diffinite 0.9.6__tar.gz → 0.10.0__tar.gz

{diffinite-0.9.6/src/diffinite.egg-info → diffinite-0.10.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: diffinite
-Version: 0.9.6
+Version: 0.10.0
 Summary: Forensic source-code comparison tool — Winnowing fingerprints and professional PDF reports for IP litigation & code audit
 Author: nash-dir
 License: Apache-2.0

{diffinite-0.9.6 → diffinite-0.10.0}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "diffinite"
-version = "0.9.6"
+version = "0.10.0"
 description = "Forensic source-code comparison tool — Winnowing fingerprints and professional PDF reports for IP litigation & code audit"
 readme = "README.md"
 license = {text = "Apache-2.0"}

{diffinite-0.9.6 → diffinite-0.10.0}/src/diffinite/cli.py

@@ -58,14 +58,6 @@ def main(argv: list[str] | None = None) -> None:
         ),
     )
 
-    # ── Output ────────────────────────────────────────────────────────
-    parser.add_argument(
-        "--output-pdf", "-o",
-        default="report.pdf",
-        help="Output PDF file path (default: report.pdf). "
-             "Ignored when any --report-* option is specified.",
-    )
-
     # ── Comparison options ────────────────────────────────────────────
     parser.add_argument(
         "--by-word",
@@ -74,7 +66,7 @@ def main(argv: list[str] | None = None) -> None:
         help="Compare by word instead of by line",
     )
     parser.add_argument(
-        "--no-comments",
+        "--strip-comments",
         action="store_true",
         default=False,
         help="Strip comments before comparison (uses 2-pass parser)",
@@ -85,7 +77,7 @@ def main(argv: list[str] | None = None) -> None:
         default=False,
         help=(
             "Collapse runs of 3+ blank lines after comment stripping. "
-            "Only effective with --no-comments. WARNING: changes line "
+            "Only effective with --strip-comments. WARNING: changes line "
             "numbers — do not use for forensic line-tracing."
         ),
     )
@@ -107,13 +99,14 @@ def main(argv: list[str] | None = None) -> None:
     )
     parser.add_argument(
         "--sort-by",
-        choices=["filename", "size", "ratio"],
+        choices=["filename", "path", "similarity", "ratio"],
         default=None,
         dest="sort_by",
         help=(
             "Sort matched file pairs in the report. "
-            "'filename' sorts by file path, 'size' by file size, "
-            "'ratio' by similarity ratio. Default: insertion order (no sort)."
+            "'filename' sorts by file basename, 'path' by full path, "
+            "'similarity' by name match score, 'ratio' by content "
+            "similarity. Default: insertion order (no sort)."
         ),
     )
     parser.add_argument(
@@ -179,7 +172,7 @@ def main(argv: list[str] | None = None) -> None:
         ),
     )
     parser.add_argument(
-        "--show-filename",
+        "--filename",
         action="store_true",
         default=False,
         help="Show the filename at the top-right of each page",
@@ -204,15 +197,35 @@ def main(argv: list[str] | None = None) -> None:
             "plain delete/add. Works in both simple and deep modes."
         ),
     )
+    parser.add_argument(
+        "--include-uncompared",
+        action=argparse.BooleanOptionalAction,
+        default=True,
+        help=(
+            "Include unmatched (uncompared) file lists in the report. "
+            "Use --no-include-uncompared to hide them (default: included)."
+        ),
+    )
+    parser.add_argument(
+        "--binary-handling",
+        choices=["exclude", "hash", "error"],
+        default="hash",
+        dest="binary_handling",
+        help=(
+            "How to handle binary (non-decodable) files: "
+            "'exclude' skips them entirely, 'hash' shows SHA-256 match "
+            "status, 'error' shows decode error (default: hash)."
+        ),
+    )
 
     # ── Report format options ─────────────────────────────────────────
     format_group = parser.add_argument_group(
         "Report Format",
         "Output format(s). Multiple can be combined. "
-        "If none specified, defaults to --output-pdf.",
+        "If none specified, defaults to PDF (report.pdf).",
     )
     format_group.add_argument(
-        "--report-pdf",
+        "--report-pdf", "-o",
         metavar="PATH",
         default=None,
         help="Generate a merged PDF report at the given path",
@@ -243,7 +256,7 @@ def main(argv: list[str] | None = None) -> None:
         "'--mode deep').",
     )
     deep_group.add_argument(
-        "--k-gram", "--kgram-size",
+        "--k-gram",
         type=int,
         default=DEFAULT_K,
         dest="k_gram",
@@ -253,7 +266,7 @@ def main(argv: list[str] | None = None) -> None:
         ),
     )
     deep_group.add_argument(
-        "--window", "--window-size",
+        "--window",
         type=int,
         default=DEFAULT_W,
         dest="window",
@@ -263,12 +276,12 @@ def main(argv: list[str] | None = None) -> None:
         ),
     )
     deep_group.add_argument(
-        "--threshold-deep", "--min-jaccard",
+        "--threshold-deep",
         type=float,
-        default=0.05,
+        default=5,
         dest="threshold_deep",
         help=(
-            "Minimum Jaccard similarity to report (default: 0.05). "
+            "Minimum Jaccard similarity 0–100 to report (default: 5). "
             "Below 5%% is considered noise."
         ),
     )
@@ -336,38 +349,45 @@ def main(argv: list[str] | None = None) -> None:
 
     args = parser.parse_args(argv)
 
