redup 0.4.30__tar.gz → 0.4.32__tar.gz

{redup-0.4.30/src/redup.egg-info → redup-0.4.32}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: redup
-Version: 0.4.30
+Version: 0.4.32
 Summary: Code duplication analyzer and refactoring planner for LLMs
 Author-email: Tom Sapletta <tom@sapletta.com>
 License-Expression: Apache-2.0
@@ -76,7 +76,7 @@ Requires-Dist: datasketch>=1.6; extra == "lsh"
 Provides-Extra: semantic
 Requires-Dist: sentence-transformers>=2.0; extra == "semantic"
 Provides-Extra: intent
-Requires-Dist: intract>=0.5.6; extra == "intent"
+Requires-Dist: intract>=0.5.7; extra == "intent"
 Provides-Extra: dev
 Requires-Dist: pytest>=7.0; extra == "dev"
 Requires-Dist: pytest-cov>=4.0; extra == "dev"
@@ -94,18 +94,16 @@ Dynamic: license-file
 [![PyPI](https://img.shields.io/pypi/v/redup)](https://pypi.org/project/redup/)
 [![License: Apache-2.0](https://img.shields.io/badge/License-Apache%202.0-blue.svg)](https://opensource.org/licenses/Apache-2.0)
 [![Python](https://img.shields.io/badge/python-3.10%2B-blue.svg)](https://python.org)
-[![Version](https://img.shields.io/badge/version-0.4.30-green.svg)](https://pypi.org/project/redup/)
+[![Version](https://img.shields.io/badge/version-0.4.32-green.svg)](https://pypi.org/project/redup/)
 
 
 ## AI Cost Tracking
 
-![PyPI](https://img.shields.io/badge/pypi-costs-blue) ![Version](https://img.shields.io/badge/version-0.4.30-blue) ![Python](https://img.shields.io/badge/python-3.9+-blue) ![License](https://img.shields.io/badge/license-Apache--2.0-green)
-![AI Cost](https://img.shields.io/badge/AI%20Cost-$31.52-orange) ![Human Time](https://img.shields.io/badge/Human%20Time-26.1h-blue) ![Model](https://img.shields.io/badge/Model-openrouter%2Fqwen%2Fqwen3--coder--next-lightgrey)
+![AI Cost](https://img.shields.io/badge/AI%20Cost-$7.50-yellow) ![AI Model](https://img.shields.io/badge/AI%20Model-openrouter%2Fdeep%2Fdeep-v4-pro-lightgrey)
 
-- 🤖 **LLM usage:** $31.5200 (74 commits)
-- 👤 **Human dev:** ~$2609 (26.1h @ $100/h, 30min dedup)
+This project uses AI-generated code. Total cost: **$7.5000** with **79** AI commits.
 
-Generated on 2026-05-31 using [openrouter/qwen/qwen3-coder-next](https://openrouter.ai/qwen/qwen3-coder-next)
+Generated on 2026-06-29 using [openrouter/deep/deep-v4-pro](https://openrouter.ai/models/openrouter/deep/deep-v4-pro)
 
 ---
 

{redup-0.4.30 → redup-0.4.32}/README.md

@@ -5,18 +5,16 @@
 [![PyPI](https://img.shields.io/pypi/v/redup)](https://pypi.org/project/redup/)
 [![License: Apache-2.0](https://img.shields.io/badge/License-Apache%202.0-blue.svg)](https://opensource.org/licenses/Apache-2.0)
 [![Python](https://img.shields.io/badge/python-3.10%2B-blue.svg)](https://python.org)
-[![Version](https://img.shields.io/badge/version-0.4.30-green.svg)](https://pypi.org/project/redup/)
+[![Version](https://img.shields.io/badge/version-0.4.32-green.svg)](https://pypi.org/project/redup/)
 
 
 ## AI Cost Tracking
 
-![PyPI](https://img.shields.io/badge/pypi-costs-blue) ![Version](https://img.shields.io/badge/version-0.4.30-blue) ![Python](https://img.shields.io/badge/python-3.9+-blue) ![License](https://img.shields.io/badge/license-Apache--2.0-green)
-![AI Cost](https://img.shields.io/badge/AI%20Cost-$31.52-orange) ![Human Time](https://img.shields.io/badge/Human%20Time-26.1h-blue) ![Model](https://img.shields.io/badge/Model-openrouter%2Fqwen%2Fqwen3--coder--next-lightgrey)
+![AI Cost](https://img.shields.io/badge/AI%20Cost-$7.50-yellow) ![AI Model](https://img.shields.io/badge/AI%20Model-openrouter%2Fdeep%2Fdeep-v4-pro-lightgrey)
 
-- 🤖 **LLM usage:** $31.5200 (74 commits)
-- 👤 **Human dev:** ~$2609 (26.1h @ $100/h, 30min dedup)
+This project uses AI-generated code. Total cost: **$7.5000** with **79** AI commits.
 
