more useful explanations of PDF failures
This commit is contained in:
parent
6f4444f67e
commit
11ea5a0d58
|
@ -316,7 +316,7 @@ class Collator:
|
|||
with fitz.open(src) as pdf:
|
||||
images = pdf_images(pdf, self.args.force_convert_pdf)
|
||||
if images is None:
|
||||
print(f'Support for weirder PDFs not yet implemented, skipping {src}')
|
||||
print(f'Failed to enumerate page images in PDF, skipping {src}')
|
||||
return None
|
||||
|
||||
self.dest.mkdir(parents=True, exist_ok=True)
|
||||
|
@ -466,10 +466,11 @@ def pdf_images(pdf, force=False):
|
|||
else:
|
||||
yield extract_image(pdf, xref)
|
||||
else:
|
||||
print(f'\nPage {idx+1}: {len(page_images)} images, {len([img for img in page_images if img["xref"] == 0])} non-xref images, {len(page.get_text("blocks"))} total objects')
|
||||
if xref_mode:
|
||||
raise ValueError
|
||||
else:
|
||||
print(f'\nGenerating pixmap for page {idx+1}')
|
||||
print(f'Generating pixmap for page {idx+1}')
|
||||
pix = page.get_pixmap(dpi=PDF_FALLBACK_DPI)
|
||||
yield { 'ext': 'png', 'image': pix.tobytes('png') }
|
||||
print(f'\x1b[2K\r{idx+1}/{pdf.page_count} pages processed...', end='')
|
||||
|
|
Loading…
Reference in a new issue