Issues 2008-10-01
 Share
The version of the browser you are using is no longer supported. Please upgrade to a supported browser.Dismiss

 
$
%
123
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
|
 
Still loading...
ABCDEFG
1
IDVotesType/MemberMilestonePriOwnerSummary/Comments
2
3
22010DEFECTtriageP4issues@sconsExecutor memoize
4
consensus
5
6
Bill
7
Brandon
8
David
9
Gary
10
GreginvalidJean sent me a private message saying that he may well have misunderstood. I replied saying that he should identify what he saw, as it may well have been something we overlooked. I see he hasn't, so get it off our plate.
11
Jim
12
Ken
13
Sohail
14
Steveninvalid
15
16
22170DEFECTtriageP4issues@sconschdir does not work with -j
17
consensusCONSENSUS: WONTFIX
18
19
Bill
20
Brandon
21
David
22
Garywontfix or futureIf someone were interested and could wrap things up with subprocess, maybe... that's why I say "... or future"
23
Gregwontfix or invalidThe processing module isn't sufficient, as it only allows very limited sharing for python tasks. On the other hand, will subprocess allow us to use chdir with real programs?
24
Jim
25
Ken
26
Sohail
27
StevenwontfixI assume he means replacing the current threaded architecture with a model where the worker slaves are in completely separate processes. If someone really wants to do the work and can show that it's a net win, I wouldn't be opposed to it, but it doesn't strike me as good use of resources.
28
29
22190FEATUREtriageP4issues@sconsAdd ability to use "$SOURCE" when specifying a target to a builder
30
consensusMOOT: Gary implemented it already
31
32
Bill
33
Brandon
34
David
35
Garyfixed for 1.1I'm surprised it doesn't work today. Yes, good idea. OK, I couldn't resist -- I implemented it. Done.
36
Greg????????Wow. A really clever idea. Could it be made to work? I have no idea of the difficulty, but if it's easy, we should do it quickly.
37
Jim
38
Ken
39
Sohail
40
Stevenfixed for 1.1garyothanks gary!
41
42
22200ENHANCEMENTtriageP4issues@sconsVariantDir Enhancement
43
consensus
44
45
Bill
46
Brandon
47
David
48
Garyworks as designedRepository maybe???
49
GreginvalidI was about to reject this out of hand, since there's an obvious path out of the thicket:

VariantDir('source', 'include')
VariantDir('generate', 'source')
VariantDir('object', 'generate')
obj = Object('generate/foo.c', CPPPATH = 'generate')
prog = Program('object/foo')

except that when I tried it to be sure, files in 'source' were deleted. Using duplicate=0 resolves the problem, but I don't like recommending a solution that causes data loss. Any idea why this would be happening? I'll add it to the man page if it can be explained.
50
Jim
51
Ken
52
Sohail
53
StevenresearchstevenknightI think we need to characterize this behavior. Once again, VariantDir() causing more trouble... I've really come around to Greg's earlier insight that it would be simpler if this were implemented with repository.
54
55
22210DEFECTtriageP4issues@sconsProg scanner should be able to find dependencies on embedded $LIBS lists
56
consensus
57
58
Bill
59
Brandon
60
David
61
GarySteven's issue; up to him. Sounds easy. 2145 looks like a similar issue to me, but not strictly a dup.
62
Gregdup 2145?It's possible to do the delayed evaluation of lists as wanted for 2145 with an external object (outside of the SConscript), but it's not obvious (and unaesthetic to boot). I'm not sure that a delayed evaluation via variables is any better, but if Steven thinks it's doable, we should schedule it.
63
Jim
64
Ken
65
Sohail
66
Steven1.2p3stevenknightMore of a nice-to-have, but it would clean up part of my config nicely. I can see how to implement it, but it changes the behavior of the Prog scanner, so it should get some shakeout during the 1.2 cycle. I don't *think* it's precisely a dup of 2145, but I think I should be able to tackle both of them at the same time.
67
68
22220PATCHtriageP4stevenknightCreated Builder for Motif Uil files
69
consensusQuestion: 1.x p4 OP. When they add suitable doc and tests, consider for next release. Who should be CC to keep an eye on the issue?
70
71
Bill
72
Brandon
73
David
74
Gary1.2?p4OP to finishAgree w/ Greg. Looks good, needs doc & testcase. Uil is their resource compiler, but I don't know anyone with it -- will have to be a blind commit, hoping the testcase passes if someone has uil.
75
Greg1.2?p4?not stevenI have no idea what a 'uil' does or why one would need it, but it's got a patch and it looks pretty simple. On the other hand, adds another Tool to the instantiation of _every_ Environment; we should be going the other way. Needs tests and documentation. If the OP can come up with those, it's a consideration for 1.2; othewise 1.x.
76
Jim
77
Ken
78
Sohail
79
Steven1.2?p4not meagree w/others
80
81
22230DEFECTtriageP4issues@sconsErroneously tries to link modules in Fortran link command.
82
consensus
83
84
Bill
85
Brandon
86
David
87
GaryresearchDavidalong w/ other Fortran stuff (he seems to understand this well, given recent ML discussions)
88
Greg2.xp2David?Maybe .mod files should be SideEffect()s? Rob seems to be having success with those in the TeX work.
89
Jim
90
Ken
91
Sohail
92
Stevenresearch or 2.xp2David
93
94
22240DEFECTtriageP4issues@sconsscons can't deal properly with VS2008 platform include path
95
consensus
96
97
Bill
98
Brandon
99
David
100
Gary1.1p2David or meMay already be fixed in vs_revamp. Will be trivial to test once I get my buildbot up (still working on vmware image for it, sorry)
Loading...
 
 
 
Sheet 1