phpMobi is a php script that can generate .mobi files from valid html files. While this was meant as an experiment, this tool works quite well and can be used to generate mobipocket files from most news articles.
IMPORTANT: Do NOT use this on a public web server. If you do, make sure the input is controlled (this script is absolutely not made to resist malicious inputs) and that the server isn't overwhelmed by the requests as this script is relatively heavy. Also, as no official documentation is available, there might be some bugs/problems in the generated files, but until now I haven't encountered any problems.
MobiPocket is an eBook format created by Mobipocket SA. This tool also uses a php readability port made by Keyvan Minoukadeh.
See index.php for an example of using this program.
Sending an online article as a download:
//Create the MOBI object
$mobi = new MOBI();
//Set the content provider
$content = new OnlineArticle("URL");
$mobi->setContentProvider($content);
//Get title and make it a 12 character long url-safe filename
$title = $mobi->getTitle();
if($title === false)
$title = "file";
$title = urlencode(str_replace(" ", "_", strtolower(substr($title, 0, 12))));
//Send the mobi file as download
$mobi->download($title.".mobi");
Using a previously generated/downloaded html file (will not download any images!):
$data = "<html>...</html>";
$options = array(
"title" => "Local document",
"author" => "Author name",
"subject" => "Subject"
);
//Create the MOBI object
$mobi = new MOBI();
//Set the data
$mobi->setData($data);
$mobi->setOptions($options);
//Save the mobi file locally
$mobi->save($options["title"].".mobi");
This code was implemented while reverse-engineering the MobiPocket format. Therefore this code absolutely isn't optimized for speed, but rather for easy changes, as getting it to produce valid files was quite fiddly.
Modular content provider system: Adding a new data source can be done by extending the ContentProvider class. See the OnlineArticle class for a simple but complete implementation of such a system.
Image support: By default, the online article downloader (and any other content provider that supports images) will download images and integrate them into the mobi file.
Partial UTF-8 support: In practice UTF-8 just works, but theoretically there might be some problems (see missing features).
Compression: This probably won't be implemented (or if it is, only to serve as a reference of the format) as PHP really isn't made for this kind of work and the compression isn't really useful anymore with current download speeds and storage space.
Different eBook types: MobiPocket supports other formats/layouts, such as newspaper-like formats. At the moment only the book layout has been implemented.
Full UTF-8 support: UTF-8 should work most of the time (it worked every time I tested it), but there might be some problems when the character is split over two "records". In practice I've never encountered this problem and special characters (and international characters, such as japanese letters) are shown without a problem on my Kindle.
This code is released under the Apache license (version 2.0)