code2llm 0.5.144__tar.gz → 0.5.146__tar.gz

{code2llm-0.5.144/code2llm.egg-info → code2llm-0.5.146}/PKG-INFO

@@ -1,11 +1,11 @@
 Metadata-Version: 2.4
 Name: code2llm
-Version: 0.5.144
+Version: 0.5.146
 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: STTS Project
 Author-email: Tom Sapletta <tom@sapletta.com>
-License-Expression: Apache-2.0
+License: Apache-2.0
 Project-URL: Homepage, https://github.com/wronai/stts
 Project-URL: Repository, https://github.com/wronai/stts
 Project-URL: Issues, https://github.com/wronai/stts/issues
@@ -54,7 +54,7 @@ Requires-Dist: pytest-xdist>=3.0; extra == "dev"
 Requires-Dist: black>=21.0; extra == "dev"
 Requires-Dist: flake8>=3.9; extra == "dev"
 Requires-Dist: mypy>=0.910; extra == "dev"
-Requires-Dist: goal>=2.1.0; extra == "dev"
+Requires-Dist: goal>=2.1.218; extra == "dev"
 Requires-Dist: costs>=0.1.20; extra == "dev"
 Requires-Dist: pfix>=0.1.60; extra == "dev"
 Dynamic: author
@@ -67,18 +67,16 @@ Dynamic: requires-python
 
 ## AI Cost Tracking
 
-![PyPI](https://img.shields.io/badge/pypi-costs-blue) ![Version](https://img.shields.io/badge/version-0.5.144-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-$7.50-orange) ![Human Time](https://img.shields.io/badge/Human%20Time-70.2h-blue) ![Model](https://img.shields.io/badge/Model-openrouter%2Fqwen%2Fqwen3--coder--next-lightgrey)
+![PyPI](https://img.shields.io/badge/pypi-costs-blue) ![Version](https://img.shields.io/badge/version-0.5.146-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-$7.50-orange) ![Human Time](https://img.shields.io/badge/Human%20Time-72.2h-blue) ![Model](https://img.shields.io/badge/Model-openrouter%2Fqwen%2Fqwen3--coder--next-lightgrey)
 
-- 🤖 **LLM usage:** $7.5000 (196 commits)
-- 👤 **Human dev:** ~$7024 (70.2h @ $100/h, 30min dedup)
+- 🤖 **LLM usage:** $7.5000 (198 commits)
+- 👤 **Human dev:** ~$7224 (72.2h @ $100/h, 30min dedup)
 
-Generated on 2026-04-25 using [openrouter/qwen/qwen3-coder-next](https://openrouter.ai/qwen/qwen3-coder-next)
+Generated on 2026-05-06 using [openrouter/qwen/qwen3-coder-next](https://openrouter.ai/qwen/qwen3-coder-next)
 
 ---
 
-
-
 This directory contains the complete analysis of your project generated by `code2llm`. Each file serves a specific purpose for understanding, refactoring, and documenting your codebase.
 
 ## 📁 Generated Files Overview
@@ -407,10 +405,10 @@ code2llm ./ -f yaml --separate-orphans
 ---
 
 **Generated by**: `code2llm ./ -f all --readme`  
-**Analysis Date**: 2026-04-25  
-**Total Functions**: 3085  
-**Total Classes**: 259  
-**Modules**: 498  
+**Analysis Date**: 2026-05-06  
+**Total Functions**: 3625  
+**Total Classes**: 250  
+**Modules**: 492  
 
 For more information about code2llm, visit: https://github.com/tom-sapletta/code2llm
 

{code2llm-0.5.144 → code2llm-0.5.146}/README.md

@@ -3,18 +3,16 @@
 
 ## AI Cost Tracking
 
-![PyPI](https://img.shields.io/badge/pypi-costs-blue) ![Version](https://img.shields.io/badge/version-0.5.144-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-$7.50-orange) ![Human Time](https://img.shields.io/badge/Human%20Time-70.2h-blue) ![Model](https://img.shields.io/badge/Model-openrouter%2Fqwen%2Fqwen3--coder--next-lightgrey)
+![PyPI](https://img.shields.io/badge/pypi-costs-blue) ![Version](https://img.shields.io/badge/version-0.5.146-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-$7.50-orange) ![Human Time](https://img.shields.io/badge/Human%20Time-72.2h-blue) ![Model](https://img.shields.io/badge/Model-openrouter%2Fqwen%2Fqwen3--coder--next-lightgrey)
 
