Support of Google Gears

Request new functionality here
Thomas
Bear Rating Trainee
Bear Rating Trainee
Posts: 2
Joined: 01 Feb 2009, 18:24

Support of Google Gears

Postby Thomas » 01 Feb 2009, 18:30

Hello,

I'm using Tiny Tiny RSS since 2 months, and I really enjoy using it.

I would like to know if it's possible to include Google Gears's support,
like in Google Reader or in Wordpress CMS.

Thank you for this great software!

User avatar
fox
^ me reading your posts ^
Posts: 6318
Joined: 27 Aug 2005, 22:53
Location: Saint-Petersburg, Russia
Contact:

Re: Support of Google Gears

Postby fox » 01 Feb 2009, 19:55

Don't think so. Three main reasons:

1) Gears' "data layer" is a javascript interface to an SQLite "database" which is about as powerful as a grep with SQL syntax added on top. tt-rss requires a lot more than that.

2) All database interaction in tt-rss is done on the backend (which is server-side php code that mostly generates HTML for frontend to display and some auxiliary XML information). It would be quite hard to replace all of the server ping-pong with some local dumb data cache and keep this cache in sync.

3) All of above is purely theoretical anyway, because I don't have time for monumental undertakings such as this.

Thomas
Bear Rating Trainee
Bear Rating Trainee
Posts: 2
Joined: 01 Feb 2009, 18:24

Re: Support of Google Gears

Postby Thomas » 01 Feb 2009, 21:48

Hey Fox,

Thank you very much for your quick & precise answer. That was just a suggest,
because at launch it takes a few seconds to load but after beeing loaded,
TT-RSS is very quick, and that what's makes it powerful and interactive
as you said with the backend :-)

Thank you again... If you need help for your project and if I can help you,
just send me a message, I'm French ;-)

User avatar
fox
^ me reading your posts ^
Posts: 6318
Joined: 27 Aug 2005, 22:53
Location: Saint-Petersburg, Russia
Contact:

Re: Support of Google Gears

Postby fox » 02 Feb 2009, 11:50

Just hack on something you'd like to improve and send me the patches. The translations also need improvement. :D

User avatar
fox
^ me reading your posts ^
Posts: 6318
Joined: 27 Aug 2005, 22:53
Location: Saint-Petersburg, Russia
Contact:

Re: Support of Google Gears

Postby fox » 02 Feb 2009, 15:51

Alright, if anyone is interested, there is some experimental stuff going on in http://tt-rss.org/tt-rss.git (google-gears-experimental branch).

User avatar
fox
^ me reading your posts ^
Posts: 6318
Joined: 27 Aug 2005, 22:53
Location: Saint-Petersburg, Russia
Contact:

Re: Support of Google Gears

Postby fox » 11 Feb 2009, 13:46

Image

Well, it was easier than I expected. I wouldn't rely on it too much, but it seems to sorta-kinda work. It downloads up to 30 days of unread articles and tries to keep unread stuff in sync (some caveats apply).

Anyway, yesterday I merged what I came up with to trunk so it's coming for 1.3.1.

InspirationDate
Bear Rating Trainee
Bear Rating Trainee
Posts: 14
Joined: 23 Feb 2008, 06:40

Re: Support of Google Gears

Postby InspirationDate » 11 Feb 2009, 19:08

Wow! Thats great. I'll have to start bringing my notebook on the train with me.

MattKatz
Bear Rating Trainee
Bear Rating Trainee
Posts: 13
Joined: 29 Oct 2008, 02:16

Re: Support of Google Gears

Postby MattKatz » 12 Feb 2009, 20:38

Hot stuff! Way to go fox.

User avatar
fox
^ me reading your posts ^
Posts: 6318
Joined: 27 Aug 2005, 22:53
Location: Saint-Petersburg, Russia
Contact:

Re: Support of Google Gears

Postby fox » 12 Feb 2009, 21:33

Latest stuff. Syncing:

Image

Done syncing:

Image

If anyone is interested, here's some info on how it currently works:

The feature is opt-in. You have to enable it in the preferences. When disabled, tt-rss uses Gears to cache some runtime stuff, but offline reading is not supported.

I tried to do without manual offline/online switch, but it presented a bit too many weird glitches for my taste. You don't have to download articles when switching offline (those are synced in background), just some javascripts, pixmaps and such - so it doesn't take that much time.

Only a subset of functions is supported in offline mode, but it should be enough for light reading. E.g. labels are displayed, but you can't edit them, tags are not supported, etc.

It also synchronizes articles' unread status between offline and online versions, although in some cases you might encounter stuff being a bit out of sync. Only unread status is synced for the time being.

It downloads up to 2000 unread articles which are not older than one month when syncing. Syncing is incremental, e.g. if it is aborted at some stage, it continues from the last successful checkpoint.

Also, please don't consider the locally stored data a perpetual archive of some sort. For example, expect it to be wiped out and populated with new stuff between version updates if I change the schema. Local data is automatically removed when it becomes older than one month.

The lack of options here is deliberate. I'd like to keep things simple and straightforward.


Return to “Feature requests”

Who is online

Users browsing this forum: No registered users and 2 guests