IrcLog2008 04 01

16:33:12 * Pankrat (n=[email protected]) has joined #scons 16:48:08 * jrandall (n=[email protected]) has joined #scons 16:50:29 * garyo-home (n=[email protected]) has joined #scons 16:55:36 <GregNoel> Hi, guys. Five more minutes to go... 16:57:22 <GregNoel> Pankrat, can you give us a preview of what you wanted to say for 235? 16:58:21 Hi, I have counterexample which does not work correctly with implicit cache activated 16:58:58 <GregNoel> And you can't add a comment? What browser are you using? 16:58:59 but I cannot post due to some error: "URL was not defined; This may indicate a bug in your browser." 16:59:23 Firefox 2.0. I had posted an issue already, which worked 16:59:27 <GregNoel> And do you have cookies enabled? 16:59:32 yes 17:00:02 <GregNoel> Can you mail it to one of us privately so we can add it? 17:00:23 yes, one moment, BTW: I'm Ludwig :) 17:00:41 <GregNoel> Guten Abend! 17:01:24 * stevenknight (n=stevenkn@nat/google/x-a4dfaed744b2a8c7) has joined #scons 17:01:31 hi all 17:01:36 <GregNoel> Hey 17:01:39 Guten Abend :) 17:01:51 <GregNoel> Oder Morgen? 17:02:03 * bdbaddog (n=[email protected]) has joined #scons 17:02:12 hey bill 17:02:22 both fits, mail has been sent 17:02:31 good evening! 17:02:35 <GregNoel> Anybody know where Bill and 17:02:46 <GregNoel> Oops, Bill's here, what about Ken? 17:02:49 hi guys 17:03:19 <GregNoel> G'day! 17:03:40 I may be a bit out of it tonight, sorry. 17:03:44 But I'm here. 17:03:45 does the wiki page link the spreadsheet? 17:03:57 <GregNoel> Yes, it should 17:04:16 oh, duh, there it is 17:04:30 So the agenda is starting with Greg's first issue list (391 issues)? 17:04:51 <GregNoel> Yes, just the first few 17:04:56 good. 17:05:26 btw. I think we can chat via google docs and it would get attached to the spreadsheet. 17:05:33 <GregNoel> 1848 is first I believe 17:06:02 <GregNoel> No, it just seems to display; there's no record (that I could find) 17:06:08 bdbaddog: that would be cool but let's try for that next time. 17:06:15 I thought we wer going through the 2002 and then 2003 bugs. which are in the spreadsheets first 17:06:17 (if there are records anyway) 17:06:33 <GregNoel> I'm recording, I hope. 17:06:40 me too I hope 17:06:45 <GregNoel> 1848? 17:06:46 me 3 17:07:21 yeah, p3 17:07:29 I though 139 was the first bug to discuss ? 17:07:50 <GregNoel> No, the first few are from the issues list; no spreadsheet. 17:08:08 <GregNoel> Just the ones with priorities or votes. 17:08:11 link's on the wiki page 17:08:16 ahh o.k. there now. 17:08:35 damn, I wash my laptop had a bigger screen right now 17:09:07 <GregNoel> It seems to me that I've done what bug is about with no problems, but I looked for it and couldn't find it. 17:09:42 <GregNoel> I wish my second screen on my desktop was working... 17:10:09 1848: sort of nagging thing that shakes confidence when people hit it 17:10:10 doesn't it cause a problem on linux because of no exe suffix? 17:10:12 anyone had time to try and reproduce 1848? 17:10:17 yes re: no .exe suffix 17:10:26 seems to me like a usual case of alias/filename conflict. 17:10:26 yes, someone should check reproducibility 17:10:40 if it's reproducible, what timeframe? 17:10:48 1.x? 17:10:59 <GregNoel> OK, or 2.x 17:11:18 2.x unless the error is really gross. 17:11:34 2.x 17:11:40 i can go with 2.x 17:11:40 done 17:11:41 how about 1966? 17:12:34 i hate wading into the configure code 17:12:35 looks like 2 issues, doc plus functional ? 17:12:38 yeah 17:12:48 maybe fix docs in 1.x, fix issue in 2.x ? 