)}")
-            sys.exit(1)
-        else:
-            print("No circular dependencies found.")
-            sys.exit(0)
-
-    # Output
-    output = {}
-    if args.format == "dot":
-        output = to_dot(results)
-    elif args.format == "mermaid":
-        output = to_mermaid(results)
-    else:
-        output = json.dumps({
-            "repos": results,
-            "cycles": cycles,
-            "summary": {
-                "total_repos": len(results),
-                "total_deps": sum(len(r["dependencies"]) for r in results.values()),
-                "cycles_found": len(cycles),
-            }
-        }, indent=2)
-
-    if args.output:
-        Path(args.output).write_text(output)
-        print(f"Written to {args.output}", file=sys.stderr)
-    else:
-        print(output)
-
-
-if __name__ == "__main__":
-    main()

scripts/diff_analyzer.py

@@ -0,0 +1,239 @@
+"""
+Diff Analyzer — Pipeline 6.1
+
+Reads PR diffs and categorizes changes: new code, deleted code, modified code, moved code.
+Produces a change summary with line counts per category.
+
+Usage:
+    from diff_analyzer import DiffAnalyzer
+    analyzer = DiffAnalyzer()
+    summary = analyzer.analyze(diff_text)
+"""
+
+from __future__ import annotations
+
+import re
+from dataclasses import dataclass, field
+from enum import Enum
+from typing import List, Optional
+
+
+class ChangeCategory(Enum):
+    """Categories of code changes in a diff hunk."""
+    ADDED = "added"
+    DELETED = "deleted"
+    MODIFIED = "modified"
+    MOVED = "moved"
+    CONTEXT = "context"
+
+
+@dataclass
+class Hunk:
+    """A single diff hunk with metadata."""
+    header: str
+    old_start: int
+    old_count: int
+    new_start: int
+    new_count: int
+    lines: List[str] = field(default_factory=list)
+    category: ChangeCategory = ChangeCategory.CONTEXT
+    old_lines: int = 0
+    new_lines: int = 0
+
+
+@dataclass
+class FileChange:
+    """Changes within a single file."""
+    path: str
+    old_path: Optional[str] = None  # For renames
+    is_new: bool = False
+    is_deleted: bool = False
+    is_renamed: bool = False
+    hunks: List[Hunk] = field(default_factory=list)
+    added_lines: int = 0
+    deleted_lines: int = 0
+    context_lines: int = 0
+
+
+@dataclass
+class ChangeSummary:
+    """Summary of all changes in a diff."""
+    files_changed: int = 0
+    files_added: int = 0
+    files_deleted: int = 0
+    files_renamed: int = 0
+    files_modified: int = 0
+    total_added: int = 0
+    total_deleted: int = 0
+    total_context: int = 0
+    hunks_added: int = 0
+    hunks_deleted: int = 0
+    hunks_modified: int = 0
+    hunks_moved: int = 0
+    file_changes: List[FileChange] = field(default_factory=list)
+
+    def to_dict(self) -> dict:
+        """Serialize to dict for JSON output."""
+        return {
+            "files_changed": self.files_changed,
+            "files_added": self.files_added,
+            "files_deleted": self.files_deleted,
+            "files_renamed": self.files_renamed,
+            "files_modified": self.files_modified,
+            "total_added": self.total_added,
+            "total_deleted": self.total_deleted,
+            "total_context": self.total_context,
+            "hunks_added": self.hunks_added,
+            "hunks_deleted": self.hunks_deleted,
+            "hunks_modified": self.hunks_modified,
+            "hunks_moved": self.hunks_moved,
+            "files": [
+                {
+                    "path": fc.path,
+                    "old_path": fc.old_path,
+                    "is_new": fc.is_new,
+                    "is_deleted": fc.is_deleted,
+                    "is_renamed": fc.is_renamed,
+                    "added": fc.added_lines,
+                    "deleted": fc.deleted_lines,
+                    "context": fc.context_lines,
+                }
+                for fc in self.file_changes
+            ],
+        }
+
+
+# Regex for unified diff headers
+_HUNK_RE = re.compile(
+    r"^@@\s+-(\d+)(?:,(\d+))?\s+\+(\d+)(?:,(\d+))?\s+@@(.*)$"
+)
+_FILE_HEADER_RE = re.compile(r"^diff --git a/(.*) b/(.*)$")
+_RENAME_RE = re.compile(r"^rename from (.+)$|^rename to (.+)$")
+
+
+class DiffAnalyzer:
+    """Parses unified diffs and categorizes changes."""
+
+    def analyze(self, diff_text: str) -> ChangeSummary:
+        """Analyze a unified diff string and return a ChangeSummary."""