-- 🤖 **LLM usage:** $7.5000 (196 commits)
-- 👤 **Human dev:** ~$7024 (70.2h @ $100/h, 30min dedup)
+- 🤖 **LLM usage:** $7.5000 (198 commits)
+- 👤 **Human dev:** ~$7224 (72.2h @ $100/h, 30min dedup)
 
-Generated on 2026-04-25 using [openrouter/qwen/qwen3-coder-next](https://openrouter.ai/qwen/qwen3-coder-next)
+Generated on 2026-05-06 using [openrouter/qwen/qwen3-coder-next](https://openrouter.ai/qwen/qwen3-coder-next)
 
 ---
 
-
-
 This directory contains the complete analysis of your project generated by `code2llm`. Each file serves a specific purpose for understanding, refactoring, and documenting your codebase.
 
 ## 📁 Generated Files Overview
@@ -343,10 +341,10 @@ code2llm ./ -f yaml --separate-orphans
 ---
 
 **Generated by**: `code2llm ./ -f all --readme`  
-**Analysis Date**: 2026-04-25  
-**Total Functions**: 3085  
-**Total Classes**: 259  
-**Modules**: 498  
+**Analysis Date**: 2026-05-06  
+**Total Functions**: 3625  
+**Total Classes**: 250  
+**Modules**: 492  
 
 For more information about code2llm, visit: https://github.com/tom-sapletta/code2llm
 

code2llm-0.5.146/VERSION

@@ -0,0 +1 @@
+0.5.146

{code2llm-0.5.144 → code2llm-0.5.146}/code2llm/__init__.py

@@ -8,7 +8,7 @@ Includes NLP Processing Pipeline for query normalization, intent matching,
 and entity resolution with multilingual support.
 """
 
-__version__ = "0.5.144"
+__version__ = "0.5.146"
 __author__ = "STTS Project"
 
 # Core analysis components (lightweight, always needed)

{code2llm-0.5.144 → code2llm-0.5.146}/code2llm/cli_analysis.py

@@ -88,6 +88,20 @@ def _build_config(args, output_dir: Path):
     config.watch = getattr(args, 'watch', False)
     # Dry-run mode (handled in orchestrator, but stored for reference)
     config.dry_run = getattr(args, 'dry_run', False)
+
+    # --fast: skip expensive analyses (vulture, centrality, DFG, communities)
+    if getattr(args, 'fast', False):
+        config.performance.fast_mode = True
+        config.performance.apply_fast_mode()
+
+    # Strategy-based performance tuning
+    strategy = getattr(args, 'strategy', 'standard')
+    if strategy == 'quick':
+        config.performance.skip_data_flow = True
+        config.performance.skip_dead_code_detection = True
+        config.performance.skip_centrality = True
+        config.performance.skip_community_detection = True
+
     return config
 
 

{code2llm-0.5.144 → code2llm-0.5.146}/code2llm/cli_exports/formats.py

@@ -2,6 +2,7 @@
 
 import os
 import sys
+import time
 from pathlib import Path
 from typing import Optional
 
@@ -13,6 +14,7 @@ from code2llm.exporters import (
     ArticleViewGenerator, HTMLDashboardGenerator,
     load_project_yaml, IndexHTMLGenerator,
 )
+from .orchestrator import _inject_generation_time as _inject_time
 
 
 def _export_evolution(args, result, output_dir: Path):
@@ -21,9 +23,12 @@ def _export_evolution(args, result, output_dir: Path):
         return
     exporter = EvolutionExporter()
     filepath = output_dir / 'evolution.toon.yaml'
+    t0 = time.monotonic()
     exporter.export(result, str(filepath))
+    elapsed = time.monotonic() - t0
+    _inject_time(filepath, elapsed)
     if args.verbose:
-        print(f"  - EVOLUTION (refactoring queue): {filepath}")
+        print(f"  - EVOLUTION (refactoring queue): {filepath} ({elapsed:.2f}s)")
 
 
 def _export_data_structures(args, result, output_dir: Path):
@@ -43,9 +48,12 @@ def _export_context_fallback(args, result, output_dir: Path, formats: list):
         return
     exporter = ContextExporter()
     filepath = output_dir / 'context.md'
+    t0 = time.monotonic()
     exporter.export(result, str(filepath))
+    elapsed = time.monotonic() - t0
+    _inject_time(filepath, elapsed)
     if args.verbose:
-        print(f"  - CONTEXT (LLM narrative): {filepath}")
+        print(f"  - CONTEXT (LLM narrative): {filepath} ({elapsed:.2f}s)")
 
