k2pdfopt: optimizes PDFs for viewing on e-readers - Page 2

youssef · 09-14-2011, 11:24 AM

doesn't work with RTL langauges like arabic, it treats the text as english and as a result an arabic sentence is broken in two line but in reverse order which make it unreadable, need to add an RTL option to the app

thomass · 09-14-2011, 12:21 PM

Quote:

Originally Posted by youssef

doesn't work with RTL langauges like arabic, it treats the text as english and as a result an arabic sentence is broken in two line but in reverse order which make it unreadable, need to add an RTL option to the app

indeed it has: http://www.willus.com/archive/k2pdfo...ght2left.shtml

youssef · 09-14-2011, 12:45 PM

Quote:

Originally Posted by thomass

indeed it has: http://www.willus.com/archive/k2pdfo...ght2left.shtml

thanks, but it fails if the document is a mix between arabic and english (i.e an RTL and LTR), the app should change the direction based on the font,
but anyways great effort although if it fails for a mixed doc

Jellby · 09-15-2011, 04:30 AM

It's complicated enough to get mixed LTR and RTL texts to display correctly, I don't think it's easy to process it backwards from image to text, but maybe something could be done with the intrinsic directionality of Unicode characters...

willus · 09-17-2011, 08:06 AM

The way k2pdfopt works--by immediately converting to bitmaps--I'm afraid auto-detecting mixed text is beyond my current bag of tricks. It would have to involve some form of OCR or some new way of parsing the PDF file. Sorry.

tannenba · 09-18-2011, 11:18 AM

Thanks for the new version, this is definitely the best PDF optimizer out there! I would say the reading experience for 75% of all the PDFs I read on my Kindle (and I read a lot) is vastly improved thanks to this program.

I found a PDF file that crashes the new version 1.25 of k2pdfopt. IIRC, conversion of this file used to work with older releases. The PDF file is here:
http://people.csail.mit.edu/dnj/publ...comparison.pdf
If I convert this file using v1.25 on Windows XP (win32) , it crashes a couple of pages into the file. If I use the "-gs" option, it completes successfully. Perhaps the new version has an updated PDF library that has a regression?

johnnyb · 09-18-2011, 08:24 PM

It works fine (and thx a lot for it)... only how am I able to append options on Mac? It seems it always interprets them as part of the file path and thus never makes use of them (justification is something I would really like to do)...

willus · 09-18-2011, 11:29 PM

Quote:

Originally Posted by tannenba

Thanks for the new version, this is definitely the best PDF optimizer out there! I would say the reading experience for 75% of all the PDFs I read on my Kindle (and I read a lot) is vastly improved thanks to this program.

I found a PDF file that crashes the new version 1.25 of k2pdfopt. IIRC, conversion of this file used to work with older releases. The PDF file is here:
http://people.csail.mit.edu/dnj/publ...comparison.pdf
If I convert this file using v1.25 on Windows XP (win32) , it crashes a couple of pages into the file. If I use the "-gs" option, it completes successfully. Perhaps the new version has an updated PDF library that has a regression?

Sorry--I can't reproduce this. That PDF converts fine on my PC. What CPU do you have and how much RAM? I haven't updated mupdf since I incorporated it, so a regression in mupdf can't be it. Have you tried the conversion on a different PC? It could be some sort of variable not being initialized correctly so that it doesn't reproduce consistently. Or an array out of bounds (which often has inconsistent results). Sorry! At least ghostscript saved the day. BTW, I've posted v1.26.

willus · 09-18-2011, 11:38 PM

Quote:

Originally Posted by johnnyb

It works fine (and thx a lot for it)... only how am I able to append options on Mac? It seems it always interprets them as part of the file path and thus never makes use of them (justification is something I would really like to do)...

You should be able to create a script like so:
1. (Start terminal shell and type in 2 - 5 below)
2. cd Desktop
3. cat > k2pdfopt_custom
./k2pdfopt -ui -j 0 -odpi 250
<Ctrl-D>
4. chmod 755 k2pdfopt_custom
5. exit
Now k2pdfopt_custom should behave like k2pdfopt but with the new command line options (which are just an example). Add the -ui so that the interactive menu still comes up.

SkyDream · 10-14-2011, 08:48 AM

Unfortunately i'm getting "format error: not a PDF or corrupted." when trying to open the result after converting with "-odpi 150" or without any option. The source PDF is about 3.6MB and the result varies between 1.5M and 300KB.

The source contains pictures, 2- and 3-columns pages.

I'm using the latest version (just downloaded) for Windows running XP SP3.

Any idea ?

willus · 10-14-2011, 09:54 PM

Quote:

Originally Posted by SkyDream

Unfortunately i'm getting "format error: not a PDF or corrupted." when trying to open the result after converting with "-odpi 150" or without any option. The source PDF is about 3.6MB and the result varies between 1.5M and 300KB.

The source contains pictures, 2- and 3-columns pages.

I'm using the latest version (just downloaded) for Windows running XP SP3.

Any idea ?

