codebook / potato /format_handlers /markdown_handler.py
davidjurgens's picture
Deploy: Potato — Codebook Annotation
aceb1b2 verified
Raw
History Blame Contribute Delete
7.46 kB
"""
Markdown Format Handler
Parses Markdown files and extracts text with source line mapping.
Supports syntax highlighting for code blocks.
Usage:
from potato.format_handlers.markdown_handler import MarkdownHandler
handler = MarkdownHandler()
output = handler.extract("document.md", {
"highlight_code": True,
"gfm": True, # GitHub Flavored Markdown
})
"""
from typing import Dict, List, Any, Optional
from pathlib import Path
import html
import logging
import re
import uuid
from .base import BaseFormatHandler, FormatOutput
from .coordinate_mapping import CoordinateMapper, CodeCoordinate, DocumentCoordinate
logger = logging.getLogger(__name__)
# Check if dependencies are available
try:
import mistune
MISTUNE_AVAILABLE = True
except ImportError:
MISTUNE_AVAILABLE = False
mistune = None
try:
from pygments import highlight
from pygments.lexers import get_lexer_by_name, guess_lexer
from pygments.formatters import HtmlFormatter
PYGMENTS_AVAILABLE = True
except ImportError:
PYGMENTS_AVAILABLE = False
class MarkdownHandler(BaseFormatHandler):
"""
Handler for Markdown files.
Uses mistune for parsing and Pygments for syntax highlighting.
Maintains line/column coordinate mappings.
"""
format_name = "markdown"
supported_extensions = [".md", ".markdown", ".mdown", ".mkd"]
description = "Markdown parsing with line/column mapping and syntax highlighting"
requires_dependencies = ["mistune"]
def get_default_options(self) -> Dict[str, Any]:
"""Get default extraction options."""
return {
"highlight_code": True,
"gfm": True, # GitHub Flavored Markdown
"include_raw_blocks": False,
"preserve_line_breaks": True,
}
def extract(
self,
file_path: str,
options: Optional[Dict[str, Any]] = None
) -> FormatOutput:
"""
Parse and render a Markdown file.
Args:
file_path: Path to the Markdown file
options: Extraction options:
- highlight_code: Syntax highlight code blocks
- gfm: Use GitHub Flavored Markdown extensions
- preserve_line_breaks: Keep original line structure
Returns:
FormatOutput with text, rendered HTML, and coordinate mappings
"""
if not MISTUNE_AVAILABLE:
raise ImportError(
"mistune is required for Markdown extraction. "
"Install with: pip install mistune"
)
opts = self.merge_options(options)
# Read source file
path = Path(file_path)
source_text = path.read_text(encoding="utf-8")
# Build line index for coordinate mapping
line_offsets = self._build_line_index(source_text)
# Parse and render
mapper = CoordinateMapper()
rendered_html = self._render_markdown(source_text, opts)
# Build coordinate mappings for each line
for line_num, (start, end) in enumerate(line_offsets, start=1):
mapper.add_mapping(
start,
end,
CodeCoordinate(line=line_num, column=1)
)
metadata = {
"format": "markdown",
"source_file": str(file_path),
"line_count": len(line_offsets),
"char_count": len(source_text),
"headings": self._extract_headings(source_text),
}
coord_dict = mapper.to_dict()
coord_dict["get_coords_for_range"] = mapper.get_coords_for_range
return FormatOutput(
text=source_text,
rendered_html=rendered_html,
coordinate_map=coord_dict,
metadata=metadata,
format_name=self.format_name,
source_path=str(file_path),
)
def _build_line_index(self, text: str) -> List[tuple]:
"""
Build an index of line start/end offsets.
Returns:
List of (start, end) tuples for each line
"""
lines = []
start = 0
for line in text.split("\n"):
end = start + len(line)
lines.append((start, end))
start = end + 1 # +1 for newline
return lines
def _render_markdown(self, text: str, opts: Dict[str, Any]) -> str:
"""
Render Markdown to HTML using mistune.
"""
# Create custom renderer with code highlighting
if opts.get("highlight_code") and PYGMENTS_AVAILABLE:
renderer = HighlightRenderer()
else:
renderer = None
# Configure mistune
if opts.get("gfm"):
# Use plugins for GFM features
md = mistune.create_markdown(
renderer=renderer,
plugins=['strikethrough', 'table', 'task_lists']
)
else:
md = mistune.create_markdown(renderer=renderer)
html_content = md(text)
# Wrap in container
return f'<div class="markdown-content">{html_content}</div>'
def _extract_headings(self, text: str) -> List[Dict[str, Any]]:
"""
Extract heading structure from Markdown source.
"""
headings = []
heading_pattern = re.compile(r'^(#{1,6})\s+(.+)$', re.MULTILINE)
for match in heading_pattern.finditer(text):
level = len(match.group(1))
title = match.group(2).strip()
headings.append({
"level": level,
"title": title,
"offset": match.start(),
"line": text[:match.start()].count("\n") + 1,
})
return headings
def extract_toc(self, file_path: str) -> List[Dict[str, Any]]:
"""
Extract table of contents from a Markdown file.
Args:
file_path: Path to the Markdown file
Returns:
List of heading entries with level, title, and line number
"""
path = Path(file_path)
source_text = path.read_text(encoding="utf-8")
return self._extract_headings(source_text)
class HighlightRenderer(mistune.HTMLRenderer if MISTUNE_AVAILABLE else object):
"""
Custom Mistune renderer with Pygments syntax highlighting.
"""
def __init__(self):
if MISTUNE_AVAILABLE:
super().__init__()
self.formatter = HtmlFormatter(cssclass="highlight") if PYGMENTS_AVAILABLE else None
def block_code(self, code: str, info: str = None) -> str:
"""
Render a code block with syntax highlighting.
"""
if not PYGMENTS_AVAILABLE or not self.formatter:
escaped = html.escape(code)
lang_attr = f' class="language-{info}"' if info else ''
return f'<pre><code{lang_attr}>{escaped}</code></pre>\n'
try:
if info:
lexer = get_lexer_by_name(info, stripall=True)
else:
lexer = guess_lexer(code)
except Exception:
# Fall back to plain text
escaped = html.escape(code)
return f'<pre><code>{escaped}</code></pre>\n'
return highlight(code, lexer, self.formatter)
def codespan(self, text: str) -> str:
"""
Render inline code.
"""
escaped = html.escape(text)
return f'<code class="inline-code">{escaped}</code>'