Issues 2008-06-23
 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
21050DEFECTtriageP4stevenknightMan page description of SetOption contradicts what is described under AddOption
4
consensus
5
6
Bill
7
Brandon
8
David
9
Gary1.0.xp4Setting your own option doesn't seem so unusual to me. I define my options up front, e.g., but later load a particular build config that may include option settings. I think user-created options should be settable, but am fine w/ doing it later. Can't be hard. I'm also OK w/ doc for now.
10
Gregdoc 1.0?Why would you want to SetOption your own option? Just change the default value in your own code. Change the doc, add a test so we won't try to do this in the future, and let it go.
11
Jim
12
Ken
13
Sohail
14
Steven1.0.xp3stevenknightMaybe change the doc for 1.0? Then fix the code as a separate post-1.0 issue.
15
16
21060PATCHtriageP4stevenknight"[Errno 21] Is a directory" when using a c++ include with the same name as a subdirectory in a sub-build
17
consensus
18
19
Bill
20
Brandon
21
David
22
Gary1.0.xp3stevenknightagree w/ Steven. Greg, I haven't seen that error myself.
23
Greg1.0.xp3stevenknightI'd worry about side effects and make sure the regression tests run clean, but this has been nagging us for some time, and if this is all it took, why wasn't it fixed sooner? Gary: Boost has boost/algorithm and a header file named <algorithm> that's a base for other calculation classes; there have been several messages in the mailing list about it.
24
Jim
25
Ken
26
Sohail
27
Steven1.0.xp3stevenknightLooks pretty good, contains tests. Good to reward the conscientious behavior by getting in sooner rather than later.
28
29
21070DEFECTtriageP4stevenknightInstall() file/dir permissions are inconsistent
30
consensus
31
32
Bill
33
Brandon
34
David
35
Gary1.xp3There should be some way to do what bsd install does: set mode, owner & group. I'm not sure we can overload Install() like that now, because everyone uses it to mean copy. Doesn't Maciej's packaging stuff have a solution for this?
36
GregWe're not consistent about the use of Install() to install files v. Copy() to move them around, either, but Install() should set not only permissions but also owner and group. Getting these right are important in many environments; doing something better by default would be a blessing. (I think default should be owner bin, group bin, mode 644 or 755 as appropriate, but I'm considered strange.) Nobody's been screaming about it because unless they're forced to think about it, people are sloppy and create products that don't always work because permissions aren't what they expect.
37
Jim
38
Ken
39
Sohail
40
Steven1.xp3???No one's been screaming about this, but is that because everyone's just using AddPostAction()?
41
42
21080DEFECTtriageP4stevenknightman page has duplicate text in description of interactive mode.
43
consensus
44
45
Bill
46
Brandon
47
David
48
Gary1.0p3bdbaddogconsensus
49
Greg1.0p3bdbaddogFix it.
50
Jim
51
Ken
52
Sohail
53
Steven1.0p3bdbaddogEasy to fix. Bill can check in himself (Bill, LMK if you need commit permission?) or anyone can do it.
54
55
21090DEFECTtriageP4stevenknightKeyError: 'im_func'
56
consensus
57
58
Bill
59
Brandon
60
David
61
Garyno idea.
62
Greg1.0.xp2BenoitThis does look like the signature stuff, so Benoit should be able to figure it out.
63
Jim
64
Ken
65
Sohail
66
Steven1.0.xp2Benoit?Not sure if this is a regression or not. Benoit touched that function signature stuff last. Might just need a try:-except: block around this. I updated the issue with a request for clarification of OS and Python versions.
67
68
21100DEFECTtriageP4stevenknightenv.Alias doesnt stick
69
consensus
70
71
Bill
72
Brandon
73
David
74
Gary1.xp3Steven: common envs for all children of the Alias? Shouldn't; SharedLibrary e.g. can have its objs built by any number of envs. If it's tricky, could defer to 2.0.
75
Greg1.xp3stevenknightReally strange corner case, but should be fixed.
76
Jim
77
Ken
78
Sohail
79
Steven1.xp3stevenknightMight be tricky. The Executor class is assuming common environments, but that doesn't hold in this case.
80
81
21110DEFECTtriageP4stevenknightInstalling SCons using easy_install in a sandbox fails.
82
consensus
83
84
Bill
85
Brandon
86
David
87
Garydup of 2051The egg is not uploaded to pypi. That's just indexing what's on scons.org. easy_install is d/ling the zip, unpacking it, and running bdist_egg on it, which builds an "egg" (a zip file with particular metadata, like a Java jar file) which can be installed in the python path. The SCons script doesn't handle this case; it looks for SCons.Script & can't find it. The bug should really be titled "Make SCons work with easy_install". In fact, this is a dup of 2051, and that one has a patch.
88
Greg???Nathan?Even after reading it, I don't know what it's doing.
89
Jim
90
Ken
91
Sohail
92
Stevenp21.0.x????"easy_install," my eye. How the heck are these packages even getting to pypi.phython.org? I'm not uploading them there... I'd love for someone to really clean up our packaging and installation...
93
94
21120PATCHtriageP4stevenknightpython setup.py bdist_rpm fails due to malformed spec file
95
consensus
96
97
Bill
98
Brandon
99
David
100
Gary1.0.xp2anyoneSeems trivial, and the RPM book is very clear that Summary is one line only.
Loading...
 
 
 
Sheet 1