Best original format for accurate conversion

[ren41]

Hi all

I decided to put some of my favourite poems into a booklet for my Kindle. I input the poems into Word, saved as a text file and then applied basic formatting - all poems in the same font and size, titles in bold, and page breaks.

Then I saved as RTF and converted to .mobi

There are quite a few formatting changes - random size changes, differences in line height in some poems, some titles are not in bold.

I thought I'd had a brainwave this morning and tried saving the file as PDF and then converting to .mobi. Much worse, with line and page breaks missing in many cases.

What would be the best way to put the file into Calibre for conversion? The only other option i can think of is to convert the RTF to HTML, make sure that's consistent and then convert to .mobi, perhaps? It seems a bit long winded though.

Does anyone have any suggestions, please?

TIA

ren

[theducks]

Originally Posted by ren41

Hi all

I decided to put some of my favourite poems into a booklet for my Kindle. I input the poems into Word, saved as a text file and then applied basic formatting - all poems in the same font and size, titles in bold, and page breaks.

Then I saved as RTF and converted to .mobi

There are quite a few formatting changes - random size changes, differences in line height in some poems, some titles are not in bold.

I thought I'd had a brainwave this morning and tried saving the file as PDF and then converting to .mobi. Much worse, with line and page breaks missing in many cases.

What would be the best way to put the file into Calibre for conversion? The only other option i can think of is to convert the RTF to HTML, make sure that's consistent and then convert to .mobi, perhaps? It seems a bit long winded though.

Does anyone have any suggestions, please?

TIA

ren

PDF is just plain BAD for conversion and small screens.

Unless you design the PDF for the screen size and shape (and don't plan on converting).
HTML (filtered) is very adjustable (and the basis of other formats)
You might wander down to the EPUB forum and read around the sticky area.
Poetry is one of the trickier layouts to get looking right.

[ren41]

Thank you. The actual poetry layout itself seems to be ok in most cases, it's the random font sizing that's annoying me most (when converting from RTF to .mobi, that is) especially as I can't see why it should be happening.

I'm developing a bit more sympathy for some of those badly converted ebooks you buy! - although the Calibre conversion does at least deal properly with accented characters!

Are you suggesting converting to ePub first would also help the formatting?

ren

[theducks]

Originally Posted by ren41

Thank you. The actual poetry layout itself seems to be ok in most cases, it's the random font sizing that's annoying me most (when converting from RTF to .mobi, that is) especially as I can't see why it should be happening.

I'm developing a bit more sympathy for some of those badly converted ebooks you buy! - although the Calibre conversion does at least deal properly with accented characters!

Are you suggesting converting to ePub first would also help the formatting?

ren

My choice is to work in EPUB (I use Sigil).
BUT you also need to be aware of what the destination format WON'T do .
It does little good to embed and use a beautiful font only to convert to some format (Mobi) that does not support embedded fonts.

(You will still have a Nifty EPUB )

[dwig]

Originally Posted by ren41

...
What would be the best way to put the file into Calibre for conversion?...

Since MOBI is an HTML based format your best bet will be HTML or some HTML-esque format (e.g. ePub). As theducks said, you need to be very aware of what formatting is supported in MOBI and what isn't.

Personally, I prefer working either in HTML using Dreamweaver and/or Notepad++, using a browser for preview with the latter, or working in ePub using Sigil as the editor. Either of these two formats will be the closest you can get to MOBI in an editable format. I then do my conversions to MOBI in calibre.

[ren41]

Hello all

Thanks very much, I'll try formatting it in HTML and see what happens. The basic font on the Kindle is fine by me, I just hate the random changes in font size etc!

Will hopefully get a chance to do a first pass sometime today.

Thanks again

ren

[ren41]

Hello all

I now have a booklet which has formatted in a consistent manner! It was a bit of a long winded process but this is what I did, in case it helps anyone else:

I started out with a Word document, into which I put all the poems. Apart from one poem, I was lucky enough to track them all down on the web. I kept line indents as per the originals. Individual lines were indented with spaces, rather than tabs.

I then saved the document as a text file and put it into Dreamweaver. I added CSS to format the paragraphs (lines) and the titles, and added the HTML. The titles were set to <h2> to trigger page breaks. I used <br /> to separate stanzas and between the title/poem and poem/author. After putting the HTML file into Calibre I converted to ePub first; I changed some of the conversion settings as follows:

In 'look & feel' I set minimum line height to 0%, checked 'remove spacing.....' and unchecked 'insert blank line....'

Then I converted the pub file to mobi.

I had wanted the author name to be indented to partway across the page, and I set this up in CSS. It worked in ePub but in mobi the names are left justified (which is my default justification). I'll go back at some point and change that by inserting spaces.

Thanks very much to everyone for their input.

ren

[ren41]

Just one quick question - I had Calibre generate a ToC which consisted of a link to just one poem, somewhere in the middle of the document. I tried several times but always got the same result. There are no hyperlinks at all in the text, and all my titles were surrounded by H2 tags so I expected to get a ToC which listed them all. Does anyone have any ideas about this, please?

TIA

ren

[ren41]

OK, I have managed to fix it now. When I took out the wild card (*) and just left the detection as h2, it worked.

ren