-Generated on 2026-05-31 using [openrouter/qwen/qwen3-coder-next](https://openrouter.ai/qwen/qwen3-coder-next)
+Generated on 2026-06-29 using [openrouter/deep/deep-v4-pro](https://openrouter.ai/models/openrouter/deep/deep-v4-pro)
 
 ---
 

{redup-0.4.30 → redup-0.4.32}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "redup"
-version = "0.4.30"
+version = "0.4.32"
 description = "Code duplication analyzer and refactoring planner for LLMs"
 readme = "README.md"
 license = "Apache-2.0"
@@ -75,7 +75,7 @@ ast = [
 ]
 lsh = ["datasketch>=1.6"]
 semantic = ["sentence-transformers>=2.0"]
-intent = ["intract>=0.5.6"]
+intent = ["intract>=0.5.7"]
 dev = [
     "pytest>=7.0",
     "pytest-cov>=4.0",
@@ -99,9 +99,6 @@ Documentation = "https://github.com/semcod/redup#readme"
 [tool.setuptools.packages.find]
 where = ["src"]
 
-[tool.uv.sources]
-intract = { path = "../intract", editable = true }
-
 [tool.ruff]
 target-version = "py310"
 line-length = 100
@@ -123,7 +120,7 @@ addopts = "-v --tb=short"
 
 [tool.pfix]
 # Self-healing Python configuration
-model = "openrouter/qwen/qwen3-coder-next"
+model = "openrouter/deep/deep-v4-pro"
 auto_apply = true
 auto_install_deps = true
 auto_restart = false
@@ -142,7 +139,7 @@ deduplicate = true
 badge = true
 update_readme = true
 readme_path = "README.md"
-default_model = "openrouter/qwen/qwen3-coder-next"
+default_model = "openrouter/deep/deep-v4-pro"
 analysis_mode = "byok"
 full_history = true
 max_commits = 500

{redup-0.4.30 → redup-0.4.32}/src/redup/__init__.py

@@ -2,7 +2,7 @@
 
 from __future__ import annotations
 
-__version__ = "0.4.30"
+__version__ = "0.4.32"
 
 # Click compatibility shim for older typer versions
 # This must run before any typer imports

{redup-0.4.30 → redup-0.4.32}/src/redup/core/config.py

@@ -118,7 +118,7 @@ def config_to_scan_config(config: dict[str, Any], path: Path) -> ScanConfig:
         min_similarity=config.get("min_similarity", scan_config.get("min_similarity", 0.85)),
         include_tests=config.get("include_tests", scan_config.get("include_tests", False)),
         lsh_enabled=lsh_config.get("enabled", True),
-        lsh_min_lines=lsh_config.get("min_lines", 50),
+        lsh_min_lines=lsh_config.get("min_lines", 12),
         lsh_threshold=lsh_config.get("threshold", 0.8),
     )
 
@@ -147,7 +147,7 @@ include_tests = false
 # Enable LSH near-duplicate detection for large blocks
 enabled = true
 # Minimum block size for LSH analysis
-min_lines = 50
+min_lines = 12
 # Similarity threshold for LSH (0.0-1.0)
 threshold = 0.8
 

{redup-0.4.30 → redup-0.4.32}/src/redup/core/hasher.py

@@ -70,7 +70,7 @@ def _ast_to_normalized_string(tree: object) -> str:
         elif isinstance(node, _ast.Attribute):
             token = "ATTR"
         elif isinstance(node, _ast.Constant):
-            token = f"CONST_{type(node.value).__name__}"
+            token = "CONST"
         elif isinstance(node, _ast.BinOp):
             token = f"BINOP_{type(node.op).__name__}"
         elif isinstance(node, _ast.Compare):

{redup-0.4.30 → redup-0.4.32}/src/redup/core/lazy_grouper.py

@@ -37,12 +37,16 @@ def find_exact_duplicates_lazy(index: HashIndex, min_lines: int = 3) -> Iterator
 
 
 def find_structural_duplicates_lazy(
-    index: HashIndex, min_lines: int = 3
+    index: HashIndex,
+    min_lines: int = 3,
+    covered_locations: set[tuple[str, int]] | None = None,
 ) -> Iterator[DuplicateGroup]:
     """Find structural duplicate groups with lazy evaluation and early exit.
 
