diffinite 0.8.0__tar.gz → 0.9.1__tar.gz

{diffinite-0.8.0/src/diffinite.egg-info → diffinite-0.9.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: diffinite
-Version: 0.8.0
+Version: 0.9.1
 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-0.8.0 → diffinite-0.9.1}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "diffinite"
-version = "0.8.0"
+version = "0.9.1"
 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-0.8.0 → diffinite-0.9.1}/src/diffinite/cli.py

@@ -95,6 +95,34 @@ def main(argv: list[str] | None = None) -> None:
         default=FUZZY_THRESHOLD,
         help=f"Fuzzy matching threshold 0–100 (default: {FUZZY_THRESHOLD})",
     )
+    parser.add_argument(
+        "--encoding",
+        default="auto",
+        help=(
+            "Source file encoding. 'auto' (default) uses charset-normalizer "
+            "auto-detection with Korean-optimized fallback (utf-8 -> euc-kr -> cp949). "
+            "Specify an explicit encoding (e.g. euc-kr, utf-8, cp949, shift_jis, "
+            "gb2312) to force-decode all files with that encoding."
+        ),
+    )
+    parser.add_argument(
+        "--sort-by",
+        choices=["filename", "size", "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)."
+        ),
+    )
+    parser.add_argument(
+        "--sort-order",
+        choices=["asc", "desc"],
+        default="asc",
+        dest="sort_order",
+        help="Sort direction (default: asc). Only effective with --sort-by.",
+    )
 
     # ── Output modes ──────────────────────────────────────────────────
     parser.add_argument(
@@ -280,6 +308,9 @@ def main(argv: list[str] | None = None) -> None:
         autojunk=not args.no_autojunk,
     )
 
+    # Resolve encoding
+    encoding = args.encoding if args.encoding.lower() != "auto" else None
+
     run_pipeline(
         dir_a=args.dir_a,
         dir_b=args.dir_b,
@@ -313,6 +344,11 @@ def main(argv: list[str] | None = None) -> None:
         report_html=args.report_html,
         report_md=args.report_md,
         report_json=args.report_json,
+        # Encoding
+        encoding=encoding,
+        # Sorting
+        sort_by=args.sort_by,
+        sort_order=args.sort_order,
     )
 
 

{diffinite-0.8.0 → diffinite-0.9.1}/src/diffinite/differ.py

@@ -38,12 +38,17 @@ logger = logging.getLogger(__name__)
 # ──────────────────────────────────────────────────────────────────────
 # 인코딩 자동 감지 파일 리더
 # ──────────────────────────────────────────────────────────────────────
-def read_file(path: str) -> Optional[str]:
+def read_file(path: str, encoding: str | None = None) -> Optional[str]:
     """파일을 읽고 인코딩을 자동 감지하여 유니코드 문자열로 반환한다.
 
     ``charset_normalizer.from_bytes()``는 BOM, 통계적 분석 등을 조합하여
     UTF-8, EUC-KR, Shift_JIS 등 다양한 인코딩을 감지한다.
 
+    Args:
+        path: 파일 경로.
+        encoding: 인코딩 지정. ``None`` 또는 ``"auto"``이면 자동 감지.
+                  ``"euc-kr"``, ``"utf-8"`` 등 지정하면 해당 인코딩으로 강제 디코딩.
+
     Returns:
         디코딩된 텍스트. 빈 파일은 ``""``, 감지 실패 시 ``None``.
 
@@ -60,16 +65,33 @@ def read_file(path: str) -> Optional[str]:
     if not raw:
         return ""
 
-    result = from_bytes(raw).best()
-    if result is None:
-        logger.warning("Could not detect encoding for %s — skipping", path)
-        return None
+    # ── Manual encoding specified ────────────────────────────────
+    if encoding and encoding.lower() not in ("auto", ""):
+        try:
+            return raw.decode(encoding)
+        except (UnicodeDecodeError, LookupError) as exc:
+            logger.error("Decoding failed for %s with encoding %s: %s",
+                         path, encoding, exc)
+            return None
 
-    try:
-        return str(result)
-    except Exception as exc:  # noqa: BLE001
-        logger.error("Decoding failed for %s (%s): %s", path, result.encoding, exc)
-        return None
+    # ── Auto-detect with charset_normalizer ──────────────────────
+    result = from_bytes(raw).best()
+    if result is not None:
+        try:
+            return str(result)
+        except Exception as exc:  # noqa: BLE001
+            logger.warning("charset_normalizer decode failed for %s (%s): %s",
+                           path, result.encoding, exc)
+
+    # ── Fallback chain (Korean-optimized) ────────────────────────
+    for fallback_enc in ("utf-8", "euc-kr", "cp949"):
+        try:
+            return raw.decode(fallback_enc)
+        except (UnicodeDecodeError, LookupError):
+            continue
+
+    logger.warning("Could not detect encoding for %s — skipping", path)
+    return None
 