If you can post the PDF files (source and result) or e-mail them to me, I'll be happy to investigate.

SkyDream · 10-15-2011, 05:37 AM

Quote:

Originally Posted by willus

If you can post the PDF files (source and result) or e-mail them to me, I'll be happy to investigate.

Hmmm..little bit tricky as i got it from a friend who got it from another one who in turn downloaded it from some other website which is located in the far east of europe...

but i'll check with some other PDF and come back to it...

SkyDream · 10-15-2011, 02:41 PM

Quote:

Originally Posted by SkyDream

but i'll check with some other PDF and come back to it...

Ok, i check with other PDFs but no problem.
Then i checked the PDF in question again on my other PC running Vista instead of XP -> no problem. Tried again under XP -> Error. Again under Vista -> Ok.

willus · 10-16-2011, 12:05 AM

Quote:

Originally Posted by SkyDream

Ok, i check with other PDFs but no problem.
Then i checked the PDF in question again on my other PC running Vista instead of XP -> no problem. Tried again under XP -> Error. Again under Vista -> Ok.

Sorry. I'd really need to see the file to track down whatever is happening in XP. I understand if you can't post it, though.

ThomasC · 10-16-2011, 02:19 AM

Quote:

Originally Posted by SkyDream

Ok, i check with other PDFs but no problem.
Then i checked the PDF in question again on my other PC running Vista instead of XP -> no problem. Tried again under XP -> Error. Again under Vista -> Ok.

You are not trying to use the win64 version (I'm using this with windows 7 no problem)?

I think there is a win32 version, probably for xp, or maybe an older xp version is available?

Cheers,
Tom

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Viewing PDFs with another font	Font	PocketBook	4	11-12-2010 08:27 AM
Viewing Textbook PDFs...	NJReader	enTourage Archive	4	08-17-2010 05:17 PM
PRS-600 Restart bug while viewing PDFs?	conundrum	Sony Reader	2	03-04-2010 08:46 PM
More on viewing pdfs	dso371	Bookeen	8	03-11-2008 07:15 PM
Viewing Untagged PDFs on Palm T\|X	Eroica	Reading and Management	3	12-10-2007 01:44 PM

09-14-2011, 11:24 AM	#16
youssef Member Posts: 15 Karma: 12 Join Date: Jan 2010 Device: JBL	doesn't work with RTL langauges like arabic, it treats the text as english and as a result an arabic sentence is broken in two line but in reverse order which make it unreadable, need to add an RTL option to the app

09-15-2011, 04:30 AM	#19
Jellby frumious Bandersnatch Posts: 7,515 Karma: 18512745 Join Date: Jan 2008 Location: Spaniard in Sweden Device: Cybook Orizon, Kobo Aura	It's complicated enough to get mixed LTR and RTL texts to display correctly, I don't think it's easy to process it backwards from image to text, but maybe something could be done with the intrinsic directionality of Unicode characters...

09-17-2011, 08:06 AM	#20
willus Fuzzball, the purple cat Posts: 1,272 Karma: 11087488 Join Date: Jun 2011 Location: California Device: iPad	The way k2pdfopt works--by immediately converting to bitmaps--I'm afraid auto-detecting mixed text is beyond my current bag of tricks. It would have to involve some form of OCR or some new way of parsing the PDF file. Sorry.

09-18-2011, 11:18 AM	#21
tannenba Member Posts: 13 Karma: 10 Join Date: Nov 2004 Device: Kindle 3 WiFi	Thanks for the new version, this is definitely the best PDF optimizer out there! I would say the reading experience for 75% of all the PDFs I read on my Kindle (and I read a lot) is vastly improved thanks to this program. I found a PDF file that crashes the new version 1.25 of k2pdfopt. IIRC, conversion of this file used to work with older releases. The PDF file is here: http://people.csail.mit.edu/dnj/publ...comparison.pdf If I convert this file using v1.25 on Windows XP (win32) , it crashes a couple of pages into the file. If I use the "-gs" option, it completes successfully. Perhaps the new version has an updated PDF library that has a regression?

09-18-2011, 08:24 PM	#22
johnnyb Cloud Reader Posts: 1,110 Karma: 4000066 Join Date: Aug 2010 Device: Kindle Oasis, Kindle Scribe, iPad Pro 11	It works fine (and thx a lot for it)... only how am I able to append options on Mac? It seems it always interprets them as part of the file path and thus never makes use of them (justification is something I would really like to do)...

10-14-2011, 08:48 AM	#25
SkyDream Wizard Posts: 1,377 Karma: 9400 Join Date: Sep 2009 Location: Europe Device: PRS-650, iPod touch 4G, iPad 3	Unfortunately i'm getting "format error: not a PDF or corrupted." when trying to open the result after converting with "-odpi 150" or without any option. The source PDF is about 3.6MB and the result varies between 1.5M and 300KB. The source contains pictures, 2- and 3-columns pages. I'm using the latest version (just downloaded) for Windows running XP SP3. Any idea ?