     Similar to find_exact_duplicates_lazy but for structural hashes.
     """
+    covered = covered_locations or set()
+
     for hash_val, blocks in index.structural.items():
         if len(blocks) < 2:
             continue
@@ -51,8 +55,18 @@ def find_structural_duplicates_lazy(
         if not _blocks_from_different_locations(blocks):
             continue
 
+        uncovered = [
+            hb
+            for hb in blocks
+            if (hb.block.file, hb.block.line_start) not in covered
+        ]
+        if len(uncovered) < 2:
+            continue
+        if not _blocks_from_different_locations(uncovered):
+            continue
+
         # Create group and check line threshold
-        group = _create_duplicate_group(hash_val, blocks, DuplicateType.STRUCTURAL)
+        group = _create_duplicate_group(hash_val, uncovered, DuplicateType.STRUCTURAL)
         if group.total_lines >= min_lines:
             yield group
 
@@ -106,11 +120,23 @@ def find_all_duplicates_lazy(
     Yields:
         DuplicateGroup objects that meet the threshold
     """
+    covered_locations: set[tuple[str, int]] = set()
+
     if include_exact:
-        yield from find_exact_duplicates_lazy(index, min_lines)
+        for group in find_exact_duplicates_lazy(index, min_lines):
+            for fragment in group.fragments:
+                covered_locations.add((fragment.file, fragment.line_start))
+            yield group
 
     if include_structural:
-        yield from find_structural_duplicates_lazy(index, min_lines)
+        for group in find_structural_duplicates_lazy(
+            index,
+            min_lines,
+            covered_locations=covered_locations,
+        ):
+            for fragment in group.fragments:
+                covered_locations.add((fragment.file, fragment.line_start))
+            yield group
 
 
 class DuplicateGroupCollector:

{redup-0.4.30 → redup-0.4.32}/src/redup/core/pipeline/duplicate_finder.py

@@ -17,6 +17,94 @@ from redup.core.models import DuplicateFragment, DuplicateGroup, DuplicateType,
 from redup.core.scanner_types import CodeBlock
 
 
+def find_fuzzy_groups(
+    all_blocks: list[CodeBlock],
+    config: ScanConfig,
+    covered_locations: set[tuple[str, int]] | None = None,
+) -> list[DuplicateGroup]:
+    """Find high-similarity function pairs missed by exact/structural hashing."""
+    from redup.core.matcher import sequence_similarity
+
+    covered = set(covered_locations or ())
+    min_lines = config.min_block_lines
+    min_similarity = config.min_similarity
+
+    candidates = [
+        block
+        for block in all_blocks
+        if block.function_name
+        and block.line_count >= min_lines
+        and (block.file, block.line_start) not in covered
+    ]
+    if len(candidates) < 2:
+        return []
+
+    candidates.sort(key=lambda block: (block.line_count, block.file, block.line_start))
+    used: set[tuple[str, int]] = set()
+    groups: list[DuplicateGroup] = []
+
+    for index, anchor in enumerate(candidates):
+        anchor_key = (anchor.file, anchor.line_start)
+        if anchor_key in used:
+            continue
+
+        cluster = [anchor]
+        for other in candidates[index + 1 :]:
+            other_key = (other.file, other.line_start)
+            if other_key in used:
+                continue
+            if abs(other.line_count - anchor.line_count) > 4:
+                break
+            if other.file == anchor.file and other.line_start == anchor.line_start:
+                continue
+            similarity = sequence_similarity(anchor.text, other.text)
+            if similarity < min_similarity:
+                continue
+            cluster.append(other)
+
+        if len(cluster) < 2:
+            continue
+
+        fragments = [
+            DuplicateFragment(
+                file=block.file,
+                line_start=block.line_start,
+                line_end=block.line_end,
+                text=block.text,
+                function_name=block.function_name,
+                class_name=block.class_name,
+            )
+            for block in cluster
+        ]
+        avg_similarity = sum(
+            sequence_similarity(cluster[0].text, block.text) for block in cluster[1:]
+        ) / (len(cluster) - 1)
+
+        groups.append(
+            DuplicateGroup(
+                id=f"F{len(groups) + 1:04d}",
+                duplicate_type=DuplicateType.FUZZY,
+                fragments=fragments,
+                similarity_score=avg_similarity,
+                normalized_hash=f"fuzzy_{anchor.file}_{anchor.line_start}",
+                normalized_name=anchor.function_name,
+            )
+        )
+        for block in cluster:
+            used.add((block.file, block.line_start))
+
+    return groups
+
+
+def _covered_locations(groups: list[DuplicateGroup]) -> set[tuple[str, int]]:
+    """Collect file/line locations already assigned to duplicate groups."""
+    covered: set[tuple[str, int]] = set()
+    for group in groups:
+        for fragment in group.fragments:
+            covered.add((fragment.file, fragment.line_start))
+    return covered
+
+
 def _finalize_duplicate_groups(
     groups: list[DuplicateGroup],
     all_blocks: list[CodeBlock],
@@ -25,6 +113,9 @@ def _finalize_duplicate_groups(
     cache: HashCache | None = None,
 ) -> list[DuplicateGroup]:
     """Attach near duplicates, sort by impact, and report timing."""
+    covered = _covered_locations(groups)
+    groups.extend(find_fuzzy_groups(all_blocks, config, covered))
+    covered = _covered_locations(groups)
     groups.extend(find_near_duplicate_groups(all_blocks, config))
     if getattr(config, "intent_enabled", False):
         groups.extend(find_intent_groups(all_blocks, config))