17:13:06 <GregNoel> good for me 17:13:16 +1 17:13:19 should we split bug into two bugs then? 17:13:27 +1 17:13:35 <GregNoel> Bill, will you do it? 17:13:46 yes. 17:13:52 <GregNoel> Next? 17:14:17 1969, looks like 17:14:25 1969: ugly problem 17:14:26 i18n 17:14:35 tip of the i18n iceberg 17:14:40 <GregNoel> Needs features not in 1.5.2; should be 2.x 17:14:50 2.x 17:14:52 (at least default tool setup can now be disabled, but that's not a good answer) 17:14:54 2.x 17:15:11 sure, 2.x 17:15:13 <GregNoel> next? 17:15:20 re; 1969 though 17:15:32 no, wait, i'll update it myself 17:15:41 he obviously didn't know you can disable the tool selection 17:16:05 personal business, brb sorry 17:16:07 keep going 17:16:30 1217 17:16:36 (how far are we going on this list, BTW?) 17:16:41 1217. anyone know the cache management stuff? 17:16:58 I think Greg said "aim for 2002 bugs, hope for 2003" to be handled. 17:16:59 that'd b e me 17:17:16 this needs a design for a mechanism, not a quick fix 17:17:16 this is not minor stuff is it? 17:17:18 2.x 17:17:21 2.x 17:17:24 <GregNoel> ok 17:17:51 <GregNoel> 235 is in our 2002 list; I propose we deal with it there. 17:18:00 ok 17:18:04 ok 17:18:17 1959 then? 17:18:57 i think 1.x, should be an easy fix, and it looks dumb if it doesn't work 17:19:03 * GregNoel stays silent, although 1959 was a very good year 17:19:35 1.x unless its messy. would be my vote. :) 17:19:54 1.x then -- can always be pushed out if it gets bad 17:20:12 <GregNoel> on to 2002 then? 17:20:17 yup. 17:21:03 139 - research 17:21:14 research 17:21:39 <GregNoel> I think we should close it until there's a need 17:22:03 hmm, now that you mention it, I'm okay with that 17:22:15 <GregNoel> wontfix? 17:22:31 if we close, will we loose a placeholder for the idea? 17:22:49 sure -- I'd love to do better than ClearCase, but if there's no compelling user demand, that's just my ego at work 17:23:08 <GregNoel> {;-} 17:23:34 <GregNoel> consensus? 17:23:41 bill, close it? 17:23:52 I guess the issue is is the bugtracker a good place to placehold ideas or should we move to a wiki page? 17:24:07 * stevenknight has quit ("This computer has gone to sleep") 17:24:15 if it's just an idea, make it future 17:24:21 <GregNoel> future p5 then, the "will get to never" stack 17:24:32 o.k. that's fine with me. 17:24:48 ok, I'm sort of here now 17:24:51 148? 17:25:19 <GregNoel> assign to Brandon for research 17:25:31 sounds good to me. 17:25:36 ok 17:25:46 <GregNoel> done 17:25:49 then on to 177 17:26:16 already mostly works like he says 17:26:30 future. I think was the concensus 17:26:34 <GregNoel> yup 17:26:38 ok by me. 17:26:45 or close it. 17:27:06 193 = gsoc? 17:27:08 <GregNoel> 193, we have a viable proposal; 17:27:20 +1 17:27:36 <GregNoel> bypass until next time? 17:27:54 or assume gsoc will get integrated in the 2.x timeframe and assign to that. 17:27:56 or after projects are accepted? 17:28:06 * stevenknight1 (n=[email protected]) has joined #scons 17:28:07 anyway, make a note in the bug 17:28:28 i'm back, had to run get the shuttle 17:28:35 194? I say wontfix 17:28:37 assign bug to soc student? 17:28:48 194 wontfix. 17:28:48 what was the consensus about where to record long-term ideas? issues or wiki? 17:28:57 future, p5 17:28:57 issues as future 17:29:11 okay 17:29:12 194 wontfis 17:29:15 wontfix 17:29:28 <GregNoel> Pankrat, yes; you'll get some too. 17:29:47 :) 17:29:52 great. 219? can o' worms. 17:30:12 yeah 17:30:30 what's the benefit of being able to do this? 17:30:55 <GregNoel> Virtual current working directory, names are more convenient. 17:31:34 kind of like a mini-SConscript? 17:31:51 <GregNoel> Um, close enough. 