# ------------------- Bookmarks / Outline ------------------- # def extract_bookmarks(pdf_path: Path, out_dir: Path): """Export the PDF's outline (bookmarks) as a JSON hierarchy.""" doc = fitz.open(str(pdf_path)) toc = doc.get_toc(simple=False) # list of [level, title, page, ...] # Turn into a nested dict for readability def build_tree(toc_entries): tree = [] stack = [(0, tree)] # (level, container) for level, title, page, *_ in toc_entries: while level <= stack[-1][0]: stack.pop() node = "title": title, "page": page, "children": [] stack[-1][1].append(node) stack.append((level, node["children"])) return tree
import argparse import json import os import re import sys from pathlib import Path from typing import List, Dict
Requirements (install via pip): pip install pdfplumber pymupdf tqdm tabula-py ocrmypdf # tabula-py needs Java; ocrmypdf needs Tesseract + poppler agnibina filetype.pdf
safe_mkdir(out_dir / "tables") # tabula can auto-detect tables across the whole doc: tables = tabula.read_pdf(str(pdf_path), pages="all", multiple_tables=True, pandas_options='dtype': str) print(f"📊 Detected len(tables) tables.") for i, df in enumerate(tables, start=1): # Try to infer the page number from the DataFrame's metadata if present # (tabula doesn’t expose page number directly; you can run per-page if you need it) csv_path = out_dir / f"tables/table_i:03d.csv" df.to_csv(csv_path, index=False) print(f" → Saved table i → csv_path")
# ------------------- Tables ------------------- # def extract_tables(pdf_path: Path, out_dir: Path): """ Uses tabula-py (Java) to pull out tables. Each table is saved as CSV under out_dir/tables/page_XX_table_YY.csv . """ try: import tabula except ImportError: print("⚠️ tabula-py not installed – skipping table extraction.") return tree)] # (level
# Optionally re-run the extraction on the OCR’d file # (You could replace the original path with ocr_output for downstream steps)
count = 0 for i in range(doc.embfile_count()): info = doc.embfile_info(i) fname = clean_filename(info["filename"]) data = doc.embfile_get(i) (att_dir / fname).write_bytes(data) count += 1 doc.close() print(f"📦 Extracted count embedded file(s).") container) for level
# ------------------- Text + Layout ------------------- # def extract_text_and_layout(pdf_path: Path, out_dir: Path) -> List[Dict]: """ Returns a list (one dict per page) with: - page_number - plain_text - list of text elements text, x0, y0, x1, y1, fontname, size """ pages_info = [] with pdfplumber.open(str(pdf_path)) as pdf: for page_num, page in enumerate(tqdm(pdf.pages, desc="Pages (text/layout)")): plain = page.extract_text() # layout objects (characters) – useful for heading detection chars = page.chars # each char already has x0, y0, x1, y1, fontname, size # Group chars into words/lines if you like, but we keep raw for flexibility pages_info.append( "page_number": page_num + 1, "text": plain, "characters": chars, ) # Save raw JSON for later inspection (out_dir / "text_layout.json").write_text(json.dumps(pages_info, indent=2, ensure_ascii=False)) return pages_info