code2llm 0.5.172__tar.gz → 0.5.173__tar.gz

{code2llm-0.5.172/code2llm.egg-info → code2llm-0.5.173}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: code2llm
-Version: 0.5.172
+Version: 0.5.173
 Summary: High-performance Python code flow analysis with optimized TOON format - CFG, DFG, call graphs, and intelligent code queries
 Home-page: https://github.com/wronai/stts
 Author: Tom Sapletta
@@ -67,11 +67,11 @@ Dynamic: requires-python
 
 ## AI Cost Tracking
 
-![PyPI](https://img.shields.io/badge/pypi-costs-blue) ![Version](https://img.shields.io/badge/version-0.5.172-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-$17.33-orange) ![Human Time](https://img.shields.io/badge/Human%20Time-92.8h-blue) ![Model](https://img.shields.io/badge/Model-openrouter%2Fdeep%2Fdeep--v4--pro-lightgrey)
+![PyPI](https://img.shields.io/badge/pypi-costs-blue) ![Version](https://img.shields.io/badge/version-0.5.173-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-$17.73-orange) ![Human Time](https://img.shields.io/badge/Human%20Time-93.8h-blue) ![Model](https://img.shields.io/badge/Model-openrouter%2Fdeep%2Fdeep--v4--pro-lightgrey)
 
-- 🤖 **LLM usage:** $17.3260 (246 commits)
-- 👤 **Human dev:** ~$9280 (92.8h @ $100/h, 30min dedup)
+- 🤖 **LLM usage:** $17.7289 (247 commits)
+- 👤 **Human dev:** ~$9380 (93.8h @ $100/h, 30min dedup)
 
 Generated on 2026-07-06 using [openrouter/deep/deep-v4-pro](https://openrouter.ai/deep/deep-v4-pro)
 

{code2llm-0.5.172 → code2llm-0.5.173}/README.md

@@ -2,11 +2,11 @@
 
 ## AI Cost Tracking
 
-![PyPI](https://img.shields.io/badge/pypi-costs-blue) ![Version](https://img.shields.io/badge/version-0.5.172-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-$17.33-orange) ![Human Time](https://img.shields.io/badge/Human%20Time-92.8h-blue) ![Model](https://img.shields.io/badge/Model-openrouter%2Fdeep%2Fdeep--v4--pro-lightgrey)
+![PyPI](https://img.shields.io/badge/pypi-costs-blue) ![Version](https://img.shields.io/badge/version-0.5.173-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-$17.73-orange) ![Human Time](https://img.shields.io/badge/Human%20Time-93.8h-blue) ![Model](https://img.shields.io/badge/Model-openrouter%2Fdeep%2Fdeep--v4--pro-lightgrey)
 
-- 🤖 **LLM usage:** $17.3260 (246 commits)
-- 👤 **Human dev:** ~$9280 (92.8h @ $100/h, 30min dedup)
+- 🤖 **LLM usage:** $17.7289 (247 commits)
+- 👤 **Human dev:** ~$9380 (93.8h @ $100/h, 30min dedup)
 
 Generated on 2026-07-06 using [openrouter/deep/deep-v4-pro](https://openrouter.ai/deep/deep-v4-pro)
 

code2llm-0.5.173/VERSION

@@ -0,0 +1 @@
+0.5.173

{code2llm-0.5.172 → code2llm-0.5.173}/code2llm/__init__.py

@@ -8,7 +8,7 @@ Includes NLP Processing Pipeline for query normalization, intent matching,
 and entity resolution with multilingual support.
 """
 
-__version__ = "0.5.172"
+__version__ = "0.5.173"
 __author__ = "STTS Project"
 