17:31:52 how about wontfix, and if someone really wants it they can contribute code 17:31:58 +1 17:32:01 no one seems to be beating down the doors for it 17:32:01 +1 17:32:05 <GregNoel> ok 17:32:15 <GregNoel> next? 17:32:26 232: reasearch, then if it's true, it's an easy fix 17:32:37 ??? I don't think it is 17:32:46 <GregNoel> assign to whom? 17:32:52 this is when a user has same header file name in more than one dir. 17:32:54 right? 17:32:55 it's replicating a quirky behavior in MSVC's preprocessor 17:33:04 steven: I don't think it's true either, never heard of that behavior 17:33:10 but can't prove it. 17:33:21 ah, if that's the case, then it is easy: INVALID 17:33:22 <GregNoel> research? 17:33:37 or just close it, it's old and no one else has complained 17:33:40 invalid. 1 guy reported it 6 years ago. 17:33:48 and never complained again? 17:33:49 i think so too. 17:33:51 <GregNoel> ok, invalid 17:34:13 243? 17:34:20 235 got lost? 17:34:21 not 235? 17:34:35 ok, 235 17:34:37 235: research 17:34:44 research. 17:34:56 I have a counter example, but could not post it 17:35:09 (I've sent it to Greg) 17:35:10 pankrat: that's exactly what's needed. 17:35:14 <GregNoel> research and report 17:35:14 excellent 17:35:47 <GregNoel> I'll add it to the bug; who should research? I can take a look. 17:35:48 Pankrat: you couldn't add it to the issue? 17:36:05 yes tigris complained 17:36:10 <GregNoel> Before you were here, he reported a problem, maybe with his browser. 17:36:16 do we need to change your project role? 17:36:19 ah 17:36:39 <GregNoel> I thought anyone could comment? 17:36:41 okay, GregNoel update and research? 17:36:47 <GregNoel> works. 17:36:54 good; now 243? 17:37:10 Greg, you said Ignores would handle this? 17:37:13 243: 1.x if it's really doc 17:37:31 <GregNoel> 1.0 if it's really doc 17:37:36 +1 17:37:40 +1 17:37:41 someone research it? 17:37:43 so that means research, then assign? 17:37:52 I'll research it. 17:38:27 <GregNoel> (Maciej uses it for his stuff; it's known to work; I'll get you a ref.) 17:38:39 <GregNoel> next? 17:38:43 oh, good 17:38:50 317 17:38:53 317: wontfix 17:39:10 <GregNoel> +1 17:39:11 actully, we should then make -d one of the "ignored for compatibility" options that show up at the top of the help 17:39:19 i'll take it 17:39:24 <GregNoel> ok 17:39:26 ok 17:39:35 <GregNoel> 1.0 then? 17:39:50 same w/ 323 I hope? 17:39:51 yeah, it's not destabilizing 17:40:10 <GregNoel> next? 17:40:11 yes re: 323, i'll take that too 17:40:35 324 2.x 17:40:39 2.x 17:40:45 ok 17:40:47 <GregNoel> I'll take it 17:41:11 325 is hard I think 17:41:29 might be, but i think it might be easy with overriding an individual Node's Decider() function 17:41:44 that's not supported by an API right now, but it's architecturally possible 17:41:50 <GregNoel> Or just a flag to ignore all dependencies 17:41:58 future? 17:42:00 future 17:42:02 future 17:42:04 <GregNoel> ok 17:42:06 Is anyone really wanting it? 17:42:09 no one's asking for it 17:42:10 <GregNoel> what priority? 17:42:19 p4? 17:42:19 <GregNoel> p3 then 17:42:21 p3, average 17:42:25 p3 17:42:45 I'd like 326 (-p, env.Dump()) 17:42:58 <GregNoel> ok, 1.x? 17:43:05 <GregNoel> or 1.0? 17:43:05 OK, give it to me. 17:43:08 1.x 17:43:10 1.x 17:43:13 it's a new feature, not 1.0. 17:43:18 <GregNoel> ok 17:43:48 327: 2.x, and i'll take it 17:43:55 <GregNoel> ok 17:43:59 unless someone else really wants in on environment stuff... 17:44:06 hey. I've gotta leave now. my comments are in the spreadsheets. 17:44:16 okay, thanks 17:44:19 <GregNoel> enjoy the sweat 17:44:36 if you have any questions about my comments which are worth waiting for shoot me an email.. otherwise enjoy the party.. 