Converting .pdf to ePub - Pages jumbled.

Xyo · 07-31-2011, 11:10 AM

^ Sorry, the title should read .pdf to .mobi. ^

Hey guys,

So for a while now I've had some books in .pdf format that I've wanted to put onto my kindle 3. I've done this with normal .pdfs, but these particular files are formatted with two columns of text per page. When I converted them to .mobi, it was jumbled, as you would expect.

So today I found out about the BRISS software that crops pdf files, and everything seemed to be going well (I now have a .pdf file of all the text in a single column, in order), but I'm still having issues.

As soon as I convert the .pdf to .mobi (using the most recent version of Calibre), everything goes wrong. The first thing I notice is that everything is there twice, like every page is duplicated. Second, bits of the text seem jumbled together somewhat. I can't quite figure out the pattern of how it was jumbled, but it's definitely messed up.

Oh and I also tried converting to .ePub, which gave the same result.

Has anyone got any idea why this is happening?

EDIT: It seems maybe the .pdf BRISS is producing is the problem. Only my pdf reader (foxit reader) seems to be able to read it normally, everything else reads it and jumbles things up.

DSpider · 07-31-2011, 12:15 PM

The Kindle 3 doesn't support ePub. In fact, Amazon itself doesn't support ePub.

It's been discussed before, PDF is an output format.

There are two types of PDF: regular and tagged. A regular PDF is what the majority of PDFs are, with objects that seem to "float" on a blank piece of paper. You can usually spot them very quickly by simply selecting the text n Adobe Reader, or whatever. If letters (or groups of letters) are separated from the next, it means each has a unique position on the page. Which makes it very difficult for converting software to process. Sometimes they have to approximate.

Tagged PDFs, however, use tags for paragraphs, formatting, etc. They're usually much easier to convert (tho far from perfect either). If you select some text and it doesn't have blank spaces between letters, then it should be ok.

Anyway, it's always better to use the source document instead of converting the PDF. Especially since it went through BRISS... Cropping software always messes up the PDF, even if you don't see it on the surface.

What you could do is OCR the PDF using ABBYY FineReader or similar, proofread it - which takes some time (basically means you read the whole thing), then save it as HTML and work your way from there (using Sigil or similar) to save it as ePub. Or Mobi. Or whatever format you want... I'd go with .docx personally, or .odt if you're using LibreOffice.

Xyo · 07-31-2011, 12:41 PM

Sorry, I meant to say .mobi in the title, not .ePub.

I see what you mean about the regular/tagged .pdfs and unfortunately all I have is a regular .pdf (no source document).

I tried putting the pdf I got out of BRISS on the kindle and it looks better than I had expected. I need to run it again to make sure every page is cropped to the same size, but it seems workable otherwise.

Thanks for your help

DSpider · 07-31-2011, 12:52 PM

Alternatively you could export the PDF as JPG images, resized specifically for the reader's screen (most have a resolution of 800 x 600).

Xyo · 07-31-2011, 01:15 PM

That's a good idea, although I think cropping them is unavoidable anyway. I think the font is only just big enough on the BRISS cropped .pdfs, so I'll just have to put up with the variance in size.

07-31-2011, 11:10 AM	#1
Xyo Junior Member Posts: 6 Karma: 10 Join Date: Jul 2011 Device: Kindle 3	Converting .pdf to ePub - Pages jumbled. ^ Sorry, the title should read .pdf to .mobi. ^ Hey guys, So for a while now I've had some books in .pdf format that I've wanted to put onto my kindle 3. I've done this with normal .pdfs, but these particular files are formatted with two columns of text per page. When I converted them to .mobi, it was jumbled, as you would expect. So today I found out about the BRISS software that crops pdf files, and everything seemed to be going well (I now have a .pdf file of all the text in a single column, in order), but I'm still having issues. As soon as I convert the .pdf to .mobi (using the most recent version of Calibre), everything goes wrong. The first thing I notice is that everything is there twice, like every page is duplicated. Second, bits of the text seem jumbled together somewhat. I can't quite figure out the pattern of how it was jumbled, but it's definitely messed up. Oh and I also tried converting to .ePub, which gave the same result. Has anyone got any idea why this is happening? EDIT: It seems maybe the .pdf BRISS is producing is the problem. Only my pdf reader (foxit reader) seems to be able to read it normally, everything else reads it and jumbles things up. Last edited by Xyo; 07-31-2011 at 12:27 PM.

07-31-2011, 12:15 PM	#2
DSpider Evangelist Posts: 450 Karma: 343115 Join Date: Nov 2009 Location: Romania Device: PW2 2014	The Kindle 3 doesn't support ePub. In fact, Amazon itself doesn't support ePub. It's been discussed before, PDF is an output format. There are two types of PDF: regular and tagged. A regular PDF is what the majority of PDFs are, with objects that seem to "float" on a blank piece of paper. You can usually spot them very quickly by simply selecting the text n Adobe Reader, or whatever. If letters (or groups of letters) are separated from the next, it means each has a unique position on the page. Which makes it very difficult for converting software to process. Sometimes they have to approximate. Tagged PDFs, however, use tags for paragraphs, formatting, etc. They're usually much easier to convert (tho far from perfect either). If you select some text and it doesn't have blank spaces between letters, then it should be ok. Anyway, it's always better to use the source document instead of converting the PDF. Especially since it went through BRISS... Cropping software always messes up the PDF, even if you don't see it on the surface. What you could do is OCR the PDF using ABBYY FineReader or similar, proofread it - which takes some time (basically means you read the whole thing), then save it as HTML and work your way from there (using Sigil or similar) to save it as ePub. Or Mobi. Or whatever format you want... I'd go with .docx personally, or .odt if you're using LibreOffice. Last edited by DSpider; 07-31-2011 at 12:21 PM.

07-31-2011, 01:15 PM	#5
Xyo Junior Member Posts: 6 Karma: 10 Join Date: Jul 2011 Device: Kindle 3	That's a good idea, although I think cropping them is unavoidable anyway. I think the font is only just big enough on the BRISS cropped .pdfs, so I'll just have to put up with the variance in size. Last edited by Xyo; 07-31-2011 at 01:17 PM.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Duplicate pages converting PDF in Mobi	karlbx	Kindle Formats	5	07-15-2011 03:42 PM
Converting CHM to PDF - pages not on 1 page	Merritt	Conversion	2	07-03-2011 11:24 AM
Images are broken across 2 pages after converting to ePub from RTF. iPhone.	vital2k	Conversion	2	05-31-2011 01:46 AM
PDF to HTML jumbled text	jeero	PDF	2	09-03-2010 04:12 AM
Converting PDF - Removing text at top of pages	halljames	Calibre	4	07-21-2009 07:00 AM

07-31-2011, 12:41 PM	#3
Xyo Junior Member Posts: 6 Karma: 10 Join Date: Jul 2011 Device: Kindle 3	Sorry, I meant to say .mobi in the title, not .ePub. I see what you mean about the regular/tagged .pdfs and unfortunately all I have is a regular .pdf (no source document). I tried putting the pdf I got out of BRISS on the kindle and it looks better than I had expected. I need to run it again to make sure every page is cropped to the same size, but it seems workable otherwise. Thanks for your help

07-31-2011, 12:52 PM	#4
DSpider Evangelist Posts: 450 Karma: 343115 Join Date: Nov 2009 Location: Romania Device: PW2 2014	Alternatively you could export the PDF as JPG images, resized specifically for the reader's screen (most have a resolution of 800 x 600).

Advert

Advert