Calibre-Kindle Conversions TOCs

[Griffin NoName]

I just converted a PDF with a Table of Conntents at the start of the PDF file, in Calibre, to a Kindle format and dumped it onto my Kindle. But the TOC is hopeless on the Kindle; it's not formatted in any way so eacb  entry scrolls over several lines.

I looked in the Calibre user guide http://manual.calibre-ebook.com/conversion.html#table-of-contents but found it less than helpful. The TOC in the PDF does not appear as metadata...........in fact in the metadata screen in Calibre I can't see any place which refers to TOC at all, like they infer in the Calibre user guide.

Anyone got any tips for how to get the PDF TOC into e-book-format sensibly, or is the only way to force autogeneration of a new TOC (which I don't think will actually produce a useful version of the TOC). ?

[Bob in a quantum-state-of-faith]

Does the PDF table of contents work, within a PDF-type program?  Not all of those do-- if the TOC is passive (not a hyperlink) then you'd have to manually create one yourself somehow.  I've done it back in HTML but that was several years back, and I'd be clueless how to do one in PDF.

As for Kindle-- it's MOBI format, and there are some tools out there which let you create your own MOBI documents from raw files, but I only experimented with a free one that has long since been abandoned, and I can't find a valid link to it anymore, or even to the one I had previously.  Unfortunately.

That being said?  If the PDF's TOC is active/hyperlinked, then perhaps an intermediate conversion to HTML/Web format, and then to MOBI via Calibre might do the trick.

[Griffin NoName]

Hmmmmmmm. I used Calibre to convert to AZW3 format which is the new Kindle format apparently. I though I was being clever not converting to MOBI as it is now considered an old format. You'd think if AZW3 is new, it would be better than MOBI and theredfore better at TOC conversion. To avoid the double conversion of intervening HTML, I'll try Calibre converting to MOBI and report back. I highly recommend Calibre, it is free and functionally rich, and dead easy to use.


EDIT

Ok. Did the MOBI, same result.

Looked back at metadata. Found a text format option (not at all obvious). It was set to "original" which means it should have worked as text format should have stayed same as original, but in fact it translates it to justified for some strange reason. But, just to see, I set it to left justified for conversion, instead of original. Reconverted to MOBI. Much improved. Page numbers at least now obiously relate to text headings which they didn't when justified. The TOC should have two levels, but doesn't so heading lines do not stand out so still all looks a bit odd. But much better. I suspect I could sort out the heading lines and indentaton for the TOC by inserting some XML which is an option, but I can't be bothered to work it all out.

[Bob in a quantum-state-of-faith]

As far as I know, AZW3 is basically MOBI with Amazon's special customized additional records (in the non-visible header part). The original MOBI was not made for DRM, as far as I know, and Amazon wanted (per it's agreements with publishers) to have some sort of copy-protection schema embedded.  Thus the AZW format.  I presume by the "3" that it's the 3rd variation of that original departure from MOBI.

Also, as far as I know, the original parent company who created MOBI is now defunct, making the original MOBI an orphaned child.  Correct me if I'm wrong about that one but that's what I'd seen somewhere or other.

But is seriously sounds like your original PDF file wasn't organized very well-- often the case, as PDF is really a sloppy "standard" without any real consistency guidelines that have to be met.  It was originally created to combine image-scanned text with user-created notes & commentary.   The original text version of PDF came later.  (that is, real computer-readable text instead of page-captures/images of pages).

I'd try spitting it out to HTML or XML and have a look at the raw text, just to see if the TOC is actually there, or what.

On the third paw, if you got it working?  Good enough. 

[Griffin NoName]

For many many years I hated PDF. It's initial purpose being to produce a printable version of documents, seemed to me to be for idiots who couldn't format their text properly. At that stage I was using Unix with gawk to format output. Or Oracle Reports. Plus in those days PDF wasn't searchable.....because it was just an image. All round irritating. Especially as users would bleat on about it. Since it's been somewhat upgraded, and because my installation of HP web page printing won't work properly I have been forced to use PDF to output some stuff off web pages' - to files rather than printer - lazy option as it's not only on the IE file menu, but also an add-in. Still don't like it though, and if I had time to work out why HP won't play good I'd prefer to use that. Research papers always download as PDF so I guess I am stuck with it: usually they are searchable but not always, also local searches mostly don't work, so for instance searching in a directory of mixed format research documents, for a particular item, the PDF files return nothing....... highly annoying and introducing very time-consuming note taking to remember what I've read in them (the PDF research papers). Of course, the not being searchable was a plus with people who wanted to put docs on the web but not let people search them, which also irritated the hell out of me. <end moan>

[Bob in a quantum-state-of-faith]

I'm right there with you with regards to PDF. 

My opinion of the parent company could not be lower-- I'd rather chew rusty razor blades, than willingly purchase an Adobe product of any kind.   Lazy developers, lazy programmers, bloat-ware, resource-hogs, and they won't uninstall cleanly.  meh.

[Griffin NoName]

Exported PDF file to MS Word - which shows the TOC is a huge mess. I went through editing it, putting tabs in so everything lined up and wasn't justified. Then saved from Word back into PDF format. No quite all lining up so will need to do some re-edits in Word and then re-save to PDF, then submit PDF to Calibre.

[Bob in a quantum-state-of-faith]

I think Calibre can directly convert Word docs to MOBI or AZW, skipping the PDF step.  I'd check it out.   

[Sibling DavidH]

I've just tries two .doc files (Word 97) and it doesn't work.  Sorry.  I had also seen something somewhere that said it did.

[Griffin NoName]

HeHe. Yes. I had already tried. Word formats (all) don't convert even though they can be added into Calibre fine and viewed. xml doesn't work either.

[Bob in a quantum-state-of-faith]

Hunh.  That feature must have been deleted then, 'cause I did use Word 2000 to create some short lists, and wanted it in MOBI form on my Kindle.  This was, maybe 3 years back?  

That's too bad, really.

However.. I wonder what GOOGLE will yield...  hmmm...

This looks quite promising, as Kindles will happily read a MOBI file directly:  Mobi Pocket Creator.  I had thought this software abandonware... happy to find it's still going.

Here's a How-To for MOBI:  http://www.ehow.com/how_5147554_convert-doc-azw.html

Edit:  here's an online converter

Edit:  here's another How To, this time from the UK http://www.gareth53.co.uk/blog/2011/05/converting-word-doc-kindle-e-book.html

[Griffin NoName]

Had to correct all of the TOC by hand (=ie. keyboard) within the Word doc, and it is no longer actually working as a TOC in Word, and I have had enough of it. So export -> pdf, pdf->>Kindle (by accident, meant to be converted)................ print too faint and small to read, and no character resizing option available, but the pdf version has actually transfered ok!  Since not actually got xray eyes, will have to go back to transfer the ePub version................

Edit ePub non transferable. Did a MOBI and this is now my Kindle version.............. it reads fine on the Kindle.

Was it worth it? Probably not.

[Bob in a quantum-state-of-faith]

But you learned several valuable things.  Right?

[Griffin NoName]

Only if I ever want to do that particular thing again.

[Sibling DavidH]

I'm sure you had a need for a TOC, but I get rid of them where I can; my Calibre is permanently set not to generate one.  Most times you can't locate bits from one anyway.