1. SCons
  2. Core
  3. SCons

Wiki

Clone wiki

SCons / BugParty / IrcLog2009-02-05

18:35:51  *      [GregNoel](GregNoel) is no longer marked as being away 
18:38:37  *      stevenknight (n=[stevenkn@c-69-181-234-150.hsd1.ca.comcast.net](mailto:stevenkn@c-69-181-234-150.hsd1.ca.comcast.net)) has joined #scons 
18:38:46  <stevenknight> hello 
18:38:52  <[GregNoel](GregNoel)>     Hi, stevenknight; Gary's not here yet 
18:39:04  <stevenknight> okay 
18:39:16  <stevenknight> bdbaddog said he'd probably show up but would be a little late too 
18:51:10  *      bdbaddog (n=[bdeegan@adsl-71-142-75-191.dsl.pltn13.pacbell.net](mailto:bdeegan@adsl-71-142-75-191.dsl.pltn13.pacbell.net)) has joined #scons 
18:51:18  <bdbaddog>     Good evening all! 
18:52:05  <[GregNoel](GregNoel)>     Hi, Bill; Steven's in and out; Gary's not here yet. 
18:52:35  <bdbaddog>     K. 
19:04:37  *      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 
19:04:59  <garyo-home>   Hi guys 
19:05:10  <[GregNoel](GregNoel)>     Hi, welcome; we're all here. 
19:05:16  <garyo-home>   great! 
19:05:30  <garyo-home>   Sorry I was late. 
19:05:34  <[GregNoel](GregNoel)>     np 
19:06:27  <garyo-home>   So, what's on the agenda this evening? 
19:06:28  <[GregNoel](GregNoel)>     Are we ready to go?  I think 2294 is up. 
19:07:40  <[GregNoel](GregNoel)>     Gary, what did you find? 
19:08:12  <garyo-home>   I just tried his testcase on the scons.org machine, and it's reproducible. 
19:08:25  <garyo-home>   Didn't go beyond that. 
19:08:19  <stevenknight> garyo-home:  disk space is fine 
19:08:40  <stevenknight> i checked our stats and we're well w/in limits now 
19:09:03  <garyo-home>   stevenknight: thanks. 
19:08:48  <garyo-home>   I could probably fix it, but I think unless it's a regression we should push it out, it's not that serious. 
19:09:13  <[GregNoel](GregNoel)>     I'm puzzled why no output shows up for me; that's really strange. 
19:09:23  <[GregNoel](GregNoel)>     but yes, I agree that it's a lower priority. 
19:09:28  <garyo-home>   Where did you try it? 
19:09:34  <[GregNoel](GregNoel)>     On my Mac. 
19:10:00  <garyo-home>   SConf is a black art to me. 
19:10:07  <[GregNoel](GregNoel)>     Ditto. 
19:10:11  <stevenknight> Ditto ditto 
19:10:25  <[GregNoel](GregNoel)>     Why do you think I want to replace it? 
19:10:34  <stevenknight> +1 to that 
19:10:41  <garyo-home>   But I bet it wouldn't be that hard to debug.  Anyway, 2.1 or so, p3? 
19:10:50  <[GregNoel](GregNoel)>     2.x p4 
19:11:02  <garyo-home>   fine w me too. 
19:11:04  <stevenknight> 2.x p4 is good by me 
19:11:16  <[GregNoel](GregNoel)>     done.  On to the schedule? 
19:11:23  <stevenknight> 2294:  2.x p4 done 
19:11:48  <garyo-home>   ok, sounds fine. 
19:11:53  <stevenknight> schedule first, then 2005q1? 
19:12:02  <[GregNoel](GregNoel)>     yes. 
19:12:05  <[GregNoel](GregNoel)>     Here's the schedule report card: 
19:12:05  <[GregNoel](GregNoel)>     The theory--- 
19:12:05  <[GregNoel](GregNoel)>     1.2 has been released, as of 21 December. 