{redup-0.4.30 → redup-0.4.32}/src/redup/mcp/server.py

@@ -10,14 +10,18 @@ from redup import __version__
 from redup.mcp.handlers import TOOL_HANDLERS
 from redup.mcp.schemas import TOOL_SCHEMA_REDUP
 
+_PROTOCOL_VERSION = "2024-11-05"
+_NOTIFICATIONS = frozenset({"notifications/initialized", "notifications/cancelled"})
 
-def handle_initialize(request_id: Any) -> dict[str, Any]:
+
+def handle_initialize(request_id: Any, params: dict[str, Any] | None = None) -> dict[str, Any]:
     """Handle MCP initialize request."""
+    client_version = (params or {}).get("protocolVersion", _PROTOCOL_VERSION)
     return {
         "jsonrpc": "2.0",
         "id": request_id,
         "result": {
-            "protocolVersion": "0.1.0",
+            "protocolVersion": client_version,
             "serverInfo": {
                 "name": "redup",
                 "version": __version__,
@@ -86,8 +90,10 @@ def handle_request(request: dict[str, Any]) -> dict[str, Any]:
     params = request.get("params", {}) or {}
     request_id = request.get("id")
 
+    if method in _NOTIFICATIONS:
+        return {}
     if method == "initialize":
-        return handle_initialize(request_id)
+        return handle_initialize(request_id, params)
     if method == "tools/list":
         return handle_tools_list(request_id)
     if method == "tools/call":
@@ -117,7 +123,8 @@ def run_server() -> None:
         try:
             request = json.loads(line)
             response = handle_request(request)
-            print(json.dumps(response), flush=True)
+            if response:
+                print(json.dumps(response), flush=True)
         except json.JSONDecodeError as exc:
             error_response = {
                 "jsonrpc": "2.0",

{redup-0.4.30 → redup-0.4.32/src/redup.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: redup
-Version: 0.4.30
+Version: 0.4.32
 Summary: Code duplication analyzer and refactoring planner for LLMs
 Author-email: Tom Sapletta <tom@sapletta.com>
 License-Expression: Apache-2.0
@@ -76,7 +76,7 @@ Requires-Dist: datasketch>=1.6; extra == "lsh"
 Provides-Extra: semantic
 Requires-Dist: sentence-transformers>=2.0; extra == "semantic"
 Provides-Extra: intent
-Requires-Dist: intract>=0.5.6; extra == "intent"
+Requires-Dist: intract>=0.5.7; extra == "intent"
 Provides-Extra: dev
 Requires-Dist: pytest>=7.0; extra == "dev"
 Requires-Dist: pytest-cov>=4.0; extra == "dev"
@@ -94,18 +94,16 @@ Dynamic: license-file
 [![PyPI](https://img.shields.io/pypi/v/redup)](https://pypi.org/project/redup/)
 [![License: Apache-2.0](https://img.shields.io/badge/License-Apache%202.0-blue.svg)](https://opensource.org/licenses/Apache-2.0)
 [![Python](https://img.shields.io/badge/python-3.10%2B-blue.svg)](https://python.org)
-[![Version](https://img.shields.io/badge/version-0.4.30-green.svg)](https://pypi.org/project/redup/)
+[![Version](https://img.shields.io/badge/version-0.4.32-green.svg)](https://pypi.org/project/redup/)
 
