|From:||Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>|
|To:||Peter Eisentraut <peter_e(at)gmx(dot)net>|
|Cc:||pgsql-hackers(at)postgresql(dot)org, Alvaro Herrera <alvherre(at)commandprompt(dot)com>, Bruce Momjian <bruce(at)momjian(dot)us>|
|Subject:||Re: Shipping documentation untarred|
|Views:||Raw Message | Whole Thread | Download mbox|
Peter Eisentraut <peter_e(at)gmx(dot)net> writes:
> I've been thinking that we could actually get rid of that build-in-srcdir
> behavior, which also occasionally puzzles vpath users with respect to gram.c
> and so on. The new behavior would be to build targets in the local directory.
> The only requirement would be that distribution tarballs be built in-tree, so
> that the stuff that is distprep'ed actually ends up in the tarball, but I
> think that should not be a problem.
I think one of the arguments for the current behavior is that the
derived files are platform-independent, so this way saves work if you
are building several sets of binaries from the same source tree.
However, I don't know of anyone actively making use of that behavior
--- and if they did want to, they'd likely use a tarball not a CVS
Having all the derived files in the build directory definitely seems
to me to reduce the complexity and surprise factor, so +1 for changing.
regards, tom lane
|Next Message||Magnus Hagander||2009-08-11 14:30:26||Re: [PATCH] "could not reattach to shared memory" on Windows|
|Previous Message||Kevin Grittner||2009-08-11 13:56:38||Re: "Hot standby"?|