 
 def _export_readme(args, result, output_dir: Path):
@@ -54,18 +62,24 @@ def _export_readme(args, result, output_dir: Path):
         return
     exporter = READMEExporter()
     filepath = output_dir / 'README.md'
+    t0 = time.monotonic()
     exporter.export(result, str(filepath))
+    elapsed = time.monotonic() - t0
+    _inject_time(filepath, elapsed)
     if args.verbose:
-        print(f"  - README (documentation): {filepath}")
+        print(f"  - README (documentation): {filepath} ({elapsed:.2f}s)")
 
 
 def _export_project_yaml(args, result, output_dir: Path):
     """Export unified project.yaml — single source of truth."""
     exporter = ProjectYAMLExporter()
     filepath = output_dir / 'project.yaml'
+    t0 = time.monotonic()
     exporter.export(result, str(filepath))
+    elapsed = time.monotonic() - t0
+    _inject_time(filepath, elapsed)
     if getattr(args, 'verbose', False):
-        print(f"  - PROJECT-YAML (single source of truth): {filepath}")
+        print(f"  - PROJECT-YAML (single source of truth): {filepath} ({elapsed:.2f}s)")
     return filepath
 
 
@@ -79,10 +93,13 @@ def _export_project_toon(args, result, output_dir: Path):
 
     exporter = ToonViewGenerator()
     filepath = output_dir / 'project.toon.yaml'
+    t0 = time.monotonic()
     exporter.generate(data, str(filepath))
+    elapsed = time.monotonic() - t0
+    _inject_time(filepath, elapsed)
 
     if getattr(args, 'verbose', False):
-        print(f"  - PROJECT-TOON (project overview): {filepath}")
+        print(f"  - PROJECT-TOON (project overview): {filepath} ({elapsed:.2f}s)")
 
     return filepath
 
@@ -129,9 +146,12 @@ def _export_simple_formats(args, result, output_dir: Path, formats):
             exporter = exporter_cls()
             # Export as plain text TOON format with .toon.yaml extension
             filepath = output_dir / filename
+            t0 = time.monotonic()
             exporter.export(result, str(filepath))
+            elapsed = time.monotonic() - t0
+            _inject_time(filepath, elapsed)
             if args.verbose:
-                print(f"  - {label}: {filepath}")
+                print(f"  - {label}: {filepath} ({elapsed:.2f}s)")
 
     # Unified project.yaml (single source of truth)
     if 'project-yaml' in formats:

{code2llm-0.5.144 → code2llm-0.5.146}/code2llm/cli_exports/orchestrator.py

@@ -7,6 +7,7 @@ Maintains backward compatibility with all existing --format values.
 import os
 import shutil
 import sys
+import time
 from pathlib import Path
 from typing import Optional, List, Dict, Any
 
@@ -280,9 +281,12 @@ def _export_registry_formats(args, result, output_dir: Path, formats: List[str])
         kwargs = _get_format_kwargs(fmt, args)
 
         try:
+            t0 = time.monotonic()
             exporter.export(result, str(filepath), **kwargs)
+            elapsed = time.monotonic() - t0
+            _inject_generation_time(filepath, elapsed)
             if args.verbose:
-                print(f"  - {label}: {filepath}")
+                print(f"  - {label}: {filepath} ({elapsed:.2f}s)")
         except Exception as e:
             if args.verbose:
                 print(f"  - {label} export failed: {e}", file=sys.stderr)
@@ -306,6 +310,53 @@ def _export_chunked(
     _chunked_impl(args, result, output_dir, source_path, formats, requested_formats)
 
