Preserving Formatting through conversion?

Ham88 · 07-12-2009, 08:10 PM

There may be a thread on this already but I couldn't find one, so here it is. I have downloaded a book that is only available in PDF, so far as I could find anyway. So I went and converted it to epub through calibre. The formatting is entirely messed up, for example all of the words with two l's in it have lost one of them (call is now cal), and all of the dialog between characters has just become one paragraph. So basically my question is there a way to preserve the formatting through conversion without having to go through the file manually and fix it?

Kostas · 07-12-2009, 08:38 PM

Hi Ham88,

Indeed, it's a strange behavior. Calibre usually gives very good results (in lrf conversion which is the format I use). Maybe your pdf source is not a "normal" text based file.
My only suggestion (I'm pretty sure other expereienced members will give you more) would be to give another try with the following online converter:
http://www.lib2go.com/

I have tested for lrfs and it gives fair results.
Good luck!

Ham88 · 07-12-2009, 08:55 PM

Lib2go hates my files, as it claims they are too big and I noticed something else its based on calibre so I would get the same results anyway, so any other ideas?

Timoleon · 07-12-2009, 08:58 PM

How about running it through soPDF first, and then letting Calibre tackle it?

Ham88 · 07-12-2009, 09:07 PM

I just looked up soPDF, I may be wrong but it appears to be a command line driven which is something I avoid because I'm incompetent with the lovely command line. But I'm currently using PDFread and am hoping this will yield the results that I want. The only problem is that this requires me to be patient, something that I lack. If this doesn't work I'll try soPDF and hope for the, best thanks for the help so far.

frabjous · 07-12-2009, 11:08 PM

I wrote a barebones GUI for SoPDF. If you go further down in the SoPDF thread, you'll find discussion of it, and a download link.

I'm not sure SoPDF tool is the right tool here, however. PDFread is a good one to try. Even though you're not using a Sony, running PDFLRF on it, and then running that through calibre may give you a good result too.

doreenjoy · 07-13-2009, 01:19 AM

I have the same problem when using Calibre to convert PDFs: all the paragraph breaks are lost, and I end up with one long long text file. I have yet to find a good PDF to LRF or PDF to ePUB converter.

Ham88 · 07-13-2009, 08:17 PM

All of the things I have tried has failed, soPDF worsened the problem as the eventual epub file was only one or two words per line on average. I think I'm going to have to read this book as a PDF on my PC unfortunately. Thanks for the help.

HarryT · 07-18-2009, 05:12 AM

Calibre is an excellent program, but PDF conversion is not one of its strengths. Try "Book Designer" - it converts PDFs as well as anything I've come across.

Unfortunately, however, PDF files sometimes simply cannot be converted well. A PDF files does not contain "text" - it has no paragraphs, lines, words, etc; just drawing instructions for individual letters. As such, it is extraordinarily difficult to convert.

DDHarriman · 07-19-2009, 06:01 AM

Hi

My advice is to OCR the PDF file, save the result in a format you will be able to edit (per example Microsoft word), proof read it and correct the errors found, create final eBook in ePub or other form your reader can handle.

Two of the best software applications for OCR are, Finereader and Omnipage.

Best regards,

frabjous · 07-19-2009, 11:06 AM

Quote:

Originally Posted by DDHarriman

Hi

My advice is to OCR the PDF file, save the result in a format you will be able to edit (per example Microsoft word), proof read it and correct the errors found, create final eBook in ePub or other form your reader can handle.

I don't think that falls under the category of "Preserving format through conversion" "without having to go through the file manually and fix it", which is what was asked for.

I continue to maintain that PDFread and/or PDRLRF>Calibre are the best ways to go to preserve the look of the original PDF, while formatting better for an e-ink device.

DDHarriman · 07-19-2009, 01:14 PM

Hi

Frabjous

You are perfectly correct, I missed that the final intent from Ham88 was to:
“(…)to preserve the formatting through conversion without having to go through the file manually and fix it?”