+    # Convert threshold-deep from 0-100 (user-facing) to 0-1 (internal)
+    min_jaccard_internal = args.threshold_deep / 100.0
+
     # Build analysis metadata (embedded in every report for transparency)
     metadata = AnalysisMetadata(
         exec_mode=args.mode,
         k=args.k_gram,
         w=args.window,
-        threshold=args.threshold_deep,
+        threshold=args.threshold_deep,  # 0-100 scale in metadata
         autojunk=not args.no_autojunk,
     )
 
     # Resolve encoding
     encoding = args.encoding if args.encoding.lower() != "auto" else None
 
+    # Resolve default PDF output if no --report-* specified
+    report_pdf = args.report_pdf
+    if report_pdf is None and args.report_html is None and args.report_md is None and args.report_json is None:
+        report_pdf = "report.pdf"
+
     run_pipeline(
         dir_a=args.dir_a,
         dir_b=args.dir_b,
         by_word=args.by_word,
-        compare_comment=not args.no_comments,
+        strip_comments=args.strip_comments,
         squash_blanks=args.squash_blanks,
-        output_pdf=args.output_pdf,
         threshold=args.threshold,
         no_merge=args.no_merge,
         show_page_number=args.page_number,
         show_file_number=args.file_number,
         show_bates_number=args.bates_number,
-        show_filename=args.show_filename,
+        show_filename=args.filename,
         collapse_identical=args.collapse_identical,
         # Execution mode & deep compare
         exec_mode=args.mode,
         workers=args.workers,
         kgram_size=args.k_gram,
         window_size=args.window,
-        min_jaccard=args.threshold_deep,
+        min_jaccard=min_jaccard_internal,
         normalize=args.normalize,
         metadata=metadata,
         # Forensic options
@@ -377,7 +397,7 @@ def main(argv: list[str] | None = None) -> None:
         embed_hash=args.embed_hash,
         bundle_path=args.bundle_path,
         # Multi-format output
-        report_pdf=args.report_pdf,
+        report_pdf=report_pdf,
         report_html=args.report_html,
         report_md=args.report_md,
         report_json=args.report_json,
@@ -388,10 +408,14 @@ def main(argv: list[str] | None = None) -> None:
         sort_order=args.sort_order,
         # Moved block detection
         detect_moved=args.detect_moved,
+        # Uncompared files
+        include_uncompared=args.include_uncompared,
         # Bates prefix/suffix
         bates_prefix=args.bates_prefix,
         bates_suffix=args.bates_suffix,
         bates_start=args.bates_start,
+        # Binary handling
+        binary_handling=args.binary_handling,
     )
 
 

{diffinite-0.9.6 → diffinite-0.10.0}/src/diffinite/models.py

@@ -98,6 +98,12 @@ class DiffResult:
     error: Optional[str] = None
     """None이 아니면 디코딩/읽기 실패 등의 에러 메시지. 이 경우 위 필드는 0/빈값."""
 
+    binary: bool = False
+    """True if file pair was detected as binary (non-decodable)."""
+
+    hash_match: Optional[bool] = None
+    """SHA-256 match status for binary files. None for text files."""
+
 
