from __future__ import annotations import ast import re from pathlib import Path from .document_tree import CPP_EXTENSIONS, FORTRAN_EXTENSIONS, PYTHON_EXTENSIONS from .markdown_utils import Heading from .text_io import DEFAULT_TEXT_PREVIEW_BYTES, read_text_with_fallback _CPP_TYPE_RE = re.compile(r"^\s*(class|struct)\s+([A-Za-z_]\w*)\b") _CPP_FUNCTION_RE = re.compile( r"^\s*(?:template\s*<[^;{}]+>\s*)?(?:[A-Za-z_~][\w:<>~*&\s,]*\s+)?([A-Za-z_~][\w:]*)\s*\([^;{}]*\)\s*(?:const\s*)?(?:noexcept\s*)?(?:->\s*[^{};]+\s*)?(?:\{|$)" ) _CPP_CONTROL_WORDS = {"if", "for", "while", "switch", "catch", "return", "sizeof"} _FORTRAN_PROGRAM_RE = re.compile(r"^\s*program\s+([A-Za-z_]\w*)\b", re.IGNORECASE) _FORTRAN_MODULE_RE = re.compile(r"^\s*module\s+(?!procedure\b)([A-Za-z_]\w*)\b", re.IGNORECASE) _FORTRAN_SUBROUTINE_RE = re.compile(r"^\s*(?:(?:recursive|pure|elemental|impure)\s+)*subroutine\s+([A-Za-z_]\w*)\b", re.IGNORECASE) _FORTRAN_FUNCTION_RE = re.compile( r"^\s*(?:(?:recursive|pure|elemental|impure)\s+)*(?:(?:type|class)\s*\([^)]*\)\s+|(?:integer|real|double\s+precision|complex|logical|character)\b(?:\s*\([^)]*\))?\s+)?function\s+([A-Za-z_]\w*)\b", re.IGNORECASE, ) _FORTRAN_TYPE_RE = re.compile(r"^\s*type\s*(?:(?:,\s*[^:]*)?::\s*|\s+)([A-Za-z_]\w*)\b", re.IGNORECASE) def extract_code_headings(path: Path, source: str | None = None) -> list[Heading]: suffix = path.suffix.lower() if source is None: source = read_text_with_fallback(path, max_bytes=DEFAULT_TEXT_PREVIEW_BYTES) if suffix in PYTHON_EXTENSIONS: return _extract_python_headings(source) if suffix in CPP_EXTENSIONS: return _extract_cpp_headings(source) if suffix in FORTRAN_EXTENSIONS: return _extract_fortran_headings(source) return [] def _extract_python_headings(source: str) -> list[Heading]: try: tree = ast.parse(source) except SyntaxError: return [] headings: list[Heading] = [] def walk(nodes: list[ast.stmt], level: int) -> None: for node in nodes: if isinstance(node, ast.ClassDef): headings.append(Heading(level=min(level, 6), title=f"class {node.name}", slug=f"line-{node.lineno}")) walk(node.body, min(level + 1, 6)) elif isinstance(node, (ast.FunctionDef, ast.AsyncFunctionDef)): prefix = "async def" if isinstance(node, ast.AsyncFunctionDef) else "def" headings.append(Heading(level=min(level, 6), title=f"{prefix} {node.name}", slug=f"line-{node.lineno}")) walk(node.body, min(level + 1, 6)) walk(list(tree.body), 1) return headings def _extract_cpp_headings(source: str) -> list[Heading]: headings: list[Heading] = [] pending_template = False for lineno, line in enumerate(source.splitlines(), start=1): stripped = line.strip() if not stripped or stripped.startswith(("//", "#")): continue if stripped.startswith("template"): pending_template = True continue type_match = _CPP_TYPE_RE.match(line) if type_match: kind, name = type_match.groups() headings.append(Heading(level=1, title=f"{kind} {name}", slug=f"line-{lineno}")) pending_template = False continue function_line = f"template <> {line}" if pending_template else line func_match = _CPP_FUNCTION_RE.match(function_line) if not func_match: pending_template = False continue name = func_match.group(1).split("::")[-1] if name in _CPP_CONTROL_WORDS: pending_template = False continue headings.append(Heading(level=2, title=f"function {func_match.group(1)}", slug=f"line-{lineno}")) pending_template = False return headings def _extract_fortran_headings(source: str) -> list[Heading]: headings: list[Heading] = [] matchers = ( ("program", _FORTRAN_PROGRAM_RE, 1), ("module", _FORTRAN_MODULE_RE, 1), ("type", _FORTRAN_TYPE_RE, 2), ("subroutine", _FORTRAN_SUBROUTINE_RE, 2), ("function", _FORTRAN_FUNCTION_RE, 2), ) for lineno, line in enumerate(source.splitlines(), start=1): stripped = line.strip() lowered = stripped.lower() if not stripped or lowered.startswith("!") or lowered.startswith("end "): continue for kind, pattern, level in matchers: match = pattern.match(line) if match: headings.append(Heading(level=level, title=f"{kind} {match.group(1)}", slug=f"line-{lineno}")) break return headings