Enforce unique doc filenames and simple wiki-links (#109)

## Summary
- Rename section index files to match their titles (tutorials.md, reference.md, how-to.md, explanation.md) so all filenames are unique
- Convert all ~47 path-based wiki-links to simple filename format across 15 files
- Update doc-filenames task to no longer skip index.md files
- Update doc-links task to reject path-based links containing '/'

This ensures all wiki-links work correctly in obsidian.nvim by making links resolvable by filename alone.

## Testing
- `mise run doc-filenames` - all unique
- `mise run doc-links` - no broken or path-based links
- `mise run doc-titles` - no duplicates

Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/109
This commit is contained in:
Erich Blume 2026-02-04 17:21:34 -08:00
commit 3da455e49c
45 changed files with 176 additions and 125 deletions

View file

@ -33,13 +33,10 @@ def main() -> int:
# Key: filename (without .md), Value: list of file paths
filenames: dict[str, list[str]] = defaultdict(list)
# Scan all markdown files (excluding zk/, changelog.d/, and index.md files)
# Scan all markdown files (excluding zk/ and changelog.d/)
for md_file in sorted(DOCS_DIR.rglob("*.md")):
if "changelog.d" in md_file.parts or "zk" in md_file.parts:
continue
# Skip index.md files - they're expected to exist in multiple directories
if md_file.name == "index.md":
continue
rel_path = str(md_file.relative_to(DOCS_DIR))
filename = md_file.stem # filename without .md

View file

@ -8,12 +8,14 @@
This script scans all markdown files in the docs/ directory (excluding
changelog.d/), extracts wiki-links, and verifies each link target
exists as a filename or path in the documentation.
exists as a unique filename in the documentation.
Wiki-link formats supported:
- [[filename]] - links to filename.md (must be unique)
- [[path/to/file]] - links to path/to/file.md (for ambiguous filenames like index)
- [[target | Display Text]] - either format with display text
- [[filename]] - links to filename.md (must be unique across all docs)
- [[target|Display Text]] - filename with display text
Path-based links (containing '/') are NOT supported to ensure all
filenames are unique and links work correctly in obsidian.nvim.
Usage: mise run doc-links
"""
@ -28,16 +30,20 @@ from rich.table import Table
DOCS_DIR = Path(__file__).parent.parent / "docs"
# Regex to match wiki-links: [[Target]] or [[Target | Display]]
WIKILINK_PATTERN = re.compile(r"\[\[([^\]|]+)(?:\s*\|\s*[^\]]+)?\]\]")
# Regex to match wiki-links: [[Target]] or [[Target|Display]]
# Captures: group(1) = target (may have spaces), group(2) = full "|Display" part if present
WIKILINK_PATTERN = re.compile(r"\[\[([^\]|]+)(\|[^\]]+)?\]\]")
# Regex to match inline code (backticks)
INLINE_CODE_PATTERN = re.compile(r"`[^`]+`")
def extract_wikilinks(file_path: Path) -> list[tuple[str, int]]:
def extract_wikilinks(file_path: Path) -> list[tuple[str, int, bool]]:
"""Extract all wiki-link targets from a markdown file with line numbers.
Returns list of (target, line_num, has_spaces) tuples.
has_spaces is True if the target or pipe separator had surrounding spaces.
Ignores wiki-links inside inline code (backticks) as these are examples.
"""
content = file_path.read_text()
@ -47,8 +53,14 @@ def extract_wikilinks(file_path: Path) -> list[tuple[str, int]]:
# Remove inline code before searching for wiki-links
line_without_code = INLINE_CODE_PATTERN.sub("", line)
for match in WIKILINK_PATTERN.finditer(line_without_code):
target = match.group(1).strip()
links.append((target, line_num))
raw_target = match.group(1)
target = raw_target.strip()
pipe_part = match.group(2) # "|Display" or None
# Check for spaces: in target, or around the pipe
has_spaces = raw_target != target
if pipe_part and (raw_target.endswith(" ") or pipe_part.startswith("| ")):
has_spaces = True
links.append((target, line_num, has_spaces))
return links
@ -90,9 +102,11 @@ def main() -> int:
if (REPO_ROOT / filename).exists():
valid_targets.add(Path(filename).stem)
# Collect all broken and ambiguous links
# Collect all broken, ambiguous, path-based, and spaced links
broken_links: list[tuple[str, int, str]] = []
ambiguous_links: list[tuple[str, int, str, list[str]]] = []
path_links: list[tuple[str, int, str]] = []
spaced_links: list[tuple[str, int, str]] = []
# Scan all markdown files for wiki-links (excluding changelog.d/)
for md_file in sorted(DOCS_DIR.rglob("*.md")):
@ -102,9 +116,15 @@ def main() -> int:
rel_path = str(md_file.relative_to(DOCS_DIR))
links = extract_wikilinks(md_file)
for target, line_num in links:
if target in ambiguous_filenames:
# Link uses an ambiguous filename - needs to use full path
for target, line_num, has_spaces in links:
if has_spaces:
# Links with spaces in target or around pipe are not allowed
spaced_links.append((rel_path, line_num, target))
elif "/" in target:
# Path-based links are not allowed - use simple filenames only
path_links.append((rel_path, line_num, target))
elif target in ambiguous_filenames:
# Link uses an ambiguous filename - needs to be renamed
ambiguous_links.append((rel_path, line_num, target, filename_counts[target]))
elif target not in valid_targets:
broken_links.append((rel_path, line_num, target))
@ -117,11 +137,45 @@ def main() -> int:
has_errors = False
if spaced_links:
has_errors = True
console.print("[bold red]Wiki-Links With Spaces Found[/bold red]")
console.print("Wiki-links must not have spaces in the target or around the pipe.")
console.print("Use [[target|Display Text]] not [[target | Display Text]].")
console.print()
table = Table(show_header=True, header_style="bold")
table.add_column("File")
table.add_column("Line", justify="right")
table.add_column("Target")
for file_path, line_num, target in spaced_links:
table.add_row(file_path, str(line_num), escape(f"[[{target}]]"))
console.print(table)
console.print()
if path_links:
has_errors = True
console.print("[bold red]Path-Based Wiki-Links Found[/bold red]")
console.print("Wiki-links must use simple filenames only (no '/' paths).")
console.print("Rename files to be unique, then use [[filename]] format.")
console.print()
table = Table(show_header=True, header_style="bold")
table.add_column("File")
table.add_column("Line", justify="right")
table.add_column("Target")
for file_path, line_num, target in path_links:
table.add_row(file_path, str(line_num), escape(f"[[{target}]]"))
console.print(table)
console.print()
if ambiguous_links:
has_errors = True
console.print("[bold red]Ambiguous Wiki-Links Found[/bold red]")
console.print("These links use filenames that exist in multiple locations.")
console.print("Use the full path instead (e.g., [[reference/index]] not [[index]]).")
console.print("Rename files to be unique across all documentation.")
console.print()
table = Table(show_header=True, header_style="bold")
table.add_column("File")