 # ──────────────────────────────────────────────────────────────────────
 # Winnowing 핑거프린트 엔트리

{diffinite-0.9.6 → diffinite-0.10.0}/src/diffinite/pdf_gen.py

@@ -82,6 +82,7 @@ table.summary th, table.summary td {
     border: 1px solid #ccc;
     padding: 5px 8px;
     text-align: left;
+    word-break: break-all;
 }
 table.summary th {
     background: #0078d4;
@@ -178,6 +179,7 @@ table.deep th, table.deep td {
     border: 1px solid #ccc;
     padding: 4px 6px;
     text-align: left;
+    word-break: break-all;
 }
 table.deep th {
     background: #6c5ce7;
@@ -212,6 +214,27 @@ table.deep tr:nth-child(even) {
 # ---------------------------------------------------------------------------
 # Helpers
 # ---------------------------------------------------------------------------
+def _break_path(path_str: str) -> str:
+    """Insert zero-width spaces after path separator symbols for line-breaking.
+
+    xhtml2pdf는 긴 파일 경로를 자동 줄바꿈하지 못하므로,
+    경로 구분자(/, \\, ., _) 뒤에 zero-width space를 삽입하여
+    자연스러운 줄바꿈 지점을 제공한다.
+
+    Args:
+        path_str: HTML-escaped 경로 문자열.
+
+    Returns:
+        줄바꿈 힌트가 삽입된 경로 문자열.
+    """
+    # HTML entity for zero-width space
+    zwsp = "​"
+    result = path_str
+    for sep in ("/", "\\", ".", "_"):
+        result = result.replace(sep, sep + zwsp)
+    return result
+
+
 def _ratio_badge(ratio: float) -> str:
     """Return an HTML badge span for a similarity ratio."""
     pct = ratio * 100
@@ -251,7 +274,7 @@ def build_hash_table_html(
             parts.append(
                 f'<tr>'
                 f'<td>{idx}</td>'
-                f'<td>{html.escape(h.rel_path)}</td>'
+                f'<td>{_break_path(html.escape(h.rel_path))}</td>'
                 f'<td style="font-family:monospace;font-size:8px">{short_hash}</td>'
                 f'<td>{h.size_bytes:,}</td>'
                 f'</tr>\n'
@@ -368,17 +391,18 @@ def build_cover_body(
     dir_a: str,
     dir_b: str,
     by_word: bool,
-    compare_comment: bool,
+    strip_comments: bool,
     *,
     deep_results: Optional[list[DeepMatchResult]] = None,
     metadata: Optional["AnalysisMetadata"] = None,
     hash_table_html: Optional[str] = None,
+    include_uncompared: bool = True,
 ) -> str:
     """Build the cover-page body fragment (no DOCTYPE/html/head wrapper)."""
     from diffinite.models import AnalysisMetadata as _AM  # avoid circular at module level
 
     unit = "word" if by_word else "line"
-    comment_mode = "included" if compare_comment else "excluded"
+    comment_mode = "stripped" if strip_comments else "included"
 
     # Analysis metadata banner (transparency)
     meta_html = ""
@@ -396,26 +420,43 @@ def build_cover_body(
 
     summary_rows = ""
     for idx, r in enumerate(results, 1):
-        badge = _ratio_badge(r.ratio)
-        err = (
-            f' <em style="color:red">({html.escape(r.error)})</em>'
-            if r.error else ""
-        )
-        summary_rows += (
-            f"<tr>"
-            f"<td>{idx}</td>"
-            f"<td>{html.escape(r.match.rel_path_a)}</td>"
-            f"<td>{html.escape(r.match.rel_path_b)}</td>"
-            f"<td>{r.match.similarity:.1f}</td>"
-            f"<td>{badge}{err}</td>"
-            f"<td style='color:green'>+{r.additions}</td>"
-            f"<td style='color:red'>-{r.deletions}</td>"
-            f"</tr>\n"
-        )
+        if r.binary:
+            if r.hash_match:
+                status = '<span class="badge badge-high">✓ Binary Match</span>'
+            else:
+                status = '<span class="badge badge-low">✗ Binary Mismatch</span>'
+            summary_rows += (
+                f"<tr>"
+                f"<td>{idx}</td>"
+                f"<td>{_break_path(html.escape(r.match.rel_path_a))}</td>"
+                f"<td>{_break_path(html.escape(r.match.rel_path_b))}</td>"
+                f"<td>{r.match.similarity:.1f}</td>"
+                f"<td>{status}</td>"
+                f"<td>—</td>"
+                f"<td>—</td>"
+                f"</tr>\n"
+            )
+        else:
+            badge = _ratio_badge(r.ratio)
+            err = (
+                f' <em style="color:red">({html.escape(r.error)})</em>'
+                if r.error else ""
+            )
+            summary_rows += (
+                f"<tr>"
+                f"<td>{idx}</td>"
+                f"<td>{_break_path(html.escape(r.match.rel_path_a))}</td>"
+                f"<td>{_break_path(html.escape(r.match.rel_path_b))}</td>"
+                f"<td>{r.match.similarity:.1f}</td>"
+                f"<td>{badge}{err}</td>"
+                f"<td style='color:green'>+{r.additions}</td>"
+                f"<td style='color:red'>-{r.deletions}</td>"
+                f"</tr>\n"
+            )
 
-    # Unmatched lists
+    # Unmatched lists (only when include_uncompared is True)
     unmatched_html = ""
-    if unmatched_a or unmatched_b:
+    if include_uncompared and (unmatched_a or unmatched_b):
         unmatched_html += "<h2>Unmatched Files</h2>\n"
         if unmatched_a:
             unmatched_html += (
@@ -423,7 +464,7 @@ def build_cover_body(
                 "<ul class='unmatched'>\n"
             )
             for f in unmatched_a:
-                unmatched_html += f"  <li>{html.escape(f)}</li>\n"
+                unmatched_html += f"  <li>{_break_path(html.escape(f))}</li>\n"
             unmatched_html += "</ul>\n"
         if unmatched_b:
             unmatched_html += (
@@ -431,7 +472,7 @@ def build_cover_body(
                 "<ul class='unmatched'>\n"
             )
             for f in unmatched_b:
-                unmatched_html += f"  <li>{html.escape(f)}</li>\n"
+                unmatched_html += f"  <li>{_break_path(html.escape(f))}</li>\n"
             unmatched_html += "</ul>\n"
 
     deep_html = ""
@@ -447,8 +488,8 @@ def build_cover_body(
                 jbadge = _ratio_badge(jaccard)
                 deep_html += (
                     f"<tr>"
-                    f"<td>{html.escape(dr.file_a)}</td>"
-                    f"<td>{html.escape(b_file)}</td>"
+                    f"<td>{_break_path(html.escape(dr.file_a))}</td>"
+                    f"<td>{_break_path(html.escape(b_file))}</td>"
                     f"<td>{shared}</td>"
                     f"<td>{jbadge}</td>"
                     f"</tr>\n"

