1. SCons
  2. Core
  3. SCons

Wiki

Clone wiki

SCons / BugParty / IrcLog2009-04-15

17:30:10  *      garyo-home (n=[chatzill@209-6-158-38.c3-0.smr-ubr1.sbo-smr.ma.cable.rcn.com](mailto:chatzill@209-6-158-38.c3-0.smr-ubr1.sbo-smr.ma.cable.rcn.com)) has joined #scons 
17:30:24  <garyo-home>   hi guys 
17:30:47  <[GregNoel](GregNoel)>     Hi, Gary. 
17:31:20  <[GregNoel](GregNoel)>     Steven's laptop went to sleep a few minutes ago, so I suspect he's on the way to the shuttle. 
17:31:24  <garyo-home>   Just waiting for Steven I think 
17:32:25  *      stevenknight (n=[stevenkn@67.218.107.200](mailto:stevenkn@67.218.107.200)) has joined #scons 
17:33:09  <garyo-home>   and here he is now.  Hi Steven! 
17:33:15  <stevenknight> hey 
17:33:25  <garyo-home>   OK to start? 
17:33:32  <stevenknight> good by me 
17:33:48  <garyo-home>   So 2393 is clearly my bug, I'll figure out what's going on there. 
17:33:54  <garyo-home>   I've reproduced it. 
17:34:08  <[GregNoel](GregNoel)>     done 
17:34:29  <stevenknight> done 
17:34:42  <[GregNoel](GregNoel)>     2394, +dmd? 
17:34:45  <garyo-home>   2394: Greg, why do you think Bill's patch won't work? 
17:35:07  <[GregNoel](GregNoel)>     It just makes sure the library doesn't occur more than once. 
17:34:59  <garyo-home>   oh wait, I see now 
17:35:03  <stevenknight> +dmd ++ 
17:35:09  <garyo-home>   Ok, I agree 
17:35:14  <[GregNoel](GregNoel)>     When? 
17:35:33  <stevenknight> 2.x p3 
17:35:24  <garyo-home>   Can Bill take it? 
17:35:39  <stevenknight> i'm okay with giving it to bill 
17:35:41  <garyo-home>   ok w/ me 
17:35:50  <[GregNoel](GregNoel)>     OK, we'll try it. 
17:35:45  <stevenknight> not sure if he'll be okay with it tho 
17:35:54  <stevenknight> the last one was just applying someone else's patch 
17:36:09  <stevenknight> so it's not like he's a dmd guy or anything 
17:36:12  <stevenknight> still, let's try it 
17:36:14  <[GregNoel](GregNoel)>     right 
17:36:19  <garyo-home>   Probably the OP will tell him that it didn't work and ask for more help. 
17:36:39  *      stevenknight takes 10x the words to say what Greg says, as usual... :-) 
17:37:00  *      [GregNoel](GregNoel) types too slowly to do otherwise 
17:37:00  <stevenknight> yeah, leave it with bill until further notice, then 
17:37:11  <[GregNoel](GregNoel)>     done 
17:37:21  <stevenknight> done 
17:37:29  <[GregNoel](GregNoel)>     2395 
17:37:33  <garyo-home>   2395: I now agree w/ you guys, need config, ask OP for patch 
17:37:43  <stevenknight> k 
17:37:53  <stevenknight> i think this should be collected with the other symlink ones 
17:37:53  <[GregNoel](GregNoel)>     no, not +symlink 
17:38:00  <stevenknight> why not? 
17:38:30  <[GregNoel](GregNoel)>     it's just an option to copytree(); it doesn't have to do with making symlinks first-class objects 
17:39:04  <stevenknight> hmm, okay 
17:39:04  <stevenknight> i guess 
17:39:10  <garyo-home>   More I think about it the more I realize neither way is better, I can make valid cases for either one.  Flag is necessary. 
17:39:12  <garyo-home>   IMHO 
17:39:53  <[GregNoel](GregNoel)>     I don't know if you can have flags on Actions, but that's the obvious solution. 