Ham88

Rephrasing, the answer to the question (cited above) you have posed is: no!

Best regards,

JSWolf · 07-19-2009, 01:18 PM

There is no such thing as a novel length PDF that will convert from PDF to any other format without errors.

frabjous · 07-19-2009, 02:28 PM

The tools I suggested convert the PDF pages to images, but then remove the margins and cut up the images into manageable chunks, and also change the file format--not that matters much if it's just a sequence of images. I'm not sure what you have in mind by "errors", but they should preserve the look more or less exactly, and there would be nothing to manually fix. The only question is whether the results would look nice enough for the purposes of reading.

JSWolf · 07-19-2009, 02:37 PM

Quote:

Originally Posted by frabjous

The tools I suggested convert the PDF pages to images, but then remove the margins and cut up the images into manageable chunks, and also change the file format--not that matters much if it's just a sequence of images. I'm not sure what you have in mind by "errors", but they should preserve the look more or less exactly, and there would be nothing to manually fix. The only question is whether the results would look nice enough for the purposes of reading.

I was meaning converting the PDF to some reflowable format and not actually images. It cannot be done without error.

07-12-2009, 08:10 PM	#1
Ham88 Zealot Posts: 134 Karma: 994 Join Date: Apr 2009 Location: Maine, United States Device: Ectaco Jetbook	Preserving Formatting through conversion? There may be a thread on this already but I couldn't find one, so here it is. I have downloaded a book that is only available in PDF, so far as I could find anyway. So I went and converted it to epub through calibre. The formatting is entirely messed up, for example all of the words with two l's in it have lost one of them (call is now cal), and all of the dialog between characters has just become one paragraph. So basically my question is there a way to preserve the formatting through conversion without having to go through the file manually and fix it?

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Preserving <br /> on epub -> txt conversion	billingd	Calibre	1	08-11-2010 06:24 AM
[KOBO] Strip existing formatting to apply my own default formatting to all books	digital_steve	Calibre	2	08-10-2010 06:34 PM
Need help formatting HTML for good conversion	ficbot	Calibre	2	04-15-2010 09:36 PM
TXT conversion to ePub or LRF - paragraph formatting	Zapped	Calibre	6	10-23-2009 05:06 PM
Preserving TOC upon conversion from Lit to Mobi	mobelby	Calibre	0	07-31-2009 07:59 AM

07-12-2009, 08:38 PM	#2
Kostas Still wondering why Posts: 253 Karma: 800 Join Date: Jun 2009 Location: Athens, Greece Device: PRS 505, (BlackBerry Bold ?)	Hi Ham88, Indeed, it's a strange behavior. Calibre usually gives very good results (in lrf conversion which is the format I use). Maybe your pdf source is not a "normal" text based file. My only suggestion (I'm pretty sure other expereienced members will give you more) would be to give another try with the following online converter: http://www.lib2go.com/ I have tested for lrfs and it gives fair results. Good luck!

07-12-2009, 08:55 PM	#3
Ham88 Zealot Posts: 134 Karma: 994 Join Date: Apr 2009 Location: Maine, United States Device: Ectaco Jetbook	Lib2go hates my files, as it claims they are too big and I noticed something else its based on calibre so I would get the same results anyway, so any other ideas?

07-12-2009, 08:58 PM	#4
Timoleon Time Enough at Last Posts: 387 Karma: 1151316 Join Date: Feb 2008 Location: New England Device: iPad 3, iPhone 5, Kindle 3, Fire, Sony PRS-350	How about running it through soPDF first, and then letting Calibre tackle it?

07-12-2009, 09:07 PM	#5
Ham88 Zealot Posts: 134 Karma: 994 Join Date: Apr 2009 Location: Maine, United States Device: Ectaco Jetbook	I just looked up soPDF, I may be wrong but it appears to be a command line driven which is something I avoid because I'm incompetent with the lovely command line. But I'm currently using PDFread and am hoping this will yield the results that I want. The only problem is that this requires me to be patient, something that I lack. If this doesn't work I'll try soPDF and hope for the, best thanks for the help so far.