19:12:05  <[GregNoel](GregNoel)>     A checkpoint was due 29 December. 
19:12:05  <[GregNoel](GregNoel)>     A checkpoint was due 12 January. 
19:12:05  <[GregNoel](GregNoel)>     A release candidate was due 19 January. 
19:12:05  <[GregNoel](GregNoel)>     1.3 was due 26 January. 
19:12:05  <[GregNoel](GregNoel)>     We should be releasing checkpoints toward 2.0. 
19:12:05  <[GregNoel](GregNoel)>     The actuality--- 
19:12:05  <[GregNoel](GregNoel)>     There is still one bug not yet cleared from 1.2. 
19:12:05  <[GregNoel](GregNoel)>     A checkpoint was released on 13 January. 
19:12:05  <[GregNoel](GregNoel)>     Another checkpoint is needed for VS revamp, yes? 
19:12:05  <[GregNoel](GregNoel)>     Some soak time (two weeks), then a release candidate. 
19:12:06  <[GregNoel](GregNoel)>     The official release of 1.3 a week later. 
19:12:06  <[GregNoel](GregNoel)>     Three weeks and change, minimum; odds are 1.3 will be in March. 
19:12:06  <[GregNoel](GregNoel)>     Comments? 
19:12:27  <stevenknight> yes re: vs_revamp checkpoint 
19:12:44  <stevenknight> [GregNoel](GregNoel):  sounds about right 
19:13:02  <garyo-home>   I'm in favor of vs_revamp asap. 
19:13:08  <bdbaddog>     +1 
19:13:25  <stevenknight> i'm fixing broken tests as fast as i can 
19:13:28  <garyo-home>   Greg: what's the 1.2 uncleared bug? 
19:13:52  <[GregNoel](GregNoel)>     Belongs to Steven; didn't look beyond that. 
19:14:38  <stevenknight> it's a scons.org bug re: adding scalable logos 
19:14:38  <garyo-home>   umm, it's 2058, scalable SVG logo for SCons. 
19:14:46  <stevenknight> after making sure they're licensed appropriately 
19:14:52  <stevenknight> i tried to reset it from 1.2 to 1.3 but couldn't 
19:15:06  <stevenknight> i presume because of the scons.org component 
19:15:20  <garyo-home>   i see what you mean. 
19:15:26  <garyo-home>   ok, minor issue anyway. 
19:15:29  <[GregNoel](GregNoel)>     huh? 
19:15:50  <garyo-home>   there's no 1.3 in the dropdown list. 
19:16:12  <[GregNoel](GregNoel)>     scons.org shouldn't have a 1.2 component 
19:16:27  <garyo-home>   milestone you mean? 
19:16:27  <stevenknight> not component, target milestone 
19:16:39  <[GregNoel](GregNoel)>     yes, milestone 
19:16:45  <stevenknight> it has three in the dropdown:  -research-, 1.2 and 1.x 
19:16:46  <stevenknight> that's all 
19:16:58  <[GregNoel](GregNoel)>     There should only be research. 
19:17:18  <[GregNoel](GregNoel)>     It's not keyed to the core release cycle. 
19:17:15  <garyo-home>   so, just redirect it to research then? 
19:17:23  <[GregNoel](GregNoel)>     yes, make it research 
19:17:23  <stevenknight> okay, i'll change it to -research-; that'll take care of that 
19:17:32  <garyo-home>   ok, back to schedule. 
19:13:29  <stevenknight> anyone else have cycles? 
19:13:44  <garyo-home>   How can I help, Steven?  I have vs2005 here. 
19:14:42  <bdbaddog>     I have some intra-build cycles. 
19:14:43  <[GregNoel](GregNoel)>     I'm trying to get the GSoC stuff in shape, also the TaskmasterNG, and then be ready to re-floor the sources; not many cycles for me, none to spare. 
