0704

脚本/generate_line_diff.py

@@ -0,0 +1,105 @@
+#!/usr/bin/env python3
+"""Generate line-by-line diff between two markdown files."""
+
+from __future__ import annotations
+
+import argparse
+import difflib
+from pathlib import Path
+
+
+def read_lines(path: Path) -> list[str]:
+    text = path.read_text(encoding="utf-8-sig")
+    return text.splitlines()
+
+
+def generate_line_diff(src: Path, dst: Path) -> str:
+    src_lines = read_lines(src)
+    dst_lines = read_lines(dst)
+    src_name = src.as_posix()
+    dst_name = dst.as_posix()
+    diff = difflib.unified_diff(
+        src_lines,
+        dst_lines,
+        fromfile=src_name,
+        tofile=dst_name,
+        lineterm="",
+    )
+    return "\n".join(diff)
+
+
+def generate_compact_line_diff(src: Path, dst: Path, context: int = 2) -> str:
+    """Generate a compact diff showing only changed lines with a little context."""
+    src_lines = read_lines(src)
+    dst_lines = read_lines(dst)
+    src_name = src.as_posix()
+    dst_name = dst.as_posix()
+
+    sm = difflib.SequenceMatcher(None, src_lines, dst_lines)
+    output = [f"--- {src_name}", f"+++ {dst_name}", ""]
+
+    for tag, i1, i2, j1, j2 in sm.get_opcodes():
+        if tag == "equal":
+            # Show only a few context lines around changes
+            if context > 0:
+                ctx = src_lines[i1:i2]
+                if len(ctx) <= context * 2 + 1:
+                    for line in ctx:
+                        output.append(f" {line}")
+                else:
+                    for line in ctx[:context]:
+                        output.append(f" {line}")
+                    output.append(" ...")
+                    for line in ctx[-context:]:
+                        output.append(f" {line}")
+        elif tag == "replace":
+            output.append(f"@@ -{i1 + 1},{i2 - i1} +{j1 + 1},{j2 - j1} @@")
+            for line in src_lines[i1:i2]:
+                output.append(f"-{line}")
+            for line in dst_lines[j1:j2]:
+                output.append(f"+{line}")
+        elif tag == "delete":
+            output.append(f"@@ -{i1 + 1},{i2 - i1} +{j1 + 1},0 @@")
+            for line in src_lines[i1:i2]:
+                output.append(f"-{line}")
+        elif tag == "insert":
+            output.append(f"@@ -{i1 + 1},0 +{j1 + 1},{j2 - j1} @@")
+            for line in dst_lines[j1:j2]:
+                output.append(f"+{line}")
+
+    return "\n".join(output)
+
+
+def main() -> int:
+    parser = argparse.ArgumentParser(description="Generate line-by-line diff for markdown files.")
+    parser.add_argument("--input-dir", type=Path, required=True, help="Directory containing source files.")
+    parser.add_argument("--output-dir", type=Path, required=True, help="Directory containing polished files.")
+    parser.add_argument("--out-diff-dir", type=Path, required=True, help="Directory to write diff files.")
+    parser.add_argument("--compact", action="store_true", help="Show only changed lines with context.")
+    parser.add_argument("--context", type=int, default=2, help="Number of context lines around changes (compact mode).")
+    args = parser.parse_args()
+
+    args.out_diff_dir.mkdir(parents=True, exist_ok=True)
+
+    for src in sorted(args.input_dir.glob("*.md")):
+        dst = args.output_dir / src.name
+        if not dst.exists():
+            print(f"Skip {src.name}: no matching polished file.")
+            continue
+
+        if args.compact:
+            diff_text = generate_compact_line_diff(src, dst, context=args.context)
+            suffix = "-compact.diff"
+        else:
+            diff_text = generate_line_diff(src, dst)
+            suffix = "-line.diff"
+
+        out_path = args.out_diff_dir / (src.stem + suffix)
+        out_path.write_text(diff_text, encoding="utf-8")
+        print(f"Wrote {out_path}")
+
+    return 0
+
+
+if __name__ == "__main__":
+    raise SystemExit(main())