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

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

scripts/gitea_issue_parser.py

@@ -1,131 +0,0 @@
-#!/usr/bin/env python3
-"""
-Gitea Issue Body Parser — Extract structured data from markdown issue bodies.
-
-Usage:
-    cat issue_body.txt | python3 scripts/gitea_issue_parser.py --stdin --pretty
-    python3 scripts/gitea_issue_parser.py --url https://forge.../api/v1/repos/.../issues/123 --pretty
-    python3 scripts/gitea_issue_parser.py body.txt --title "Fix thing (#42)" --labels pipeline extraction
-"""
-
-import argparse
-import json
-import re
-import sys
-from typing import Dict, List, Any, Optional
-
-
-def parse_issue_body(body: str, title: str = "", labels: List[str] = None) -> Dict[str, Any]:
-    """Parse a Gitea issue markdown body into structured JSON.
-    
-    Extracted fields:
-    - title: Issue title
-    - context: Background/description section
-    - criteria[]: Acceptance criteria (checkboxes or numbered lists)
-    - labels[]: Issue labels
-    - epic_ref: Parent/epic issue reference (from "Closes #N" or title)
-    - sections{}: All ## sections as key-value pairs
-    """
-    result = {
-        "title": title,
-        "context": "",
-        "criteria": [],
-        "labels": labels or [],
-        "epic_ref": None,
-        "sections": {},
-    }
-    
-    if not body:
-        return result
-    
-    # Extract epic reference from title or body
-    epic_patterns = [
-        r"(?:closes|fixes|addresses|refs?)\s+#(\d+)",
-        r"#(\d+)",
-    ]
-    for pattern in epic_patterns:
-        match = re.search(pattern, (title + " " + body).lower())
-        if match:
-            result["epic_ref"] = int(match.group(1))
-            break
-    
-    # Parse ## sections
-    section_pattern = r"^##\s+(.+?)$\n((?:^(?!##\s).*$\n?)*)"
-    for match in re.finditer(section_pattern, body, re.MULTILINE):
-        section_name = match.group(1).strip().lower().replace(" ", "_")
-        section_content = match.group(2).strip()
-        result["sections"][section_name] = section_content
-    
-    # Extract acceptance criteria (checkboxes)
-    checkbox_pattern = r"^\s*-\s*\[([ xX])\]\s*(.+)$"
-    for match in re.finditer(checkbox_pattern, body, re.MULTILINE):
-        checked = match.group(1).lower() == "x"
-        text = match.group(2).strip()
-        result["criteria"].append({"text": text, "checked": checked})
-    
-    # If no checkboxes, try numbered lists in "Acceptance Criteria" or "Criteria" section
-    if not result["criteria"]:
-        for section_name in ["acceptance_criteria", "criteria", "acceptance criteria"]:
-            if section_name in result["sections"]:
-                numbered = r"^\s*\d+\.\s*(.+)$"
-                for match in re.finditer(numbered, result["sections"][section_name], re.MULTILINE):
-                    result["criteria"].append({"text": match.group(1).strip(), "checked": False})
-                break
-    
-    # Extract context (first section or first paragraph before any ## heading)
-    first_heading = body.find("## ")
-    if first_heading > 0:
-        context_text = body[:first_heading].strip()
-    else:
-        context_text = body.split("\n\n")[0].strip()
-    # Clean up: remove "## Context" or "## Problem" header if present
-    context_text = re.sub(r"^#+\s*\w+\s*\n?", "", context_text).strip()
-    result["context"] = context_text[:500]  # Cap at 500 chars
-    
-    return result
-
-
-def fetch_issue_from_url(url: str) -> Dict[str, Any]:
-    """Fetch an issue from a Gitea API URL and parse it."""
-    import urllib.request
-    req = urllib.request.Request(url, headers={"Accept": "application/json"})
-    with urllib.request.urlopen(req) as resp:
-        data = json.loads(resp.read())
-    
-    return parse_issue_body(
-        body=data.get("body", ""),
-        title=data.get("title", ""),
-        labels=[l["name"] for l in data.get("labels", [])]
-    )
-
-
-def main():
-    parser = argparse.ArgumentParser(description="Parse Gitea issue markdown into structured JSON")
-    parser.add_argument("file", nargs="?", help="Issue body file (or use --stdin)")
-    parser.add_argument("--stdin", action="store_true", help="Read from stdin")
-    parser.add_argument("--url", help="Gitea API URL to fetch issue from")
-    parser.add_argument("--title", default="", help="Issue title")
-    parser.add_argument("--labels", nargs="*", default=[], help="Issue labels")
-    parser.add_argument("--pretty", action="store_true", help="Pretty-print JSON output")
-    
-    args = parser.parse_args()
-    
-    if args.url:
-        result = fetch_issue_from_url(args.url)
-    elif args.stdin:
-        body = sys.stdin.read()
-        result = parse_issue_body(body, args.title, args.labels)
-    elif args.file:
-        with open(args.file) as f:
-            body = f.read()
-        result = parse_issue_body(body, args.title, args.labels)
-    else:
-        parser.print_help()
-        sys.exit(1)
-    
-    indent = 2 if args.pretty else None
-    print(json.dumps(result, indent=indent))
-
-
-if __name__ == "__main__":
-    main()

