Spaces:
Running
Running
# PDF_Ingestion_Lib.py | |
######################################### | |
# Library to hold functions for ingesting PDF files.# | |
# | |
#################### | |
# Function List | |
# | |
# 1. convert_pdf_to_markdown(pdf_path) | |
# 2. ingest_pdf_file(file_path, title=None, author=None, keywords=None): | |
# 3. | |
# | |
# | |
#################### | |
# Import necessary libraries | |
from datetime import datetime | |
import logging | |
import os | |
import re | |
import shutil | |
import tempfile | |
# | |
# Import External Libs | |
import pymupdf | |
import pymupdf4llm | |
from docling.document_converter import DocumentConverter | |
# | |
# Import Local | |
from App_Function_Libraries.DB.DB_Manager import add_media_with_keywords | |
from App_Function_Libraries.Metrics.metrics_logger import log_counter, log_histogram | |
# | |
# Constants | |
MAX_FILE_SIZE_MB = 50 | |
CONVERSION_TIMEOUT_SECONDS = 300 | |
# | |
####################################################################################################################### | |
# Function Definitions | |
# | |
def extract_text_and_format_from_pdf(pdf_path): | |
""" | |
Extract text from a PDF file and convert it to Markdown, preserving formatting. | |
""" | |
try: | |
log_counter("pdf_text_extraction_attempt", labels={"file_path": pdf_path}) | |
start_time = datetime.now() | |
markdown_text = "" | |
with pymupdf.open(pdf_path) as doc: | |
for page_num, page in enumerate(doc, 1): | |
markdown_text += f"## Page {page_num}\n\n" | |
blocks = page.get_text("dict")["blocks"] | |
current_paragraph = "" | |
for block in blocks: | |
if block["type"] == 0: # Text block | |
for line in block["lines"]: | |
line_text = "" | |
for span in line["spans"]: | |
text = span["text"] | |
font_size = span["size"] | |
font_flags = span["flags"] | |
# Apply formatting based on font size and flags | |
if font_size > 20: | |
text = f"# {text}" | |
elif font_size > 16: | |
text = f"## {text}" | |
elif font_size > 14: | |
text = f"### {text}" | |
if font_flags & 2 ** 0: # Bold | |
text = f"**{text}**" | |
if font_flags & 2 ** 1: # Italic | |
text = f"*{text}*" | |
line_text += text + " " | |
# Remove hyphens at the end of lines | |
line_text = line_text.rstrip() | |
if line_text.endswith('-'): | |
line_text = line_text[:-1] | |
else: | |
line_text += " " | |
current_paragraph += line_text | |
# End of block, add paragraph | |
if current_paragraph: | |
# Remove extra spaces | |
current_paragraph = re.sub(r'\s+', ' ', current_paragraph).strip() | |
markdown_text += current_paragraph + "\n\n" | |
current_paragraph = "" | |
elif block["type"] == 1: # Image block | |
markdown_text += "[Image]\n\n" | |
markdown_text += "\n---\n\n" # Page separator | |
# Clean up hyphenated words | |
markdown_text = re.sub(r'(\w+)-\s*\n(\w+)', r'\1\2', markdown_text) | |
end_time = datetime.now() | |
processing_time = (end_time - start_time).total_seconds() | |
log_histogram("pdf_text_extraction_duration", processing_time, labels={"file_path": pdf_path}) | |
log_counter("pdf_text_extraction_success", labels={"file_path": pdf_path}) | |
return markdown_text | |
except Exception as e: | |
logging.error(f"Error extracting text and formatting from PDF: {str(e)}") | |
log_counter("pdf_text_extraction_error", labels={"file_path": pdf_path, "error": str(e)}) | |
raise | |
def pymupdf4llm_parse_pdf(pdf_path): | |
""" | |
Extract text from a PDF file and convert it to Markdown, preserving formatting. | |
""" | |
try: | |
log_counter("pdf_text_extraction_attempt", labels={"file_path": pdf_path}) | |
start_time = datetime.now() | |
markdown_text = pymupdf4llm.to_markdown(pdf_path) | |
end_time = datetime.now() | |
processing_time = (end_time - start_time).total_seconds() | |
log_histogram("pdf_text_extraction_duration", processing_time, labels={"file_path": pdf_path}) | |
log_counter("pdf_text_extraction_success", labels={"file_path": pdf_path}) | |
return markdown_text | |
except Exception as e: | |
logging.error(f"Error extracting text and formatting from PDF: {str(e)}") | |
log_counter("pdf_text_extraction_error", labels={"file_path": pdf_path, "error": str(e)}) | |
raise | |
def extract_metadata_from_pdf(pdf_path): | |
""" | |
Extract metadata from a PDF file using PyMuPDF. | |