+        summary = ChangeSummary()
+        lines = diff_text.splitlines(keepends=False)
+
+        current_file: Optional[FileChange] = None
+        current_hunk: Optional[Hunk] = None
+        old_path: Optional[str] = None
+        new_path: Optional[str] = None
+
+        for line in lines:
+            # File header
+            m = _FILE_HEADER_RE.match(line)
+            if m:
+                # Save previous file
+                if current_file:
+                    self._classify_file(current_file)
+                    summary.file_changes.append(current_file)
+
+                old_path = m.group(1)
+                new_path = m.group(2)
+                current_file = FileChange(path=new_path, old_path=old_path)
+                current_hunk = None
+                continue
+
+            if current_file is None:
+                continue
+
+            # Detect new/deleted file markers
+            if line.startswith("new file mode"):
+                current_file.is_new = True
+                continue
+            if line.startswith("deleted file mode"):
+                current_file.is_deleted = True
+                continue
+
+            # Detect renames
+            rm = _RENAME_RE.match(line)
+            if rm:
+                current_file.is_renamed = True
+                continue
+
+            # Hunk header
+            hm = _HUNK_RE.match(line)
+            if hm:
+                if current_hunk:
+                    self._classify_hunk(current_hunk)
+                    current_file.hunks.append(current_hunk)
+
+                current_hunk = Hunk(
+                    header=line,
+                    old_start=int(hm.group(1)),
+                    old_count=int(hm.group(2) or 1),
+                    new_start=int(hm.group(3)),
+                    new_count=int(hm.group(4) or 1),
+                )
+                continue
+
+            if current_hunk is None:
+                continue
+
+            # Hunk content
+            current_hunk.lines.append(line)
+
+            if line.startswith("+"):
+                current_hunk.new_lines += 1
+                current_file.added_lines += 1
+            elif line.startswith("-"):
+                current_hunk.old_lines += 1
+                current_file.deleted_lines += 1
+            elif line.startswith(" "):
+                current_file.context_lines += 1
+
+        # Finalize last hunk and file
+        if current_hunk:
+            self._classify_hunk(current_hunk)
+            if current_file:
+                current_file.hunks.append(current_hunk)
+        if current_file:
+            self._classify_file(current_file)
+            summary.file_changes.append(current_file)
+
+        # Aggregate
+        summary.files_changed = len(summary.file_changes)
+        for fc in summary.file_changes:
+            summary.total_added += fc.added_lines
+            summary.total_deleted += fc.deleted_lines
+            summary.total_context += fc.context_lines
+            if fc.is_new:
+                summary.files_added += 1
+            elif fc.is_deleted:
+                summary.files_deleted += 1
+            elif fc.is_renamed:
+                summary.files_renamed += 1
+            else:
+                summary.files_modified += 1
+
+            for h in fc.hunks:
+                if h.category == ChangeCategory.ADDED:
+                    summary.hunks_added += 1
+                elif h.category == ChangeCategory.DELETED:
+                    summary.hunks_deleted += 1
+                elif h.category == ChangeCategory.MODIFIED:
+                    summary.hunks_modified += 1
+                elif h.category == ChangeCategory.MOVED:
+                    summary.hunks_moved += 1
+
+        return summary
+
+    def _classify_hunk(self, hunk: Hunk) -> None:
+        """Classify a hunk based on its add/delete ratio."""
+        if hunk.new_lines > 0 and hunk.old_lines == 0:
+            hunk.category = ChangeCategory.ADDED
+        elif hunk.old_lines > 0 and hunk.new_lines == 0:
+            hunk.category = ChangeCategory.DELETED
+        elif hunk.new_lines > 0 and hunk.old_lines > 0:
+            hunk.category = ChangeCategory.MODIFIED
+        else:
+            hunk.category = ChangeCategory.CONTEXT
+
+    def _classify_file(self, fc: FileChange) -> None:
+        """Final file classification (renames already detected via headers)."""
+        pass

tests/test_diff_analyzer.py

@@ -0,0 +1,224 @@
+"""Tests for diff_analyzer module."""
+
+import sys
+import os
+sys.path.insert(0, os.path.join(os.path.dirname(__file__), '..', 'scripts'))
+
+from diff_analyzer import DiffAnalyzer, ChangeCategory
+
+
+def test_parse_simple_addition():
+    diff = """diff --git a/foo.py b/foo.py
+new file mode 100644
+--- /dev/null
++++ b/foo.py
+@@ -0,0 +1,3 @@
++def hello():
++    return "world"
++# end
+"""
+    analyzer = DiffAnalyzer()
+    summary = analyzer.analyze(diff)
+
+    assert summary.files_changed == 1
+    assert summary.files_added == 1
