LRF output

[kovidgoyal]

You don't actually need --ignore-tables, since it automatically ignores table markup when cells are alrge. --ignore-tables causes it to ignore *all* table markup.

[moneytoo]

Hi kovidgoyal,
thanks for all the useful tools. Im just trying to convert some web pages to see how it can handle the charset I use.

I used this command

html2lrf -o "c:\Program Files\libprs500\test\test.lrf" --link-levels=0 --serif-family="C:\Windows\Fonts, Times New Roman" "c:\Program Files\libprs500\test\default8391.html"

But all I get is a LRF file which cant be read neither the Sony eBook Library or the LRF viewer.

The LRF viewer prints this:

LRFParseError: Bad tag ID 00 at 1701355
Failed to render document
Detailed traceback:
Traceback (most recent call last):
  File "main.py", line 47, in run
  File "libprs500\ebooks\lrf\parser.pyo", line 38, in parse
  File "libprs500\ebooks\lrf\parser.pyo", line 62, in _parse_objects
  File "libprs500\ebooks\lrf\parser.pyo", line 70, in _parse_object
  File "libprs500\ebooks\lrf\objects.pyo", line 1102, in get_object
  File "libprs500\ebooks\lrf\objects.pyo", line 79, in __init__
  File "libprs500\ebooks\lrf\tags.pyo", line 202, in __init__
LRFParseError: Bad tag ID 00 at 1701355

If I delete all the CSS files, html2lrf creates right file. But I think it shouldnt create LRF which cant be read. Havent tried put it on the Reader because I dont have any warranty outside the US (I know it has reset button but Im too scared :-) ).

EDIT: I have just tryid downloading the site using web2disk and then converting but I got same results. (The site is http://www.mobilmania.cz/default.aspx?textver=1)

[kovidgoyal]

Fixed in svn. Will be in next release.

[balok]

Hi kovidgoyal,

The --font-delta option seems to affect both line spacing and font size. Is there a way to change only the line spacing, without affecting font size? If not, could you incorporate this ability into a future version?

Thanks!

[kovidgoyal]

The next version will have support for the CSS property line-height. Use

html2lrf --override-css "body {line-height: 200%}" myfile.html

[balok]

Originally Posted by kovidgoyal

The next version will have support for the CSS property line-height. Use

html2lrf --override-css "body {line-height: 200%}" myfile.html

Will this also work for decreasing the line spacing, by using negative values?

[kovidgoyal]

Use percentages smaller than 100%

[balok]

Originally Posted by kovidgoyal

Use percentages smaller than 100%

Of course, thanks.

[aapezzuto]

Is there any way to get a multipage demo of this, with 3-4 html files in a zip, and that spine elements file? maybe even an image or two so I can see how all that is suppose to work?

[kovidgoyal]

You dont need to do anything special for images. As for the spine, just get any lit file and unpack it using clit, look at the .opf file for an example of the spine in action.

[beartard]

I've got a new question (I think) about html2lrf's capabilities.

I've been trying to convert a book with a great deal of endnotes. In the text where the endnote number is superscripted, I'm making the following tag:

<a name="g1"></a><a href="#f1">[1]</a>

Theoretically this should define the current spot as "g1" (for "go-back" when I want to return from the endnote page) and take me to the spot I've named as "f1" (footnote number one). On the endnote page, I reverse the process, giving myself a link back to the spot I came from when following the endnote link.

When I run the html through html2lrf, it doesn't seem to be able to pick up on an empty "name" anchor. My browser seems to handle it just fine. Am I doing something wrong or is this just an example where I should wrap the "name" anchor around some text?

[kovidgoyal]

Hmm that shouldn't happen. Can you attach a test case that demonstrates this behavior.

[beartard]

I'll send you one of the books I"m working on via e-mail if you like. It's still under copyright, so I won't post it here.

[Icarusbop]

This is great work - thanks for all your effort, this program may well convert me from buying a Gen3 to getting the new 505...

Can I have some clarification on zip files please.

I assume I use zip files for sending more than 1 file to the converter (multiple html's for example), I can also add an opf file that then tells your software in which order to add the multiple files.
I read earlier that the 'Spine' area of the opf in used for the file order - are the rest of the areas in the opf file ignored? (the 'manifest' area for example)

Can I use this sytem to insert a cover image? I presume I can create an html file containing a link to the cover image I want, add both the file and my image to the zip, then add the name of the html file to the beginning of the 'spine' area of the opf file. I have tried this but it does not work: is there a maximum size (in dimensions or Kb) that the cover image I am using can be?

Thanks for all your great work!!

[kovidgoyal]

To have the cover detected from the zip file you can do the following:

In the <guide> element have a <reference> element lije that shown below

<reference type="cover" href="relative path to cover file" />

Also note that the converter is recursive, if you give it a single HTML file, it will follow all links in that file to other files on your hard disk.