 
 ## AI Cost Tracking
 
-![PyPI](https://img.shields.io/badge/pypi-costs-blue) ![Version](https://img.shields.io/badge/version-0.4.30-blue) ![Python](https://img.shields.io/badge/python-3.9+-blue) ![License](https://img.shields.io/badge/license-Apache--2.0-green)
-![AI Cost](https://img.shields.io/badge/AI%20Cost-$31.52-orange) ![Human Time](https://img.shields.io/badge/Human%20Time-26.1h-blue) ![Model](https://img.shields.io/badge/Model-openrouter%2Fqwen%2Fqwen3--coder--next-lightgrey)
+![AI Cost](https://img.shields.io/badge/AI%20Cost-$7.50-yellow) ![AI Model](https://img.shields.io/badge/AI%20Model-openrouter%2Fdeep%2Fdeep-v4-pro-lightgrey)
 
-- 🤖 **LLM usage:** $31.5200 (74 commits)
-- 👤 **Human dev:** ~$2609 (26.1h @ $100/h, 30min dedup)
+This project uses AI-generated code. Total cost: **$7.5000** with **79** AI commits.
 
-Generated on 2026-05-31 using [openrouter/qwen/qwen3-coder-next](https://openrouter.ai/qwen/qwen3-coder-next)
+Generated on 2026-06-29 using [openrouter/deep/deep-v4-pro](https://openrouter.ai/models/openrouter/deep/deep-v4-pro)
 
 ---
 

{redup-0.4.30 → redup-0.4.32}/src/redup.egg-info/requires.txt

@@ -61,7 +61,7 @@ pybloom-live>=4.0
 rapidfuzz>=3.0
 
 [intent]
-intract>=0.5.6
+intract>=0.5.7
 
 [llm]
 litellm>=1.0

{redup-0.4.30 → redup-0.4.32}/tests/test_hasher.py

@@ -33,6 +33,20 @@ def test_structural_hash_ignores_literals():
     assert hash_block_structural(a) == hash_block_structural(b)
 
 
+def test_structural_hash_env_var_readers():
+    a = '''def read_a() -> str:
+    raw = os.environ.get("VAR_A", "").strip()
+    if raw.isdigit():
+        return raw
+    return "28"'''
+    b = '''def read_b() -> str:
+    raw = os.environ.get("VAR_B", "").strip()
+    if raw.isdigit():
+        return raw
+    return "29"'''
+    assert hash_block_structural(a) == hash_block_structural(b)
+
+
 def test_structural_hash_different_structure():
     a = "x = 1\ny = 2"
     b = "for i in range(10):\n    print(i)"

{redup-0.4.30 → redup-0.4.32}/tests/test_mcp_server.py

@@ -80,6 +80,7 @@ def _analyze_test_project(root: Path):
 
 def test_initialize_and_tools_list() -> None:
     init_response = handle_request({"jsonrpc": "2.0", "method": "initialize", "id": 1})
+    assert init_response["result"]["protocolVersion"] == "2024-11-05"
     assert init_response["result"]["serverInfo"]["name"] == "redup"
     assert init_response["result"]["serverInfo"]["version"]
 

{redup-0.4.30 → redup-0.4.32}/tests/test_pipeline.py

@@ -4,8 +4,10 @@ import sqlite3
 import tempfile
 from pathlib import Path
 
-from redup.core.models import ScanConfig
+from redup.core.models import DuplicateType, ScanConfig
 from redup.core.pipeline import analyze, analyze_optimized
+from redup.core.pipeline.duplicate_finder import find_fuzzy_groups
+from redup.core.scanner import CodeBlock
 
 
 def _create_test_project(root: Path) -> None:
@@ -157,3 +159,39 @@ def test_analyze_optimized_stores_incremental_cache():
             file_count = conn.execute("SELECT COUNT(*) FROM file_hashes").fetchone()[0]
 
         assert file_count >= 1
+
+
+def test_find_fuzzy_groups_detects_renamed_env_readers():
+    blocks = [
+        CodeBlock(
+            file="a.py",
+            line_start=1,
+            line_end=6,
+            text=(
+                'def read_enter() -> str:\n'
+                '    raw = os.environ.get("ENTER", "").strip()\n'
+                '    if raw.isdigit():\n'
+                '        return raw\n'
+                '    return "28"\n'
+            ),
+            function_name="read_enter",
+        ),
+        CodeBlock(
+            file="b.py",
+            line_start=1,
+            line_end=6,
+            text=(
+                'def read_ctrl() -> str:\n'
+                '    raw = os.environ.get("CTRL", "").strip()\n'
+                '    if raw.isdigit():\n'
+                '        return raw\n'
+                '    return "29"\n'
+            ),
+            function_name="read_ctrl",
+        ),
+    ]
+    config = ScanConfig(min_block_lines=3, min_similarity=0.85)
+    groups = find_fuzzy_groups(blocks, config)
+    assert len(groups) == 1
+    assert groups[0].duplicate_type == DuplicateType.FUZZY
+    assert groups[0].occurrences == 2