""" | |
try: | |
log_counter("pdf_metadata_extraction_attempt", labels={"file_path": pdf_path}) | |
with pymupdf.open(pdf_path) as doc: | |
metadata = doc.metadata | |
log_counter("pdf_metadata_extraction_success", labels={"file_path": pdf_path}) | |
return metadata | |
except Exception as e: | |
logging.error(f"Error extracting metadata from PDF: {str(e)}") | |
log_counter("pdf_metadata_extraction_error", labels={"file_path": pdf_path, "error": str(e)}) | |
return {} | |
def process_and_ingest_pdf(file, title, author, keywords, parser='pymupdf4llm'): | |
if file is None: | |
log_counter("pdf_ingestion_error", labels={"error": "No file uploaded"}) | |
return "Please select a PDF file to upload." | |
try: | |
log_counter("pdf_ingestion_attempt", labels={"file_name": file.name}) | |
start_time = datetime.now() | |
# Create a temporary directory | |
with tempfile.TemporaryDirectory() as temp_dir: | |
# Create a path for the temporary PDF file | |
temp_path = os.path.join(temp_dir, "temp.pdf") | |
# Copy the contents of the uploaded file to the temporary file | |
shutil.copy(file.name, temp_path) | |
if parser == 'pymupdf': | |
# Extract text and convert to Markdown | |
markdown_text = extract_text_and_format_from_pdf(temp_path) | |
elif parser == 'pymupdf4llm': | |
# Extract text and convert to Markdown | |
markdown_text = pymupdf4llm_parse_pdf(temp_path) | |
elif parser == 'docling': | |
# Extract text and convert to Markdown using Docling | |
converter = DocumentConverter() | |
parsed_pdf = converter.convert(temp_path) | |
markdown_text = parsed_pdf.document.export_to_markdown() | |
# Extract metadata from PDF | |
metadata = extract_metadata_from_pdf(temp_path) | |
# Use metadata for title and author if not provided | |
if not title: | |
title = metadata.get('title', os.path.splitext(os.path.basename(file.name))[0]) | |
if not author: | |
author = metadata.get('author', 'Unknown') | |
# If keywords are not provided, use a default keyword | |
if not keywords: | |
keywords = 'pdf_file,markdown_converted' | |
else: | |
keywords = f'pdf_file,markdown_converted,{keywords}' | |
# Add metadata-based keywords | |
if 'subject' in metadata: | |
keywords += f",{metadata['subject']}" | |
# Add the PDF content to the database | |
add_media_with_keywords( | |
url=file.name, | |
title=title, | |
media_type='document', | |
content=markdown_text, | |
keywords=keywords, | |
prompt='No prompt for PDF files', | |
summary='No summary for PDF files', | |
transcription_model='None', | |
author=author, | |
ingestion_date=datetime.now().strftime('%Y-%m-%d') | |
) | |
end_time = datetime.now() | |
processing_time = (end_time - start_time).total_seconds() | |
log_histogram("pdf_ingestion_duration", processing_time, labels={"file_name": file.name}) | |
log_counter("pdf_ingestion_success", labels={"file_name": file.name}) | |
return f"PDF file '{title}' by {author} ingested successfully and converted to Markdown." | |
except Exception as e: | |
logging.error(f"Error ingesting PDF file: {str(e)}") | |
log_counter("pdf_ingestion_error", labels={"file_name": file.name, "error": str(e)}) | |
return f"Error ingesting PDF file: {str(e)}" | |
def process_and_cleanup_pdf(file, title, author, keywords, parser='pymupdf4llm'): | |
if file is None: | |
log_counter("pdf_processing_error", labels={"error": "No file uploaded"}) | |
return "No file uploaded. Please upload a PDF file." | |
try: | |
log_counter("pdf_processing_attempt", labels={"file_name": file.name}) | |
start_time = datetime.now() | |
result = process_and_ingest_pdf(file, title, author, keywords, parser) | |
end_time = datetime.now() | |
processing_time = (end_time - start_time).total_seconds() | |
log_histogram("pdf_processing_duration", processing_time, labels={"file_name": file.name}) | |
log_counter("pdf_processing_success", labels={"file_name": file.name}) | |
return result | |
except Exception as e: | |
logging.error(f"Error in processing and cleanup: {str(e)}") | |
log_counter("pdf_processing_error", labels={"file_name": file.name, "error": str(e)}) | |
return f"Error: {str(e)}" | |
# | |
# End of PDF_Ingestion_Lib.py | |
####################################################################################################################### | |