{diffinite-0.9.6 → diffinite-0.10.0}/src/diffinite/pipeline.py

@@ -36,19 +36,20 @@ import json
 import logging
 import os
 import tempfile
-from pathlib import Path
+from pathlib import Path, PurePosixPath
 
 from diffinite.collector import collect_files, match_files, FUZZY_THRESHOLD
 from diffinite.deep_compare import run_deep_compare
 from diffinite.differ import compute_diff, generate_html_diff, read_file
 from diffinite.evidence import (
+    _sha256_file,
     compute_file_hashes,
     create_evidence_bundle,
     write_manifest,
 )
 from diffinite.fingerprint import DEFAULT_K, DEFAULT_W
 from diffinite.models import AnalysisMetadata, DiffResult, DeepMatchResult
-from diffinite.parser import strip_comments
+from diffinite.parser import strip_comments as _strip_comments_fn
 from diffinite.pdf_gen import (
     _html_wrap,
     add_bates_numbers,
@@ -113,15 +114,16 @@ def _generate_markdown_report(
     dir_a: str,
     dir_b: str,
     by_word: bool,
-    compare_comment: bool,
+    strip_comments: bool,
     deep_results: list[DeepMatchResult] | None,
     output_path: str,
     *,
     metadata: AnalysisMetadata | None = None,
+    include_uncompared: bool = True,
 ) -> None:
     """Generate a Markdown summary report."""
     unit = "word" if by_word else "line"
-    comment_mode = "included" if compare_comment else "excluded"
+    comment_mode = "stripped" if strip_comments else "included"
 
     lines: list[str] = []
     lines.append("# Diffinite — Source Code Diff Report\n")
@@ -141,16 +143,24 @@ def _generate_markdown_report(
     lines.append("| # | File A | File B | Name Sim. | Match | +Added | −Deleted |")
     lines.append("|---|--------|--------|:---------:|:-----:|:------:|:--------:|")
     for idx, r in enumerate(results, 1):
-        pct = r.ratio * 100
-        err = f" ⚠ {r.error}" if r.error else ""
-        lines.append(
-            f"| {idx} | `{r.match.rel_path_a}` | `{r.match.rel_path_b}` "
-            f"| {r.match.similarity:.1f} | {pct:.1f}%{err} "
-            f"| +{r.additions} | −{r.deletions} |"
-        )
+        if r.binary:
+            status = "✓ Match" if r.hash_match else "✗ Mismatch"
+            lines.append(
+                f"| {idx} | `{r.match.rel_path_a}` | `{r.match.rel_path_b}` "
+                f"| {r.match.similarity:.1f} | [Binary: {status}] "
+                f"| — | — |"
+            )
+        else:
+            pct = r.ratio * 100
+            err = f" ⚠ {r.error}" if r.error else ""
+            lines.append(
+                f"| {idx} | `{r.match.rel_path_a}` | `{r.match.rel_path_b}` "
+                f"| {r.match.similarity:.1f} | {pct:.1f}%{err} "
+                f"| +{r.additions} | −{r.deletions} |"
+            )
 
     # Unmatched
-    if unmatched_a or unmatched_b:
+    if include_uncompared and (unmatched_a or unmatched_b):
         lines.append("\n## Unmatched Files\n")
         if unmatched_a:
             lines.append(f"### Only in A (`{dir_a}`)\n")
@@ -188,11 +198,12 @@ def _generate_json_report(
     dir_a: str,
     dir_b: str,
     by_word: bool,
-    compare_comment: bool,
+    strip_comments: bool,
     deep_results: list[DeepMatchResult] | None,
     output_path: str,
     *,
     metadata: AnalysisMetadata | None = None,
+    include_uncompared: bool = True,
 ) -> None:
     """Generate a JSON report for programmatic consumption.
 
@@ -201,7 +212,7 @@ def _generate_json_report(
     re-running the pipeline.
     """
     unit = "word" if by_word else "line"
-    comment_mode = "included" if compare_comment else "excluded"
+    comment_mode = "stripped" if strip_comments else "included"
 
     meta_dict = None
     if metadata is not None:
@@ -215,7 +226,7 @@ def _generate_json_report(
 
     result_list = []
     for r in results:
-        result_list.append({
+        entry = {
             "file_a": r.match.rel_path_a,
             "file_b": r.match.rel_path_b,
             "name_similarity": r.match.similarity,
@@ -224,7 +235,11 @@ def _generate_json_report(
             "deletions": r.deletions,
             "html_diff": r.html_diff,
             "error": r.error,
-        })
+            "binary": r.binary,
+        }
+        if r.binary:
+            entry["hash_match"] = r.hash_match
+        result_list.append(entry)
 