 
 # ──────────────────────────────────────────────────────────────────────

diffinite-0.9.1/src/diffinite/languages/data.py

@@ -0,0 +1,64 @@
+"""Data / query languages: SQL, YAML, TOML."""
+
+from diffinite.languages._spec import LangSpec
+from diffinite.languages._registry import register
+from diffinite.models import CommentSpec
+
+# ── SQL ───────────────────────────────────────────────────────────
+register(LangSpec(
+    name="SQL",
+    extensions=(".sql", ".ddl", ".dml", ".pks", ".pkb", ".plsql", ".tsql"),
+    comment=CommentSpec(line_markers=("--",), block_start="/*", block_end="*/"),
+    keywords=frozenset({
+        # DML
+        "select", "from", "where", "insert", "update", "delete", "merge",
+        "into", "values", "set", "returning",
+        # DDL
+        "create", "drop", "alter", "table", "index", "view", "schema",
+        "database", "sequence", "trigger", "procedure", "function",
+        "column", "constraint", "primary", "foreign", "key", "references",
+        "unique", "check", "default", "auto_increment", "identity",
+        "cascade", "restrict", "truncate", "rename", "replace",
+        # Joins
+        "join", "inner", "outer", "left", "right", "cross", "full",
+        "natural", "on", "using",
+        # Operators & conditions
+        "and", "or", "not", "in", "like", "ilike", "between",
+        "is", "null", "true", "false", "case", "when", "then", "else",
+        # Aggregates & ordering
+        "order", "by", "group", "having", "limit", "offset", "fetch",
+        "as", "distinct", "union", "intersect", "except", "all", "exists",
+        "count", "sum", "avg", "min", "max",
+        # Window functions
+        "over", "partition", "row_number", "rank", "dense_rank",
+        "lead", "lag", "first_value", "last_value", "ntile",
+        "rows", "range", "unbounded", "preceding", "following", "current",
+        # Transactions & control flow
+        "begin", "end", "commit", "rollback", "savepoint", "transaction",
+        # DCL
+        "grant", "revoke", "deny",
+        # Subquery & CTE
+        "with", "recursive",
+        # Types
+        "int", "integer", "varchar", "char", "text", "boolean", "date",
+        "timestamp", "decimal", "numeric", "float", "double", "blob", "clob",
+        # Procedural (PL/SQL, T-SQL)
+        "declare", "variable", "cursor", "open", "close", "fetch",
+        "loop", "while", "for", "if", "elseif", "elsif", "return",
+        "exec", "execute", "call", "raise", "exception", "handler",
+    }),
+))
+
+# ── YAML ──────────────────────────────────────────────────────────
+register(LangSpec(
+    name="YAML",
+    extensions=(".yaml", ".yml"),
+    comment=CommentSpec(line_markers=("#",), block_start=None, block_end=None),
+))
+
+# ── TOML ──────────────────────────────────────────────────────────
+register(LangSpec(
+    name="TOML",
+    extensions=(".toml",),
+    comment=CommentSpec(line_markers=("#",), block_start=None, block_end=None),
+))

{diffinite-0.8.0 → diffinite-0.9.1}/src/diffinite/pipeline.py

@@ -377,6 +377,11 @@ def run_pipeline(
     # Evidence integrity
     embed_hash: bool = False,
     bundle_path: str | None = None,
+    # Encoding
+    encoding: str | None = None,
+    # Sorting
+    sort_by: str | None = None,
+    sort_order: str = "asc",
 ) -> None:
     """Execute the full diff-to-report pipeline.
 
@@ -448,8 +453,8 @@ def run_pipeline(
         abs_b = str(root_b / m.rel_path_b)
         ext = Path(m.rel_path_a).suffix.lower()
 
-        text_a = read_file(abs_a)
-        text_b = read_file(abs_b)
+        text_a = read_file(abs_a, encoding=encoding)
+        text_b = read_file(abs_b, encoding=encoding)
 
         if text_a is None or text_b is None:
             results.append(DiffResult(
@@ -491,8 +496,8 @@ def run_pipeline(
         abs_b = str(root_b / m.rel_path_b)
         ext = Path(m.rel_path_a).suffix.lower()
 
-        text_a = read_file(abs_a)
-        text_b = read_file(abs_b)
+        text_a = read_file(abs_a, encoding=encoding)
+        text_b = read_file(abs_b, encoding=encoding)
         if text_a is None or text_b is None:
             continue
         if not compare_comment:
@@ -519,6 +524,22 @@ def run_pipeline(
 
     total_files = len(results)
 
+    # ── Sort results ──────────────────────────────────────────────
+    if sort_by:
+        reverse = sort_order == "desc"
+        if sort_by == "filename":
+            results.sort(key=lambda r: r.match.rel_path_a.lower(), 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)
     deep_results = None
     if exec_mode == "deep":

{diffinite-0.8.0 → diffinite-0.9.1/src/diffinite.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: diffinite
-Version: 0.8.0
+Version: 0.9.1
 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-0.8.0/src/diffinite/languages/data.py

@@ -1,36 +0,0 @@
-"""Data / query languages: SQL, YAML, TOML."""
-
-from diffinite.languages._spec import LangSpec
-from diffinite.languages._registry import register
-from diffinite.models import CommentSpec
-
-# ── SQL ───────────────────────────────────────────────────────────
-register(LangSpec(
-    name="SQL",
-    extensions=(".sql",),
-    comment=CommentSpec(line_markers=("--",), block_start="/*", block_end="*/"),
-    keywords=frozenset({
-        "select", "from", "where", "insert", "update", "delete",
-        "create", "drop", "alter", "table", "index", "view",
-        "join", "inner", "outer", "left", "right", "on",
-        "and", "or", "not", "in", "like", "between",
-        "order", "by", "group", "having", "limit", "offset",
-        "as", "distinct", "union", "all", "exists",
-        "null", "true", "false", "is",
-        "begin", "end", "commit", "rollback", "transaction",
-    }),
-))
-
-# ── YAML ──────────────────────────────────────────────────────────
-register(LangSpec(
-    name="YAML",
-    extensions=(".yaml", ".yml"),
-    comment=CommentSpec(line_markers=("#",), block_start=None, block_end=None),
-))
-
-# ── TOML ──────────────────────────────────────────────────────────
-register(LangSpec(
-    name="TOML",
-    extensions=(".toml",),
-    comment=CommentSpec(line_markers=("#",), block_start=None, block_end=None),
-))