19:18:05  <garyo-home>   Steven: we need to clear out some failing tests, right?  I saw some that looked like regex expectation mismatches, maybe I can look at those. 
19:18:08  <stevenknight> done 
19:18:23  <stevenknight> garyo-home:  that sounds good 
19:18:46  <stevenknight> probably the most help right now is to at least check out vs_revamp 
19:18:56  <stevenknight> and run all the tests on any Windows system available 
19:18:59  <garyo-home>   Are those the main hurdles to getting vs_revamp in? 
19:19:08  <stevenknight> that and doc 
19:19:13  <bdbaddog>     ok. any special setups needed to do that, just qmtest right? 
19:19:20  <stevenknight> and some tests that should be written, but we can probably let that slide 
19:19:30  <stevenknight> you don't even absolutely have to have qmtest 
19:19:30  <[GregNoel](GregNoel)>     you don't even need QMTest 
19:19:40  <stevenknight> runtest.py knows how to fall back and execute the tests without it 
19:19:52  <bdbaddog>     ok, but runs better with qmtest? 
19:19:51  <stevenknight> QMTest is actually another thing I wouldn't mind getting rid of 
19:20:05  <stevenknight> it doesn't have enough mindshare and we're not using any of its theoretic potential 
19:20:17  <stevenknight> but that's a whole 'nother discussion... 
19:20:24  <[GregNoel](GregNoel)>     yes, later 
19:20:32  <garyo-home>   interesting opinion: I was thinking of using it at work as a test fwk. 
19:20:39  <garyo-home>   ok, I'll defer that q. 
19:20:36  <bdbaddog>     well besides wanting to get rid of it, will tests work better if it's installed? 
19:20:43  <garyo-home>   I'll do vs2005 and vs2003 by tomorrow and report the results, does that help? 
19:20:53  <stevenknight> garyo-home:  yes, very much 
19:20:58  <[GregNoel](GregNoel)>     garyo-home ++ 
19:20:55  <bdbaddog>     I have a VS2008express machine. 
19:21:04  <stevenknight> bdbaddog++ too 
19:21:33  <garyo-home>   my buildbot crashed, sorry, I'm restarting it now. 
19:22:00  <stevenknight> good news is that vs_revamp does look like it clears up the few remaining test failures i had on trunk 
19:22:09  <stevenknight> so we should have green buildbots on Windows once it lands 
19:21:50  <bdbaddog>     Steven u have room for the dual quad core? 
19:22:22  <stevenknight> bdbaddog:  ooh, tempting 
19:22:23  <stevenknight> i might 
19:22:26  <stevenknight> how big is the case? 
19:22:37  <bdbaddog>     hmm more or less full size tower. 
19:23:11  <stevenknight> does it have a wifi card?  it wouldn't be close to my router 
19:23:13  <bdbaddog>     I can send you dimensions if you need them..;) dell poweredge sc1430 I think. 
19:23:16  <bdbaddog>     no wifi. 
19:23:20  <bdbaddog>     can get one. 
19:23:21  <stevenknight> i could rustle one up i think 
19:23:36  <stevenknight> let's sync up off line 
19:24:31  <bdbaddog>     k 
19:24:18  <stevenknight> schedule: see if we can land vs_revamp on trunk by early next week? 
19:24:43  <garyo-home>   great w/ me. 
19:24:54  <[GregNoel](GregNoel)>     fine by me, although I don't use it 
19:24:56  <garyo-home>   I will see if I can try it on our real build too. 
19:25:28  <stevenknight> cool 
19:25:53  <bdbaddog>     does this end up with the big delays that David mentioned on startup? 
19:25:57  <stevenknight> just went and got my main windows laptop 
19:26:10  <stevenknight> i have three test failures right now 
19:26:15  <stevenknight> that don't show up on my system at work: 
19:26:30  <stevenknight> src\engine\SCons\Tool\msvsTests.py 
19:27:05  <stevenknight> --- issues w/retrieving default visual studio version on my home system 
19:27:18  <garyo-home>   msvsTests.py fails in the same way for me here. 
