Kindle conversion problem

cheektocheek · 02-11-2011, 01:30 PM

I try to convert my pdf's to .mobi with the calibre program and I have noticed the following very irritating problems.

1) After conversion the index numbers get messed up and I can't jump to a certain page. Instead of page 15 I get page 150 after conversion.

2)In landscape mode on some pages the last line of letters gets eaten up.

Any solution to these problems? How can i convert pdf's "perfectly" ?

Help!

silasgreenback · 02-11-2011, 02:37 PM

Quote:

Originally Posted by cheektocheek

Any solution to these problems? How can i convert pdf's "perfectly" ?

Unless that pdf has lots of images or special formatting you want to keep intact, you may want to try converting the pdf to txt first and then work with that.

If there are artifacts like page numbers or "converted by ABC", etc. left over, the find & replace function should be helpful, either with Calibre or a text editor of your choice.

Also, welcome to MobileRead!

TenaciousBadger · 02-11-2011, 03:09 PM

Quote:

Originally Posted by cheektocheek

Any solution to these problems? How can i convert pdf's "perfectly" ?

Help!

Unfortunately there is no easy answer to your question. Most conversions require some additional editing but from PDF to... umm, ahemm... anything? there;s extra work involved, especially if the PDF you're starting from is already messed up. (honestly, even if it's not, you have to clean it up a bit)

my suggestion would be to convert the PDF to RTF, using Calibre (if you can't "save as..." a DOC) and prettify it from there.

i use Open Office to do make them as perfect as possible and reconvert to MOBI after lots of trials and tribulations.

tomsem · 02-11-2011, 04:03 PM

Other people prefer to convert PDF to HTML and do the cleanup there. Of course that requires some HTML/CSS knowledge. But if you want it to be perfect, there's probably no other way to achieve it.

Also with HTML you can automate a lot of the cleanup with scripts or macros. The problem is figuring out which issues a particular file has. It will all depend on how the PDF was baked.

cheektocheek · 02-11-2011, 08:30 PM

What exactly i need to clean up from the file once its html?

It sounds as too much hassle just to fix 1 file and make it appear properly. I had assumed kindle can display pdf's properly with no problems. Well it does but everything looks too small. The 150% zoom option is too much. You really need around 120% it seems to me.

FF2 · 02-12-2011, 12:29 AM

The other option is to use the FONT button to rotate the pdf 90 degrees. That way you can see/read left to right and possibly not have to scroll but you see less page up/down.

chyron8472 · 02-12-2011, 01:22 AM

Quote:

Originally Posted by cheektocheek

What exactly i need to clean up from the file once its html?

It sounds as too much hassle just to fix 1 file and make it appear properly.

First, I suggest Sigil for editing. You can use it to edit HTML and non-DRM ePub books.

Second, It's not the Kindle's fault for some PDFs to not look exactly right sometimes. It's the PDF file format's fault. As quoted from the Calibre user manual on converting files:

Quote:

from http://calibre-ebook.com/user_manual...-pdf-documents
PDF documents are one of the worst formats to convert from. They are a fixed page size and text placement format. Meaning, it is very difficult to determine where one paragraph ends and another begins. Calibre will try to unwrap paragraphs using a configurable, Line Un-Wrapping Factor. [...]

Also, they often have headers and footers as part of the document that will become included with the text. Use the Search and Replace panel to remove headers and footers to mitigate this issue. If the headers and footers are not removed from the text it can throw off the paragraph unwrapping. [...]

Some limitations of PDF input (when converting in Calibre) are:

* Complex, multi-column, and image based documents are not supported.
* Extraction of vector images and tables from within the document is also not supported.
* Some PDFs use special glyphs to represent ll or ff or fi, etc. Conversion of these may or may not work depending on just how they are represented internally in the PDF.
* Some PDFs store their images upside down with a rotation instruction, calibre currently doesn’t support that instruction, so the images will be rotated in the output as well.
* Links and Tables of Contents are not supported

To re-iterate PDF is a really, really bad format to use as input. If you absolutely must use PDF, then be prepared for an output ranging anywhere from decent to unusable, depending on the input PDF.

cybmole · 02-12-2011, 12:17 PM

Quote:

Originally Posted by cheektocheek

What exactly i need to clean up from the file once its html?

