| Git Clone URL: | https://aur.archlinux.org/k2pdfopt.git (read-only, click to copy) |
|---|---|
| Package Base: | k2pdfopt |
| Description: | Tool for optimizing PDF/DJVU files for mobile e-readers and smartphones. |
| Upstream URL: | http://www.willus.com/k2pdfopt |
| Keywords: | convert djvu ebook kindle ocr pdf reflow |
| Licenses: | GPL3 |
| Submitter: | FacuTuesca |
| Maintainer: | gilcu3 |
| Last Packager: | gilcu3 |
| Votes: | 49 |
| Popularity: | 0.000005 |
| First Submitted: | 2014-03-14 21:46 (UTC) |
| Last Updated: | 2024-05-28 15:01 (UTC) |
I just updated to package to the best of my abilities, still it is very buggy and only some functionality works. In my tests this is also the case with the debian package. If anyone is able to fix it please let me know. For the moment the only solution to get full functionality is to download the pre-built binary from willus.
PS: kudos to the debian developer for fixing the build with mupdf, currently it works on my end. Only tesseract support remains disabled.
I have the same problem as Dingisoul.
There's a bug in aur's k2pdfopt but not in the executable k2pdfopt directly from willus.com. Both version is 2.53. The output PDF files ends far before the EOF. I have tried the -p options, but didn't work. All the pdfs I'm using are conference papers. I don't know why , so asking for help here. This is one input file and the output file generated by aur's k2pdfopt with default options output file
Thanks for updating this package ;). k2pdfopt's build system is weird.
Tried to update but I can't compile even the basic version of his website of 2.53 (missing djvu.h). On the debian download page, there is a 2.52 version with more patches, although I haven't tried them yet.
Applying patches from Debian sid I still get error from linker:
/usr/bin/ld: willuslib/libwilluslib.a(ocrtess.c.o): in function `ocrtess_set_logfile':
ocrtess.c:(.text+0x1a4): undefined reference to `tess_capi_set_logfile'
/usr/bin/ld: willuslib/libwilluslib.a(ocrtess.c.o): in function `ocrtess_debug_message':
ocrtess.c:(.text+0x1b4): undefined reference to `tess_capi_debug_message'
/usr/bin/ld: willuslib/libwilluslib.a(ocrtess.c.o): in function `ocrtess_init':
ocrtess.c:(.text+0x1192): undefined reference to `tess_capi_init'
/usr/bin/ld: willuslib/libwilluslib.a(ocrtess.c.o): in function `ocrtess_end':
ocrtess.c:(.text+0x11e4): undefined reference to `tess_capi_end'
/usr/bin/ld: willuslib/libwilluslib.a(ocrtess.c.o): in function `ocrtess_ocrwords_from_bmp8':
ocrtess.c:(.text+0x1575): undefined reference to `tess_capi_get_ocr_multiword'
/usr/bin/ld: willuslib/libwilluslib.a(ocrtess.c.o): in function `ocrtess_from_bmp8':
ocrtess.c:(.text+0x1a84): undefined reference to `tess_capi_get_ocr'
collect2: error: ld returned 1 exit status
probably some kind of conflict with modified tesseract. I don't really have time to fix this so if someone wants to step in.. I am disowning this now. Thanks.
How to build this package with ocr?
I can no longer maintain this package. Up for adoption.
Pinned Comments
gilcu3 commented on 2024-05-19 09:29 (UTC) (edited on 2024-05-28 15:03 (UTC) by gilcu3)
I just updated to package to the best of my abilities, still it is very buggy and only some functionality works. In my tests this is also the case with the debian package. If anyone is able to fix it please let me know. For the moment the only solution to get full functionality is to download the pre-built binary from willus.
PS: kudos to the debian developer for fixing the build with
mupdf, currently it works on my end. Onlytesseractsupport remains disabled.