17:39:21  <stevenknight> i see the technical distinction 
17:39:32  <stevenknight> but not sure if users care 
17:39:40  <stevenknight> okay, not +symlink 
17:39:48  <stevenknight> ask OP to contribute 
17:40:11  <[GregNoel](GregNoel)>     yes, agreed, ask OP 
17:40:23  <stevenknight> do we need any more categorization? 
17:40:46  <garyo-home>   i don't think so 
17:40:50  <[GregNoel](GregNoel)>     ditto 
17:40:52  <stevenknight> okay, done 
17:41:00  <[GregNoel](GregNoel)>     2396 
17:41:22  <[GregNoel](GregNoel)>     Dearth of comments on this one... {;-} 
17:41:48  <garyo-home>   Greg: your only reservation is the string join, right? 
17:41:51  <stevenknight> had trouble looking at the patch 
17:41:51  <[GregNoel](GregNoel)>     Ah, I see, wrong row... 
17:42:01  <garyo-home>   (sorry, just fixed my comment) 
17:42:14  <stevenknight> and bad network lag w/Firefox on the bus now... 
17:42:19  <[GregNoel](GregNoel)>     yes, the string join; I don't know if this can run on 2.2 or before. 
17:42:41  <garyo-home>   I think since it's a patch to a previous patch we should get it in sooner. 
17:43:37  <stevenknight> agreed re: sooner if we broke something in a recent patch 
17:42:48  <stevenknight> string joins are usually easy to convert to old syntax... 
17:43:04  <garyo-home>   Sure, trivial. 
17:43:13  <[GregNoel](GregNoel)>     +Easy for sure 
17:43:20  <stevenknight> so 1.3 p2 ? 
17:43:30  <[GregNoel](GregNoel)>     works; who? 
17:44:43  <garyo-home>   I'm heading out soon for a week so I'm not the right person for this one.  I'd commit it if there were a test though. 
17:44:59  <stevenknight> give it to me 
17:45:05  <[GregNoel](GregNoel)>     done 
17:45:28  <[GregNoel](GregNoel)>     (Although I hate to interfere with 1.3-RC.) 
17:45:33  <stevenknight> 2397:  consensus 
17:45:41  <garyo-home>   ok 
17:45:48  <[GregNoel](GregNoel)>     done 
17:46:00  <garyo-home>   2398 consensus too? 
17:46:06  <[GregNoel](GregNoel)>     2398, also consensus 
17:47:04  <[GregNoel](GregNoel)>     2399, Gary, I agree with your comment, which is why I said it's not obvious to find... 
17:47:25  <garyo-home>   I could take this one as long as it's not high priority. 
17:47:37  <garyo-home>   I have a Mac and understand that part of the code. 
17:47:59  <[GregNoel](GregNoel)>     For 2.x p3, I'm not going to worry about who gets it. 
17:48:18  <garyo-home>   ok, good. 
17:48:25  <garyo-home>   esp. if it's +Easy 
17:48:20  <stevenknight> is there any more useful information the OP could provide? 
17:49:05  <garyo-home>   Steven: could ask for small testcase, but otoh I haven't tried it. 
17:49:38  <[GregNoel](GregNoel)>     I'll mention that in the note assigning it to you 
17:49:49  <stevenknight> sounds good 
17:49:51  <stevenknight> done 
17:49:51  <garyo-home>   I think it'll either be easy to repro, or we'll have to get more info (in which case it could be user error) 
17:50:32  <[GregNoel](GregNoel)>     On to 2004? 
17:50:48  <stevenknight> yeah 
17:50:53  <stevenknight> i didn't get to the spreadsheet, though 
17:50:59  <garyo-home>   sorry me neither 
17:51:08  <[GregNoel](GregNoel)>     I'm only half-way 
17:51:15  <garyo-home>   743 is wontfix though. 
17:51:20  <stevenknight> let's make what progress we can 
17:51:22  <stevenknight> 743:  wontfix 
17:51:38  <[GregNoel](GregNoel)>     done 
17:51:43  <stevenknight> the part of me that wants to do *everything* Make can do doesn't like it 
17:51:51  <stevenknight> but it's definitely not high enough priority 
17:51:59  <garyo-home>   not nearly high enough! 
