Calibre's new features - ePub support and automated news download!

[astrodad]

Two of the top features that calibre users have requested for the wonderful calibre application are about to be implemented! Or should I say could be implemented if we, the MobileRead community stand behind Kovid and support his efforts to update our favorite ebook manager.

Based on the feedback at this thread, we've created two Chipin pages that will allow calibre users to contribute to the effort.

1. Support for ePub. The goal here is to make the ePub format the base file format for calibre. This is based on a gamble that the ePub format will eventually become the "standard" for new devices. Therefore, this would be the first important step to making calibre work with multiple devices.

To support this effort, please visit:
http://calibre.chipin.com/epub-support-for-calibre

2. The automated downloader. Many calibre users download news from news sites or magazine sites using the site fetching capabilities of calibre. This enhancement request will provide for a way to automatically download and store the fetches into the calibre library.

To support this effort, please visit:
http://calibre.chipin.com/automated-...er-for-calibre

Please participate in this effort by assisting in whichever feature you want most (or both!) using the links provided.

Thanks again Kovid for a wonderful application and for continuing to improve it!

[edembowski]

Just donated, it looks like we're up to 160 already! Hopefully, ePub will become the standard format that Calibre stores data, but having the conversion utilities help greatly.

- Ed

[kovidgoyal]

Note that calibre doesn't mandate a standard format. You can store data in as many formats as you like. The choice of a standard format to store data in is entirely upto the user.

[pilotbob]

Originally Posted by kovidgoyal

Note that calibre doesn't mandate a standard format. You can store data in as many formats as you like. The choice of a standard format to store data in is entirely upto the user.

Right, but there is no choice of what the prefered "Convert To" format is... the default now is always to LRF, understandable since this was originally called libprs500.

BOb

[astrodad]

Good question Bob! I would think that if we stipulate that ePub be the standard format, then we should have a "Convert To" option.

[kovidgoyal]

To clarify:

calibre does not and will not ever have a "standard format"

Adding epub support to calibre means two things:

1) Implementing the *2lrf converters
2) Allowing the user to set epub as the desired output format instead of LRF in the GUI

[astrodad]

Thanks for clearing that up. So how many converters will be necessary? Will there be LIT2EPUB or a MOBI2EUB?

[kovidgoyal]

All that's necessary is html2epub, the rest follow automagically from that

[mattepiu]

Originally Posted by kovidgoyal

All that's necessary is html2epub, the rest follow automagically from that

Sorry, I don't think so.
Even being mainly xhtml, Epub format has a TOC ond other really useful
features that wouldn't be used that way.
For a preliminary python script to convert oeb (lit just through convertlit)
to epub, there's my site here,
however a real interface is needed if you want to add a toc and the chapter titles in your book aren't bolded.
(I'm slowly working for a version using pyqt to give that interface, don't expect it too soon)

Another choiche is instead nakee work here.

[kovidgoyal]

An html2epub converter I write will handle TOC/metadata, just as my html2lrf converter does. When i say html2epub, I certainly don't mean a script to package up HTML into a zip file.

A true html2epub would include support for things like overriding CSS, detecting chapters, following links recursively, detecting metadata, reading/writing OPF, NCX, rationalizing font sizes and so on.

Oh and nakee and I are co-operating.

[mattepiu]

Originally Posted by kovidgoyal

An html2epub converter I write will handle TOC/metadata, just as my html2lrf converter does. When i say html2epub, I certainly don't mean a script to package up HTML into a zip file.

A true html2epub would include support for things like overriding CSS, detecting chapters, following links recursively, detecting metadata, reading/writing OPF, NCX, rationalizing font sizes and so on.

Oh and nakee and I are co-operating.

Good to know! however I was referring to the fact that "detecting" chapters
isn't enough (ebooks and expecially lit are dishomogeneus), so some sort
of interface is needed for manual correction (thus my change from pure
script to pyQt). I have a proof of concept version of my script working this
way but it's still flawed for multiple htmls files (as epub format allows
for example one xhtml file per chapter...)

[kovidgoyal]

Actually detecting chapters works pretty well. By pretty well I mean "it does the right thing" in 90% of cases without any user input. When it comes to writing a GUI to edit ebooks, I'd prefer to write a GUI that that works of a lightweight markup format rather than epub as light weight markup has all the features most ebook creators/editors are ever going to need. The GUI would accept HTML, turn it into light weight markup, and then edit it and turn it back into html.

[wallcraft]

Originally Posted by edembowski

it looks like we're up to 160 already!

Stuck at $160 since 06-19. Now that the PRS-505 reads ePub I hope others will Chipin at

http://calibre.chipin.com/epub-support-for-calibre

[seajewel]

i will chip in some money towards this effort. I'd been meaning to, anyways, just because Calibre is that awesome, but now with the advent of actual epub support on the Reader, and reports that epub is faster (less time to format) I would really love to see epub be an output format for calibre. i'll deal with it in a couple of days, though.. in the middle of finishing up my job and moving. and not very much, i'm afraid, but i'll do what i can in appreciation.

[kovidgoyal]

Well I appreciate all the support and calibre will be getting epub support reagrdless of whether we hit the target or not.