 # Core analysis components (lightweight, always needed)

{code2llm-0.5.172 → code2llm-0.5.173}/code2llm/analysis/smells.py

@@ -131,18 +131,34 @@ class SmellDetector:
         return smells
 
     def _detect_data_clumps(self) -> List[CodeSmell]:
-        """Detect 3+ variables frequently passed together."""
+        """Detect 3+ variables frequently passed together.
+
+        Grouped by (file, argument-set), not argument-set alone: a common
+        constructor shape like `(self, project_root, config)` is a
+        conventional pattern reused independently by unrelated classes
+        across unrelated files (any two classes that each take a project
+        root and a config object) -- that's coincidence, not evidence the
+        two classes should share a bundled parameter object. Grouping by
+        bare argument-set alone falsely flagged such coincidences as data
+        clumps (reproduced: `wup.anomaly_detector.AnomalyDetector.__init__`
+        and `wup.testql_monitor.TestQLMonitor.__init__` -- two unrelated
+        classes in unrelated files -- flagged as clumped on `project_root,
+        self, config`). A genuine data clump -- args that should become one
+        object -- is a same-file (usually same-class) phenomenon; scoping
+        by file preserves that signal while dropping cross-file
+        coincidental-signature noise.
+        """
         smells = []
         # Simplified: find functions with same 3+ arguments
-        arg_sets = {}  # frozenset(args) -> List[func_names]
+        arg_sets = {}  # (file, frozenset(args)) -> List[func_names]
         for func_name, func_info in self.result.functions.items():
             if len(func_info.args) >= 3:
-                args = frozenset(func_info.args)
-                if args not in arg_sets:
-                    arg_sets[args] = []
-                arg_sets[args].append(func_name)
+                key = (func_info.file, frozenset(func_info.args))
+                if key not in arg_sets:
+                    arg_sets[key] = []
+                arg_sets[key].append(func_name)
 
-        for args, funcs in arg_sets.items():
+        for (file, args), funcs in arg_sets.items():
             if len(funcs) >= 2:
                 for func_name in funcs:
                     func_info = self.result.functions[func_name]

{code2llm-0.5.172 → code2llm-0.5.173}/code2llm/nlp/__init__.py

@@ -4,7 +4,7 @@ Provides query normalization, intent matching, and entity resolution
 with multilingual support and fuzzy matching.
 """
 
-__version__ = "0.5.172"
+__version__ = "0.5.173"
 
 from .pipeline import NLPPipeline
 from .normalization import QueryNormalizer

{code2llm-0.5.172 → code2llm-0.5.173/code2llm.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: code2llm
-Version: 0.5.172
+Version: 0.5.173
 Summary: High-performance Python code flow analysis with optimized TOON format - CFG, DFG, call graphs, and intelligent code queries
 Home-page: https://github.com/wronai/stts
 Author: Tom Sapletta
@@ -67,11 +67,11 @@ Dynamic: requires-python
 
 ## AI Cost Tracking
 
-![PyPI](https://img.shields.io/badge/pypi-costs-blue) ![Version](https://img.shields.io/badge/version-0.5.172-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-$17.33-orange) ![Human Time](https://img.shields.io/badge/Human%20Time-92.8h-blue) ![Model](https://img.shields.io/badge/Model-openrouter%2Fdeep%2Fdeep--v4--pro-lightgrey)
+![PyPI](https://img.shields.io/badge/pypi-costs-blue) ![Version](https://img.shields.io/badge/version-0.5.173-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-$17.73-orange) ![Human Time](https://img.shields.io/badge/Human%20Time-93.8h-blue) ![Model](https://img.shields.io/badge/Model-openrouter%2Fdeep%2Fdeep--v4--pro-lightgrey)
 
-- 🤖 **LLM usage:** $17.3260 (246 commits)
-- 👤 **Human dev:** ~$9280 (92.8h @ $100/h, 30min dedup)
+- 🤖 **LLM usage:** $17.7289 (247 commits)
+- 👤 **Human dev:** ~$9380 (93.8h @ $100/h, 30min dedup)
 
 Generated on 2026-07-06 using [openrouter/deep/deep-v4-pro](https://openrouter.ai/deep/deep-v4-pro)
 

{code2llm-0.5.172 → code2llm-0.5.173}/code2llm.egg-info/SOURCES.txt

@@ -202,5 +202,6 @@ tests/test_project_toon_export.py
 tests/test_prompt_engine.py
 tests/test_prompt_txt.py
 tests/test_refactoring_engine.py
+tests/test_smells_data_clumps.py
 tests/test_smells_shotgun_surgery.py
 tests/test_toon_v2.py