     deep_list = None
     if deep_results is not None:
@@ -251,13 +266,13 @@ def _generate_json_report(
         "comment_mode": comment_mode,
         "summary": {
             "matched_pairs": len(results),
-            "unmatched_a": len(unmatched_a),
-            "unmatched_b": len(unmatched_b),
+            "unmatched_a_count": len(unmatched_a),
+            "unmatched_b_count": len(unmatched_b),
         },
         "results": result_list,
         "deep_results": deep_list,
-        "unmatched_a": unmatched_a,
-        "unmatched_b": unmatched_b,
+        "unmatched_a": unmatched_a if include_uncompared else [],
+        "unmatched_b": unmatched_b if include_uncompared else [],
     }
 
     out = Path(output_path)
@@ -276,21 +291,23 @@ def _generate_html_report(
     dir_a: str,
     dir_b: str,
     by_word: bool,
-    compare_comment: bool,
+    strip_comments: bool,
     deep_results: list[DeepMatchResult] | None,
     output_path: str,
     ln_col_width: int = 28,
     *,
     metadata: AnalysisMetadata | None = None,
     hash_table_html: str | None = None,
+    include_uncompared: bool = True,
 ) -> None:
     """Generate a standalone HTML report with all diffs inline."""
     cover_html_body = build_cover_body(
         results, unmatched_a, unmatched_b,
-        dir_a, dir_b, by_word, compare_comment,
+        dir_a, dir_b, by_word, strip_comments,
         deep_results=deep_results,
         metadata=metadata,
         hash_table_html=hash_table_html,
+        include_uncompared=include_uncompared,
     )
 
     # Append all inline diffs
@@ -346,9 +363,8 @@ def run_pipeline(
     dir_a: str,
     dir_b: str,
     by_word: bool = False,
-    compare_comment: bool = True,
+    strip_comments: bool = False,
     squash_blanks: bool = False,
-    output_pdf: str = "report.pdf",
     threshold: float = FUZZY_THRESHOLD,
     *,
     no_merge: bool = False,
@@ -384,10 +400,14 @@ def run_pipeline(
     sort_order: str = "asc",
     # Moved block detection
     detect_moved: bool = False,
+    # Uncompared files
+    include_uncompared: bool = True,
     # Bates prefix/suffix
     bates_prefix: str = "",
     bates_suffix: str = "",
     bates_start: int = 1,
+    # Binary handling
+    binary_handling: str = "hash",
 ) -> None:
     """Execute the full diff-to-report pipeline.
 
@@ -410,7 +430,7 @@ def run_pipeline(
     """
     # Determine effective output paths
     if report_pdf is None and report_html is None and report_md is None and report_json is None:
-        report_pdf = output_pdf
+        report_pdf = "report.pdf"
 
     # Build default metadata if caller didn't provide one
     if metadata is None:
@@ -463,15 +483,30 @@ def run_pipeline(
         text_b = read_file(abs_b, encoding=encoding)
 
         if text_a is None or text_b is None:
-            results.append(DiffResult(
-                match=m, ratio=0.0, additions=0, deletions=0,
-                html_diff="", error="Could not decode one or both files",
-            ))
+            if binary_handling == "exclude":
+                continue
+            elif binary_handling == "hash":
+                hash_a = _sha256_file(str(root_a / m.rel_path_a))
+                hash_b = _sha256_file(str(root_b / m.rel_path_b))
+                hash_match = hash_a == hash_b
+                results.append(DiffResult(
+                    match=m,
+                    ratio=1.0 if hash_match else 0.0,
+                    additions=0, deletions=0,
+                    html_diff="",
+                    binary=True,
+                    hash_match=hash_match,
+                ))
+            else:  # "error"
+                results.append(DiffResult(
+                    match=m, ratio=0.0, additions=0, deletions=0,
+                    html_diff="", error="Could not decode one or both files",
+                ))
             continue
 
-        if not compare_comment:
-            text_a = strip_comments(text_a, ext, squash_blanks=squash_blanks)
-            text_b = strip_comments(text_b, ext, squash_blanks=squash_blanks)
+        if strip_comments:
+            text_a = _strip_comments_fn(text_a, ext, squash_blanks=squash_blanks)
+            text_b = _strip_comments_fn(text_b, ext, squash_blanks=squash_blanks)
 
         all_line_counts.append(text_a.count("\n") + 1)
         all_line_counts.append(text_b.count("\n") + 1)
@@ -493,11 +528,11 @@ def run_pipeline(
                 ln_col_width, max(all_line_counts) if all_line_counts else 0)
 
     # Generate HTML diffs with unified column width
-    for m_idx, m in enumerate(matches):
-        r = results[m_idx]
-        if r.error:
+    for r_idx, r in enumerate(results):
+        if r.error or r.binary:
             continue
 