19:27:24  <stevenknight> test\[KeyboardInterrupt](KeyboardInterrupt).py 
19:27:31  <[GregNoel](GregNoel)>     ouch 
19:27:51  <stevenknight> -- not sure of cause yet 
19:28:01  <stevenknight> test\option-j.py 
19:28:01  <[GregNoel](GregNoel)>     timing.... 
19:28:15  <garyo-home>   [KeyboardInterrupt](KeyboardInterrupt) fails reliably for me too here. 
19:28:17  <[GregNoel](GregNoel)>     also timing.... 
19:28:35  <stevenknight> possibly timing, not sure 
19:28:34  <garyo-home>   but it's not timing, it's looking for the mssdk. (?) 
19:28:50  <stevenknight> sorry, test\option-j.py isn't timing 
19:29:05  <stevenknight> it's in the part where it's testing fallback on Python versions that don't have threading 
19:29:16  <stevenknight> which it does by dropping in a threading.py that just raises [ImportError](ImportError) 
19:29:16  <[GregNoel](GregNoel)>     ah 
19:29:16  <garyo-home>   option-j fails w/ same stack trace as [KeyboardInterrupt](KeyboardInterrupt) for me too. 
19:29:37  <stevenknight> this interferes with the vs_revamp code that uses subprocess to try to execute the .bat script 
19:29:45  <bdbaddog>     do you guys have the activestate python, or python.org python? 
19:29:46  <stevenknight> vcvars32.bat 
19:29:49  <bdbaddog>     and which versions? 
19:30:02  <garyo-home>   python.org, 2.5.2 for me 
19:30:07  <stevenknight> python.org, 2.4.3 
19:30:17  <bdbaddog>     try swapping in activestate? 
19:30:40  <stevenknight> that won't help test\option-j.py 
19:30:41  <bdbaddog>     which module is it trying to import? 
19:30:49  <garyo-home>   could do, but it needs to work on python.org too. 
19:30:54  <stevenknight> threading.py 
19:31:19  <garyo-home>   Steven: I don't think my failures are the same.  I'm seeing mssdk is not defined. 
19:31:29  <stevenknight> this raises another issue, which is that vs_revamp's use of subprocess to get the info 
19:32:01  <bdbaddog>     import threading works find on activestate python 2.5.2 
19:32:06  <stevenknight> means we won't work on Windows Pythons that aren't compiled with threading 
19:32:14  <stevenknight> bdbaddog:  it works fine for us, too 
19:32:24  <bdbaddog>     ahh o.k. 
19:32:14  <[GregNoel](GregNoel)>     (Note that there's a function in Action that wraps subprocess, if you can use it.) 
19:32:32  <stevenknight> the issue is that test\option-j.py tries to force a test of falling back to working w/out threading 
19:32:42  <stevenknight> by faking out the import with our own "module" 
19:32:40  <garyo-home>   Steven: my failure is at [http://pastebin.com/d3a1654fd](http://pastebin.com/d3a1654fd), is that same as yours or not? 
19:33:16  <stevenknight> garyo-home:  d'oh! 
19:33:20  <garyo-home>   import threading works for me too... ? 
19:33:22  <stevenknight> no, those are other errors 
19:33:47  <stevenknight> i was working them this morning, but didn't get a chance to check them in before leaving for work 
19:33:51  <stevenknight> hang on... 
19:34:34  <stevenknight> garyo-home:  update and try again 
19:34:50  <garyo-home>   ok... 
19:35:16  <garyo-home>   option-j: now fails with your error ('threading' is not defined) 
19:35:16  <stevenknight> i redid the SDK searching to add a new Tool/mssdk.py tool for independent discovery of the SDKs 
19:35:25  <stevenknight> yeah 
19:35:32  <stevenknight> i think we have to at least rewrite that test 
19:35:53  <stevenknight> and decide that we're okay with requiring threading when SCons is run on Wiindows 
19:36:30  <garyo-home>   Well, I'm personally OK w/ it... 
