Bad PDF to ePub Conversion

xxx666yyy777 · 12-10-2010, 02:14 PM

Hi,

I have been trying to convert several pdfs to the epub format (for iPad, all other settings default). As a result, the pdf page breaks are lost, page numbers, where the old pdf page breaks used to be have been inserted into the resulting ePub, however the page numbers are NOT aligned with the ePub document page breaks. So, i now have different page breaks, than the original pdf, and pager numbers in the middle of pages/paragraphs.

Am I missing something?

Thx.

ldolse · 12-10-2010, 02:44 PM

PDF conversion automatically removes all the original page breaks. To remove the page numbers/footers, etc you need to use the remove header/footer regex option under Structure detection. You need to write a regular expression to do this, because every pdf is bit different with respect to page numbers/headers/footers.

Starson17 · 12-10-2010, 03:22 PM

Quote:

Originally Posted by xxx666yyy777

Hi,
....
Am I missing something?

Yes, but it's not your fault. PDF is based on postScript - a great printer language, and a lousy ebook format.

Quote:

pdf page breaks are lost

Page breaks in a pdf are positioned when there's not enough room on the paper to display the next line of text. They are in the middle of sentences, etc. so you don't normally want them in an ebook. Your breaks weren't "lost" they were removed. Calibre can often put in the breaks you want by looking for certain words, like "Chapter."

Quote:

page numbers, where the old pdf page breaks used to be have been inserted into the resulting ePub,

These weren't "inserted." They were there from the start. PDF format makes no distinction between text in the header, footer or page number and regular text in the body of the book. In fact, PDF has no real concept of a sentence. It just puts characters at specified locations on the paper. Converters, like Calibre have to try to reconstruct the sentence by looking for the period or capital letters.

Quote:

however the page numbers are NOT aligned with the ePub document page breaks.

They wouldn't be. The page numbers in the text of the PDF corresponded to the paper size. You have to do some work to remove them.

Quote:

So, i now have different page breaks, than the original pdf, and pager numbers in the middle of pages/paragraphs.

PDF is a lousy ebook format. Your best option is not to start with a PDF. If you have no choice, then your second best option is to read the PDF and not convert. If you have to convert, then you're stuck trying to fix up the PDF in the conversion by setting the remove header/footer options and adjusting the unwrap factor.

Good luck.

JSWolf · 12-10-2010, 03:25 PM

Quote:

Originally Posted by xxx666yyy777

Hi,

I have been trying to convert several pdfs to the epub format (for iPad, all other settings default). As a result, the pdf page breaks are lost, page numbers, where the old pdf page breaks used to be have been inserted into the resulting ePub, however the page numbers are NOT aligned with the ePub document page breaks. So, i now have different page breaks, than the original pdf, and pager numbers in the middle of pages/paragraphs.

Am I missing something?

Thx.

The problem is that PDF is not a good format to convert from. It cannot be converted to anything else without errors.So really, treat PDF like it diesn't exist and life will be a lot easier to live.

xxx666yyy777 · 12-10-2010, 03:30 PM

Thanks for all the great responses. I understand now...!

Thx.

leday · 12-10-2010, 04:02 PM

Quote:

Originally Posted by ldolse

PDF conversion automatically removes all the original page breaks. To remove the page numbers/footers, etc you need to use the remove header/footer regex option under Structure detection. You need to write a regular expression to do this, because every pdf is bit different with respect to page numbers/headers/footers.

How do you write the expression? It seems complicated to me when I look?

Manichean · 12-10-2010, 05:07 PM

Quote:

Originally Posted by leday

How do you write the expression? It seems complicated to me when I look?

They seem a little scary at first, but are not really that complicated to write, reading them is a different matter. There's a tutorial available in the manual.

leday · 12-10-2010, 06:15 PM

Thanks. This really helps. Now, the only problem I have is that after the conversion, it is closing up words every so often, is in the example below. Is there any way to keep it from doing this as I am having to go through TONS of words to separate them.

“Right. That’s it. I officially call an end to today. I’mgoing home and going to bed until it’s over.” I rolled overonto my side and propped myself up on one hand. A pairof shoes appeared next to me, attached to a woman’s legs.I followed the legs up to the rest of the person.

He turned away, his voice as smooth and polished as it had been the first time I’d met him. “You will tell me with whom you are working, or I will break your body, corruptyour soul, and banish you to an eternity of torment.”
Every inch of my body broke out into a terrified coldsweat as I frantically looked around the room, desperatefor some way to escape, or something I could do todistract Ariton long enough to get away

ldolse · 12-10-2010, 11:21 PM

Words running into each other is probably unique to that pdf, though it's not a totally uncommon problem. As Starson17 was saying, it's all postscript, the pdf itself isn't aware of 'words' per-se, just a bunch of letters. When those strings are converted to text it's likely there is too little spacing or some other odd bit of formatting which prevents the conversion engine from recognizing the space and retain it.

leday · 12-11-2010, 02:00 AM

OK thanks. On looking further at several of my pdf file conversions it does seem that the problem is indeed unique to the particular pdf files that I have, and is there BEFORE conversion to epub.