+        m = r.match
         abs_a = str(root_a / m.rel_path_a)
         abs_b = str(root_b / m.rel_path_b)
         ext = Path(m.rel_path_a).suffix.lower()
@@ -506,9 +541,9 @@ def run_pipeline(
         text_b = read_file(abs_b, encoding=encoding)
         if text_a is None or text_b is None:
             continue
-        if not compare_comment:
-            text_a = strip_comments(text_a, ext, squash_blanks=squash_blanks)
-            text_b = strip_comments(text_b, ext, squash_blanks=squash_blanks)
+        if strip_comments:
+            text_a = _strip_comments_fn(text_a, ext, squash_blanks=squash_blanks)
+            text_b = _strip_comments_fn(text_b, ext, squash_blanks=squash_blanks)
 
         html_diff = generate_html_diff(
             text_a, text_b,
@@ -522,7 +557,7 @@ def run_pipeline(
             by_word=by_word,
             detect_moved=detect_moved,
         )
-        results[m_idx] = DiffResult(
+        results[r_idx] = DiffResult(
             match=r.match,
             ratio=r.ratio,
             additions=r.additions,
@@ -536,16 +571,16 @@ def run_pipeline(
     if sort_by:
         reverse = sort_order == "desc"
         if sort_by == "filename":
+            results.sort(
+                key=lambda r: PurePosixPath(r.match.rel_path_a).name.lower(),
+                reverse=reverse,
+            )
+        elif sort_by == "path":
             results.sort(key=lambda r: r.match.rel_path_a.lower(), reverse=reverse)
+        elif sort_by == "similarity":
+            results.sort(key=lambda r: r.match.similarity, reverse=reverse)
         elif sort_by == "ratio":
             results.sort(key=lambda r: r.ratio, reverse=reverse)
-        elif sort_by == "size":
-            def _file_size(r: DiffResult) -> int:
-                try:
-                    return os.path.getsize(str(root_a / r.match.rel_path_a))
-                except OSError:
-                    return 0
-            results.sort(key=_file_size, reverse=reverse)
         logger.info("  Sorted by %s (%s)", sort_by, sort_order)
 
     # Deep Compare (only in deep mode)
@@ -569,9 +604,10 @@ def run_pipeline(
         logger.info("Generating JSON report …")
         _generate_json_report(
             results, unmatched_a, unmatched_b,
-            dir_a, dir_b, by_word, compare_comment,
+            dir_a, dir_b, by_word, strip_comments,
             deep_results, report_json,
             metadata=metadata,
+            include_uncompared=include_uncompared,
         )
 
     # Markdown report
@@ -579,9 +615,10 @@ def run_pipeline(
         logger.info("Generating Markdown report …")
         _generate_markdown_report(
             results, unmatched_a, unmatched_b,
-            dir_a, dir_b, by_word, compare_comment,
+            dir_a, dir_b, by_word, strip_comments,
             deep_results, report_md,
             metadata=metadata,
+            include_uncompared=include_uncompared,
         )
 
     # HTML report
@@ -589,10 +626,11 @@ def run_pipeline(
         logger.info("Generating HTML report …")
         _generate_html_report(
             results, unmatched_a, unmatched_b,
-            dir_a, dir_b, by_word, compare_comment,
+            dir_a, dir_b, by_word, strip_comments,
             deep_results, report_html, ln_col_width,
             metadata=metadata,
             hash_table_html=hash_table_html,
+            include_uncompared=include_uncompared,
         )
 
     # PDF report
@@ -600,7 +638,7 @@ def run_pipeline(
         logger.info("Generating PDF report (divide-and-conquer) …")
         _generate_pdf_report(
             results, unmatched_a, unmatched_b,
-            dir_a, dir_b, by_word, compare_comment,
+            dir_a, dir_b, by_word, strip_comments,
             deep_results, report_pdf,
             no_merge=no_merge,
             show_page_number=show_page_number,
@@ -614,6 +652,7 @@ def run_pipeline(
             bates_prefix=bates_prefix,
             bates_suffix=bates_suffix,
             bates_start=bates_start,
+            include_uncompared=include_uncompared,
         )
 
     logger.info("Done (reports) ✓")
@@ -658,7 +697,7 @@ def _generate_pdf_report(
     dir_a: str,
     dir_b: str,
     by_word: bool,
-    compare_comment: bool,
+    strip_comments: bool,
     deep_results: list[DeepMatchResult] | None,
     output_pdf: str,
     *,
@@ -674,6 +713,7 @@ def _generate_pdf_report(
     bates_prefix: str = "",
     bates_suffix: str = "",
     bates_start: int = 1,
+    include_uncompared: bool = True,
 ) -> None:
     """Generate PDF report with divide-and-conquer merging."""
     if no_merge:
@@ -686,10 +726,11 @@ def _generate_pdf_report(
         # (1) Cover page
         cover_body = build_cover_body(
             results, unmatched_a, unmatched_b,
-            dir_a, dir_b, by_word, compare_comment,
+            dir_a, dir_b, by_word, strip_comments,
             deep_results=deep_results,
             metadata=metadata,
             hash_table_html=hash_table_html,
+            include_uncompared=include_uncompared,
         )
         cover_html = _html_wrap("Diffinite — Cover", cover_body)
         if no_merge:
@@ -700,9 +741,43 @@ def _generate_pdf_report(
         if cover_ok:
             logger.info("  Cover page → OK")
 