 
+def _inject_generation_time(filepath: Path, elapsed: float) -> None:
+    """Inject generation time comment into the second line of an exported file."""
+    try:
+        path = Path(filepath)
+        if not path.exists():
+            return
+        suffix = path.suffix.lower()
+        name = path.name.lower()
+
+        # Only inject into text-based files
+        if suffix not in ('.yaml', '.yml', '.md', '.txt', '.mmd', '.html', '.json', '.export'):
+            return
+
+        content = path.read_text(encoding='utf-8')
+        if not content:
+            return
+
+        tag = f"generated in {elapsed:.2f}s"
+
+        if suffix in ('.yaml', '.yml', '.mmd', '.export', '.txt') or name.endswith('.toon.yaml'):
+            # YAML/Mermaid/text: insert '# generated in X.XXs' after first line
+            lines = content.split('\n', 1)
+            if len(lines) == 2:
+                content = f"{lines[0]}\n# {tag}\n{lines[1]}"
+            else:
+                content = f"{lines[0]}\n# {tag}\n"
+        elif suffix == '.md':
+            # Markdown: insert HTML comment after first line
+            lines = content.split('\n', 1)
+            if len(lines) == 2:
+                content = f"{lines[0]}\n<!-- {tag} -->\n{lines[1]}"
+            else:
+                content = f"{lines[0]}\n<!-- {tag} -->\n"
+        elif suffix == '.html':
+            # HTML: insert comment after <!DOCTYPE or <html>
+            content = content.replace('\n', f'\n<!-- {tag} -->\n', 1)
+        elif suffix == '.json':
+            # JSON doesn't support comments — skip
+            return
+        else:
+            return
+
+        path.write_text(content, encoding='utf-8')
+    except Exception:
+        pass  # Never fail the export pipeline for a comment
+
+
 # Backward-compatible aliases
 _export_single_project = _export_single
 _export_chunked_results = _export_chunked

{code2llm-0.5.144 → code2llm-0.5.146}/code2llm/cli_exports/orchestrator_handlers.py

@@ -4,6 +4,7 @@ This module contains all the individual export handler functions
 that were extracted from orchestrator.py to reduce its size.
 """
 
+import time
 from pathlib import Path
 from typing import Any, Dict, List
 
@@ -20,10 +21,14 @@ from .orchestrator_constants import FORMAT_LABELS
 
 def _export_mermaid(args, result, output_dir: Path):
     """Export mermaid diagrams."""
+    from .orchestrator import _inject_generation_time
+
     exporter = MermaidExporter()
     include_examples = getattr(args, 'flow_include_examples', False)
 
     # Core diagrams
+    mmd_files = ['flow.mmd', 'calls.mmd', 'compact_flow.mmd']
+    t0 = time.monotonic()
     exporter.export_flow_compact(result, str(output_dir / 'flow.mmd'), include_examples)
     exporter.export_call_graph(result, str(output_dir / 'calls.mmd'))
     exporter.export_compact(result, str(output_dir / 'compact_flow.mmd'))
@@ -31,21 +36,27 @@ def _export_mermaid(args, result, output_dir: Path):
     # Optional detailed diagrams
     if getattr(args, 'flow_detail', False):
         exporter.export_flow_detailed(result, str(output_dir / 'flow_detailed.mmd'), include_examples)
+        mmd_files.append('flow_detailed.mmd')
     if getattr(args, 'flow_full', False):
         exporter.export_flow_full(result, str(output_dir / 'flow_full.mmd'), include_examples)
+        mmd_files.append('flow_full.mmd')
+    elapsed_mmd = time.monotonic() - t0
+
+    # Inject timing into each .mmd file
+    for mf in mmd_files:
+        _inject_generation_time(output_dir / mf, elapsed_mmd)
 
     # Also export calls.yaml/toon
     yaml_exporter = YAMLExporter()
+    t0 = time.monotonic()
     yaml_exporter.export_calls(result, str(output_dir / 'calls.yaml'))
     yaml_exporter.export_calls_toon(result, str(output_dir / 'calls.toon.yaml'))
+    elapsed_calls = time.monotonic() - t0
+    _inject_generation_time(output_dir / 'calls.toon.yaml', elapsed_calls)
 
     if args.verbose:
-        files = ['flow.mmd', 'calls.mmd', 'compact_flow.mmd', 'calls.yaml']
-        if getattr(args, 'flow_detail', False):
-            files.append('flow_detailed.mmd')
-        if getattr(args, 'flow_full', False):
-            files.append('flow_full.mmd')
-        print(f"  - Mermaid: {output_dir}/*.mmd ({', '.join(files)})")
+        files = mmd_files + ['calls.yaml']
+        print(f"  - Mermaid: {output_dir}/*.mmd ({', '.join(files)}) ({elapsed_mmd:.2f}s)")
 
