It sounds good, i'm not around tonight tho.. Drop me an email with an irc channel and can prob do it tomorrow evening or one of the days this week if thats not too late. Would be good to get the pointer stuff properly hammered down, could make some nice stuff possible.
ricardo lafuente wrote: > a quickie -- as i have to deliver the final proofs of a book tonight --, > but i'd be inclined to think we could use the python-cwiid for the > wiimote part. I did some basic stuff with it some time ago and it works > decently, apart from the fact that it crashes often using 2 wiimotes > (which is a shortcoming of the cwiid lib itself, not Python). > only concern would be to keep the commands compatible with the Nodebox > Wiinode lib; but then, maybe we could just scratch that and make our own > method names in case it makes stuff quicker to implement. > let me know if you feel like hacking together on this, would meeting up > on irc suit you? > Stuart Axon wrote: >> I did see that, it uses all the mac apis, and I might well look at >> implementing the api when it comes >> to doing wiimote. I like the idea of a general pointing api that other >> devices plug into, so we don't >> need to change code to support other pointing devices. >> >> (shoebot pointing api) >> | | >> (nodebox api) (wii node api ?) >> >> >> Basically the nodebox api is a little limited, (MOUSEX, MOUSEY, buttondown) >> but needed for compatability. >> I think we can do better (at least support other buttons and wheel for a >> start). >> >> My code is a little experimental at the mo. >> >> >> The most cludgy bit I had to look at was the variables that update every >> frame; as their not re-imported into >> the namespace every frame you just get the first value. >> >> I tried changing them to properties, but the namespace import just gets the >> value the first time. >> >> I reckon properties are the way to go though - (we don't want to re-import >> all vars by namespace every frame, >> it seems expensive). >> >> >> The variables could be properties in the bot, and the setter of each, >> updates it in the namespace of the sketch - >> the bit I'm trying to get my head round is how we can just generate all >> those, so the properties don't need to >> be manually created for all bot vars. >> >> >> >> ----- Original Message ---- >> From: ricardo lafuente <[email protected]> >> To: [email protected] >> Sent: Friday, January 9, 2009 3:22:57 PM >> Subject: Re: [shoebot-devel] Mouse working, mecurial not >> >> i realised i totally missed this thread when it appeared; >> have you seen http://nodebox.net/code/index.php/WiiNode ? >> >> Stuart Axon wrote: >> >>> Well, I've really broken my mecurial now; it just crashes... but I have >>> submitted a bug report. >>> >>> In the meantime heres the mouse support as a diff. >>> >>> The example from nodebox works: >>> http://nodebox.net/code/index.php/Interaction >>> >>> I built this thinking of multiple mouse support + got a bit confused along >>> the way, so possibly >>> it might be over-engineered. The bit I'm least sure about is if I put the >>> data into the namespace >>> in the correct way. >>> >>> I think for shoebot we could probably have a mouse api that is a bit less >>> basic than MOUSEX, MOUSEY, >>> mousedown - I already mentioned multiple pointers, at the very least >>> multiple buttons would be nice. >>> >>> If I get some time, I might look into adding support for the wiimote using >>> one of the libraries out there, I >>> guess the nodebox api would be a global pointer, while there'd be some way >>> to access individual >>> pointers. >>> >>> >>> >>> >>> ----- Original Message ---- >>> From: Paulo Silva <[email protected]> >>> To: [email protected] >>> Sent: Thursday, December 25, 2008 12:09:09 AM >>> Subject: Re: [shoebot-devel] Mouse working, mecurial not >>> >>> i also had errors from mercurial, but only when trying to download >>> into a long directory path (like in the 10th directory level with >>> folders with around 20 characters each) - i had to use a temporary >>> directory like '/mnt/sda4/shoebot/' just for it - maybe it causes the >>> same when uploading, and btw i only had problems when downloading - >>> like me, please report all bugs to mercurial developers... >>> >>> cheers! =) >>> >>> On 12/24/08, Stuart Axon <[email protected]> wrote: >>> >>> >>>> For some reason I'm getting a funny error trying to push my changes. >>>> Possibly I did the wrong thing; I just pulled the animation branch and >>>> started working there. >>>> >>>> hg says: >>>> >>>> [] hg push http://freehg.org/u/stuaxo/shoebot-animation/ >>>> pushing to http://freehg.org/u/stuaxo/shoebot-animation/ >>>> searching for changes >>>> abort: error: Software caused connection abort >>>> >>>> >>>> >>>> _______________________________________________ >>>> Shoebot-devel mailing list >>>> [email protected] >>>> http://lists.tinkerhouse.net/listinfo.cgi/shoebot-devel-tinkerhouse.net >>>> >>>> >>>> >>> _______________________________________________ >>> Shoebot-devel mailing list >>> [email protected] >>> http://lists.tinkerhouse.net/listinfo.cgi/shoebot-devel-tinkerhouse.net >>> >>> >>> >>> >>> ------------------------------------------------------------------------ >>> >>> _______________________________________________ >>> Shoebot-devel mailing list >>> [email protected] >>> http://lists.tinkerhouse.net/listinfo.cgi/shoebot-devel-tinkerhouse.net >>> >>> >> >> _______________________________________________ >> Shoebot-devel mailing list >> [email protected] >> http://lists.tinkerhouse.net/listinfo.cgi/shoebot-devel-tinkerhouse.net >> >> >> >> >> _______________________________________________ >> Shoebot-devel mailing list >> [email protected] >> http://lists.tinkerhouse.net/listinfo.cgi/shoebot-devel-tinkerhouse.net >> >> > _______________________________________________ > Shoebot-devel mailing list > [email protected] > http://lists.tinkerhouse.net/listinfo.cgi/shoebot-devel-tinkerhouse.net _______________________________________________ Shoebot-devel mailing list [email protected] http://lists.tinkerhouse.net/listinfo.cgi/shoebot-devel-tinkerhouse.net