+        # ── Pre-flight: warn about large diffs that may slow PDF ────
+        # 500 KB of HTML ≈ 500+ source lines in side-by-side diff table.
+        # xhtml2pdf layout becomes noticeably slow above this threshold.
+        _LARGE_DIFF_BYTES = 500_000
+        large_files = [
+            (i, r) for i, r in enumerate(results, 1)
+            if not r.error and not r.binary and len(r.html_diff) > _LARGE_DIFF_BYTES
+        ]
+        if large_files:
+            logger.warning(
+                "⚠ %d file(s) have large diffs — PDF rendering may be "
+                "slow or hang:", len(large_files),
+            )
+            for i, r in large_files:
+                size_kb = len(r.html_diff) / 1024
+                logger.warning(
+                    "  %d. %s (%.0f KB HTML)",
+                    i, r.match.rel_path_a, size_kb,
+                )
+            logger.warning(
+                "  Consider: --collapse-identical (shrink diffs), "
+                "--no-merge (split PDFs), or --report-html (fast export)."
+            )
+
         # (2) Per-file diff pages
         diff_pdf_pairs: list[tuple[str, DiffResult]] = []
         for idx, r in enumerate(results, 1):
+            # Per-file warning for large diffs
+            if (not r.error and not r.binary
+                    and len(r.html_diff) > _LARGE_DIFF_BYTES):
+                size_kb = len(r.html_diff) / 1024
+                logger.warning(
+                    "⚠ Rendering PDF %d/%d (%s, %.0f KB) — "
+                    "this may take a while…",
+                    idx, len(results), r.match.rel_path_a, size_kb,
+                )
+
             diff_html = build_diff_page_html(
                 r, idx, unit,
                 show_page_number=show_page_number,

{diffinite-0.9.6 → diffinite-0.10.0/src/diffinite.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: diffinite
-Version: 0.9.6
+Version: 0.10.0
 Summary: Forensic source-code comparison tool — Winnowing fingerprints and professional PDF reports for IP litigation & code audit
 Author: nash-dir
 License: Apache-2.0

{diffinite-0.9.6 → diffinite-0.10.0}/tests/test_cli.py

@@ -69,7 +69,7 @@ class TestDeepCompareArgs:
             "--mode", "deep",
             "--k-gram", "5",
             "--window", "3",
-            "--threshold-deep", "0.20",
+            "--threshold-deep", "20",
         ])
 
 
@@ -84,7 +84,7 @@ class TestAnnotationsAndReportFlags:
             "-o", str(tmp_path / "out.pdf"),
             "--collapse-identical",
             "--page-number", "--file-number",
-            "--bates-number", "--show-filename",
+            "--bates-number", "--filename",
         ])
 
     def test_threshold_accepts_value(self, tmp_path):
@@ -107,3 +107,41 @@ class TestAnnotationsAndReportFlags:
         ])
         from pathlib import Path
         assert Path(json_path).exists()
+
+
+class TestIncludeUncomparedFlag:
+    """Verify --include-uncompared / --no-include-uncompared flags."""
+
+    def test_include_uncompared_default_true(self, tmp_path):
+        """Default behavior includes uncompared files."""
+        d_a = tmp_path / "a"; d_a.mkdir()
+        d_b = tmp_path / "b"; d_b.mkdir()
+        (d_a / "only_a.py").write_text("x = 1\n", encoding="utf-8")
+        json_path = str(tmp_path / "out.json")
+        main([
+            str(d_a), str(d_b),
+            "--report-json", json_path,
+        ])
+        import json
+        from pathlib import Path
+        data = json.loads(Path(json_path).read_text(encoding="utf-8"))
+        assert "only_a.py" in data["unmatched_a"]
+
+    def test_no_include_uncompared_excludes(self, tmp_path):
+        """--no-include-uncompared excludes unmatched file lists."""
+        d_a = tmp_path / "a"; d_a.mkdir()
+        d_b = tmp_path / "b"; d_b.mkdir()
+        (d_a / "only_a.py").write_text("x = 1\n", encoding="utf-8")
+        json_path = str(tmp_path / "out.json")
+        main([
+            str(d_a), str(d_b),
+            "--report-json", json_path,
+            "--no-include-uncompared",
+        ])
+        import json
+        from pathlib import Path
+        data = json.loads(Path(json_path).read_text(encoding="utf-8"))
+        assert data["unmatched_a"] == []
+        assert data["unmatched_b"] == []
+        # Summary counts should still show the real values
+        assert data["summary"]["unmatched_a_count"] == 1

{diffinite-0.9.6 → diffinite-0.10.0}/tests/test_pdf_gen.py

@@ -3,7 +3,7 @@
 import pytest
 
 from diffinite.models import DiffResult, FileMatch, DeepMatchResult
-from diffinite.pdf_gen import build_cover_body, build_diff_page_html
+from diffinite.pdf_gen import build_cover_body, build_diff_page_html, _break_path
 
