[Scons-dev] [PATCH] scons soname on OpenBSD

[Stefan Sperling]

On Fri, Sep 06, 2013 at 10:04:07PM -0400, Gary Oberbrunner wrote:

> I don't have any problem with this conceptually.  The

> sys.platform.startswith() would be better as a function perhaps

> (is_openbsd() or maybe just is_bsd()).  Is this also true for freebsd for

> instance?


I don't know if this applies to FreeBSD, but I wouldn't count on it.
The various BSDs can be very different in such matters.

I do know that FreeBSD's ports system wasn't as strict about shared
library versioning years ago when I used it, but that might have
changed since. If I were you, I would assume FreeBSD is fine with
whatever scons is doing, unless someone complains or sends a patch.


> And a more flexible way of handling the multi-part version

> numbers would be welcome, perhaps as a separate patch.


I agree, but I don't think I have the time to do that. I'm not
sure what the implications are for scons code base, and I don't
really know the code base. So if you're asking me to do that,
I'm afraid I'll have to decline.

I'm just trying to provide a drive-by fix for a scons issue that I
have to deal with in order to provide an up-to-date port of Subversion
for OpenBSD (a problem I only have because serf, a dependency of
Subversion, has switched to using scons exclusively -- else, I wouldn't
be here).


> If you can submit this as a mercurial patch, ideally with a testcase (SCons

> uses TDD), we should be able to work it in.


Pardon my ignorance: I sent the output of 'hg diff'.
Is a "mercurial patch" something else?

What kind of test case do you expect? Something that checks linker
invokation command lines and fails if a soname is used on OpenBSD?

Can you point me at docs for writing test cases?