BookDesigner and advanced HTML files

[Pablo]

I am trying to load the HTML version of "Sailing alone around the world" by Joshua Slocum into BookDesigner, but am losing images and parts of the text in the process.
My guess is BD doesn't understand some advaced formatting features, like inline styles. I've tried loading with Word and saving as HTML-filtered and then loading with BD, but that didn't help either.
Any ideas?
Thanks

[nrapallo]

My problem using BD has always been I could never get satisfactory results when I start with a well coded (and elegant) .html ebook, namely most Project Gutenberg ebooks!

I solely convert from .html (with images) and have never been able to achieve good results using BD as others have done so easily when starting with text files as their source.

But, if your wish is to have Sony PRS-505 formatted ebooks from that or other Project Gutenberg ebooks, why not try my freeware GuteBook (see the link in my sig below) to convert it for you (the way you like it).

I converted that PG Etext #6317 to my reader's .imp format as well as a Sony .epub and .lrf at the same time. I, however, did rerun my ebook creation using the GuteBook supplied "rebuild-6317.bat" batch file to correct a minor issue, namely the captions below images didn't start on a new line underneath the images as expected, so I inserted a "<br />" before every caption in the source file "6317-h.htm".

To rerun that conversion for one ebook format just double-click that batch file. If you created more than one ebook formatted ebook i.e. .epub and .lrf, then you need to edit the "rebuild-6317.bat" file to insert "start /w " in front of each invocation of "ebook-convert" therein. This is necessary since after I created GuteBook, calibre now closes the ebook-convert DOS window when it's done and my batch file unexpectedly terminates early. This is the accepted work-around for this issue.

If you want I can upload my .epub or .lrf ebook creation here for you to see how it looks...

[Pablo]

Nick,
First of all, thank you very much for your kind and detailed answer.

I never tried your utility for just one reason: it uses calibre to generate the ebooks.

I have nothing against calibre, it is in fact a great library management program and I use it on a daily basis, but I don't think it is the best tool for ebook creation.

People who just want a usable file may find it useful, but I am also interested in producing clean and simple ePubs and learning a bit in the process. Calibre creates a plethora of classes which results in an unnecessary long and complicated style sheet and a cluttered text body.

You may wonder why I still use BD, instead of just going to Sigil (which, by the way, imports this file beautifully). Well, I find BD's "Element Browser", "Book corrector" and "Notes and Links" tools easier and faster to use. On top of that, I wrote a simple utility that converts the hideous "html0" format to a simple html with css ready to be imported with Sigil, taking care of smart quotes, chapter breaks, indents and the like in the process. This rather complicated combination has produced the best results for me.

Please do upload your ePub version of "Sailing alone....", I will take a look at it. Maybe you should upload it to MR's library, I think it's an interesting book (though I haven't read it yet).

When I find the time, I will also try your utility; right now I don't have a computer in which I can install it.

Thanks again
Pablo

[nrapallo]

Yes, it uses calibre via its command line utility, ebook-convert. It does add a lot of classes, but I have learned to live with that!

The resulting epub is attached for your perusal. I also attach another version that has been tweaked/fixed i.e. removed larger images that were hyperlinked to (this works well when reading on a computer monitor, but not so when reading on a 6" device) and improved the toc.ncx.

I find that calibre can properly convert most .html and images ebooks, but it may be necessary to "tweak" that epub after conversion. This is usually to "fix" some use of the CSS classes (i.e remove page-break in unwanted places/classes or add center to a tag).

I also use ePubFixer (found in our ePub formats forum) to clean up the calibre generated TOC entries (located in the toc.ncx) as I have done below with my conversion of PG Etext #6317.

All in all, the conversion took about 1 to 2 minutes and then another 10 to 20 minutes to tweak things to my (and most people's) liking...

The beauty of GuteBook is that YOU get to download and customize your own version of that PG book, just the way YOU like it (after some brute-force fixing/tweaking afterward).

Enjoy!

[Pablo]

Thank you very much, got them. I will have a look at them.
What is the reason for the huge difference in size between both epub versions?

[nrapallo]

Originally Posted by Pablo

Thank you very much, got them. I will have a look at them.
What is the reason for the huge difference in size between both epub versions?

The first one includes two images (smaller one and larger one) for each illustration as found in the original PG offering.

It allows the user to click the image (href's to images are NOT well supported on ebook devices) to see an enlarged version.

The second "fixed" version has those larger images removed as the smaller ones fit/render well on ebook devices.

[Pablo]

After failing miserably in all my attempts to load the PG html file, I changed the extension from ".htm" to ".html0" and..... THE FILE LOADED OK!!!

THIS IS GREAT NEWS!!!!

[veldskoen2]

Hi...I'm not sure if I'm at the right place, but here's the problem: I'm looking for an automatic converter that will convert a word or pdf file into epub. I've used mobipocket creator, but I lose all the indents and drop caps. Also all centered text gets shifted to the left. Images also end up in the wrong place. Any suggestions?

[nrapallo]

Originally Posted by veldskoen2

Hi...I'm not sure if I'm at the right place, but here's the problem: I'm looking for an automatic converter that will convert a word or pdf file into epub.

For word to .epub, try the BookCreator Tool available in this forum as a sticky here.

For .pdf, I'm afraid very little can be done when it's used as a source to convert to anything but .txt files. Mobipocket Creator works well if the .pdf was created from a word document and not scanned images. Try calibre, which has it's own subforum here at mobileread.com.