     # PNG generation
     _export_mermaid_pngs(args, output_dir)
@@ -99,28 +110,38 @@ def _export_data_structures(args, result, output_dir: Path):
 def _export_project_toon(args, result, output_dir: Path):
     """Export project.toon.yaml from project.yaml data."""
     from ..exporters.project_yaml_exporter import ProjectYAMLExporter
+    from .orchestrator import _inject_generation_time
 
     project_yaml_exporter = ProjectYAMLExporter()
     prev_evolution = load_previous_evolution(output_dir / 'project.yaml')
     data = project_yaml_exporter._build_project_yaml(result, prev_evolution)
 
     generator = ToonViewGenerator()
-    generator.generate(data, str(output_dir / 'project.toon.yaml'))
+    filepath = output_dir / 'project.toon.yaml'
+    t0 = time.monotonic()
+    generator.generate(data, str(filepath))
+    elapsed = time.monotonic() - t0
+    _inject_generation_time(filepath, elapsed)
 
     if args.verbose:
-        print(f"  - PROJECT-TOON (project overview): {output_dir / 'project.toon.yaml'}")
+        print(f"  - PROJECT-TOON (project overview): {filepath} ({elapsed:.2f}s)")
 
 
 def _export_readme(args, result, output_dir: Path):
     """Export README.md."""
     if getattr(args, 'no_readme', False):
         return
+    from .orchestrator import _inject_generation_time
     exporter_cls = get_exporter('readme')
     if exporter_cls:
         exporter = exporter_cls()
-        exporter.export(result, str(output_dir / 'README.md'))
+        filepath = output_dir / 'README.md'
+        t0 = time.monotonic()
+        exporter.export(result, str(filepath))
+        elapsed = time.monotonic() - t0
+        _inject_generation_time(filepath, elapsed)
         if args.verbose:
-            print(f"  - README (documentation): {output_dir / 'README.md'}")
+            print(f"  - README (documentation): {filepath} ({elapsed:.2f}s)")
 
 
 def _export_index_html(args, output_dir: Path):
@@ -128,10 +149,14 @@ def _export_index_html(args, output_dir: Path):
     if 'all' not in getattr(args, 'format', ''):
         return
     try:
+        from .orchestrator import _inject_generation_time
         generator = IndexHTMLGenerator(output_dir)
+        t0 = time.monotonic()
         index_path = generator.generate()
+        elapsed = time.monotonic() - t0
+        _inject_generation_time(index_path, elapsed)
         if args.verbose:
-            print(f"  - INDEX (file browser): {index_path}")
+            print(f"  - INDEX (file browser): {index_path} ({elapsed:.2f}s)")
     except Exception as e:
         if args.verbose:
             print(f"  - INDEX generation failed: {e}", file=__import__('sys').stderr)

{code2llm-0.5.144 → code2llm-0.5.146}/code2llm/cli_exports/prompt.py

@@ -1,6 +1,7 @@
 """Prompt generation — prompt.txt for LLM consumption (regular and chunked)."""
 
 import sys
+import time
 from pathlib import Path
 from typing import List, Optional, Tuple
 
@@ -30,9 +31,13 @@ def _export_prompt_txt(args, output_dir: Path, formats: list[str], source_path:
     lines.extend(_build_prompt_footer(chunked=False, file_analysis=file_analysis))
 
     prompt_path = output_dir / 'prompt.txt'
+    t0 = time.monotonic()
     prompt_path.write_text("\n".join(lines) + "\n", encoding='utf-8')
+    elapsed = time.monotonic() - t0
+    from .orchestrator import _inject_generation_time
+    _inject_generation_time(prompt_path, elapsed)
     if args.verbose:
-        print(f"  - PROMPT: {prompt_path}")
+        print(f"  - PROMPT: {prompt_path} ({elapsed:.2f}s)")
 
 
 def _export_chunked_prompt_txt(args, output_dir: Path, formats: list[str], source_path: Optional[Path] = None, subprojects: list = None) -> None:

{code2llm-0.5.144 → code2llm-0.5.146}/code2llm/cli_parser.py

@@ -159,6 +159,12 @@ Strategy Options (--strategy):
         help='Auto-detect changed files and only re-analyze those (faster subsequent runs)'
     )
 
+    parser.add_argument(
+        '--fast',
+        action='store_true',
+        help='Skip expensive analyses (vulture, centrality, DFG, communities) for faster runs'
+    )
+
     parser.add_argument(
         '--strategy',
         choices=['quick', 'standard', 'deep'],

