code2llm 0.5.176__tar.gz → 0.5.183__tar.gz

{code2llm-0.5.176 → code2llm-0.5.183}/MANIFEST.in

@@ -2,3 +2,4 @@ include VERSION
 include README.md
 include LICENSE
 recursive-include code2llm *.py
+recursive-include code2llm/templates *.md

{code2llm-0.5.176/code2llm.egg-info → code2llm-0.5.183}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: code2llm
-Version: 0.5.176
+Version: 0.5.183
 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
@@ -36,7 +36,7 @@ Requires-Dist: psutil>=5.9
 Requires-Dist: code2logic
 Requires-Dist: vulture>=2.10
 Requires-Dist: tiktoken>=0.5
-Requires-Dist: tree-sitter>=0.21
+Requires-Dist: tree-sitter>=0.25
 Requires-Dist: tree-sitter-python>=0.21
 Requires-Dist: tree-sitter-javascript>=0.21
 Requires-Dist: tree-sitter-typescript>=0.21
@@ -48,6 +48,14 @@ Requires-Dist: tree-sitter-cpp>=0.22
 Requires-Dist: tree-sitter-c-sharp>=0.21
 Requires-Dist: tree-sitter-php>=0.22
 Requires-Dist: tree-sitter-ruby>=0.21
+Provides-Extra: native
+Requires-Dist: code2llm-rust>=0.1.0; extra == "native"
+Provides-Extra: graph
+Requires-Dist: code2graph>=0.1.0; extra == "graph"
+Provides-Extra: toon
+Requires-Dist: code2toon>=0.1.0; extra == "toon"
+Provides-Extra: diagrams
+Requires-Dist: code2flow>=0.1.0; extra == "diagrams"
 Provides-Extra: dev
 Requires-Dist: pytest>=6.2; extra == "dev"
 Requires-Dist: pytest-cov>=2.12; extra == "dev"
@@ -55,9 +63,11 @@ 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>=2.1.218; extra == "dev"
 Requires-Dist: costs>=0.1.20; extra == "dev"
-Requires-Dist: pfix>=0.1.60; extra == "dev"
+Requires-Dist: code2llm-rust>=0.1.0; extra == "dev"
+Requires-Dist: code2graph>=0.1.0; extra == "dev"
+Requires-Dist: code2toon>=0.1.0; extra == "dev"
+Requires-Dist: code2flow>=0.1.0; extra == "dev"
 Dynamic: author
 Dynamic: home-page
 Dynamic: license-file
@@ -67,13 +77,13 @@ Dynamic: requires-python
 
 ## AI Cost Tracking
 
-![PyPI](https://img.shields.io/badge/pypi-costs-blue) ![Version](https://img.shields.io/badge/version-0.5.176-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-$8.32-orange) ![Human Time](https://img.shields.io/badge/Human%20Time-105.3h-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-0.5.183-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-$0.57-orange) ![Human Time](https://img.shields.io/badge/Human%20Time-140.3h-blue) ![Model](https://img.shields.io/badge/Model-openrouter%2Fdeep%2Fdeep--v4--pro-lightgrey)
 
-- 🤖 **LLM usage:** $8.3176 (263 commits)
-- 👤 **Human dev:** ~$10535 (105.3h @ $100/h, 30min dedup)
+- 🤖 **LLM usage:** $0.5710 (336 commits)
+- 👤 **Human dev:** ~$14034 (140.3h @ $100/h, 30min dedup)
 
-Generated on 2026-08-06 using [openrouter/deep/deep-v4-pro](https://openrouter.ai/deep/deep-v4-pro)
+Generated on 2026-10-18 using [openrouter/deep/deep-v4-pro](https://openrouter.ai/deep/deep-v4-pro)
 
 ---
 
@@ -418,3 +428,7 @@ For more information about code2llm, visit: https://github.com/tom-sapletta/code
 ## License
 
 Licensed under Apache-2.0.
+
+## Dependency maintenance
+
+See [dependency updates and Python tool groups](docs/dependencies.md) for locked tests, daily updates and freshness checks.

