Add source files: models and parser

src/gdiffer/parser.py

@@ -0,0 +1,203 @@
+"""Diff parser for unified git diff format."""
+
+import re
+from typing import Optional
+
+from gdiffer.models import DiffFile, DiffHunk
+
+
+class DiffParser:
+    """Parser for unified diff format (as produced by git diff)."""
+
+    HUNK_PATTERN = re.compile(r'^@@ -(\d+),?(\d*) \+(\d+),?(\d*) @@')
+
+    def __init__(self):
+        self.files: list[DiffFile] = []
+        self.errors: list[str] = []
+
+    def parse(self, diff_content: str) -> list[DiffFile]:
+        self.files = []
+        self.errors = []
+
+        if not diff_content.strip():
+            return self.files
+
+        lines = diff_content.splitlines()
+        self._parse_lines(lines)
+        return self.files
+
+    def _parse_lines(self, lines: list[str]) -> None:
+        i = 0
+        n = len(lines)
+
+        while i < n:
+            line = lines[i].rstrip('\n')
+
+            if line.startswith('diff --git'):
+                file_obj = self._parse_file(lines, i)
+                if file_obj:
+                    self.files.append(file_obj)
+                    i += 1
+                    continue
+
+            i += 1
+
+    def _parse_file(self, lines: list[str], start: int) -> Optional[DiffFile]:
+        if start >= len(lines):
+            return None
+
+        first_line = lines[start]
+        if not first_line.startswith('diff --git'):
+            return None
+
+        parts = first_line.split(' ', 3)
+        if len(parts) < 4:
+            return None
+
+        old_path = parts[2][2:] if len(parts) > 2 else ''
+        new_path = parts[3][2:] if len(parts) > 3 else old_path
+
+        if old_path.startswith('a/'):
+            old_path = old_path[2:]
+        if new_path.startswith('b/'):
+            new_path = new_path[2:]
+
+        file_obj = DiffFile(old_path=old_path, new_path=new_path)
+
+        i = start + 1
+        n = len(lines)
+
+        while i < n:
+            line = lines[i].rstrip('\n')
+
+            if line.startswith('new file mode '):
+                file_obj.new_file_mode = line.split()[-1]
+                file_obj.change_type = "add"
+                i += 1
+                continue
+
+            if line.startswith('deleted file mode '):
+                file_obj.deleted_file_mode = line.split()[-1]
+                file_obj.change_type = "delete"
+                i += 1
+                continue
+
+            if line.startswith('similarity index '):
+                file_obj.similarity_index = line.split()[-1].rstrip('%')
+                i += 1
+                continue
+
+            if line.startswith('rename from '):
+                file_obj.rename_from = line[12:]
+                i += 1
+                continue
+
+            if line.startswith('rename to '):
+                file_obj.rename_to = line[10:]
+                file_obj.change_type = "rename"
+                i += 1
+                continue
+
+            if line.startswith('---'):
+                i += 1
+                continue
+
+            if line.startswith('+++'):
+                i += 1
+                continue
+
+            if line.startswith('@@'):
+                hunk, consumed = self._parse_hunk(lines, i)
+                if hunk:
+                    file_obj.hunks.append(hunk)
+                    i += consumed
+                    continue
+
+            if line.startswith('diff --git'):
+                break
+
+            i += 1
+
+        return file_obj
+
+    def _parse_hunk(self, lines: list[str], start: int) -> tuple[Optional[DiffHunk], int]:
+        if start >= len(lines):
+            return None, 0
+
+        line = lines[start].rstrip('\n')
+        match = self.HUNK_PATTERN.match(line)
+
+        if not match:
+            return None, 0
+
+        old_start = int(match.group(1))
+        old_lines = int(match.group(2)) if match.group(2) else 1
+        new_start = int(match.group(3))
+        new_lines = int(match.group(4)) if match.group(4) else 1
+
+        hunk = DiffHunk(
+            old_start=old_start,
+            old_lines=old_lines,
+            new_start=new_start,
+            new_lines=new_lines
+        )
+
+        i = start + 1
+        n = len(lines)
+        old_lines_collected = 0
+        new_lines_collected = 0
+
+        old_content = []
+        new_content = []
+
+        while i < n:
+            line = lines[i].rstrip('\n')
+
+            if line.startswith('@@'):
+                break
+
+            if line.startswith('diff --git'):
+                break
+
+            if line.startswith('---'):
+                break
+
+            if line.startswith('+++'):
+                break
+
+            if old_lines_collected >= old_lines and new_lines_collected >= new_lines:
+                break
+
+            if line.startswith('+') and not line.startswith('+++'):
+                new_content.append(line)
+                new_lines_collected += 1
+            elif line.startswith('-') and not line.startswith('---'):
+                old_content.append(line)
+                old_lines_collected += 1
+            elif line.startswith(' ') or line == '':
+                old_content.append(line)
+                new_content.append(line)
+                old_lines_collected += 1
+                new_lines_collected += 1
+            else:
+                break
+
+            i += 1
+
+        hunk.old_lines_content = old_content
+        hunk.new_lines_content = new_content
+
+        return hunk, i - start
+
+
+def parse_diff(diff_content: str) -> list[DiffFile]:
+    """Parse diff content and return list of DiffFile objects."""
+    parser = DiffParser()
+    return parser.parse(diff_content)
+
+
+def parse_diff_from_file(filepath: str) -> list[DiffFile]:
+    """Read a diff file and parse its contents."""
+    with open(filepath, 'r') as f:
+        content = f.read()
+    return parse_diff(content)