{code2llm-0.5.144 → code2llm-0.5.146}/code2llm/core/__init__.py

@@ -9,7 +9,7 @@ from .models import (
 __all__ = [
     'ProjectAnalyzer',
     'StreamingAnalyzer',
-    'IncrementalAnalyzer',
+    'StreamingIncrementalAnalyzer',
     'ScanStrategy',
     'SmartPrioritizer',
     'STRATEGY_QUICK',
@@ -43,7 +43,7 @@ def __getattr__(name):
         return locals()[name]
     
     _streaming_names = {
-        'StreamingAnalyzer', 'IncrementalAnalyzer', 'ScanStrategy',
+        'StreamingAnalyzer', 'StreamingIncrementalAnalyzer', 'ScanStrategy',
         'SmartPrioritizer', 'STRATEGY_QUICK', 'STRATEGY_STANDARD', 'STRATEGY_DEEP',
     }
     if name in _streaming_names:

{code2llm-0.5.144 → code2llm-0.5.146}/code2llm/core/analyzer.py

@@ -121,10 +121,16 @@ class ProjectAnalyzer:
             return None, [], files
 
     def _run_analysis(self, files_to_analyze: List[Tuple[str, str]]) -> List[Dict]:
-        """Analyze files in parallel or sequentially depending on config."""
+        """Analyze files in parallel or sequentially depending on config.
+
+        Parallel mode has significant startup/pickle overhead.  It is only
+        beneficial for larger file sets (threshold: 30 files).
+        """
         if not files_to_analyze:
             return []
-        if self.config.performance.parallel_enabled and len(files_to_analyze) > 1:
+        _PARALLEL_THRESHOLD = 30
+        if (self.config.performance.parallel_enabled
+                and len(files_to_analyze) > _PARALLEL_THRESHOLD):
             return self._analyze_parallel(files_to_analyze)
         return self._analyze_sequential(files_to_analyze)
 
@@ -197,39 +203,61 @@ class ProjectAnalyzer:
         if self.config.verbose:
             self._print_summary(merged)
     
+    @staticmethod
+    def _should_collect_file(
+        filename: str,
+        ext_set: set,
+        filename_set_lower: frozenset,
+        filename_prefixes_lower: tuple,
+    ) -> bool:
+        """Return True if *filename* matches a known language extension,
+        well-known filename (Dockerfile, Makefile, ...), or filename prefix
+        (Dockerfile.dev, Makefile.am).
+        """
+        filename_lower = filename.lower()
+        suffix = os.path.splitext(filename)[1].lower()
+        return (
+            suffix in ext_set
+            or filename_lower in filename_set_lower
+            or filename_lower.startswith(filename_prefixes_lower)
+        )
+
+    @staticmethod
+    def _compute_module_name(
+        rel: str, filename: str, project_name: str
+    ) -> str:
+        """Derive a Pythonic module name from a relative file path."""
+        parts = rel.replace('\\', '/').split('/')
+        dir_parts = parts[:-1]
+        init_names = frozenset({'__init__.py', 'index.js', 'index.ts', 'mod.rs', 'lib.rs'})
+        if filename in init_names:
+            return '.'.join(dir_parts) if dir_parts else project_name
+        stem = os.path.splitext(filename)[0]
+        return '.'.join(dir_parts + [stem]) if dir_parts else stem
+
     def _collect_files(self, project_path: Path) -> List[Tuple[str, str]]:
         """Collect all source files with their module names for all supported languages.
-        
+
         Uses a single os.walk traversal with early directory pruning instead of
         separate rglob calls per extension (~40x speedup on large repos).
         """
         files = []
-        ext_set = set(ALL_EXTENSIONS)  # O(1) lookup
-        # Filename lookup uses a case-insensitive set. We lowercase once
-        # here so the hot loop only does a single lower() per file.
+        ext_set = set(ALL_EXTENSIONS)
         filename_set_lower = frozenset(n.lower() for n in ALL_FILENAMES)
         filename_prefixes_lower = tuple(p.lower() for p in LANGUAGE_FILENAME_PREFIXES)
-        init_names = frozenset({'__init__.py', 'index.js', 'index.ts', 'mod.rs', 'lib.rs'})
-        seen = set()  # guard against duplicate paths (e.g. .h in both c and cpp lists)
+        seen = set()
         project_str = str(project_path)
+        project_name = project_path.name
 