{code2llm-0.5.176 → code2llm-0.5.183}/README.md

@@ -2,13 +2,13 @@
 
 ## AI Cost Tracking
 
-![PyPI](https://img.shields.io/badge/pypi-costs-blue) ![Version](https://img.shields.io/badge/version-0.5.176-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-$8.32-orange) ![Human Time](https://img.shields.io/badge/Human%20Time-105.3h-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-0.5.183-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-$0.57-orange) ![Human Time](https://img.shields.io/badge/Human%20Time-140.3h-blue) ![Model](https://img.shields.io/badge/Model-openrouter%2Fdeep%2Fdeep--v4--pro-lightgrey)
 
-- 🤖 **LLM usage:** $8.3176 (263 commits)
-- 👤 **Human dev:** ~$10535 (105.3h @ $100/h, 30min dedup)
+- 🤖 **LLM usage:** $0.5710 (336 commits)
+- 👤 **Human dev:** ~$14034 (140.3h @ $100/h, 30min dedup)
 
-Generated on 2026-08-06 using [openrouter/deep/deep-v4-pro](https://openrouter.ai/deep/deep-v4-pro)
+Generated on 2026-10-18 using [openrouter/deep/deep-v4-pro](https://openrouter.ai/deep/deep-v4-pro)
 
 ---
 
@@ -353,3 +353,7 @@ For more information about code2llm, visit: https://github.com/tom-sapletta/code
 ## License
 
 Licensed under Apache-2.0.
+
+## Dependency maintenance
+
+See [dependency updates and Python tool groups](docs/dependencies.md) for locked tests, daily updates and freshness checks.

code2llm-0.5.183/VERSION

@@ -0,0 +1 @@
+0.5.183

{code2llm-0.5.176 → code2llm-0.5.183}/code2llm/__init__.py

@@ -8,7 +8,7 @@ Includes NLP Processing Pipeline for query normalization, intent matching,
 and entity resolution with multilingual support.
 """
 
-__version__ = "0.5.176"
+__version__ = "0.5.183"
 __author__ = "STTS Project"
 
 # Core analysis components (lightweight, always needed)

{code2llm-0.5.176 → code2llm-0.5.183}/code2llm/analysis/__init__.py

@@ -12,6 +12,10 @@ __all__ = [
     "PipelineDetector",
     "PipelineResolver",
     "PipelineClassifier",
+    "calculate_complexity_regex",
+    "calculate_python_complexity",
+    "compute_cyclomatic_complexity",
+    "extract_function_body",
 ]
 
 
@@ -29,6 +33,10 @@ def __getattr__(name):
         "PipelineDetector": ".pipeline_detector",
         "PipelineResolver": ".pipeline_resolver",
         "PipelineClassifier": ".pipeline_classifier",
+        "calculate_complexity_regex": ".complexity",
+        "calculate_python_complexity": ".complexity",
+        "compute_cyclomatic_complexity": ".complexity",
+        "extract_function_body": ".complexity",
     }
     if name in _imports:
         import importlib

code2llm-0.5.183/code2llm/analysis/call_graph.py

@@ -0,0 +1,16 @@
+"""Call graph extractor using AST.
+
+Backward-compatibility facade delegating to code2llm.analysis.call_graph_engine.
+"""
+
+from __future__ import annotations
+
+from code2llm.analysis.call_graph_engine import (
+    CallGraphExtractor,
+    calculate_call_metrics,
+)
+
+__all__ = [
+    "CallGraphExtractor",
+    "calculate_call_metrics",
+]