19:36:36  <stevenknight> since it seems threading is enabled in the python.org and activestate versions, that should be okay 
19:37:09  <stevenknight> and people don't build custom Pythons on Windows as readily as they do for other platforms 
19:37:19  <stevenknight> okay, I'll add a release note 
19:37:37  <stevenknight> and a mention of it in the installation chapter of the User's Guide, too 
19:37:50  <stevenknight> garyo-home:  do you also get a test\[KeyboardInterrupt](KeyboardInterrupt).py failure after update? 
19:38:20  <garyo-home>   Good plan.  now [KeyboardInterrupt](KeyboardInterrupt) passes for me, I'm rerunning it to see if it passes twice, yes it passes both times. 
19:38:32  <stevenknight> okay, i need to dig on that one 
19:38:37  <stevenknight> it passes on my work system, too 
19:39:09  <stevenknight> [GregNoel](GregNoel)'s suggestion of timing is distinctly possible, my home system is kind of slow and under-powered 
19:39:27  <garyo-home>   Mine's zippy 
19:40:38  <stevenknight> [GregNoel](GregNoel):  re: wrapping subprocess 
19:40:42  <stevenknight> a la Action.py 
19:40:51  <[GregNoel](GregNoel)>     yes? 
19:41:22  <stevenknight> hang on, interrupt... 
19:41:46  *      [GregNoel](GregNoel) is hanging, but the rope is beginning to dig into his neck... 
19:42:08  <garyo-home>   funny. 
19:42:44  <stevenknight> yeah, that might be what we have to do 
19:43:17  <stevenknight> the hassle is that it's an import inside subprocess 
19:43:37  <stevenknight> and i was really hoping we'd be able to migrate to using subprocess directly 
19:43:43  <stevenknight> instead of Yet Another wrapper of our own devising 
19:43:59  <bdbaddog>     for 2.0? 
19:44:15  <stevenknight> threading requirement for 2.0?  maybe 
19:44:27  <stevenknight> well, except that vs_revamp is what needs this 
19:44:42  <stevenknight> and that's the 1.3 pacing item 
19:45:11  <[GregNoel](GregNoel)>     (what's an import inside subprocess?  I'm missing something) 
19:45:19  <stevenknight> import threading 
19:45:47  <[GregNoel](GregNoel)>     wait a sec... 
19:46:02  <stevenknight> on Windows, it uses threads to handle the async communication in things like _communicate() 
19:46:41  <stevenknight> hmm.... 
19:46:49  <stevenknight> hmmm.... 
19:46:53  <[GregNoel](GregNoel)>     I was checking to see if there was a name change, but not... 
19:47:01  <stevenknight> hmmmm..... 
19:47:35  <[GregNoel](GregNoel)>     Go for more m&m's! 
19:47:29  <stevenknight> i may be able to sidestep this 
19:47:40  <stevenknight> the threading is only used inside communicate() 
19:48:14  <stevenknight> but we don't need to actually use communicate() to get the output from vcvars32.bat 
19:48:27  <stevenknight> because we're not feeding anything to stdin 
19:48:35  <stevenknight> so I think it's actually overkill 
19:48:41  <stevenknight> okay, we're beating this to death 
19:48:45  <bdbaddog>     :) 
19:48:47  <stevenknight> i have a plan on test\option-j.py 
19:49:08  <stevenknight> i also have to tackle test\[KeyboardInterrupt](KeyboardInterrupt).py since it only shows up on my system 
19:49:24  <stevenknight> garyo-home, you can repro the msvsTests.py failures? 
19:49:40  <stevenknight> even after update? 
19:49:45  <garyo-home>   yes.  Want me to look into the cause? 
19:49:59  <stevenknight> yeah, if you could tackle that, it'd be a definite help 
19:50:29  <garyo-home>   ok, will do. 
