Spaces:
Running
Running
File size: 3,001 Bytes
cc01c7a 8f970d0 cc01c7a | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 | import re
import requests
import io
import logging
logger = logging.getLogger(__name__)
def generate_latex(markdown_text: str) -> str:
"""Converts basic markdown with MathJax into a raw LaTeX document string."""
tex = markdown_text
# Protect math blocks temporarily so regex doesn't mess up math internals
math_blocks = []
def save_math(match):
math_blocks.append(match.group(0))
return f"__MATH_{len(math_blocks)-1}__"
tex = re.sub(r'\$\$.*?\$\$', save_math, tex, flags=re.DOTALL)
tex = re.sub(r'\$.*?\$', save_math, tex)
# Markdown -> LaTeX conversions
tex = re.sub(r'\*\*(.*?)\*\*', r'\\textbf{\1}', tex)
tex = re.sub(r'\*(.*?)\*', r'\\textit{\1}', tex)
tex = re.sub(r'^### (.*)$', r'\\subsubsection*{\1}', tex, flags=re.MULTILINE)
tex = re.sub(r'^## (.*)$', r'\\subsection*{\1}', tex, flags=re.MULTILINE)
tex = re.sub(r'^# (.*)$', r'\\section*{\1}', tex, flags=re.MULTILINE)
# Simple list handling
tex = re.sub(r'^- (.*)$', r'\\item \1', tex, flags=re.MULTILINE)
if "\\item" in tex:
# Extremely naive list wrapping
tex = re.sub(r'((\\item.*\n?)+)', r'\\begin{itemize}\n\1\\end{itemize}\n', tex)
# Replace empty lines with double escape for spacing if needed
# Actually, empty lines in LaTeX are good for paragraph breaks.
# Restore math blocks
for i, block in enumerate(math_blocks):
tex = tex.replace(f"__MATH_{i}__", block)
preamble = r"""\documentclass[12pt,a4paper]{article}
\usepackage[utf8]{inputenc}
\usepackage[T1]{fontenc}
\usepackage{amsmath, amssymb, amsfonts}
\usepackage{geometry}
\geometry{a4paper, margin=1in}
\usepackage{hyperref}
\usepackage{xcolor}
\title{MathMinds Generated Report}
\author{MathMinds AI Assistant}
\date{\today}
\begin{document}
\maketitle
"""
return preamble + tex + "\n\n\\end{document}"
def compile_pdf(tex_content: str) -> bytes:
"""Compiles a complete LaTeX string to a PDF using public latexonline.cc API."""
try:
import urllib.parse
encoded_tex = urllib.parse.quote(tex_content)
url = f"https://latexonline.cc/compile?text={encoded_tex}"
# If the encoded URL is massive, the server might reject the GET request.
# Fallback to the POST /data endpoint with a multipart file upload.
if len(url) > 8000:
logger.info("URL too long, falling back to POST /data multipart")
r = requests.post(
"https://latexonline.cc/data?command=pdflatex",
files={"file": ("document.tex", tex_content)},
timeout=45
)
else:
r = requests.get(url, timeout=45)
if r.status_code == 200:
return r.content
else:
logger.error(f"LaTeX compile failed: {r.status_code} - {r.text[:200]}")
except Exception as e:
logger.error(f"LaTeX API exception: {e}")
return b""
|