code2llm-0.5.183/code2llm/analysis/call_graph_engine.py

@@ -0,0 +1,448 @@
+"""Call graph extraction and symbol call resolution engine.
+
+Atomized module supporting:
+- Multi-language call extraction via regex and native Rust acceleration (code2llm-fast-core)
+- Python AST call graph extraction with caller/callee resolution
+- Call graph metrics computation (fan-in, fan-out, called_by, call edges)
+"""
+
+from __future__ import annotations
+
+import ast
+import json
+import re
+import subprocess
+from pathlib import Path
+from typing import Any, Dict, List, Optional
+
+from code2llm.analysis.complexity import _find_rust_binary, extract_function_body
+from code2llm.analysis.utils import ast_unparse, qualified_name
+from code2llm.core.config import Config
+from code2llm.core.models import AnalysisResult, FlowEdge
+
+CALL_PATTERN_C_FAMILY = re.compile(
+    r"(?<!\bfunction\b\s)"
+    r"(?<!\bclass\b\s)"
+    r"\b([a-zA-Z_]\w*)\s*\("
+    r"|"
+    r"(?:this|self)\s*\.\s*(\w+)\s*\("
+    r"|"
+    r"\b(\w+)\s*\.\s*(\w+)\s*\("
+)
+
+_CALL_KEYWORDS = frozenset(
+    {
+        "if",
+        "for",
+        "while",
+        "switch",
+        "catch",
+        "return",
+        "throw",
+        "new",
+        "typeof",
+        "instanceof",
+        "import",
+        "export",
+        "require",
+        "console",
+        "super",
+        "class",
+        "function",
+        "async",
+        "await",
+        "delete",
+        "void",
+        "case",
+        "default",
+    }
+)
+
+
+def _resolve_call(
+    simple_call: str,
+    func_qname: str,
+    module_name: str,
+    known_simple: Dict[str, List[str]],
+    calls_seen: set[str],
+    func_info: Any,
+) -> None:
+    """Resolve a single call name and append to func_info.calls if novel."""
+    if simple_call in known_simple:
+        candidates = known_simple[simple_call]
+        my_module = func_qname.rsplit(".", 1)[0]
+        resolved = next(
+            (c for c in candidates if c.rsplit(".", 1)[0] == my_module),
+            candidates[0],
+        )
+        if resolved != func_qname and resolved not in calls_seen:
+            func_info.calls.append(resolved)
+            calls_seen.add(resolved)
+    else:
+        ext_name = f"{module_name}.{simple_call}"
+        if ext_name not in calls_seen:
+            func_info.calls.append(ext_name)
+            calls_seen.add(ext_name)
+
+
+def _rust_batch_calls(
+    lines: list[int],
+    file_path: str | None = None,
+    content: str | None = None,
+) -> dict[int, list[str]] | None:
+    """Call native Rust extension or binary for fast batch call extraction."""
+    if not lines:
+        return None
+
+    # 1. Native PyO3 extension (zero-subprocess in-memory execution)
+    from code2llm.analysis.native_bridge import native_batch_calls
+
+    resolved_content = content
+    if resolved_content is None and file_path and Path(file_path).is_file():
+        try:
+            resolved_content = Path(file_path).read_text(encoding="utf-8", errors="ignore")
+        except Exception:
+            resolved_content = None
+
+    if resolved_content is not None:
+        native_res = native_batch_calls(resolved_content, lines)
+        if native_res is not None:
+            return native_res
+
+    # 2. Secondary fallback: CLI binary
+    binary = _find_rust_binary()
+    if not binary:
+        return None
+
+    lines_arg = ",".join(str(l) for l in lines)
+    cmd = [str(binary), "--calls", "--lines", lines_arg]
+
+    try:
+        if file_path and Path(file_path).is_file():
+            cmd.extend(["--file", str(file_path)])
+            proc = subprocess.run(
+                cmd,
+                capture_output=True,
+                text=True,
+                check=True,
+                timeout=5,
+            )
+        elif content is not None:
+            proc = subprocess.run(
+                cmd,
+                input=content,
+                capture_output=True,
+                text=True,
+                check=True,
+                timeout=5,
+            )
+        else:
+            return None
+
+        data = json.loads(proc.stdout.strip())
+        result: dict[int, list[str]] = {}
+        for item in data:
+            result[item["line"]] = item.get("calls", [])
+        return result
+    except Exception:
+        return None
+
+
+def extract_calls_regex(
+    content: str,
+    module_name: str,
+    result: Dict[str, Any],
+    file_path: str | None = None,
+) -> None:
+    """Extract function calls from function bodies using regex or native Rust acceleration."""
+    functions = result.get("functions", {})
+    if not functions:
+        return
+
+    known_simple: Dict[str, List[str]] = {}
+    for qname in functions:
+        simple = qname.rsplit(".", 1)[-1]
+        known_simple.setdefault(simple, []).append(qname)
+
+    lines = [f.line for f in functions.values() if getattr(f, "line", None)]
+    rust_calls = _rust_batch_calls(lines, file_path=file_path, content=content)
+