 
 # ---------------------------------------------------------------------------
@@ -37,7 +37,7 @@ def _cover(results=None, *, deep_results=None):
         dir_a="dir_a",
         dir_b="dir_b",
         by_word=False,
-        compare_comment=True,
+        strip_comments=False,
         deep_results=deep_results,
     )
 
@@ -55,8 +55,9 @@ class TestBuildCoverHtml:
 
     def test_contains_file_names(self):
         html = _cover()
-        assert "handler.java" in html
-        assert "looper.java" in html
+        # _break_path inserts ​ after path separators (., /, \, _)
+        assert "handler." in html
+        assert "looper." in html
 
     def test_contains_ratio(self):
         html = _cover()
@@ -79,10 +80,10 @@ class TestBuildCoverHtml:
             dir_a="left",
             dir_b="right",
             by_word=False,
-            compare_comment=True,
+            strip_comments=False,
         )
-        assert "orphan_a.py" in html
-        assert "orphan_b.py" in html
+        # _break_path inserts ​ after separators, so check partial strings
+        assert "orphan" in html
 
     def test_deep_results_without_channels(self):
         deep = [
@@ -92,8 +93,8 @@ class TestBuildCoverHtml:
             ),
         ]
         html = _cover(deep_results=deep)
-        assert "foo.py" in html
-        assert "bar.py" in html
+        assert "foo." in html
+        assert "bar." in html
 
     def test_deep_results_display(self):
         deep = [
@@ -104,8 +105,8 @@ class TestBuildCoverHtml:
             ),
         ]
         html = _cover(deep_results=deep)
-        assert "foo.py" in html
-        assert "bar.py" in html
+        assert "foo." in html
+        assert "bar." in html
         assert "50" in html  # shared hashes
 
 
@@ -155,3 +156,67 @@ class TestBuildDiffPageHtml:
             show_filename=True,
         )
         assert "annotated.py" in html
+
+
+# ---------------------------------------------------------------------------
+# _break_path tests
+# ---------------------------------------------------------------------------
+class TestBreakPath:
+    """Verify _break_path inserts zero-width spaces at path separators."""
+
+    def test_slash(self):
+        result = _break_path("src/main/java")
+        assert "src/​main/​java" == result
+
+    def test_backslash(self):
+        result = _break_path("src\\main\\java")
+        assert "src\\​main\\​java" == result
+
+    def test_dot(self):
+        result = _break_path("handler.java")
+        assert "handler.​java" == result
+
+    def test_underscore(self):
+        result = _break_path("my_file_name")
+        assert "my_​file_​name" == result
+
+    def test_combined(self):
+        result = _break_path("src/com/example/my_handler.java")
+        assert "​" in result
+
+    def test_empty(self):
+        assert _break_path("") == ""
+
+
+# ---------------------------------------------------------------------------
+# include_uncompared tests
+# ---------------------------------------------------------------------------
+class TestIncludeUncompared:
+    """Verify include_uncompared parameter on build_cover_body."""
+
+    def test_excludes_unmatched_when_false(self):
+        html = build_cover_body(
+            _make_results(),
+            unmatched_a=["orphan_a.py"],
+            unmatched_b=["orphan_b.py"],
+            dir_a="left",
+            dir_b="right",
+            by_word=False,
+            strip_comments=False,
+            include_uncompared=False,
+        )
+        assert "orphan" not in html
+        assert "Unmatched Files" not in html
+
+    def test_includes_unmatched_by_default(self):
+        html = build_cover_body(
+            _make_results(),
+            unmatched_a=["orphan_a.py"],
+            unmatched_b=["orphan_b.py"],
+            dir_a="left",
+            dir_b="right",
+            by_word=False,
+            strip_comments=False,
+        )
+        assert "orphan" in html
+        assert "Unmatched Files" in html

{diffinite-0.9.6 → diffinite-0.10.0}/tests/test_pipeline.py

@@ -23,9 +23,9 @@ class TestPipelineE2E:
         run_pipeline(
             dir_a=EXAMPLE_LEFT,
             dir_b=EXAMPLE_RIGHT,
-            output_pdf=output,
+            report_pdf=output,
             by_word=False,
-            compare_comment=False,
+            strip_comments=True,
         )
         assert Path(output).exists()
         assert Path(output).stat().st_size > 0
@@ -37,7 +37,7 @@ class TestPipelineE2E:
         run_pipeline(
             dir_a=EXAMPLE_LEFT,
             dir_b=EXAMPLE_RIGHT,
-            output_pdf=output,
+            report_pdf=output,
             exec_mode="deep",
             workers=2,
             kgram_size=5,
@@ -53,7 +53,7 @@ class TestPipelineE2E:
         run_pipeline(
             dir_a=EXAMPLE_LEFT,
             dir_b=EXAMPLE_RIGHT,
-            output_pdf=output,
+            report_pdf=output,
             no_merge=True,
         )
         files_dir = tmp_path / "individual_files"