It sounds as too much hassle just to fix 1 file and make it appear properly. I had assumed kindle can display pdf's properly with no problems. Well it does but everything looks too small. The 150% zoom option is too much. You really need around 120% it seems to me.

email the pdf to your free kindle address with CONVERT in the email subject line.

Amazon will convert it better than you can, probably. & will send it back to your Kindle.

if you are still not happy, give up & find a non-pdf source

PS try this.

open pdf in adobe reader. on the edit tab select - copy file to clipboard. then paste clipboard contents into word.

that gets the text out of the pdf, into something that you can edit with.
if you like what you see, take it from there - ie continue to tidy, then save as rich text format, or save as filtered html , then add the saved file to calibre & convert that to mobi for Kindle. ( or you can send the .rtf to your kindle address

TenaciousBadger · 02-12-2011, 02:17 PM

any way you look at it and any method you choose, it still is a long journey to the perfect mobi, with a little bit of learning involved in the process (whether it is html/css or, the reason i love Open Office, regular expressions that allow you to make all changes in one go)
here's a wiki about them regular expressions if you're interested

in a nutshell, it's like the find function in Word, you find and then replace.
^$ - will take care of empty lines
^(tab space) - will take care of spaces before the first line of the paragraph
(tab space)$ - will take care of spaces at the end of the paragraph
^chapter .*$ - will find you all your chapters (as long as they have the word "chapter" in there)
^[0-9]* - will find all the numbers at the beginning of a sentence/paragraph (especially useful when you don't have the word "chapter" anywhere in there)
[a-z|;|.]$ - is a long story about repairing broken paragraphs

--as always a big thanks to tt

silasgreenback · 02-13-2011, 01:31 AM

Quote:

Originally Posted by TenaciousBadger

...any way you look at it and any method you choose, it still is a long journey to the perfect mobi, with a little bit of learning involved in the process (whether it is html/css or, the reason i love Open Office, regular expressions that allow you to make all changes in one go)

I've done a lot of formatting with text files in E-book Tidy and it's been really valuable, but after reading your post, I can't wait to start tinkering around with Open Office!

It really is fun learning new tricks and this seems like it'll be a blast! (Not hyperbole, seriously.)

cybmole · 02-13-2011, 03:00 AM

if it's text you want - then the above posts give a range of workable options.

where Kindle & similar are useless is for PDFs that are mostly image & that have lots of text wrapped around images, or where use of color is necessary to convey meaning - like a game strategy guide or a technical how-to manual.

with those it is a case of waiting for the next gen A4 size + colour reader. ( Hmm- don't apple do one already , for folks with more money than sense ? )

Valentino · 02-13-2011, 03:01 AM

The main way I convert pdf to mobi/epub is to use pdftohtml -xml, then use pdfreflow on the xml output to get rid of the headers, footers and page numbers and obviously reflow the text.
It turns out 99.9% perfect, and is so easy.

I'm not sure that there is a windows equivalent, if that is what you use.

02-11-2011, 01:30 PM	#1
cheektocheek Junior Member Posts: 2 Karma: 10 Join Date: Feb 2011 Device: kindle	Kindle conversion problem I try to convert my pdf's to .mobi with the calibre program and I have noticed the following very irritating problems. 1) After conversion the index numbers get messed up and I can't jump to a certain page. Instead of page 15 I get page 150 after conversion. 2)In landscape mode on some pages the last line of letters gets eaten up. Any solution to these problems? How can i convert pdf's "perfectly" ? Help!