Sigh....pdf is rather a pain isn't it....

mrmikel · 12-11-2010, 07:42 AM

PDF is a pain to be sure. These problems are why any epub conversion for instance may need to be re-edited in Sigil or other program that can do search and replace and spelling checks to make up for words run together. I am working on book from the US Army Center for Military History that has bad problems with words run together. It also suffers from captions put in graphically and textually both and the fancy captions are overlaid on the bottom of pictures with a black haze. Fine to print, heck to convert.

12-10-2010, 02:14 PM	#1
xxx666yyy777 Junior Member Posts: 6 Karma: 10 Join Date: Dec 2010 Device: iPad	Bad PDF to ePub Conversion Hi, I have been trying to convert several pdfs to the epub format (for iPad, all other settings default). As a result, the pdf page breaks are lost, page numbers, where the old pdf page breaks used to be have been inserted into the resulting ePub, however the page numbers are NOT aligned with the ePub document page breaks. So, i now have different page breaks, than the original pdf, and pager numbers in the middle of pages/paragraphs. Am I missing something? Thx.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
PDF to EPUB Conversion	LuchoResto	General Discussions	1	11-19-2010 04:54 PM
pdf to epub conversion	Storyowner	Calibre	3	11-03-2010 08:01 AM
PDF to EPUB conversion	jfontana	Calibre	2	03-17-2010 03:09 AM
pdf to epub conversion	mediax	Sigil	16	11-19-2009 03:48 PM
Help with conversion from PDF to EPUB	Fizz	Calibre	5	10-25-2009 11:48 AM

12-10-2010, 02:44 PM	#2
ldolse Wizard Posts: 1,337 Karma: 123455 Join Date: Apr 2009 Location: Malaysia Device: PRS-650, iPhone	PDF conversion automatically removes all the original page breaks. To remove the page numbers/footers, etc you need to use the remove header/footer regex option under Structure detection. You need to write a regular expression to do this, because every pdf is bit different with respect to page numbers/headers/footers.

12-10-2010, 03:30 PM	#5
xxx666yyy777 Junior Member Posts: 6 Karma: 10 Join Date: Dec 2010 Device: iPad	Thanks for all the great responses. I understand now...! Thx.

12-10-2010, 06:15 PM	#8
leday Groupie Posts: 171 Karma: 400 Join Date: Jun 2009 Device: Sony PRS-700, Nook Color	Thanks. This really helps. Now, the only problem I have is that after the conversion, it is closing up words every so often, is in the example below. Is there any way to keep it from doing this as I am having to go through TONS of words to separate them. “Right. That’s it. I officially call an end to today. I’mgoing home and going to bed until it’s over.” I rolled overonto my side and propped myself up on one hand. A pairof shoes appeared next to me, attached to a woman’s legs.I followed the legs up to the rest of the person. He turned away, his voice as smooth and polished as it had been the first time I’d met him. “You will tell me with whom you are working, or I will break your body, corruptyour soul, and banish you to an eternity of torment.” Every inch of my body broke out into a terrified coldsweat as I frantically looked around the room, desperatefor some way to escape, or something I could do todistract Ariton long enough to get away

12-10-2010, 11:21 PM	#9
ldolse Wizard Posts: 1,337 Karma: 123455 Join Date: Apr 2009 Location: Malaysia Device: PRS-650, iPhone	Words running into each other is probably unique to that pdf, though it's not a totally uncommon problem. As Starson17 was saying, it's all postscript, the pdf itself isn't aware of 'words' per-se, just a bunch of letters. When those strings are converted to text it's likely there is too little spacing or some other odd bit of formatting which prevents the conversion engine from recognizing the space and retain it.

12-11-2010, 02:00 AM	#10
leday Groupie Posts: 171 Karma: 400 Join Date: Jun 2009 Device: Sony PRS-700, Nook Color	OK thanks. On looking further at several of my pdf file conversions it does seem that the problem is indeed unique to the particular pdf files that I have, and is there BEFORE conversion to epub. Sigh....pdf is rather a pain isn't it....

12-11-2010, 07:42 AM	#11
mrmikel Color me gone Posts: 2,089 Karma: 1445295 Join Date: Apr 2008 Location: Central Oregon Coast Device: PRS-300	PDF is a pain to be sure. These problems are why any epub conversion for instance may need to be re-edited in Sigil or other program that can do search and replace and spelling checks to make up for words run together. I am working on book from the US Army Center for Military History that has bad problems with words run together. It also suffers from captions put in graphically and textually both and the fancy captions are overlaid on the bottom of pictures with a black haze. Fine to print, heck to convert.

Advert

Advert