k2pdfopt: optimizes PDFs for viewing on e-readers

[willus]

Originally Posted by chrox

It's turned on in koreader by default since there is no good default spacing value for both Western text and CJK text. ... I would rather suggest leaving this option turned off by default until better detecting algorithm comes up.

Okay--I'll leave it off by default then, but it will be able to be turned on with a command-line option. Thanks for the reply.

[wlaf]

I have used this program and it's really great, and I want to appreciate willus for creating such a great application.
But rather than not being open-source, I have problems with downloading it, because it does not provide a simple link and it provides a combination of Captcha/downloader script, and because of the slow connection of the website (maybe not with every single ISP and country), my download remains incomplete. I want to ask here to have the latest version of program attached here (or in some mirrors), so users can download it more easily.

[willus]

Originally Posted by wlaf

... I have problems with downloading it, because it does not provide a simple link and it provides a combination of Captcha/downloader script, and because of the slow connection of the website (maybe not with every single ISP and country), my download remains incomplete. I want to ask here to have the latest version of program attached here (or in some mirrors), so users can download it more easily.

Can you help me understand why my capcha method causes problems for a slow connection?

[wlaf]

Originally Posted by willus

Can you help me understand why my capcha method causes problems for a slow connection?

I can not continue my incomplete download, and several retries lead to the same incomplete download.

Implementing a resumable download script needs some tweaking:
http://stackoverflow.com/questions/1...-send-the-file

[willus]

Originally Posted by wlaf

I can not continue my incomplete download, and several retries lead to the same incomplete download.

Implementing a resumable download script needs some tweaking:
http://stackoverflow.com/questions/1...-send-the-file

Thank you for the link. I have implemented the resumable download features in my k2pdfopt download script based on your link, but I don't have a way to test it. Do you have a way to test it?

[wlaf]

Originally Posted by willus

Thank you for the link. I have implemented the resumable download features in my k2pdfopt download script based on your link, but I don't have a way to test it. Do you have a way to test it?

Right now, the resume does not work, but now the file size is visible in the browser when I am downloading, that means the implementation is partly OK! I hope you will have time to complete the implementation of resumable download. Thanks.

[willus]

The crop box option (-cbox) is more powerful now, KOReader patches have been applied, and other minor bug fixes and enhancments have been incorporated. See the web site for details.

[willus]

This is a mostly cosmetic change to some of the source files, but it does correct an incorrect win32 build from v2.13. Thanks to Dirk Thierbach for some of the source modifications. See the web site for details.

[sage79]

I notice that from few version your software creates the pdf bookmark. There is a bug in kobo firmware that block pdf bookmarked with the frase "this book is drm protected"

[willus]

Originally Posted by sage79

I notice that from few version your software creates the pdf bookmark. There is a bug in kobo firmware that block pdf bookmarked with the frase "this book is drm protected"

Bookmarks in the k2pdfopt output can be turned off with the -toc- command-line option (if you are using the Windows GUI, put that into the "additional options" box).

[ittiandro]

Originally Posted by thomass

K2pdfopt optimizes PDF/DJVU files for mobile e-readers (e.g. the Kindle) and smartphones. It works well on multi-column PDF/DJVU files. K2pdfopt is open source.

I have a number of image-based( scanned) PDF files which I am unable to open ( or they open extremely slowly) in my Android pad , unless I convert them to Epub, but then I can no longer control the page layout and the fonts, as I need to because the page size and the fonts have decreased in the conversion.
So I came across this k2pdfopt soiftware.

I tried it and, indeed, on the plus side, now I can finally open the file almost instantly, whereas before it would take ages to open( if it opened!). But there is a big drawback: the k2pdfopt PDF files become too huge: from the original 13 MB to 100( ONE HUNDRED!)MB!!! And I have 50 of these books to convert!

It seems I am caught in a double bind: either I can finally open and read the PDF file( with k2pdfopt) with full control but then the file file is too huge or, if I keep the smaller EPUB, file I cannot control the page layout and the fonts, as I can easily do with all the regular EPUB files.
I hear that there is a way around by converting the PDF file to TEXT or HTML and then back to PDF or EPUB, but I am a bit fuzzy on this.

How can I decrease the size of the k2pdfopt converted files? I tried the Acrobat Reader " SAVE AS " options : save as REDUCED, or OPTIMIZED PDF, but it won't work. Any comment or help on this?

Thanks

Ittiandro

[willus]

Originally Posted by ittiandro

How can I decrease the size of the k2pdfopt converted files?

Please see the third question in the k2pdfopt FAQ. If that doesn't help, please post an example of one of your source documents (e.g. the 13 MB one) you are converting (or PM me if you don't want to post it) and the options you are using to convert it and I'll see if I can help you get the size down.

[terry2010]

thanks for your great work!

I used this ,but have small problems.


small words become very big







I use k2pdfopt like this

k2pdfopt -ui- -p 30-40 -as -x -sm -gtw 0.03937 test.pdf $*

how can I fix it ? thank you very much

[terry2010]

thanks for your great work!

I used this ,but have small problems.


small words become very big







I use k2pdfopt like this

k2pdfopt -ui- -p 30-40 -as -x -sm -gtw 0.03937 test.pdf $*

how can I fix it ? thank you very much

[ittiandro]

Originally Posted by willus

Please see the third question in the k2pdfopt FAQ. If that doesn't help, please post an example of one of your source documents (e.g. the 13 MB one) you are converting (or PM me if you don't want to post it) and the options you are using to convert it and I'll see if I can help you get the size down.

Your software is in many respects great ( v 2.14): it solved my major problem, by allowing to open my books very fast, once converted.
As to the large size, I can probably live with it, but now I seem to have another problem cropping up: the software often crashes in the midst of the conversion for no known reasons. It works 50% of the time !
Concerning the settings, I only checked the box for color output and OCR ( Tesseract). The rest are unchecked. I don't know what " reflow text", " native PDF output" and " OCR Tesseract " do, but I am hesitant to check them, because as they are they do not affect the conversion ( when it doesn't crash!). I was afraid that by checking " native PDF" and unchecking " OCR", I'd fall back into the problem I had before using your software, i.e. scanned PDF images too slow( or impossible) to open and not allowing font control or other.
Perhaps you can explain briefly what " native PDF", " reflow text" and " OCR Tesseract" do and give me some hints as to why the software crashes so often?

Thanks

Ittiandro