k2pdfopt: optimizes PDFs for viewing on e-readers - Page 135

Quoth · 02-21-2023, 06:51 AM

I discovered a PDF (version 9 acrobat?) that wouldn't work properly in the Kobo Sage or Kobo Elipsa Crop. Also on Kobo you could only change page using the naviagtion slider, neither tap, swipe or buttons would change page. Actual text, so I might convert to epub if I can export it sensibly. It does import to LO Draw, but using that to format other than paper size or margins is tedious.

Anyhow, cropping it before importing to Calibre sorted it, though it got a little larger. After crop you could page through it.

It's not OCR text, but just text. A copy and paste in any desktop program does give the text in the right order, but each line break in PDF is a new line and so it can't be reflowed, the paragraph breaks same as newlines.

Can K2pdfopt export the text with only newline at paragraphs rather than the end of every line in the PDF? It's not an OCR layer, but only text.

j.p.s · 02-21-2023, 03:04 PM

Adding asciidoc or markdown markup should be almost trivial. Then newlines are ignored except for empty lines, which indicate a new paragraph.

Extremely long lines are editor unfriendly and are annoying, so I actually prefer text exports that haven't stripped newlines.

willus · 02-21-2023, 04:15 PM

The information I get about a PDF file from the muPDF library is simply a list of characters and what their X,Y positions on the page are. There is no information to indicate either new-line or a paragraph--you have to infer this solely from the character positions, so I'd have to deduce from the line spacings or indentation if a paragraph was indended, which will be quite error prone. It's probably easier to hand edit the unicode text out of k2pdfopt (-ocrout option).

Quoth · 02-22-2023, 09:47 AM

Paragraphs have either more indent or more space or start with a bolded word (all caps too usually) or are numbered list items. Only the ALL CAPS and list numbers survive text copy.

Rats. PDFs are evil to do anything with except print or read on a giant screen

Quoth · 02-22-2023, 10:09 AM

Actually the k2pdfout with -ocrout option put two newlines at paragraph breaks, but I'd only got one with a copy/paste.

So replaced all \n\n with ¬
replaced all \n with a "space"
replaced ¬ with \n

(But I wasn't using Windows notepad

)

Now only needs a day's work to make into an epub. Almost looks sane in LO Writer.

Quoth · 02-22-2023, 11:11 AM

First pass edit done, now proof later as an epub and edit back annotations.
Thanks, @willus.

willus · 02-25-2023, 12:38 PM

Quote:

Originally Posted by Quoth

First pass edit done, now proof later as an epub and edit back annotations.
Thanks, @willus.

Glad you had some success.

Quoth · 02-25-2023, 03:30 PM

Only 20 minutes ago finished major edit/format/style in LO Writer and conversion of extra docx save in Calibre to epub2.

Still, beats typing it up from looking at the paperback!

bamboohouses · 03-06-2023, 11:58 AM

I'm having problems downloading and installing k2pdfopt. I'm using a mac but the download always stalls, I click re-load and it downloads but when I install it I get the following:
Last login: Mon Mar 6 16:54:47 on ttys000
Davids-MacBook-Air:~ david$ /Users/david/Desktop/k2pdfopt ; exit;
/Users/david/Desktop/k2pdfopt: line 1: syntax error near unexpected token `<'
/Users/david/Desktop/k2pdfopt: line 1: `<html><head><title>Willus.com's K2pdfopt Download Page</title>'
logout
Saving session...
...copying shared history...
...saving history...truncating history files...
...completed.

[Process completed]

Any suggestions?

bamboohouses · 03-07-2023, 01:51 AM

I am wondering if k2pdfopt isn't downloading properly as I have to reload it or whether it isn't running correctly. I'm not sure what the syntax error near the unexpected token ' <' meant and how I could correct it?

I'd be very grateful for any suggestions or advice as k2pdfopt looks exactly what I'm looking for and there doesn't seem to be anything better as far as I can find.