17:52:14  <stevenknight> 744:  i'll get the patch off sourceforge 
17:52:35  <[GregNoel](GregNoel)>     I'll assign it to you; assign it back when the patch is there. 
17:52:48  <stevenknight> done 
17:53:34  <stevenknight> 765:  2.x, p2, stevenknight, packaging 
17:53:47  <stevenknight> note re: someone else taking it if we find a better packaging guru 
17:53:49  <garyo-home>   765: could be research to see if it still happens 
17:54:07  <stevenknight> garyo-home:  good point 
17:54:07  <[GregNoel](GregNoel)>     point 
17:54:31  <[GregNoel](GregNoel)>     I like research, I guess 
17:54:36  <stevenknight> okay, research 
17:54:42  <[GregNoel](GregNoel)>     done 
17:54:42  <stevenknight> more likely it'll get off the plate sooner 
17:55:08  <stevenknight> 767:  yeah, already fixed 
17:55:25  <stevenknight> when we changed the default behavior w/the sig refactoring 
17:55:25  <[GregNoel](GregNoel)>     FIXED, then? 
17:55:26  <garyo-home>   ok 
17:55:32  <stevenknight> yes 
17:55:35  <[GregNoel](GregNoel)>     done 
17:56:26  <[GregNoel](GregNoel)>     768 
17:56:31  <stevenknight> (trying to pull it up) 
17:56:32  <garyo-home>   768 is really asking for Install with perms I think. 
17:56:42  <[GregNoel](GregNoel)>     yes 
17:57:06  <garyo-home>   That's a useful suggestion and there's a wrapper builder on the wiki for it. 
17:57:34  <garyo-home>   I recommend closing this one and adding an enh req for Install with perms. 
17:57:40  <garyo-home>   (unless there's already one) 
17:57:45  <stevenknight> seems like some other bug in our voluminous list should cover this topic more generally 
17:58:01  <stevenknight> i like closing and opening an Install-with-perms issue 
17:58:10  <[GregNoel](GregNoel)>     That would be 'automake' compatibility. 
17:58:19  <garyo-home>   yes, in a way 
17:59:34  <[GregNoel](GregNoel)>     wontfix?  invalid?  something else? 
18:00:01  <garyo-home>   doesn't much matter as long as there's a comment explaining why.  wontfix. 
18:00:09  <stevenknight> wontfix 
18:00:12  <[GregNoel](GregNoel)>     done 
18:00:36  <[GregNoel](GregNoel)>     780 
18:00:49  <garyo-home>   invalid. 
18:00:54  <stevenknight> 780:  actually, iirc, this was deceptively difficult when I tried to look at it awhile ago 
18:01:19  <stevenknight> the obvious solutions had side effects due to our tangled web of builder subtypes 
18:01:19  <garyo-home>   (Also, I fixed the EEXIST mkdir error that would have occurred if he had made a proper Action here.) 
18:01:59  <stevenknight> i could be misremembering though 
18:02:20  <garyo-home>   I believe it 
18:02:46  <stevenknight> how about 2.x p3 stevenknight, then? 
18:03:23  *      [GregNoel](GregNoel) will defer to the experts 
18:03:25  <stevenknight> or research, to characterize whether it is or isn't easy 
18:03:26  <garyo-home>   to fix the doc? 
18:04:08  <stevenknight> or to fix the doc 
18:04:10  <stevenknight> yeah, research, me 
18:05:06  <[GregNoel](GregNoel)>     Gary concurs? 
18:05:13  <garyo-home>   sure. 
18:05:16  <[GregNoel](GregNoel)>     done 
18:05:21  <stevenknight> done 
18:05:30  <garyo-home>   790: we'd need more info or a mingw expert to fix this. 
18:06:06  *      [GregNoel](GregNoel) will defer to the experts again 
18:06:40  *      stevenknight can't get firefox to stop going non-responsive 
18:07:23  <stevenknight> gary, any thoughts?  I can't pull up anything right now 
18:07:53  <garyo-home>   I don't think any of us know enough to write a patch for this issue, we'd need more info. 
18:08:13  *      stevenknight tries shutting down firefox to start from scratch 
18:08:54  <stevenknight> is the OP identifiable as an active user? 