19:50:37  <stevenknight> also both you and bdbaddog check if any other tests fail on any windows systems you have available 
19:50:50  <bdbaddog>     ok. 
19:50:56  <bdbaddog>     on vs_revamp right? 
19:51:02  <[GregNoel](GregNoel)>     (Hmmm...  If communicate() is called with only one pipe, it acts synchronously; no threading.) 
19:51:52  <stevenknight> aha 
19:52:03  <stevenknight> bdbaddog:  yes, branches/vs_revamp 
19:53:01  <stevenknight> if we land it and release a checkpoint next week, then a 1.3 release candidate two weeks later, right? 
19:53:12  <[GregNoel](GregNoel)>     yes 
19:53:47  <[GregNoel](GregNoel)>     week of 23 Feb. 
19:53:49  <stevenknight> so best bet, assuming Monday, yes, that puts us at...  February 30th! 
19:54:00  <stevenknight> for 1.3 
19:54:04  <[GregNoel](GregNoel)>     yes 
19:54:25  <stevenknight> okay, that sounds like the plan of record, then 
19:54:24  <[GregNoel](GregNoel)>     We'll slide, but it's a nice theory. 
19:54:30  <stevenknight> right 
19:54:43  <stevenknight> on to 2005q1 then? 
19:54:58  <[GregNoel](GregNoel)>     I'm ready; 1064? 
19:55:26  <[GregNoel](GregNoel)>     Can't put any issues in 1.x any more, 2.x p4? 
19:55:40  <stevenknight> 1064:  yes, 2.x p4 
19:55:51  <[GregNoel](GregNoel)>     works 
19:55:51  <stevenknight> (I filled these out a long time ago) 
19:56:12  <garyo-home>   ok 
19:56:30  <stevenknight> 1065:  2.x p4 ? 
19:56:36  <garyo-home>   same. 
19:56:37  <[GregNoel](GregNoel)>     done 
19:56:59  <stevenknight> 1074: 
19:57:15  <stevenknight> i'll go with 2.x p4 tbd on this, too 
19:57:23  <[GregNoel](GregNoel)>     done 
19:57:29  <garyo-home>   ok 
19:57:46  <stevenknight> 1078: 
19:57:48  <stevenknight> sconf_revamp++ 
19:57:58  <[GregNoel](GregNoel)>     1078, 3.x p3 
19:58:07  <stevenknight> done 
19:58:18  <stevenknight> that'll give us nice incentive to move along from 2.0 to 3.0... :-) 
19:58:59  <[GregNoel](GregNoel)>     There's 100+ issues in 2.x; it'll take a while. 
19:58:42  <stevenknight> btw, [GregNoel](GregNoel), thanks for bolding the appropriate rows in the spreadsheet 
19:58:55  <stevenknight> really helps the navigability 
19:59:02  <[GregNoel](GregNoel)>     np 
19:59:31  <[GregNoel](GregNoel)>     1093 
19:59:49  <stevenknight> "subst" keyword 
20:00:12  <garyo-home>   ok w/ that, prob 2.x? 
20:00:25  <stevenknight> I'll add the keyword 
20:00:30  <[GregNoel](GregNoel)>     Sigh, I have a partial writeup of a wiki page discussing this, but it's nowhere near internally consistent enough to put out. 
20:00:58  <stevenknight> 2.x sounds right 
20:01:28  <stevenknight> p2? 
20:01:33  <[GregNoel](GregNoel)>     I'll buy 2.x p2 
20:01:48  <garyo-home>   sold 
20:01:57  <stevenknight> done 
20:02:02  <stevenknight> 1094: 
20:02:12  <garyo-home>   1094 looks like Greg to dup it and untangle 
20:02:14  <stevenknight> dup [GregNoel](GregNoel) 
20:02:21  <[GregNoel](GregNoel)>     Gary has a comment about the subst keyword later; we'll get to it below. 
