Option `--output-type pdf` causes PDF to be blank (with wrong geometry)

With following PDF file: test.pdf

If you do ocrmypdf --output-type pdf test.pdf foo.pdf, the file foo.pdf is 10 times smaller than the original, and is blank when you open it. Remove the option --output-type pdf and the problem is gone.

I noted this:

$ pdfimages foo.pdf img
$ identify -format "%@" img-000.pbm
identify-im6.q16: geometry does not contain image `img-000.pbm' @ warning/attribute.c/GetImageBoundingBox/247.
0x0+2023+2866

The geometry 0x0 is obviously problematic.

$ ocrmypdf --version
9.4.0+dfsg
$ tesseract --version
tesseract 4.1.1
 leptonica-1.79.0
  libgif 5.1.9 : libjpeg 6b (libjpeg-turbo 1.5.2) : libpng 1.6.37 : libtiff 4.1.0 : zlib 1.2.11 : libwebp 0.6.1 : libopenjp2 2.3.1
 Found AVX2
 Found AVX
 Found FMA
 Found SSE
 Found libarchive 3.4.0 zlib/1.2.11 liblzma/5.2.4 bz2lib/1.0.8 liblz4/1.9.1 libzstd/1.4.3

[gnewell3]

Hi I can confirm this issue has been resolved.

Tested with the test pdf provided on ocrmypdf:v13.6.2 and it gives no issues. All perfect crisp OCR with text layer 😉

Believe can be closed 😃!

@jbarlow83

[jbarlow83]

@fnoy You have an old version of ocrmypdf.

@damnms What version of ocrmypdf do you have?

It’s quite possible this issue was fixed by #732 or commit 49734d which is in v10.2.1+