Re: How to properly modify the code in pq.c for smsd

[Jan Derfinak]

On Mon, 7 Aug 2006, "Zdeněk Habala" wrote:

> Dear all, but I think, it is mainly for Jan Derfinak, I have troubles with
> compilation of the smsd daemon. I wanted to change the behaviour of the
> smsd daemon (using postgresql) when it is looking for new sms-es to be
> send. Originaly it looks for all not processed sms-es. And it could be a
> great number like 1000 or more. Because I don't want to process all of
> them in one transaction. So I thought that when I change the SQL statement
> in pq.c to something like SELECT * FROM OUTBOX WHERE NOT processed LIMIT
> 10, and it will do the trick. But I don't know, how to compile the smsd
> with the pq.c. I recompiled the smsd, but it is not using the pq.c. I
> think that there is something with the libpq.so, but i have no idea, how
> properly I should do all the compiling.

Smsd dynamically loads DB modules on runtime. You don't need to recompile
smsd when you change DB module. Only specify path and name of the module to
smsd.

Compilation is desribed in README file:
gettextize                # only for CVS copies
autoconf                  # only for CVS copies
autoheader                # only for CVS copies
./configure
[ If you use latest Red Hat with new (0.11) gettext version ]
[ run autogen.sh with configure options instead of the      ]
[ above commands. It will do all needed things.             ]
cd smsd
vi Makefile (edit paths in "DB Modules" section)
make
make libpq.la OR make libmysql.la OR make libfile.la
make install

Please send you patch here, when it will be done. I think it is useful.