18:09:27  *      garyo-home goes to check 
18:09:45  <garyo-home>   nope. 
18:09:53  <garyo-home>   pre-Tigris I guess 
18:10:20  <stevenknight> either assign it as research to someone (me) or else defer to next time 
18:10:41  <stevenknight> i may be able to get advice from co-workers 
18:10:42  <[GregNoel](GregNoel)>     or close it?  Five years old... 
18:10:56  <garyo-home>   tempting indeed. 
18:10:33  <garyo-home>   do you have mingw? 
18:10:58  <stevenknight> no 
18:11:00  <stevenknight> i try to keep my Windows systems vanilla, 
18:12:15  <stevenknight> defer this one to next time and let's move on? 
18:12:20  <[GregNoel](GregNoel)>     done 
18:12:29  <[GregNoel](GregNoel)>     797, what he _wants_ to do is bind shannon.so to the wrapper library, but what he's _trying_ to do is put shannon.so inside the wrapper library. 
18:12:29  <[GregNoel](GregNoel)>     I know the latter is bogus and I don't think we have any handy way to do the former. 
18:13:01  <garyo-home>   Right, the workaround is as always attributes.shared=1, which is not ideal 
18:13:21  <stevenknight> damn, shuttle stop < 1 min. 
18:13:40  <[GregNoel](GregNoel)>     Damn, I wanted to ask about schedule... 
18:13:52  <stevenknight> i thought our answer was setting STATIC_AND_SHARED_ARE_THE_SAME 
18:13:57  <stevenknight> or whatever that big long variable name is 
18:14:22  <garyo-home>   steven: that's not quite the same, he's adding a .so as a source, and that's not tagged. 
18:14:31  <stevenknight> ah 
18:14:44  <stevenknight> Greg, i can IRC later tonight 
18:14:49  <stevenknight> or we can try for tomorrow 
18:15:12  <[GregNoel](GregNoel)>     good, when?  Pizza is arriving for me, but I'm free after that. 
18:15:06  <stevenknight> send a time that's convenient for you in email 
18:15:06  <garyo-home>   I'm out next week, but actually I *can* do 2 wks from now. 
18:15:22  <garyo-home>   (wait, you're asking about release schedule, not bugparty schedule) 
18:15:33  <stevenknight> gotta go...  i'll check in when i get home 
18:15:34  *      stevenknight has quit ("This computer has gone to sleep") 
18:15:44  <[GregNoel](GregNoel)>     What I can do is put all the issues together... 
18:15:45  <garyo-home>   I have to bow out too, but let me know what transpires. 
18:15:57  <[GregNoel](GregNoel)>     ... and everybody is gone... 
18:15:58  <garyo-home>   ok... 
18:16:02  <garyo-home>   no I'm still here 
18:16:12  <[GregNoel](GregNoel)>     going, going, .... 
18:16:09  <garyo-home>   what were you saying? 
18:16:28  <garyo-home>   you can put all the issues together and then what? 
18:16:53  <[GregNoel](GregNoel)>     We're so close to killing the backlog that I can just make all the old issues show up in the current stack for next time. 
18:17:27  <garyo-home>   That makes good sense to me. 
18:17:28  <[GregNoel](GregNoel)>     If you're going to be here in two weeks, that would give < 30 issues 
18:17:39  <[GregNoel](GregNoel)>     A lot for one sitting, but possible. 
18:17:48  <garyo-home>   I will be here.  (Thought I had a school meeting but it was changed.) 
18:18:13  <[GregNoel](GregNoel)>     OK, I'll make the next meeting in two weeks. 
18:18:06  <garyo-home>   If I can get to the spreadsheet that'll help of course 
18:18:37  <[GregNoel](GregNoel)>     And the spreadsheet is always good to do... 
18:18:45  <garyo-home>   :-) 
18:18:26  <garyo-home>   Enjoy your pizza. 
18:18:51  <[GregNoel](GregNoel)>     Thanks, I will.  CUL 
18:18:55  <garyo-home>   bye 
18:18:59  *      garyo-home has quit ("[ChatZilla](ChatZilla) 0.9.84 [Firefox 3.0.8/2009032609]") 

Updated