[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Nmh-workers] [nmh-1.2] a serious bug in the nmh configure script

From: pmaydell
Subject: Re: [Nmh-workers] [nmh-1.2] a serious bug in the nmh configure script
Date: Fri, 23 Dec 2005 23:19:48 +0000

Joel Reicher wrote:
>The more I look at what's on NetBSD the more I think this is Berkeley DB
>version 1, it's just that the headers aren't in a db1 subdir. I think
>the check should be for gdbm/ndbm.h first, then ndbm.h, then db.h, then

OK. I now have some autoconf changes I'm testing where we test
a combination of header and library (by checking that
#define DB_DBM_HSEARCH 1
#include header
int main(void) { dbm_open("",0,0); }
compiles and links).

So what order do we want to test in? (I've added a couple of
configure options so you can now override with a header/lib if
we get it wrong.)

>From the webpage quoted earlier I suggest:

db.h -ldb  # Berkeley DB2
ndbm.h -ldb # Berkeley DB1
ndbm.h -ldb1 # Berkeley DB1
ndbm.h -lndbm # classic NDBM
db1/ndbm.h -ldb1 # Berkeley DB1 as shipped on glibc2.1 systems
gdbm/ndbm.h -lgdbm # GNU gdbm as shipped on glibc2.1 systems
gdbm/ndbm.h -lgdbm_compat -lgdbm # newer GNU gdbm
ndbm.h -lgdbm # GNU gdbm
ndbm.h -lgdbm_compat -lgdbm # newer GNU gdbm

based on my personal prejudices and assuming this order doesn't
mess up the BSDs. The current configure script tries -ldbm but
I have no idea what that's for.

-- PMM

reply via email to

[Prev in Thread] Current Thread [Next in Thread]