Feature Request For Exporting DB Info

[RoninTech]

Hi Kovid,

Great software you've got here! I'm just setting up my ipod Touch with the Stanza ePub reader and am using Calibre to convert my books to epub format. Stanza has a feature whereby you can access your library online right from the program by simply serving an xml file and your epub books from a webserver and then telling stanza the URL of your online catalog. The syntax of the XML file is pretty straight forward. I was going to whip up a script that would use the info in the Calibre database to generate this XML file and simply mount the Calibre ebook storage folder along with the XML file on my web server so we could access our library from anywhere.

After searching this forum I came across the calibredb command along with its --list feature. The only thing its missing for me to generate this XML file is for it to be able to dump the relative path and filename within the Calibre ebook folder of each books book and image files.

Is this info accessible by calibredb and if so could these 2 extra fields be added to the --list command?

I love it when I stumble across gems of the internet like this. Thanks again for a wonderful piece of software!

Cheers,
CrazyIvan

[kovidgoyal]

Sure open a ticket, and I'll add the functionality for the next release.

[RoninTech]

Done. Thanks!

[RoninTech]

Hi Kovid,

I just tried the new version of calibredb and see the new field for the books file path ( ) but not the one for the associated cover image file. I've updated the ticket with a comment to reflect this.

I've also added another ticket, , to request a CSV option for calibredb's list output. Currently it uses spaces to separate the fields, which can also appear in the field data making it hard to parse out. Adding a --csv option to use commas to separate the fields, including the headings, rather than spaces would greatly simplify parsing. I also threw in a request for an "all" field that would list them all.

Cheers,
Crazy

[RoninTech]

Something just dawned on me. Is the cover image file _always_ "cover.jpg"?



Cheers,
Crazy

[kovidgoyal]

yeah cover is always named the same.

[RoninTech]

As I wait for the CSV option update to calibredb I had a thought . Rather than me running an external script using the new features in calibredb would it be out of line for me to ask for a feature request for an option for Calibre to generate an atom feed XML file for its catalog? This would mean any iPhone/Touch Stanza user could access their Calibre ebook collection on their own device by having the Calibre library on a web server along with this generated XML atom feed. Stanza's catalog atom feed format is pretty straight forward. Here's an example I whipped up that works on my iTouch:

PHP Code:

<feed xmlns="http://www.w3.org/2005/Atom" xmlns:dc="http://purl.org/dc/elements/1.1/">
  <title>Calibre Generated Stanza Catalog</title>
  <id>tag:mywebserver,2008-10-12:/ebooks/stanza.xml</id>
  <updated>2008-10-12T00:44:20+00:00</updated>
  <link rel="self" type="application/atom+xml" href="http://mywebserver/ebooks/stanza.xml"/>
  <author>
    <name>Crazy</name>
    <uri>http://mywebserver</uri>
    <email>librarian@mydomain.com</email>
  </author>
  <entry>
    <title>20000 Leagues Under The Sea</title>
    <content type="xhtml">
      <div xmlns="http://www.w3.org/1999/xhtml"> Published: 1870 Subject: Novels Language: en</div>
    </content>
    <id>tag:mywebserver,2008-10-12:/ebooks/Jules%20Verne/20000%20Leagues%20Under%20the%20Sea%20(2)/20000%20Leagues%20Under%20the%20Sea%20-%20Jules%20Verne.epub</id>
    <author>
      <name>Verne, Jules</name>
    </author>
    <updated>2008-10-12T00:44:20+00:00</updated>
    <link type="application/epub+zip" href="http://mywebserver/ebooks/Jules%20Verne/20000%20Leagues%20Under%20the%20Sea%20(2)/20000%20Leagues%20Under%20the%20Sea%20-%20Jules%20Verne.epub"/>
    <link rel="x-stanza-cover-image-thumbnail" type="image/jpg" href="http://mywebserver/ebooks/Jules%20Verne/20000%20Leagues%20Under%20the%20Sea%20(2)/cover.jpg"/>
    <link rel="x-stanza-cover-image" type="image/jpg" href="http://mywebserver/ebooks/Jules%20Verne/20000%20Leagues%20Under%20the%20Sea%20(2)/cover.jpg"/>
  </entry>
  <entry>
    <title>The Time Machine</title>
    <content type="xhtml">
      <div xmlns="http://www.w3.org/1999/xhtml"> Published: 1895 Subject: Novels Language: en</div>
    </content>
    <id>tag:mywebserver,2008-10-12:/ebooks/H%20G%20Wells/The%20Time%20Machine%20(8)/The%20Time%20Machine%20-%20H%20G%20Wells.epub</id>
    <author>
      <name>Wells, H G</name>
    </author>
    <updated>2008-10-12T00:55:20+00:00</updated>
    <link type="application/epub+zip" href="http://mywebserver/ebooks/H%20G%20Wells/The%20Time%20Machine%20(8)/The%20Time%20Machine%20-%20H%20G%20Wells.epub"/>
    <link rel="x-stanza-cover-image-thumbnail" type="image/jpg" href="http://mywebserver/ebooks/H%20G%20Wells/The%20Time%20Machine%20(8)/cover.jpg"/>
    <link rel="x-stanza-cover-image" type="image/jpg" href="http://mywebserver/ebooks/H%20G%20Wells/The%20Time%20Machine%20(8)/cover.jpg"/>
  </entry>
</feed> 


The majority of the info in this xml file would be available to Calibre as it sits except for the web server base URL and the catalog owner's name and email address. The only other thing would be to replace the spaces in the url's with %20's. I think more info can be provided by an entry's content->div but I haven't played with it yet. Let me know what you think about supporting iphone/Touch's in Calibre via this atom feed route and if you agree I'll add a feature request. Otherwise I'll do this as I originally planned, in an external script via the updated calibredb.

It ends up looking like the attached ipod Touch screenshot.

Crazy

[RoninTech]

Sorry, I'm new to this forum and have a newbie question. I made a post yesterday that said it needed moderator approval. I'm assuming that was because it had a link to a screenshot of Stanza running on my touch and/or some xml code inline (using the PHP tag). How long do posts generally take to be approved in this forum? I'm wondering if something went wrong and the post is lost in the ether and would like to know whether to repost or not.

Thanks,
Crazy

[Nate the great]

I just now saw the post.

[kovidgoyal]

Are you saying you want calibre to run a webserver?

[RoninTech]

Originally Posted by Nate the great

I just now saw the post.

Thanks Nate.

Originally Posted by kovidgoyal

Are you saying you want calibre to run a webserver?

Not personally Kovid, I use a Linux box with Apache to run my home automation stuff so I planned to simply samba mount my Windows Calibre library on my Linux webserver so I can access all of my books in Stanza on my ipod Touch. All that's missing to do this is the XML file describing the location of the epubs and their cover images.

Now, that being said, if there's a nice tight web server module that could be slapped into Calibre it might prove useful to some that have wirelessly enabled reading devices. There seems to be quite a few apps out there these days that include their own web front end in order to increase the apps accessibility.

For starters though, since there are so manY free web servers available to folks on all the major platforms, just having help generating this atom feed xml file would be a great help.

Crazy

[kovidgoyal]

Generating an XML file from calibredb is pretty trivial, open a ticket for it, but I have a lot of other things in my queue at the moment, and I don't have a iPhone, so if you want this done quickly you're better off doing it externally.

[RoninTech]

Sounds good. I'll post the script once I have it working to help out anyone else in the meantime. Thanks again for your help!