20:02:46  <[GregNoel](GregNoel)>     1094 2.x p4 
20:02:59  <stevenknight> 1094:  done 
20:03:26  <stevenknight> 1098:  i like the i18n keyword 
20:03:35  <[GregNoel](GregNoel)>     1098 is nasty 
20:03:39  <stevenknight> or unicode, whichever 
20:03:42  <stevenknight> agreed re: nasty 
20:04:00  <bdbaddog>     4.x 
20:04:56  <garyo-home>   unicode is more correct 
20:05:16  <[GregNoel](GregNoel)>     yes, unicode is not the same as i18n 
20:05:03  <stevenknight> 1098:  future, unicode keyword 
20:05:45  <[GregNoel](GregNoel)>     I hate to push it out that far, but I can't think of how to deal with it, either. 
20:06:07  <stevenknight> agreed on both counts 
20:06:06  <garyo-home>   GSoC? 
20:06:20  <[GregNoel](GregNoel)>     garyo-home, hmmm... 
20:06:30  <garyo-home>   just a thought. 
20:06:57  <stevenknight> hey, if the right brilliant student shows up 
20:06:59  <[GregNoel](GregNoel)>     I'll put it on my GSoC list and see if I can figure something out. 
20:07:20  <stevenknight> i think we could write it up to sound like an attractive tough problem to solve 
20:08:10  <[GregNoel](GregNoel)>     We get so few students, I'd put other things as a higher priority, but it might be OK. 
20:07:58  <stevenknight> have we reached consensus? 
20:08:27  <[GregNoel](GregNoel)>     no consensus I see; defer until next time? 
20:08:34  <stevenknight> okay 
20:09:06  <bdbaddog>     I'd say make it future.  We've not had a lot of request for this have we? 
20:09:14  <bdbaddog>     unicode that is. 
20:09:37  <stevenknight> bdbaddog:  we can go into more depth next week 
20:09:46  <bdbaddog>     k. 
20:09:47  <[GregNoel](GregNoel)>     It's because a filename has unicode characters in it. 
20:09:01  <stevenknight> 1107:  2.1 p3 
20:09:04  <stevenknight> and split into parts? 
20:10:08  <[GregNoel](GregNoel)>     1107, I'll untangle it. 
20:10:03  <garyo-home>   What is the point of 1107? 
20:10:30  <[GregNoel](GregNoel)>     gcc -include xxx.h on the command line. 
20:10:39  <garyo-home>   oh, got it. 
20:11:18  <[GregNoel](GregNoel)>     1107, I'll untangle it? 
20:11:35  <bdbaddog>     +1 
20:11:36  <garyo-home>   agree, you do that, then 2.x p3 for the parts? 
20:11:46  <stevenknight> i'm okay with that 
20:11:51  <stevenknight> or re-triaging the parts separately next week 
20:11:50  <[GregNoel](GregNoel)>     Or do we want me to try untangling it and review it next time? 
20:12:00  <garyo-home>   that's fine w/ me too. 
20:12:05  <bdbaddog>     +1 on untangle and retriage 
20:12:08  <[GregNoel](GregNoel)>     OK, I'll do that 
20:12:28  <[GregNoel](GregNoel)>     1116, no clue 
20:13:01  <garyo-home>   1116, it thinks there's going to be a .exp file but there isn't one. 
20:13:21  <garyo-home>   emitter bug in mingw, or mingw not clearing out some var from msvc most likely 
20:13:44  <[GregNoel](GregNoel)>     (Here's Gary's comment about what the subst keyword should cover.) 
20:14:45  <garyo-home>   1116, not a huge prob because of easy workaround (you don't typically install all the files resulting from a [SharedLibrary](SharedLibrary) call anyway) 
20:14:54  <garyo-home>   2.x p4 tbd? 
20:15:06  <[GregNoel](GregNoel)>     done 
20:15:09  <stevenknight> done 
20:15:26  <[GregNoel](GregNoel)>     last one, 1126 
20:15:31  <stevenknight> 1120? 
