Files

192 lines
5.6 KiB
Python

#!/usr/bin/env python3
from __future__ import annotations
import argparse
import re
import sys
from dataclasses import dataclass, field
from pathlib import Path
from typing import Iterator, Optional, Sequence
EXCLUDED_DIRECTORIES = {".git", "build", "dist", "node_modules", "tmp"}
FENCE_PATTERN = re.compile(r"^\s*(`{3,}|~{3,})")
INLINE_CODE_PATTERN = re.compile(r"`[^`]*`")
INLINE_LINK_PATTERN = re.compile(r"!?\[[^]]*\]\(([^)]*)\)")
REFERENCE_LINK_PATTERN = re.compile(
r"^\s*\[[^]]+\]:\s*(?:<([^>]+)>|([^\s]+))"
)
@dataclass(frozen=True)
class MarkdownLink:
line: int
target: str
@dataclass(frozen=True)
class BrokenLink:
source: Path
line: int
target: str
resolved_target: Path
@dataclass
class LinkCheckResult:
markdown_files: int = 0
total_links: int = 0
valid_links: int = 0
skipped_links: int = 0
broken_links: list[BrokenLink] = field(default_factory=list)
def extract_links(text: str) -> Iterator[MarkdownLink]:
fence_character: Optional[str] = None
for line_number, raw_line in enumerate(text.splitlines(), start=1):
fence_match = FENCE_PATTERN.match(raw_line)
if fence_match:
marker_character = fence_match.group(1)[0]
if fence_character is None:
fence_character = marker_character
elif marker_character == fence_character:
fence_character = None
continue
if fence_character is not None:
continue
line = INLINE_CODE_PATTERN.sub("", raw_line)
for match in INLINE_LINK_PATTERN.finditer(line):
yield MarkdownLink(line_number, match.group(1).strip())
reference_match = REFERENCE_LINK_PATTERN.match(line)
if reference_match:
target = reference_match.group(1) or reference_match.group(2)
yield MarkdownLink(line_number, target.strip())
def iter_markdown_files(root: Path) -> Iterator[Path]:
for path in sorted(root.rglob("*.md")):
relative_path = path.relative_to(root)
if path.name.endswith(".template.md"):
continue
if any(part in EXCLUDED_DIRECTORIES for part in relative_path.parts[:-1]):
continue
if path.is_file():
yield path
def _local_target(target: str) -> Optional[str]:
target = target.strip()
if target.startswith("<") and target.endswith(">"):
target = target[1:-1].strip()
target = target.split("#", 1)[0]
if not target:
return None
lower_target = target.lower()
if lower_target.startswith(("http://", "https://", "mailto:")):
return None
if target.startswith("/en/docs/"):
return None
return target
def _resolve_target(root: Path, source: Path, target: str) -> Path:
if target.startswith("/"):
return (root / target.lstrip("/")).resolve()
return (source.parent / target).resolve()
def check_repository(root: Path) -> LinkCheckResult:
root = root.resolve()
markdown_files = list(iter_markdown_files(root))
result = LinkCheckResult(markdown_files=len(markdown_files))
for source in markdown_files:
text = source.read_text(encoding="utf-8")
for link in extract_links(text):
result.total_links += 1
local_target = _local_target(link.target)
if local_target is None:
result.skipped_links += 1
continue
resolved_target = _resolve_target(root, source, local_target)
if resolved_target.exists():
result.valid_links += 1
continue
result.broken_links.append(
BrokenLink(source, link.line, link.target, resolved_target)
)
return result
def _display_path(path: Path, root: Path) -> str:
try:
return path.relative_to(root).as_posix()
except ValueError:
return str(path)
def print_report(result: LinkCheckResult, root: Path) -> None:
root = root.resolve()
print(f"Playbook root: {root}")
print(f"Markdown files: {result.markdown_files}")
for broken in result.broken_links:
source = _display_path(broken.source, root)
target = _display_path(broken.resolved_target, root)
print(f"BROKEN: {source}:{broken.line}")
print(f" Link: {broken.target}")
print(f" Target: {target}")
print(f"Total links: {result.total_links}")
print(f"Valid links: {result.valid_links}")
print(f"Skipped links: {result.skipped_links}")
print(f"Broken links: {len(result.broken_links)}")
def parse_args(argv: Optional[Sequence[str]] = None) -> argparse.Namespace:
parser = argparse.ArgumentParser(description="Check local links in Markdown files.")
parser.add_argument(
"--root",
type=Path,
default=Path(__file__).resolve().parents[2],
help="repository root to scan (default: inferred from this script)",
)
return parser.parse_args(argv)
def _configure_output() -> None:
for stream in (sys.stdout, sys.stderr):
reconfigure = getattr(stream, "reconfigure", None)
if reconfigure is not None:
reconfigure(errors="backslashreplace")
def main(argv: Optional[Sequence[str]] = None) -> int:
_configure_output()
args = parse_args(argv)
root = args.root.resolve()
if not root.is_dir():
print(f"ERROR: root directory does not exist: {root}")
return 2
result = check_repository(root)
print_report(result, root)
if result.broken_links:
print("Document link check failed.")
return 1
print("All document links passed.")
return 0
if __name__ == "__main__":
raise SystemExit(main())