+    assert summary.files_modified == 0
+    assert summary.total_added == 3
+    assert summary.total_deleted == 0
+    assert summary.hunks_added == 1
+    assert len(summary.file_changes) == 1
+    assert summary.file_changes[0].is_new is True
+    assert summary.file_changes[0].path == "foo.py"
+
+
+def test_parse_simple_deletion():
+    diff = """diff --git a/old.py b/old.py
+deleted file mode 100644
+--- a/old.py
++++ /dev/null
+@@ -1,2 +0,0 @@
+-x = 1
+-y = 2
+"""
+    analyzer = DiffAnalyzer()
+    summary = analyzer.analyze(diff)
+
+    assert summary.files_changed == 1
+    assert summary.files_deleted == 1
+    assert summary.total_deleted == 2
+    assert summary.total_added == 0
+    assert summary.hunks_deleted == 1
+    assert summary.file_changes[0].is_deleted is True
+
+
+def test_parse_modification():
+    diff = """diff --git a/bar.py b/bar.py
+--- a/bar.py
++++ b/bar.py
+@@ -10,3 +10,4 @@ def foo():
+     existing()
+-    old_call()
++    new_call()
++    extra_step()
+     return
+"""
+    analyzer = DiffAnalyzer()
+    summary = analyzer.analyze(diff)
+
+    assert summary.files_changed == 1
+    assert summary.files_modified == 1
+    assert summary.total_added == 2   # +new_call(), +extra_step()
+    assert summary.total_deleted == 1  # -old_call()
+    assert summary.total_context == 2  # 2 context lines
+    assert summary.hunks_modified == 1
+
+
+def test_parse_multiple_files():
+    diff = """diff --git a/a.py b/a.py
+--- a/a.py
++++ b/a.py
+@@ -1,1 +1,2 @@
+ existing
++added
+diff --git a/b.py b/b.py
+new file mode 100644
+--- /dev/null
++++ b/b.py
+@@ -0,0 +1,1 @@
++new file
+diff --git a/c.py b/c.py
+deleted file mode 100644
+--- a/c.py
++++ /dev/null
+@@ -1,1 +0,0 @@
+-gone
+"""
+    analyzer = DiffAnalyzer()
+    summary = analyzer.analyze(diff)
+
+    assert summary.files_changed == 3
+    assert summary.files_added == 1
+    assert summary.files_deleted == 1
+    assert summary.files_modified == 1
+    assert summary.total_added == 2
+    assert summary.total_deleted == 1
+
+
+def test_parse_rename():
+    diff = """diff --git a/old_name.py b/new_name.py
+rename from old_name.py
+rename to new_name.py
+--- a/old_name.py
++++ b/new_name.py
+@@ -1,1 +1,1 @@
+-old_func()
++new_func()
+"""
+    analyzer = DiffAnalyzer()
+    summary = analyzer.analyze(diff)
+
+    assert summary.files_changed == 1
+    assert summary.files_renamed == 1
+    assert summary.file_changes[0].is_renamed is True
+    assert summary.file_changes[0].old_path == "old_name.py"
+    assert summary.file_changes[0].path == "new_name.py"
+
+
+def test_parse_mixed_hunks():
+    """A file with one add hunk and one delete hunk."""
+    diff = """diff --git a/mixed.py b/mixed.py
+--- a/mixed.py
++++ b/mixed.py
+@@ -5,0 +6,2 @@
++new_line_1
++new_line_2
+@@ -20,2 +22,0 @@
+-removed_1
+-removed_2
+"""
+    analyzer = DiffAnalyzer()
+    summary = analyzer.analyze(diff)
+
+    assert summary.files_changed == 1
+    assert summary.hunks_added == 1
+    assert summary.hunks_deleted == 1
+    assert summary.total_added == 2
+    assert summary.total_deleted == 2
+
+
+def test_empty_diff():
+    analyzer = DiffAnalyzer()
+    summary = analyzer.analyze("")
+
+    assert summary.files_changed == 0
+    assert summary.total_added == 0
+    assert summary.total_deleted == 0
+
+
+def test_to_dict():
+    diff = """diff --git a/test.py b/test.py
+new file mode 100644
+--- /dev/null
++++ b/test.py
+@@ -0,0 +1,2 @@
++line1
++line2
+"""
+    analyzer = DiffAnalyzer()
+    summary = analyzer.analyze(diff)
+    d = summary.to_dict()
+
+    assert d["files_changed"] == 1
+    assert d["files_added"] == 1
+    assert d["total_added"] == 2
+    assert d["total_deleted"] == 0
+    assert len(d["files"]) == 1
+    assert d["files"][0]["path"] == "test.py"
+    assert d["files"][0]["is_new"] is True
+
+
+def test_context_only_hunk():
+    """A hunk with only context lines (rare but possible)."""
+    diff = """diff --git a/noop.py b/noop.py
+--- a/noop.py
++++ b/noop.py
+@@ -5,3 +5,3 @@
+ context1
+ context2
+ context3
+"""
+    analyzer = DiffAnalyzer()
+    summary = analyzer.analyze(diff)
+
+    assert summary.total_context == 3
+    assert summary.total_added == 0
+    assert summary.total_deleted == 0
+
+
+def test_binary_files_skipped():
+    """Binary file diffs have no content lines — just headers."""
+    diff = """diff --git a/image.png b/image.png
+--- a/image.png
++++ b/image.png
+Binary files a/image.png and b/image.png differ
+"""
+    analyzer = DiffAnalyzer()
+    summary = analyzer.analyze(diff)
+
+    assert summary.files_changed == 1
+    assert summary.total_added == 0
+    assert summary.total_deleted == 0
+
+
+if __name__ == "__main__":
+    test_parse_simple_addition()
+    test_parse_simple_deletion()
+    test_parse_modification()
+    test_parse_multiple_files()
+    test_parse_rename()
+    test_parse_mixed_hunks()
+    test_empty_diff()
+    test_to_dict()
+    test_context_only_hunk()
+    test_binary_files_skipped()
+    print("All 10 tests passed.")