yujunjun/olmocr
1
0
Fork 0
mirror of https://github.com/allenai/olmocr.git synced 2025-10-12 00:32:45 +00:00
Code Issues Packages Projects Releases Wiki Activity

Fix for iabooks

Browse Source
This commit is contained in:
Jake Poznanski 2025-08-16 17:26:51 +00:00
parent 0a9c82927f
commit cffbb82b0b
1 changed files with 11 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
olmocr/data/process_openai_batch_results.py
View File

@ -9,6 +9,7 @@ that mirrors the original structure with side-by-side PDF and MD files.
import argparse import argparse
import json import json
import shutil import shutil
import re
from pathlib import Path from pathlib import Path
from typing import Dict, Any, Optional from typing import Dict, Any, Optional
from concurrent.futures import ThreadPoolExecutor, as_completed from concurrent.futures import ThreadPoolExecutor, as_completed
@ -111,6 +112,15 @@ def process_single_result(
if not original_pdf_path.exists(): if not original_pdf_path.exists():
print(f"Warning: Original PDF not found: {original_pdf_path}") print(f"Warning: Original PDF not found: {original_pdf_path}")
original_pdf_path = str(original_pdf_path)
pattern = r'(.+?)(-\d+)\.pdf$'
replacement = r'\1.pdf\2.pdf'
original_pdf_path = Path(re.sub(pattern, replacement, original_pdf_path))
if not original_pdf_path.exists():
print(f"Error: Original PDF not found: {original_pdf_path}")
return False return False
# Create output paths # Create output paths