02-11-2011, 04:03 PM	#4
tomsem Grand Sorcerer Posts: 6,478 Karma: 26425959 Join Date: Apr 2009 Location: USA Device: iPhone 15PM, Kindle Scribe, iPad mini 6, PocketBook InkPad Color 3	Other people prefer to convert PDF to HTML and do the cleanup there. Of course that requires some HTML/CSS knowledge. But if you want it to be perfect, there's probably no other way to achieve it. Also with HTML you can automate a lot of the cleanup with scripts or macros. The problem is figuring out which issues a particular file has. It will all depend on how the PDF was baked. Last edited by tomsem; 02-11-2011 at 04:06 PM.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Comic conversion problem (CBZ to PDF) on Kindle DX Graphite	kindle-dxg-owner	Calibre	10	12-15-2010 08:58 AM
conversion problem	Alexandra Irini	Introduce Yourself	5	11-22-2010 11:33 AM
Conversion problem from PRC to anything	EricLandes	Calibre	2	02-12-2010 12:13 PM
conversion problem?	mountainman80	ePub	8	01-29-2010 11:54 PM
Conversion Problem	tonyt	Calibre	5	06-13-2009 05:04 PM

02-11-2011, 08:30 PM	#5
cheektocheek Junior Member Posts: 2 Karma: 10 Join Date: Feb 2011 Device: kindle	What exactly i need to clean up from the file once its html? It sounds as too much hassle just to fix 1 file and make it appear properly. I had assumed kindle can display pdf's properly with no problems. Well it does but everything looks too small. The 150% zoom option is too much. You really need around 120% it seems to me.

02-12-2011, 12:29 AM	#6
FF2 Wizard Posts: 1,105 Karma: 1025784 Join Date: Oct 2010 Device: WiFi Kindle3	The other option is to use the FONT button to rotate the pdf 90 degrees. That way you can see/read left to right and possibly not have to scroll but you see less page up/down.

02-12-2011, 02:17 PM	#9
TenaciousBadger Mrawr? Posts: 1,109 Karma: 15039064 Join Date: Aug 2010 Device: kindle 3 wifi	any way you look at it and any method you choose, it still is a long journey to the perfect mobi, with a little bit of learning involved in the process (whether it is html/css or, the reason i love Open Office, regular expressions that allow you to make all changes in one go) here's a wiki about them regular expressions if you're interested in a nutshell, it's like the find function in Word, you find and then replace. ^$ - will take care of empty lines ^(tab space) - will take care of spaces before the first line of the paragraph (tab space)$ - will take care of spaces at the end of the paragraph ^chapter .$ - will find you all your chapters (as long as they have the word "chapter" in there) ^[0-9] - will find all the numbers at the beginning of a sentence/paragraph (especially useful when you don't have the word "chapter" anywhere in there) [a-z\|;\|.]$ - is a long story about repairing broken paragraphs --as always a big thanks to tt

02-13-2011, 03:00 AM	#11
cybmole Wizard Posts: 3,720 Karma: 1759970 Join Date: Sep 2010 Device: none	if it's text you want - then the above posts give a range of workable options. where Kindle & similar are useless is for PDFs that are mostly image & that have lots of text wrapped around images, or where use of color is necessary to convey meaning - like a game strategy guide or a technical how-to manual. with those it is a case of waiting for the next gen A4 size + colour reader. ( Hmm- don't apple do one already , for folks with more money than sense ? )

02-13-2011, 03:01 AM	#12
Valentino Connoisseur Posts: 86 Karma: 999999 Join Date: Dec 2010 Device: some.	The main way I convert pdf to mobi/epub is to use pdftohtml -xml, then use pdfreflow on the xml output to get rid of the headers, footers and page numbers and obviously reflow the text. It turns out 99.9% perfect, and is so easy. I'm not sure that there is a windows equivalent, if that is what you use.

Advert

Advert