Many thanks in anticipation.

taddymack · 03-07-2023, 10:07 AM

You're downloading a source code of the download page, not the program.
The first line of the source page reads:
<html><head><title>Willus.com's K2pdfopt Download Page</title>
Fill in the captcha and click on the relevant software (v2.54 mac OSX x86 64-bit ?).

bamboohouses · 03-07-2023, 12:00 PM

Thanks for explaining that. One step closer! However I am clicking on the relevant software and that what is downloading so I'm not quite sure what is going on. Very frustrating.

taddymack · 03-07-2023, 02:09 PM

It seems you have to right click on the rectangle (e.g. Mac OSX...) or control-click (when on Mac) on the rectangle and then select Save as... which saves the page, not the program.
Or try a different browser or wait till Willus responds.

bamboohouses · 03-07-2023, 03:26 PM

Great! Good idea - I think a different browser might be better. I was wondering if perhaps the captcha is timing out before the download has fully completed? I'll try your suggestions - Thanks.

willus · 03-07-2023, 10:16 PM

Maybe the video on the mac help page will help?

02-21-2023, 06:51 AM	#2011
Quoth the rook, bossing Never. Posts: 11,171 Karma: 85874891 Join Date: Jun 2017 Location: Ireland Device: All 4 Kinds: epub eink, Kindle, android eink, NxtPaper11	I discovered a PDF (version 9 acrobat?) that wouldn't work properly in the Kobo Sage or Kobo Elipsa Crop. Also on Kobo you could only change page using the naviagtion slider, neither tap, swipe or buttons would change page. Actual text, so I might convert to epub if I can export it sensibly. It does import to LO Draw, but using that to format other than paper size or margins is tedious. Anyhow, cropping it before importing to Calibre sorted it, though it got a little larger. After crop you could page through it. It's not OCR text, but just text. A copy and paste in any desktop program does give the text in the right order, but each line break in PDF is a new line and so it can't be reflowed, the paragraph breaks same as newlines. Can K2pdfopt export the text with only newline at paragraphs rather than the end of every line in the PDF? It's not an OCR layer, but only text. Last edited by Quoth; 02-21-2023 at 06:53 AM.

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

02-21-2023, 03:04 PM	#2012
j.p.s Grand Sorcerer Posts: 5,285 Karma: 98804578 Join Date: Apr 2011 Device: pb360	Adding asciidoc or markdown markup should be almost trivial. Then newlines are ignored except for empty lines, which indicate a new paragraph. Extremely long lines are editor unfriendly and are annoying, so I actually prefer text exports that haven't stripped newlines.

02-21-2023, 04:15 PM	#2013
willus Fuzzball, the purple cat Posts: 1,273 Karma: 11087488 Join Date: Jun 2011 Location: California Device: iPad	The information I get about a PDF file from the muPDF library is simply a list of characters and what their X,Y positions on the page are. There is no information to indicate either new-line or a paragraph--you have to infer this solely from the character positions, so I'd have to deduce from the line spacings or indentation if a paragraph was indended, which will be quite error prone. It's probably easier to hand edit the unicode text out of k2pdfopt (-ocrout option).

02-22-2023, 09:47 AM	#2014
Quoth the rook, bossing Never. Posts: 11,171 Karma: 85874891 Join Date: Jun 2017 Location: Ireland Device: All 4 Kinds: epub eink, Kindle, android eink, NxtPaper11	Paragraphs have either more indent or more space or start with a bolded word (all caps too usually) or are numbered list items. Only the ALL CAPS and list numbers survive text copy. Rats. PDFs are evil to do anything with except print or read on a giant screen

02-22-2023, 10:09 AM	#2015
Quoth the rook, bossing Never. Posts: 11,171 Karma: 85874891 Join Date: Jun 2017 Location: Ireland Device: All 4 Kinds: epub eink, Kindle, android eink, NxtPaper11	Actually the k2pdfout with -ocrout option put two newlines at paragraph breaks, but I'd only got one with a copy/paste. So replaced all \n\n with ¬ replaced all \n with a "space" replaced ¬ with \n (But I wasn't using Windows notepad ) Now only needs a day's work to make into an epub. Almost looks sane in LO Writer.

