Matterify

Extract and aggregate YAML frontmatter from Markdown files, with optional SHA-256 hashes and file statistics.

Features

• Recursive Markdown discovery with configurable directory exclusions
• YAML frontmatter extraction with structured ok/illegal status reporting
• Optional SHA-256 file hashes and file stats (size, mtime, atime)
• Parallel scan workers for faster processing on larger vaults/projects

Quick Start

pip install matterify
matterify ./docs -o output.json

Installation

# Using uv (recommended)
uv add matterify

# Or with pip
pip install matterify

CLI Usage

matterify DIRECTORY [OPTIONS]

DIRECTORY must exist and is scanned recursively for .md and .markdown files.

Options:

• --version - Show version information and exit
• --debug - Enable debug logging
• -o, --output PATH - Write JSON to file instead of stdout (if omitted, outputs to stdout)
• --n-procs INT - Worker process count (default: auto-detect CPU cores)
• -v, --verbose - Show progress and summary
• -e, --exclude TEXT - Glob patterns to exclude (e.g., **/.git, **/__pycache__)
• -i, --include PATH - Additional file paths to include in scan (repeatable)
• --hash / --no-hash - Enable/disable SHA-256 hash computation
• --stats / --no-stats - Enable/disable file statistics (size, modified time, access time)
• --frontmatter / --no-frontmatter - Enable/disable YAML frontmatter extraction
• --help - Show command help and exit

When --no-frontmatter is used, metadata fields files_with_frontmatter and files_without_frontmatter are null.

Examples:

# Output to stdout (JSON)
matterify ./docs

# Output to file
matterify ./docs -o output.json

# Verbose output
matterify ./docs --verbose

# Disable hashes and file stats
matterify ./docs --no-hash --no-stats

# Hash + stats only (skip YAML parsing)
matterify ./docs --no-frontmatter

# Exclude directories using glob patterns
matterify ./docs -e '**/build' -e '**/.cache'

# Include additional files (any extension)
matterify ./docs -i notes.txt -i ../shared/changelog.txt

# Full help
matterify --help

Python API

Public Functions

from pathlib import Path
from matterify import (
    scan_directory,
)

scan_directory

Scan directory and aggregate frontmatter using parallel workers. Returns a ScanResults dataclass.

from pathlib import Path
from matterify import scan_directory

result = scan_directory(Path("./docs"))

# ScanResults contains:
# - result.metadata: ScanMetadata with scan statistics
# - result.files: list of file entries with extraction results

# Access metadata
print(result.metadata.total_files)
print(result.metadata.files_with_frontmatter)
print(result.metadata.scan_duration_seconds)

# Access files
for entry in result.files:
    print(entry.file_path, entry.status)
    print(entry.stats.file_size if entry.stats else None)

Custom data callback

You can pass a callback function to inject custom data into each file entry. The callback receives the raw file content as a string and should return any value or None. The result is stored in the custom_data field of each FileEntry.

from pathlib import Path
from matterify import scan_directory

def count_words(content: str) -> object:
    return {"word_count": len(content.split())}

result = scan_directory(Path("./docs"), callback=count_words)

for entry in result.files:
    if entry.custom_data:
        print(entry.file_path, entry.custom_data["word_count"])

Important: The callback must be a module-level function (picklable for multiprocessing), not a lambda or closure.

Public Types

from matterify import (
    FileEntry,
    ScanMetadata,
    ScanResults,
)

# FileEntry: extracted frontmatter from a single file
entry: FileEntry

# ScanMetadata: summary statistics about a scan
metadata: ScanMetadata

# ScanResults: holds metadata and file entries
result: ScanResults

JSON Output Structure

When using CLI (stdout or --output), the payload has this shape:

{
  "metadata": {
    "root": "/path/to/docs",
    "total_files": 10,
    "files_with_frontmatter": 8,
    "files_without_frontmatter": 2,
    "errors": 0,
    "scan_duration_seconds": 0.523,
    "avg_duration_per_file_ms": 52.3,
    "throughput_files_per_second": 19.1
  },
  "files": [
    {
      "file_path": "getting-started.md",
      "frontmatter": {
        "title": "Getting Started",
        "date": "2024-01-15",
        "tags": ["guide", "tutorial"]
      },
      "status": "ok",
      "error": null,
      "stats": {
        "file_size": 1234,
        "modified_time": "2024-01-15T10:30:00",
        "access_time": "2024-01-15T10:30:00"
      },
      "file_hash": "abc123..."
    }
  ]
}

status is either "ok" or "illegal".

Default Exclusions

The following glob patterns are excluded from scanning by default:

• **/.git - Git repositories
• **/.obsidian - Obsidian vault settings
• **/__pycache__ - Python bytecode cache
• **/.venv - Python virtual environments
• **/venv - Python virtual environments
• **/node_modules - Node.js dependencies
• **/.mypy_cache - MyPy type checker cache
• **/.pytest_cache - Pytest cache
• **/.ruff_cache - Ruff linter cache

The **/ prefix matches directories at any depth. Use -e or --exclude to add custom exclusion patterns.

Development

# Install with dev dependencies
uv sync --all-extras

# Run tests
uv run pytest

# Format and lint
uv run ruff format src/ tests/
uv run ruff check src/ tests/

# Type check
uv run mypy src/

License

MIT License - see LICENSE file for details.