# html_template = """
#
#
#
#
#
#
#
# {content}
#
#
# """
html_template = """
{content}
"""
import re
import markdown
from html.parser import HTMLParser
from weasyprint import HTML
class TOCHtmlRestructurer(HTMLParser):
def __init__(self):
super().__init__()
self.output = []
self.level_stack = [] # Tracks nesting hierarchy
self.parent_count = 0
self.child_count = 0
self.capture_text = False
self.current_item_text = ""
self.current_href = None # Tracks anchor references dynamically
def handle_starttag(self, tag, attrs):
if tag in ['ol', 'ul']:
if not self.level_stack:
self.level_stack.append('parent')
else:
self.level_stack.append('child')
elif tag == 'li':
self.capture_text = True
self.current_item_text = ""
self.current_href = None
elif tag == 'a' and self.capture_text:
# Capture the link anchor destination safely
attrs_dict = dict(attrs)
self.current_href = attrs_dict.get('href')
def handle_endtag(self, tag):
if tag in ['ol', 'ul']:
if self.level_stack:
self.level_stack.pop()
elif tag == 'li':
self.capture_text = False
self.process_accumulated_item()
def handle_data(self, data):
if self.capture_text:
self.current_item_text += data
def process_accumulated_item(self):
# 1. Clean out stray prefix digits and list flags safely
clean = self.current_item_text.strip()
clean = re.sub(r'^([\d\.\s\-•]*\d+\.\d+|[\d\.\s\-•]*\d+\.)\s*', '', clean)
clean = re.sub(r'^\d+\s+(?=[A-Za-z])', '', clean)
clean = re.sub(r'^[\s\-•]*', '', clean).strip()
if not clean:
return
# 2. Check current depth level stack state
current_state = self.level_stack[-1] if self.level_stack else 'parent'
# 3. Rebuild the text as a clickable link if an anchor was present
if self.current_href:
display_text = f'{clean}'
else:
display_text = clean
# 4. Generate structured HTML outputs with calculated taxonomy counters
if current_state == 'parent':
self.parent_count += 1
self.child_count = 0
self.output.append(f'{self.parent_count}. {display_text}
')
else:
self.child_count += 1
self.output.append(f'{self.parent_count}.{self.child_count} {display_text}
')
def extract_and_fix_toc_blocks(html_content):
"""
Finds the compiled HTML lists block, intercepts it, clears
numbering artifacts structurally via HTML parsing tree mechanics.
"""
match = re.search(r'(<(?:ol|ul)>[\s\S]*?)', html_content)
if not match:
return html_content
raw_toc_html = match.group(1)
parser = TOCHtmlRestructurer()
parser.feed(raw_toc_html)
new_toc_html = '\n
Table of Contents
\n'
new_toc_html += '\n'.join(parser.output)
new_toc_html += '\n'
return html_content.replace(raw_toc_html, new_toc_html, 1)
def export_to_pdf(inputText, filename):
"""
Processes plain dynamic markdown, captures structural compilation output,
restructures TOC nodes downstream safely, and outputs a high-fidelity PDF.
"""
html_content = markdown.markdown(inputText, extensions=['tables', 'fenced_code'])
final_body_content = extract_and_fix_toc_blocks(html_content)
final_html = html_template.format(content=final_body_content)
HTML(string=final_html).write_pdf(filename)