02-22-2023, 11:11 AM	#2016
Quoth the rook, bossing Never. Posts: 11,171 Karma: 85874891 Join Date: Jun 2017 Location: Ireland Device: All 4 Kinds: epub eink, Kindle, android eink, NxtPaper11	First pass edit done, now proof later as an epub and edit back annotations. Thanks, @willus.

02-25-2023, 03:30 PM	#2018
Quoth the rook, bossing Never. Posts: 11,171 Karma: 85874891 Join Date: Jun 2017 Location: Ireland Device: All 4 Kinds: epub eink, Kindle, android eink, NxtPaper11	Only 20 minutes ago finished major edit/format/style in LO Writer and conversion of extra docx save in Calibre to epub2. Still, beats typing it up from looking at the paperback!

03-06-2023, 11:58 AM	#2019
bamboohouses Junior Member Posts: 6 Karma: 10 Join Date: Mar 2023 Device: kindle oasis	I'm having problems downloading and installing k2pdfopt. I'm using a mac but the download always stalls, I click re-load and it downloads but when I install it I get the following: Last login: Mon Mar 6 16:54:47 on ttys000 Davids-MacBook-Air:~ david$ /Users/david/Desktop/k2pdfopt ; exit; /Users/david/Desktop/k2pdfopt: line 1: syntax error near unexpected token `<' /Users/david/Desktop/k2pdfopt: line 1: `<html><head><title>Willus.com's K2pdfopt Download Page</title>' logout Saving session... ...copying shared history... ...saving history...truncating history files... ...completed. [Process completed] Any suggestions?

03-07-2023, 01:51 AM	#2020
bamboohouses Junior Member Posts: 6 Karma: 10 Join Date: Mar 2023 Device: kindle oasis	I am wondering if k2pdfopt isn't downloading properly as I have to reload it or whether it isn't running correctly. I'm not sure what the syntax error near the unexpected token ' <' meant and how I could correct it? I'd be very grateful for any suggestions or advice as k2pdfopt looks exactly what I'm looking for and there doesn't seem to be anything better as far as I can find. Many thanks in anticipation.

03-07-2023, 10:07 AM	#2021
taddymack Member Posts: 15 Karma: 50592 Join Date: Dec 2021 Device: Kobo Libra 2	You're downloading a source code of the download page, not the program. The first line of the source page reads: <html><head><title>Willus.com's K2pdfopt Download Page</title> Fill in the captcha and click on the relevant software (v2.54 mac OSX x86 64-bit ?).

03-07-2023, 12:00 PM	#2022
bamboohouses Junior Member Posts: 6 Karma: 10 Join Date: Mar 2023 Device: kindle oasis	Thanks for explaining that. One step closer! However I am clicking on the relevant software and that what is downloading so I'm not quite sure what is going on. Very frustrating.

Advert

Advert

03-07-2023, 02:09 PM	#2023
taddymack Member Posts: 15 Karma: 50592 Join Date: Dec 2021 Device: Kobo Libra 2	It seems you have to right click on the rectangle (e.g. Mac OSX...) or control-click (when on Mac) on the rectangle and then select Save as... which saves the page, not the program. Or try a different browser or wait till Willus responds.

03-07-2023, 03:26 PM	#2024
bamboohouses Junior Member Posts: 6 Karma: 10 Join Date: Mar 2023 Device: kindle oasis	Great! Good idea - I think a different browser might be better. I was wondering if perhaps the captcha is timing out before the download has fully completed? I'll try your suggestions - Thanks.

03-07-2023, 10:16 PM	#2025
willus Fuzzball, the purple cat Posts: 1,273 Karma: 11087488 Join Date: Jun 2011 Location: California Device: iPad	Maybe the video on the mac help page will help?