         for dirpath, dirnames, filenames in os.walk(project_str, topdown=True):
-            # Prune skipped directories in-place so os.walk won't descend into them
             dirnames[:] = [
                 d for d in dirnames
                 if not self.file_filter.should_skip_dir(d)
             ]
 
             for filename in filenames:
-                filename_lower = filename.lower()
-                suffix = os.path.splitext(filename)[1].lower()
-                # Accept by extension, well-known filename
-                # (Dockerfile, Makefile, Jenkinsfile, ...), or known
-                # prefix (Dockerfile.dev, Dockerfile.prod, Makefile.am).
-                if (
-                    suffix not in ext_set
-                    and filename_lower not in filename_set_lower
-                    and not filename_lower.startswith(filename_prefixes_lower)
+                if not self._should_collect_file(
+                    filename, ext_set, filename_set_lower, filename_prefixes_lower
                 ):
                     continue
 
@@ -241,17 +269,8 @@ class ProjectAnalyzer:
                 if not self.file_filter.should_process(file_str):
                     continue
 
-                # Calculate module name from relative path
                 rel = os.path.relpath(file_str, project_str)
-                parts = rel.replace('\\', '/').split('/')
-                dir_parts = parts[:-1]  # everything before filename
-
-                if filename in init_names:
-                    module_name = '.'.join(dir_parts) if dir_parts else project_path.name
-                else:
-                    stem = os.path.splitext(filename)[0]
-                    module_name = '.'.join(dir_parts + [stem]) if dir_parts else stem
-
+                module_name = self._compute_module_name(rel, filename, project_name)
                 files.append((file_str, module_name))
 
         return files

{code2llm-0.5.144 → code2llm-0.5.146}/code2llm/core/config.py

@@ -81,6 +81,10 @@ class PerformanceConfig:
     fast_mode: bool = False
     skip_data_flow: bool = False
     skip_pattern_detection: bool = False
+    skip_refactoring_analysis: bool = False
+    skip_dead_code_detection: bool = True
+    skip_centrality: bool = False
+    skip_community_detection: bool = False
 
     def get_workers(self) -> int:
         """Get effective worker count (auto-detect if set to 0)."""
@@ -88,6 +92,16 @@ class PerformanceConfig:
             return _get_optimal_workers(default=4)
         return self.parallel_workers
 
+    def apply_fast_mode(self) -> None:
+        """Apply fast_mode overrides — skip expensive analyses."""
+        if not self.fast_mode:
+            return
+        self.skip_data_flow = True
+        self.skip_pattern_detection = True
+        self.skip_dead_code_detection = True
+        self.skip_centrality = True
+        self.skip_community_detection = True
+
 
 @dataclass
 class FilterConfig:

{code2llm-0.5.144 → code2llm-0.5.146}/code2llm/core/file_analyzer.py

@@ -124,8 +124,9 @@ class FileAnalyzer:
         # Calculate complexity with radon
         self._calculate_complexity(content, file_path, result)
         
-        # Deep Analysis for refactoring
-        self._perform_deep_analysis(tree, module_name, file_path, result)
+        # Deep Analysis for refactoring (skip when data flow not needed)
+        if not self.config.performance.skip_data_flow:
+            self._perform_deep_analysis(tree, module_name, file_path, result)
 
         self.stats['files_processed'] += 1
         return result
@@ -164,12 +165,18 @@ class FileAnalyzer:
             result['mutations'] = dfg_res.mutations
             result['data_flows'] = dfg_res.data_flows
             
-            # Update function calls from CG extractor which is more robust
+            # Update function calls from CG extractor which is more robust.
+            # Use set-based merge to avoid duplicate call entries (ast.walk
+            # in _process_function already extracts calls).
             cg_ext = CallGraphExtractor(self.config)
             cg_res = cg_ext.extract(tree, module_name, file_path)
             for func_name, cg_func in cg_res.functions.items():
                 if func_name in result['functions']:
-                    result['functions'][func_name].calls.extend(list(cg_func.calls))
+                    existing = set(result['functions'][func_name].calls)
+                    for c in cg_func.calls:
+                        if c not in existing:
+                            result['functions'][func_name].calls.append(c)
+                            existing.add(c)
         except Exception as e:
             if self.config.verbose:
                 print(f"Error in deep analysis for {file_path}: {e}")