17:44:39 329: what is -w? 17:45:16 make -w tells it to print the "Entering/Exiting directory" messages 17:45:28 <GregNoel> not needed 17:45:32 the only time we do that is if they specify -C on the command line 17:45:43 give it to me, i'll move it to "ignored for compatibility" with the others 17:45:49 <GregNoel> ok 17:46:35 322: consensus seems good, 1.x and Jim Randall 17:46:42 aye 17:46:42 <GregNoel> yes 17:46:50 332 i meant 17:46:54 <GregNoel> Hi, Jim 17:47:05 hello! 17:47:11 hey jim 17:47:17 <GregNoel> that was quick; next? 17:47:21 336: wontfix... 17:47:33 <GregNoel> yes 17:47:43 341: wontfix... 17:47:48 (hey, we're on a roll here...) 17:48:19 <GregNoel> wontfix, aye 17:48:28 342: fixed 17:48:32 <GregNoel> the spreadsheet helps... 17:48:38 <GregNoel> yes 17:48:53 yeah 17:49:11 343: future, at a minimum 17:49:15 <GregNoel> 343, RANLIB 17:49:31 i don't think it's really just RANLIB 17:49:39 <GregNoel> Question: does SCons automatically apply RANLIB for those platforms that need it? Or does the user have to code something? 17:50:44 hang on, let me check 17:51:14 we just set it up in the Tool/ar.py module 17:51:38 <GregNoel> It needs to be applied; that's 1.x 17:51:42 and only if we detect 'ranlib' installed independent from 'ar' 17:51:57 okay, so there are two parts to the issue here 17:52:13 make RANLIB independent from ar: 1.x 17:52:21 GCCTOOLCHAIN stuff: future 17:52:23 ??? 17:52:26 <GregNoel> +1 17:53:04 <GregNoel> Or just drop the GCC alternate toolchain stuff; no user need 17:53:24 okay, i can live with that 17:53:46 i can take RANLIB, I guess 17:53:54 <GregNoel> I'll mark it up. 17:54:01 <GregNoel> next? 17:54:26 344: 1.x, mine 17:54:33 <GregNoel> What priority? 17:54:55 p2, i think 17:54:59 <GregNoel> done 17:55:23 347: wontfix 17:55:24 <GregNoel> 347, 349, close 17:55:30 yes, yes 17:55:45 353 close 17:56:18 <GregNoel> yes, no ego permitted {;-} 17:56:26 :-) 17:56:31 356: wontfix 17:56:40 <GregNoel> yes 17:56:44 374: wontfix 17:56:58 <GregNoel> yes 17:57:10 and that's all for 2002... 17:57:23 <GregNoel> next 2003 17:57:29 <GregNoel> 397? 17:57:37 wow, i step out of the room and you're on to 2003 already! 17:57:47 <GregNoel> got to be quick 17:58:08 i didn't pre-scan these... 17:58:13 nor me, sorry 17:58:24 <GregNoel> homework... 17:58:30 greg, is the issue link you sent sorted in this order? 17:58:47 <GregNoel> uh, not quite, but close enough 17:58:52 <GregNoel> two issues are out of order 17:58:58 397 looks like future to me 17:59:08 or wontfix 17:59:10 <GregNoel> I accidently sorted them by issue id 17:59:22 397 wontfix: I prefer repos as they are 17:59:31 justme 17:59:56 402 then? 18:00:12 we should do something about this one, it's bit me. 18:00:15 :-) 18:00:16 397: wontfix 18:00:42 402: give it to me, I'm revamping the Windows toolchain support 18:00:49 2.x 18:00:54 p2 18:00:55 agree 18:00:55 <GregNoel> done 18:01:19 409: irix is dead, i say wontfix. 18:01:26 409: wontfix 18:01:29 <GregNoel> yes 18:01:31 Besides, parallel builds on IRIX have never really worked right. 18:02:16 416: i say wontfix 18:02:31 it's interesting though 18:02:38 future? 18:02:38 <GregNoel> Interesting, yes 18:02:44 strikes me as the sort of nice-sounding idea that probably has lots of unintended side effects due to statefulness 18:03:02 i can live with future 18:03:15 <GregNoel> There's an associated bug report with a model for doing it 18:03:28 <GregNoel> but it is intended for advanced users 18:03:39 <GregNoel> future is fine 18:03:40 greg: where? 18:03:44 ok, future 18:03:46 which bug report? am i missing a link? 18:04:07 <GregNoel> Isn't there one at the bottom of the bug? 18:04:31 sorry, that. Yes, that's what's interesting. 18:05:12 bottom of 416? 18:05:16 <GregNoel> Oops, no link. wait. 18:06:45 anyway, it's going to end up future. How about 433? 18:06:50 <GregNoel> 1933 18:07:09 huh? 18:07:20 1933 is the associated bug report to 416 18:07:24 <