[3:02]<Dantman> Heh, a single more message and my unread messages counter for the MongoDB mailing list will reach 2000 [4:20]* Dantman whistles and celebrates...... ^_^ my MongoDB mailing list unread messages counter has now hit 2000 [10:09]<kuya> ashb: http://dpaste.com/179270/ [10:09]<kuya> ashb: are you meant to be able to create an OID? [11:04]<ashb> kuya: in theory, yeah [11:04]<ashb> oh tho you can't find based ona a string [11:05]<ashb> so L13-14 shouldn't work [11:05]<ashb> L18 should work somehow tho. [11:05]<kuya> i get create OID failed probably: Error: Object is not native [11:07]<kuya> also, slightly different topic... when you insert an object, shouldnt it get an _id property? [11:07]<kuya> i dont think its possible to get back the id of the object you inserted [11:07]<ashb> yea the DB will, but you only see it when you refetch it [11:07]<ashb> might be a cmd to see it. let me check [11:08]<kuya> how can i fetch it without the id? :] [11:10]<Dantman> MongoDB driver? [11:11]<ashb> yeah for flusspferd [11:11]<Dantman> Every proper MongoDB driver creates ObjectId's in the driver, it doesn't expect the server to do that. [11:11]<ashb> oh, really? [11:11]<ashb> didn't know that [11:11]<Dantman> mhmm [11:11]<kuya> i never got that impression from any docs... [11:12]<ashb> i'm not seeing that in the C++ client lib [11:12]<kuya> python driver returns a new obj id when you insert [11:12]<Dantman> They also usually either return the id and/or modify the object you gave them [11:17]<ashb> maybe get last error? [11:18]<ashb> hmmm nope. [11:18]<Dantman> get last error? [11:19]<ashb> kuya: hmmm the valueOf() ctor should work. [11:21]<kuya> oid = new OID.valueOf('blah') ? [11:21]<ashb> yeah [11:21]<ashb> er no [11:21]<ashb> just new OID('blah') [11:22]<kuya> it doesnt :( [11:22]<kuya> i tried that in that dpaste post [11:22]<ashb> hmm. [11:22]<ashb> bug then [11:23]<kuya> thats what gives me the object is not native error [11:23]<kuya> a quick google gave me no real clues to that but i presume its spidermonkey stuff [11:24]<ashb> flusspferd stuff actually [11:24]<kuya> ah ok [11:27]<ashb> hrmmmm. even new m.OID() doesn't work [11:37]<ashb> kuya: hrmm. not quite sure whats going on here either [11:37]<ashb> my head isn't in the right place to track it down either :( [11:39]<kuya> np ashb :) [11:39]<kuya> i wasnt expecting a fix now :) [11:44]<ashb> kuya: as a work around _id doesn't have to be an OID [11:44]<ashb> it can be anything unique [11:44]<kuya> oh right [11:44]<kuya> i can probably work with that till its fixed [11:44]<ashb> which is why i haven't noticed it - i'm using strings [12:14]<kuya> sillyness: http://dpaste.com/179287/ :D [13:01]<ashb> hmmmm [13:01]<ashb> heh [13:32]<kuya> :) [13:32]<kuya> im sure there are some advantages [13:33]<kuya> but its a bit weird passing objects and arrays around [13:33]<kuya> js {} -> python != dict [13:33]<kuya> py [] -> js != Array