20:15:52  <stevenknight> i think 1120 is what you meant re: gary's subst comment? 
20:15:53  <[GregNoel](GregNoel)>     Oops, I was one ahead 
20:16:13  <stevenknight> 1120:  2.x p3 subst keyword? 
20:16:15  <garyo-home>   Yes.  1120, patch is pretty simple and non-intrusive 
20:16:35  <garyo-home>   It only does anything if the cmd ends with \ 
20:16:43  <[GregNoel](GregNoel)>     Is Jim still around?  Who should contact him? 
20:17:03  <stevenknight> either you or me 
20:17:14  <[GregNoel](GregNoel)>     hmmm, you? 
20:17:41  <stevenknight> okay 
20:17:47  <[GregNoel](GregNoel)>     I wrote him around Christmas and got no reply 
20:18:17  <stevenknight> i'll check, and if i get no reply, too, then we should probably reassign 
20:18:26  <[GregNoel](GregNoel)>     concur 
20:18:27  <stevenknight> actually, that's true for a number of people we haven't heard from 
20:18:45  <stevenknight> but that's yet *another* topic for another time 
20:18:53  <[GregNoel](GregNoel)>     yes, I just heard from Ludwig; he's busy but hoping to spend some time on SCons 
20:19:03  <garyo-home>   Greg: good news. 
20:18:42  <garyo-home>   1120: Anyone could do this one.  Even the testcase should be easy. 
20:19:16  <stevenknight> garyo-home:  would 1120 be good for your spare-half-hour back burner? 
20:19:35  <garyo-home>   sure, why not.  It's windows after all :-/ 
20:19:42  <stevenknight> okay 
20:19:59  <[GregNoel](GregNoel)>     done; what milestone and priority? 
20:20:02  <stevenknight> 1120:  2.x p3 garyo-home subst keyword 
20:20:07  <[GregNoel](GregNoel)>     done 
20:20:11  <garyo-home>   good 
20:20:17  <[GregNoel](GregNoel)>     now the last one... 
20:20:33  <stevenknight> 1126:  agree w/Greg's plan 
20:20:48  <stevenknight> merge w/2079 
20:20:48  <[GregNoel](GregNoel)>     done, and that's all 
20:20:57  <stevenknight> okay, i have to run 
20:21:02  <garyo-home>   good job guys! 
20:21:09  <bdbaddog>     :) 
20:21:17  <stevenknight> great that we're now well down below 50 to classify... 
20:21:19  <[GregNoel](GregNoel)>     Great job; we only have 2004 left now 
20:21:22  <garyo-home>   ok, I'll run my tests & get you results tomorrow. 
20:21:27  <bdbaddog>     ditto 
20:21:33  <stevenknight> thanks 
20:21:33  <bdbaddog>     running now. 
20:21:35  <[GregNoel](GregNoel)>     G'night all 
20:21:39  <garyo-home>   night. 
20:21:39  <bdbaddog>     night 
20:21:46  *      garyo-home has quit ("[ChatZilla](ChatZilla) 0.9.84 [Firefox 3.0.6/2009011913]") 
20:21:48  *      [GregNoel](GregNoel) has been marked as being away 
20:21:49  <stevenknight> [GregNoel](GregNoel):  are you planning to add 2004 spreadsheets? 
20:22:10  <[GregNoel](GregNoel)>     Already have; info just not pushed over to scons.org yet. 
20:22:22  <stevenknight> okay, i'll look for those 
20:22:30  <stevenknight> thanks 
20:22:34  <stevenknight> i have to run myself... 
20:22:47  *      stevenknight has quit ("Leaving") 
20:32:55  *      bdbaddog (n=[bdeegan@adsl-71-142-75-191.dsl.pltn13.pacbell.net](mailto:bdeegan@adsl-71-142-75-191.dsl.pltn13.pacbell.net)) has left #scons 

Updated