Bookit plugin for Firefox

[beowulf573]

Originally Posted by =X=

Beowulf, will you be saving of HTML off recursively too? That is the nice feature of web2lrf. If you are then I say go for it, taking advantage of Netscape is a good idea.


Comment on the proxy. The proxy information is located in the pref.js for Firefox. I've noticed your Plugin data is stored there too. If there is an API where you can read/write to the pref.js file. Then you should be able to get the proxy info.
You would have to make one check first before setting the proxy to calibe and that is if the proxy is enabled or disabled.

I'd rather not recreate web2lrf features in javascript if I don't have too. Too much work and too much duplication.

Actually, I can't just read the pref information. There are several plugins to override the default behavior; for example determining which proxy to used based on a regexp or to use a random proxy. I can query the proxy interface to determine what proxy would be used for a given url, however the low level library used by calibre only supports http proxies. If you're using SOCKS you're out of luck.

[ronaldo]

I have just read all thread and got a lot of information, Thanks guys.
Now I am going to other sections of forums.

[dsuden]

Ultimately some documentation will be useful for this tool, since it is becoming so nicely sophisticated. I would be willing to help with this if that's useful, though I will also need some guidance describing some of the features.

Dane

[=X=]

Originally Posted by beowulf573

I'd rather not recreate web2lrf features in javascript if I don't have too. Too much work and too much duplication.

I completely understand. Why re-invent the wheel. My personal feel then is to stick with using the web2lrf. (Or have a toggle button on your options page where the user can specify which to use [1]web2lrf or [2]html2lrf)

Originally Posted by beowulf573

Actually, I can't just read the pref information. There are several plugins to override the default behavior; for example determining which proxy to used based on a regexp or to use a random proxy. I can query the proxy interface to determine what proxy would be used for a given url, however the low level library used by calibre only supports http proxies. If you're using SOCKS you're out of luck.

Yes as long as your able to poin to the proxy that should be fine for me. My corp. uses SOCKS, but I've been able to determine what is the actual proxy used by the company.

Thank you
=X=

[Peto]

Originally Posted by dsuden

Ultimately some documentation will be useful for this tool, since it is becoming so nicely sophisticated. I would be willing to help with this if that's useful, though I will also need some guidance describing some of the features.

Dane

I could help translating the program and the documentation into Spanish.

[jotheman]

Hi there,

I played around with the clipmarks plugin for firefox today. It has a really nice and easy way for selecting the content of a page that one wants to save. Only problem is the 1000-character-limit, makes it practically unusable for my purposes.

I would love to just select (wether asisted by the plugin or just doing a normal selection) the part of the page I want to convert into an eBook, click a button and get a good result.

Please think about the "selection to eBook" feature I suggested earlier.

Thanks,


jo.

[beowulf573]

Originally Posted by jotheman

I would love to just select (wether asisted by the plugin or just doing a normal selection) the part of the page I want to convert into an eBook, click a button and get a good result.

I'm definitely going to implement it, in fact that's what made me think about moving the whole plugin to use html2lrf. I think I'll put the latter on hold for now and work on...

• Cleaning up the create with options/no options feature.
• Documentation
• Selection to ebook

I should have a new release with these out this weekend sometime, unless my wife adds more items to my chore list.

[jotheman]

That's great news!

[dsuden]

Thanks Beowulf...looking forward to it!

[beowulf573]

I've put a new version up that has the selection to LRF file support. Since several folks seem to really want it, I've put it on my site but not yet in the update feed, so you'll need to install it manually.

I've not tested it yet in Firefox 3.0, so this is for folks who just can't wait.

Version 0.3.6

Sometime tonight I'll fix a few more bugs and do some more testing and then I'll put 0.3.7 out and add it to the update feed.

To use the new feature, select some stuff, bring up the context menu and select "Selection To LRF". There's also a Hide Element menu item. If there's some stuff in the middle of the selection you don't want to see in your final document, right click on it and select Hide Element. This just sets the visible attribute to false, if you refresh your page everything resets. Also this doesn't work when building a LRF file from the entire page using the normal process.

[sdoane57]

Just tried 0.3.6 in Firefox 3.0. Works GREAT! Thanks, Beowulf573!

[beowulf573]

Cool, thanks. I just updated the feed with 0.3.7. I added code to default the current title in the create dialog (and selection) to be the document title. And fixed a few minor bugs.

[kovidgoyal]

Since bookit seems to be maturing nicely, I'm planning to add a link to it on the calibre download page, is there a preferred URL?

[beowulf573]

This is where I'm hosting it for now. Once it's stable and I'm not changing it every day I may submit it to the mozilla add-ons site.

[jotheman]

Dear beowulf573,

thanks for the quick integration of "selection to LRF"! I tested it immediately and it works really nicely most of the time.

My first suggestion would have been to set the tilte of the current page as document title automatically - but you've already done that since i downloaded the experimental version! Great!

Anyway, that also seems to be part of one of the problems I have discovered: BookIt won't create files with chinese titles. Nothing happens if there is chinese in the title. If I change the it to something without chinese chars in it, it seems to work flawlessly.

Another thing is including images. Hasn't worked for any website I tried it on yet... (www.dw-world.de/chinese for example)
And not all thext that's bold on the webpage will end up bold in the LRF.

Cheers,


jo.