{code2llm-0.5.172 → code2llm-0.5.173}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "code2llm"
-version = "0.5.172"
+version = "0.5.173"
 description = "High-performance Python code flow analysis with optimized TOON format - CFG, DFG, call graphs, and intelligent code queries"
 readme = "README.md"
 requires-python = ">=3.10"

code2llm-0.5.173/tests/test_smells_data_clumps.py

@@ -0,0 +1,83 @@
+"""Tests for SmellDetector._detect_data_clumps.
+
+2026-07-06 incident (same root cause as the Shotgun Surgery fix): grouping
+functions by their argument-name set alone, ignoring file, meant a common
+constructor shape like `(self, project_root, config)` -- conventionally
+reused independently by unrelated classes across unrelated files -- falsely
+triggered "Data Clump". Reproduced live:
+`wup.anomaly_detector.AnomalyDetector.__init__` and
+`wup.testql_monitor.TestQLMonitor.__init__` are two unrelated classes in
+unrelated files, flagged as clumped on `project_root, self, config` purely
+because both happen to take a project root and a config object.
+"""
+
+from __future__ import annotations
+
+from code2llm.analysis.smells import SmellDetector
+from code2llm.core.models import AnalysisResult, FunctionInfo
+
+
+def _function(name: str, file: str, args: list[str], line: int = 1) -> FunctionInfo:
+    return FunctionInfo(
+        name=name, qualified_name=f"{file}::{name}", file=file, line=line, args=args
+    )
+
+
+def test_same_args_across_unrelated_files_is_not_a_data_clump():
+    """The core regression: two unrelated classes in unrelated files each
+    taking (self, project_root, config) must NOT be flagged."""
+    result = AnalysisResult()
+    result.functions["a::AnomalyDetector.__init__"] = _function(
+        "AnomalyDetector.__init__", "wup/anomaly_detector.py", ["self", "project_root", "config"]
+    )
+    result.functions["b::TestQLMonitor.__init__"] = _function(
+        "TestQLMonitor.__init__", "wup/testql_monitor.py", ["self", "project_root", "config"]
+    )
+
+    smells = SmellDetector(result)._detect_data_clumps()
+
+    assert smells == []
+
+
+def test_genuine_same_file_data_clump_still_detected():
+    """Two methods in the SAME file sharing a 3+ arg signature is the real
+    smell and must still be caught."""
+    result = AnalysisResult()
+    file = "core/session.py"
+    result.functions["core.session::start"] = _function(
+        "start", file, ["self", "user_id", "token", "scope"]
+    )
+    result.functions["core.session::refresh"] = _function(
+        "refresh", file, ["self", "user_id", "token", "scope"]
+    )
+
+    smells = SmellDetector(result)._detect_data_clumps()
+
+    assert len(smells) == 2
+    assert all(s.name == "Data Clump: self, user_id, token, scope"
+               or set(s.context["clump"]) == {"self", "user_id", "token", "scope"}
+               for s in smells)
+    assert all(s.file == file for s in smells)
+
+
+def test_single_function_with_matching_args_in_other_file_not_flagged():
+    """Only one function per file matches the arg-set -- no clump within
+    either file, and cross-file coincidence must not count."""
+    result = AnalysisResult()
+    result.functions["a::f"] = _function("f", "a.py", ["self", "x", "y", "z"])
+    result.functions["b::g"] = _function("g", "b.py", ["self", "x", "y", "z"])
+
+    smells = SmellDetector(result)._detect_data_clumps()
+
+    assert smells == []
+
+
+def test_fewer_than_three_args_never_counted():
+    result = AnalysisResult()
+    file = "core/session.py"
+    result.functions["core.session::start"] = _function("start", file, ["self", "user_id"])
+    result.functions["core.session::refresh"] = _function("refresh", file, ["self", "user_id"])
+
+    smells = SmellDetector(result)._detect_data_clumps()
+
+    assert smells == []

code2llm-0.5.172/VERSION

@@ -1 +0,0 @@
-0.5.172