Re: [PATCHES] Reduce dependancies of postmaster (without --as-needed)

2005-11-28 Thread Martijn van Oosterhout
On Sun, Nov 27, 2005 at 06:30:59PM -0300, Alvaro Herrera wrote: Martijn van Oosterhout wrote: Attached is a patch which applies this filtering to the backend and has the same results as linking with --as-needed. I basically took the filter list of libpq and altered it as follows:

Re: [PATCHES] Reduce dependancies of postmaster (without --as-needed)

2005-11-28 Thread Alvaro Herrera
Martijn van Oosterhout wrote: On Sun, Nov 27, 2005 at 06:30:59PM -0300, Alvaro Herrera wrote: Martijn van Oosterhout wrote: Attached is a patch which applies this filtering to the backend and has the same results as linking with --as-needed. I basically took the filter list of libpq

Re: [PATCHES] Case Conversion Fix for MB Chars

2005-11-28 Thread Volkan YAZICI
On 11/27/05, Volkan YAZICI [EMAIL PROTECTED] wrote: Tests made on an i686 with a 2.6.12.5 kernel. Here's a short list of cases I tried with both latin5 and unicode charsets: - lower/upper functions with Turkish characters. - ILIKE matches with both lower and upper case Turkish characters.

Re: [PATCHES] Reduce dependancies of postmaster (without --as-needed)

2005-11-28 Thread Tom Lane
Martijn van Oosterhout kleptog@svana.org writes: I've been thinking about -lnsl and -lresolv and it occurred to me that maybe on some platforms they are needed. However, on Linux they are 100% redundant. There -lnsl provides functions for NIS and YP and -lresolv provides functions for making

Re: [PATCHES] Install pg_regress by default

2005-11-28 Thread Alvaro Herrera
Peter Eisentraut wrote: Alvaro Herrera wrote: This patch allows pg_regress to be installed by default. This was proposed awhile back but never done, any objections if I commit this? If you do, there should also be documentation, such as a reference page, about how to use it. Applied

Re: [PATCHES] Reduce dependancies of postmaster (without --as-needed)

2005-11-28 Thread Martijn van Oosterhout
On Mon, Nov 28, 2005 at 10:18:08AM -0500, Tom Lane wrote: Pulling those out is just not a good idea; we'd never have included them in the first place if they weren't needed on some platforms. A lot of these system libraries are very hard to test for in a reasonable way. For instance, IIRC the

Re: [PATCHES] Reduce dependancies of postmaster (without --as-needed)

2005-11-28 Thread Martijn van Oosterhout
On Mon, Nov 28, 2005 at 01:58:09PM -0500, Tom Lane wrote: I have no problem with trying to make configure more selective about which libraries we need at all. That's an orthogonal problem from what the backend makefile should try to filter out, though. With respect to system libraries, I

Re: [PATCHES] Reduce dependancies of postmaster (without --as-needed)

2005-11-28 Thread Tom Lane
Martijn van Oosterhout kleptog@svana.org writes: It bothers me that we don't even *know* the dependancies or even why they're there. Those autoconf lines have been there ever since autoconf use was started. They were added with only the comment detect these libraries. If -lnsl or -lresolv is

Re: [PATCHES] Reduce dependancies of postmaster (without --as-needed)

2005-11-28 Thread Larry Rosenman
On Nov 28 2005, Martijn van Oosterhout wrote: On Mon, Nov 28, 2005 at 10:18:08AM -0500, Tom Lane wrote: Pulling those out is just not a good idea; we'd never have included them in the first place if they weren't needed on some platforms. A lot of these system libraries are very hard to test

Re: [PATCHES] Reduce dependancies of postmaster (without --as-needed)

2005-11-28 Thread Andrew Dunstan
Martijn van Oosterhout wrote: Regarding the other, this is the kind of thing the buildfarm would be good for. Would it be possible to arrange for each buildfarm machine to execute the following after successful completion and capture the output? It only needs to be run once. cd

Re: [PATCHES] Reduce dependancies of postmaster (without --as-needed)

2005-11-28 Thread Tom Lane
Larry Rosenman ler@lerctr.org writes: -lnsl is needed on SVR4 derivatives, like Solaris and UnixWare. it is the network services library. You'll needed it for ANY socket based code on these platforms. Is there any specific function symbol we can test for in that library? If it exports

Re: [PATCHES] Reduce dependancies of postmaster (without --as-needed)

2005-11-28 Thread Tom Lane
Martijn van Oosterhout kleptog@svana.org writes: Regarding the restrictions for backend libs, consider the attached patch. It just filters out anything readline related. Threading LIBS are not in the LIBS variable at all. Sorry, forgot the attachment. Applied in a modified format --- I see

Re: [PATCHES] Reduce dependancies of postmaster (without --as-needed)

2005-11-28 Thread Larry Rosenman
On Nov 28 2005, Tom Lane wrote: Larry Rosenman ler@lerctr.org writes: -lnsl is needed on SVR4 derivatives, like Solaris and UnixWare. it is the network services library. You'll needed it for ANY socket based code on these platforms. Is there any specific function symbol we can test for

Re: [PATCHES] Reduce dependancies of postmaster (without --as-needed)

2005-11-28 Thread Larry Rosenman
On Nov 28 2005, Tom Lane wrote: Larry Rosenman ler@lerctr.org writes: -lnsl is needed on SVR4 derivatives, like Solaris and UnixWare. it is the network services library. libsocket requires libnsl: [1] NEEDED /usr/lib/libnsl.so.1 Hmmm ... but given that, is it needed to

Re: [PATCHES] Reduce dependancies of postmaster (without --as-needed)

2005-11-28 Thread Martijn van Oosterhout
On Mon, Nov 28, 2005 at 05:31:55PM -0500, Tom Lane wrote: Larry Rosenman ler@lerctr.org writes: -lnsl is needed on SVR4 derivatives, like Solaris and UnixWare. it is the network services library. libsocket requires libnsl: [1] NEEDED /usr/lib/libnsl.so.1 Hmmm ... but

Re: [PATCHES] Install pg_regress by default

2005-11-28 Thread Alvaro Herrera
Peter Eisentraut wrote: Alvaro Herrera wrote: This patch allows pg_regress to be installed by default. This was proposed awhile back but never done, any objections if I commit this? If you do, there should also be documentation, such as a reference page, about how to use it. Ok,

[PATCHES] psql, tab completion additions

2005-11-28 Thread Sergey E. Koposov
Hello All, 1) I'm proposing a patch to do the DROP FUNCTION argument tab completion. Now, the arguments of the drop function can be tab completed. for example drop function strpos ( press tab drop FUNCTION strpos (text, text) or: wsdb=# drop FUNCTION length ( bit)bytea)

Re: [PATCHES] Reduce dependancies of postmaster (without --as-needed)

2005-11-28 Thread Tom Lane
Martijn van Oosterhout kleptog@svana.org writes: It bothers me that we don't even *know* the dependancies or even why they're there. Those autoconf lines have been there ever since autoconf use was started. Not all of them. I dug a bit in the cvs history, and found that -lresolv was added