scripts/test_gitea_issue_parser.py

@@ -1,109 +0,0 @@
-#!/usr/bin/env python3
-"""Tests for scripts/gitea_issue_parser.py"""
-
-import sys
-import os
-sys.path.insert(0, os.path.dirname(__file__) or ".")
-
-# Import from sibling
-import importlib.util
-spec = importlib.util.spec_from_file_location("parser", os.path.join(os.path.dirname(__file__) or ".", "gitea_issue_parser.py"))
-mod = importlib.util.module_from_spec(spec)
-spec.loader.exec_module(mod)
-parse_issue_body = mod.parse_issue_body
-
-
-def test_basic_parsing():
-    body = """## Context
-
-This is the background info.
-
-## Acceptance Criteria
-
-- [ ] First criterion
-- [x] Second criterion (done)
-
-## What to build
-
-Some description.
-"""
-    result = parse_issue_body(body, title="Test (#42)", labels=["bug"])
-    assert result["title"] == "Test (#42)"
-    assert result["labels"] == ["bug"]
-    assert result["epic_ref"] == 42
-    assert len(result["criteria"]) == 2
-    assert result["criteria"][0]["text"] == "First criterion"
-    assert result["criteria"][0]["checked"] == False
-    assert result["criteria"][1]["checked"] == True
-    assert "context" in result["sections"]
-    print("PASS: test_basic_parsing")
-
-
-def test_numbered_criteria():
-    body = """## Acceptance Criteria
-
-1. First item
-2. Second item
-3. Third item
-"""
-    result = parse_issue_body(body)
-    assert len(result["criteria"]) == 3
-    assert result["criteria"][0]["text"] == "First item"
-    print("PASS: test_numbered_criteria")
-
-
-def test_epic_ref_from_body():
-    body = "Closes #123\n\nSome description."
-    result = parse_issue_body(body)
-    assert result["epic_ref"] == 123
-    print("PASS: test_epic_ref_from_body")
-
-
-def test_empty_body():
-    result = parse_issue_body("")
-    assert result["criteria"] == []
-    assert result["context"] == ""
-    assert result["sections"] == {}
-    print("PASS: test_empty_body")
-
-
-def test_no_sections():
-    body = "Just a plain issue body with no headings."
-    result = parse_issue_body(body)
-    assert result["context"] == "Just a plain issue body with no headings."
-    print("PASS: test_no_sections")
-
-
-def test_multiple_sections():
-    body = """## Problem
-
-Something is broken.
-
-## Fix
-
-Do this instead.
-
-## Notes
-
-Additional info.
-"""
-    result = parse_issue_body(body)
-    assert "problem" in result["sections"]
-    assert "fix" in result["sections"]
-    assert "notes" in result["sections"]
-    assert "Something is broken" in result["sections"]["problem"]
-    print("PASS: test_multiple_sections")
-
-
-def run_all():
-    test_basic_parsing()
-    test_numbered_criteria()
-    test_epic_ref_from_body()
-    test_empty_body()
-    test_no_sections()
-    test_multiple_sections()
-    print("\nAll 6 tests passed!")
-
-
-if __name__ == "__main__":
-    run_all()

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.")