Caching cutlef.py
parent
ebc7a9aadc
commit
38ea0069af
11
cutleft.py
11
cutleft.py
|
|
@ -1,4 +1,5 @@
|
|||
import sys
|
||||
from functools import lru_cache
|
||||
import os
|
||||
import time
|
||||
import json # Added for schema output
|
||||
|
|
@ -82,14 +83,20 @@ def stitch_images(image_list):
|
|||
|
||||
return combined
|
||||
|
||||
@lru_cache(maxsize=3)
|
||||
def get_pdf_pages(filename):
|
||||
"""Caches the heavy PDF rendering step for the current and next files."""
|
||||
pdf_path = os.path.join(INPUT_DIR, filename)
|
||||
return convert_from_path(pdf_path)
|
||||
|
||||
def process_single_pdf(filename, shift_offset=0, max_per_file=5):
|
||||
"""
|
||||
Converts PDF to stitched images.
|
||||
Returns a tuple: (preview_image_resized, list_of_split_images, schema_dict)
|
||||
"""
|
||||
pdf_path = os.path.join(INPUT_DIR, filename)
|
||||
# pdf_path = os.path.join(INPUT_DIR, filename)
|
||||
try:
|
||||
pages = convert_from_path(pdf_path)
|
||||
pages = get_pdf_pages(filename)
|
||||
cropped_images = []
|
||||
|
||||
for img in pages:
|
||||
|
|
|
|||
Loading…
Reference in New Issue