Issues 2010-02-02
 Share
The version of the browser you are using is no longer supported. Please upgrade to a supported browser.Dismiss

 
View only
 
 
Still loading...
ABCDEFG
1
IDVotesType/MemberMilestonePriOwnerSummary/Comments
2
3
Steven's former research issues need to be parceled out (left over from last time)
4
5
19100DEFECT-research-P1issues@sconsRecursive flag ignored in dictionary based scanners
6
Steven to add test case and partial fix then assign to Gary 2.x p4
7
8
Bill
9
Brandon
10
David
11
Gary2.xp3stevenThe doc isn't clear on the intended behavior here. I'd hesitate to say it's definitely a bug, but I'm not sure how he can get the behavior he wants so I guess it should be classified as a bug. I say assign it to Steven for 2.x; he can say whether it's working as intended and if not, sketch out the fix and/or write a test with the correct behavior and anyone (e.g. me) could implement it.
12
Greg
13
Jim
14
Ken
15
Sohail
16
Steven2.xp3?i thought i had a fix, but it broke some other tests (yay for regression tests) and needs work. i'll upload a patch to give someone a headstart
17
18
23610DEFECT-research-P2issues@sconsclass Tool: self.options is not a callable entity
19
consensus: anytime p4 Steven
20
21
Bill
22
Brandon
23
David
24
Garyanytimep4stevenbug says Steven to doc and mark WONTFIX.
25
Greganytimep4stevenbug says Steven to doc and mark WONTFIX.
26
Jim
27
Ken
28
Sohail
29
Stevenanytimep4skagree
30
31
7800DEFECT-research-P3issues@sconsNon-Builder BUILDERS value doesn't generate error
32
Steven to add test case and partial fix then assign to Gary 2.x p4
33
34
Bill
35
Brandon
36
David
37
Garyanytimep4 +EasyWhether doc or code, it's not large.
38
Gregdunno +EasyWhether doc or code, it's not large.
39
Jim
40
Ken
41
Sohail
42
Steven2.xp4?I had a partial fix which I just dusted off and will upload as a patch. it's not as easy as it seems at first glance, we actually have some unit tests that rely on setting the ".builder" attribute to an integer as a cheap way to unit test behavior of various methods without creating a real builder object... :-(
43
44
11870PATCH-research-P3issues@sconsAllow spaces in executable name when temp file is used
45
consensus: 2.x p3 +subst
46
47
Bill
48
Brandon
49
David
50
Gary2.xp3 +substright, it's not +VS. I can see how the original patch was not all that great. I think this is really the same as subst since it's about quoting what gets passed into or through the tempfile.
51
Greg +VSOr maybe a new keyword +subprocess? If we pass a list of strings to subprocess, it will quote them correctly if it has to run the command via a shell.
52
Jim
53
Ken
54
Sohail
55
Steven2.xp3 +substGary's in the ballpark, although it's kind of in the interaction between subst() and command execution, not purely a subst() thing itself
56
57
17450DEFECT-research-P3issues@sconsscons --clean' doesn't remove *.ilk files
58
consensus: 2.1 p4 Gary +VS
59
60
Bill
61
Brandon
62
David
63
Garyp4 +VScould just add .ilk as side effect to MSVS link.
64
Greg +VS
65
Jim
66
Ken
67
Sohail
68
Steven +VS
69
70
18830PATCH-research-P3issues@scons[win] Add SCons to App Paths
71
consensus: dup of new ticket (2.1 p2 Lukas w/ Gary as QA) to integrate new DOS installer
72
73
Bill
74
Brandon
75
David
76
Gary2.1p2garyoThe "new" Windows installer will do this, iirc. Just need to get it integrated.
77
Greg
78
Jim
79
Ken
80
Sohail
81
Steven-research-p2?garyo
82
83
19450DEFECT-research-P3issues@sconsscons should not cache implicit dependnecy found from a directory scanner
84
consensus: 2.x p2 Ludwig w/ Steven as QA
85
86
BillLudwig?I like the option to flag a Node as non-implicit-cacheable and then just automatically turn it on for directory nodes. Maybe Ludwig could lead a thread on the mailing list?
87
Brandon
88
David
89
Gary2.xp2Steven -> LudwigLudwig already proposed 4 solutions. Steven needs to look at them and pick one. (I personally like #1 and #3.) I think this should be p2 since it happens to me and my teammates regularly, and they're getting into the habit of blowing away the .sconsign and starting again which I don't like.
90
GregLudwig?I like the option to flag a Node as non-implicit-cacheable and then just automatically turn it on for directory nodes. Maybe Ludwig could lead a thread on the mailing list?
91
Jim
92
Ken
93
Sohail
94
Steven2.xp2Ludwig, or SKAgree w/Gary re: #1 and #3 being the most attractive. I think there's a way to get rid of --implicit-cache entirely, so I don't think it makes sense to put much time into making it work for this case.
95
96
20960DEFECT-research-P4issues@sconsVariantDir and Configure do not play well ?
97
consensus: 2.x p3 +sconf_revamp
98
99
Bill +sconf_revamp
100
Brandon
Loading...
 
 
 
Sheet 1