07-12-2009, 11:08 PM	#6
frabjous Wizard Posts: 1,213 Karma: 12890 Join Date: Feb 2009 Location: Amherst, Massachusetts, USA Device: Sony PRS-505	I wrote a barebones GUI for SoPDF. If you go further down in the SoPDF thread, you'll find discussion of it, and a download link. I'm not sure SoPDF tool is the right tool here, however. PDFread is a good one to try. Even though you're not using a Sony, running PDFLRF on it, and then running that through calibre may give you a good result too.

07-13-2009, 01:19 AM	#7
doreenjoy 01000100 01001010 Posts: 1,889 Karma: 2400000 Join Date: Mar 2009 Device: Polyamorous	I have the same problem when using Calibre to convert PDFs: all the paragraph breaks are lost, and I end up with one long long text file. I have yet to find a good PDF to LRF or PDF to ePUB converter.

07-13-2009, 08:17 PM	#8
Ham88 Zealot Posts: 134 Karma: 994 Join Date: Apr 2009 Location: Maine, United States Device: Ectaco Jetbook	All of the things I have tried has failed, soPDF worsened the problem as the eventual epub file was only one or two words per line on average. I think I'm going to have to read this book as a PDF on my PC unfortunately. Thanks for the help.

07-18-2009, 05:12 AM	#9
HarryT eBook Enthusiast Posts: 85,544 Karma: 93383043 Join Date: Nov 2006 Location: UK Device: Kindle Oasis 2, iPad Pro 10.5", iPhone 6	Calibre is an excellent program, but PDF conversion is not one of its strengths. Try "Book Designer" - it converts PDFs as well as anything I've come across. Unfortunately, however, PDF files sometimes simply cannot be converted well. A PDF files does not contain "text" - it has no paragraphs, lines, words, etc; just drawing instructions for individual letters. As such, it is extraordinarily difficult to convert.

07-19-2009, 06:01 AM	#10
DDHarriman Guru Posts: 860 Karma: 4380 Join Date: Feb 2008 Location: Almada, Portugal Device: Cybook Gen3, Sony PRS 505, Kindle DXG and Samsung Galaxy Note	Hi My advice is to OCR the PDF file, save the result in a format you will be able to edit (per example Microsoft word), proof read it and correct the errors found, create final eBook in ePub or other form your reader can handle. Two of the best software applications for OCR are, Finereader and Omnipage. Best regards,

07-19-2009, 01:14 PM	#12
DDHarriman Guru Posts: 860 Karma: 4380 Join Date: Feb 2008 Location: Almada, Portugal Device: Cybook Gen3, Sony PRS 505, Kindle DXG and Samsung Galaxy Note	Hi Frabjous You are perfectly correct, I missed that the final intent from Ham88 was to: “(…)to preserve the formatting through conversion without having to go through the file manually and fix it?” Ham88 Rephrasing, the answer to the question (cited above) you have posed is: no! Best regards,

Advert

Advert

07-19-2009, 01:18 PM	#13
JSWolf Resident Curmudgeon Posts: 73,957 Karma: 128903250 Join Date: Nov 2006 Location: Roslindale, Massachusetts Device: Kobo Libra 2, Kobo Aura H2O, PRS-650, PRS-T1, nook STR, PW3	There is no such thing as a novel length PDF that will convert from PDF to any other format without errors.

07-19-2009, 02:28 PM	#14
frabjous Wizard Posts: 1,213 Karma: 12890 Join Date: Feb 2009 Location: Amherst, Massachusetts, USA Device: Sony PRS-505	The tools I suggested convert the PDF pages to images, but then remove the margins and cut up the images into manageable chunks, and also change the file format--not that matters much if it's just a sequence of images. I'm not sure what you have in mind by "errors", but they should preserve the look more or less exactly, and there would be nothing to manually fix. The only question is whether the results would look nice enough for the purposes of reading.