+    for func_qname, func_info in functions.items():
+        line = getattr(func_info, "line", None)
+        body = extract_function_body(content, line or 1)
+        if not body:
+            continue
+        calls_seen: set[str] = set()
+
+        if rust_calls and line in rust_calls:
+            for simple_call in rust_calls[line]:
+                _resolve_call(
+                    simple_call,
+                    func_qname,
+                    module_name,
+                    known_simple,
+                    calls_seen,
+                    func_info,
+                )
+        else:
+            for m in CALL_PATTERN_C_FAMILY.finditer(body):
+                simple_call = m.group(1) or m.group(2) or m.group(4)
+                if not simple_call or simple_call in _CALL_KEYWORDS:
+                    continue
+                _resolve_call(
+                    simple_call,
+                    func_qname,
+                    module_name,
+                    known_simple,
+                    calls_seen,
+                    func_info,
+                )
+
+
+def calculate_call_metrics(
+    functions: Dict[str, Any],
+    metrics: Dict[str, Any] | None = None,
+) -> Dict[str, Dict[str, Any]]:
+    """Calculate fan-in, fan-out and populate called_by for all functions."""
+    from code2llm.analysis.native_bridge import native_calculate_call_metrics
+
+    res_metrics = metrics if metrics is not None else {}
+
+    # Fast path: native Rust
+    fn_tuples = []
+    for name, fi in functions.items():
+        calls = list(getattr(fi, "calls", []))
+        called_by = list(getattr(fi, "called_by", []))
+        raw_cc = getattr(fi, "complexity", 1.0)
+        cc_val = float(raw_cc.get("cyclomatic", 1.0) if isinstance(raw_cc, dict) else (raw_cc or 1.0))
+        fn_tuples.append((name, calls, called_by, cc_val))
+
+    native_res = native_calculate_call_metrics(fn_tuples)
+    if native_res is not None:
+        computed_metrics, called_by_map = native_res
+        for name, callers in called_by_map.items():
+            if name in functions:
+                fi = functions[name]
+                if hasattr(fi, "called_by"):
+                    fi.called_by = callers
+        for name, (fan_in, fan_out, cc) in computed_metrics.items():
+            res_metrics[name] = {
+                "fan_in": fan_in,
+                "fan_out": fan_out,
+                "complexity": cc,
+            }
+        return res_metrics
+
+    # Fallback: Python implementation
+    for caller_name, caller_info in functions.items():
+        calls = getattr(caller_info, "calls", [])
+        for callee_name in calls:
+            if callee_name in functions:
+                callee_info = functions[callee_name]
+                called_by = getattr(callee_info, "called_by", None)
+                if called_by is not None and caller_name not in called_by:
+                    called_by.append(caller_name)
+
+    for func_name, func_info in functions.items():
+        calls = getattr(func_info, "calls", [])
+        called_by = getattr(func_info, "called_by", [])
+        fan_out = len(set(calls))
+        fan_in = len(set(called_by))
+        res_metrics[func_name] = {
+            "fan_in": fan_in,
+            "fan_out": fan_out,
+            "complexity": getattr(func_info, "complexity", 1),
+        }
+
+    return res_metrics
+
+
+class CallGraphExtractor(ast.NodeVisitor):
+    """Extract call graph from AST with fast resolution and optional astroid inference."""
+
+    def __init__(self, config: Config):
+        self.config = config
+        self.result = AnalysisResult()
+        self.module_name = ""
+        self.file_path = ""
+
+        # Context
+        self.function_stack: List[str] = []
+        self.class_stack: List[str] = []
+        self.imports: Dict[str, str] = {}
+        self.astroid_tree = None
+
+    def extract(
+        self, tree: ast.AST, module_name: str, file_path: str
+    ) -> AnalysisResult:
+        """Extract call graph from AST."""
+        self.result = AnalysisResult()
+        self.module_name = module_name
+        self.file_path = file_path
+        self.function_stack = []
+        self.class_stack = []
+        self.imports = {}
+
+        # Lazy astroid tree import (only when not skipped)
+        if not getattr(self.config.performance, "skip_astroid_inference", True):
+            try:
+                import astroid as _astroid
+
+                self.astroid_tree = _astroid.MANAGER.ast_from_file(file_path)
+            except Exception:
+                self.astroid_tree = None
+        else:
+            self.astroid_tree = None
+
+        self.visit(tree)
+        self._calculate_metrics()
+        return self.result
+
+    def _calculate_metrics(self) -> None:
+        """Calculate fan-in and fan-out metrics (backward compatibility)."""
+        calculate_call_metrics(self.result.functions, self.result.metrics)
+
+    def visit_Import(self, node: ast.Import) -> None:
+        """Track imports."""
+        for alias in node.names:
