diff options
| author | Bryan Newbold <bnewbold@archive.org> | 2020-06-25 17:23:06 -0700 |
|---|---|---|
| committer | Bryan Newbold <bnewbold@archive.org> | 2020-06-25 17:23:06 -0700 |
| commit | 585e0ba76687ae2872faed88bd1edc5b051136ca (patch) | |
| tree | 332d1915873717bfa440e872dcce457c71c91346 | |
| parent | b150a62569a972b2719da71403b744bafa4f3fb6 (diff) | |
| download | sandcrawler-585e0ba76687ae2872faed88bd1edc5b051136ca.tar.gz sandcrawler-585e0ba76687ae2872faed88bd1edc5b051136ca.zip | |
poppler: correct RGBA buffer endian-ness
| -rw-r--r-- | python/sandcrawler/pdfextract.py | 2 | ||||
| -rwxr-xr-x | python/scripts/pdf_thumbnail.py | 2 |
2 files changed, 2 insertions, 2 deletions
diff --git a/python/sandcrawler/pdfextract.py b/python/sandcrawler/pdfextract.py index 4606632..6a78a0a 100644 --- a/python/sandcrawler/pdfextract.py +++ b/python/sandcrawler/pdfextract.py @@ -158,7 +158,7 @@ def process_pdf(blob: bytes, thumb_size=(180,300), thumb_type="JPEG") -> PdfExtr renderer = poppler.PageRenderer() try: full_img = renderer.render_page(page0) - img = Image.frombuffer("RGBA", (full_img.width, full_img.height), full_img.data, 'raw', "RGBA", 0, 1) + img = Image.frombuffer("RGBA", (full_img.width, full_img.height), full_img.data, 'raw', "BGRA", 0, 1) img.thumbnail(thumb_size, Image.BICUBIC) buf = BytesIO() img.save(buf, thumb_type) diff --git a/python/scripts/pdf_thumbnail.py b/python/scripts/pdf_thumbnail.py index e093dc3..af08db6 100755 --- a/python/scripts/pdf_thumbnail.py +++ b/python/scripts/pdf_thumbnail.py @@ -22,7 +22,7 @@ def run(inpath, outpath): renderer = poppler.PageRenderer() full_page = renderer.render_page(page) - img = Image.frombuffer("RGBA", (full_page.width, full_page.height), full_page.data, 'raw', "RGBA", 0, 1) + img = Image.frombuffer("RGBA", (full_page.width, full_page.height), full_page.data, 'raw', "BGRA", 0, 1) img.thumbnail((180,300), Image.BICUBIC) #img.thumbnail((360,600), Image.BICUBIC) img.save(outpath) |