+            name = alias.asname if alias.asname else alias.name
+            self.imports[name] = alias.name
+            if hasattr(self.result, "imports"):
+                self.result.imports[name] = alias.name
+
+    def visit_ImportFrom(self, node: ast.ImportFrom) -> None:
+        """Track from imports."""
+        module = node.module or ""
+        for alias in node.names:
+            name = alias.asname if alias.asname else alias.name
+            full_name = f"{module}.{alias.name}" if module else alias.name
+            self.imports[name] = full_name
+            if hasattr(self.result, "imports"):
+                self.result.imports[name] = full_name
+
+    def visit_ClassDef(self, node: ast.ClassDef) -> None:
+        """Visit class definition."""
+        self.class_stack.append(node.name)
+
+        self.result.classes[node.name] = {
+            "file": self.file_path,
+            "line": node.lineno,
+            "methods": [m.name for m in node.body if isinstance(m, ast.FunctionDef)],
+            "bases": [self._expr_to_str(b) for b in node.bases],
+        }
+
+        for stmt in node.body:
+            self.visit(stmt)
+
+        self.class_stack.pop()
+
+    def visit_FunctionDef(self, node: ast.FunctionDef) -> None:
+        """Visit function definition and track calls within it."""
+        func_name = qualified_name(self.module_name, self.class_stack, node.name)
+        self.function_stack.append(func_name)
+
+        for stmt in node.body:
+            self.visit(stmt)
+
+        self.function_stack.pop()
+
+    def visit_AsyncFunctionDef(self, node: ast.AsyncFunctionDef) -> None:
+        """Visit async function."""
+        self.visit_FunctionDef(node)
+
+    def visit_Call(self, node: ast.Call) -> None:
+        """Track function calls."""
+        if not self.function_stack:
+            self.generic_visit(node)
+            return
+
+        caller = self.function_stack[-1]
+        callee = self._resolve_call(node.func)
+
+        # If ast-based resolution returned None or None.sth, try astroid if loaded
+        if (not callee or "None." in callee) and self.astroid_tree:
+            astroid_callee = self._resolve_with_astroid(node)
+            if astroid_callee:
+                callee = astroid_callee
+
+        if callee and caller in self.result.functions:
+            self.result.functions[caller].calls.append(callee)
+
+            edge = FlowEdge(
+                source=str(-1),
+                target=str(-1),
+                edge_type="call",
+                label=f"{caller}->{callee}",
+            )
+            if hasattr(self.result, "call_edges"):
+                self.result.call_edges.append(edge)
+            else:
+                self.result.edges.append(edge)
+
+        self.generic_visit(node)
+
+    def _resolve_call(self, node: ast.AST) -> Optional[str]:
+        """Resolve a call to its full name."""
+        if isinstance(node, ast.Name):
+            if node.id in self.imports:
+                return self.imports[node.id]
+            return f"{self.module_name}.{node.id}"
+
+        if isinstance(node, ast.Attribute):
+            parts: List[str] = []
+            current = node
+
+            while isinstance(current, ast.Attribute):
+                parts.append(current.attr)
+                current = current.value
+
+            if isinstance(current, ast.Name):
+                parts.append(current.id)
+                parts.reverse()
+
+                root = parts[0]
+                if root in self.imports:
+                    return f"{self.imports[root]}.{'.'.join(parts[1:])}"
+
+                if root in ("self", "cls") and self.class_stack:
+                    return f"{self.module_name}.{self.class_stack[-1]}.{'.'.join(parts[1:])}"
+
+                return f"{self.module_name}.{'.'.join(parts)}"
+
+        return None
+
+    def _resolve_with_astroid(self, node: ast.Call) -> Optional[str]:
+        """Use astroid to infer the call target."""
+        if not self.astroid_tree:
+            return None
+
+        try:
+            import astroid as _astroid
+
+            for astroid_node in self.astroid_tree.nodes_of_class(_astroid.Call):
+                if (
+                    astroid_node.lineno == node.lineno
+                    and astroid_node.col_offset == node.col_offset
+                ):
+                    inferred = astroid_node.func.infer()
+                    for target in inferred:
+                        if hasattr(target, "qname"):
+                            return target.qname()
+                    break
+        except Exception:
+            pass
+        return None
+
+    def _expr_to_str(self, node: ast.AST) -> str:
+        """Unparse an AST expression node to a source string."""
+        return ast_unparse(node, default_none="")
+
+
+__all__ = [
+    "CALL_PATTERN_C_FAMILY",
+    "CallGraphExtractor",
+    "calculate_call_metrics",
+    "extract_calls_regex",
+    "_resolve_call",
+    "_rust_batch_calls",
+]