phpagiconf does not install it's DB on sqlite

The module is still non-functional, since the function: phpagiconf_add contains a bad SQL command set. This should be fixed. (I still have no working code)

The attached code fixes the phpagiconf_add. I am still not comitting, since i would like someone to review my code.

diegolocal@turing:~/Proj/freepbx/freepbx-modules/branches/2.3/phpagiconf$ svn diff
Index: functions.inc.php
===================================================================
--- functions.inc.php   (revision 3931)
+++ functions.inc.php   (working copy)
@@ -52,7 +52,20 @@
        } else {
                $asmanager = array ($amp_conf['AMPDBUSER'], $amp_conf['AMPDBPASS']);
        }
-       $results = sql("INSERT INTO phpagiconf SET `debug`=$p_debug, error_handler=$p_error_handler, err_email='$p_err_email', hostname='$p_hostname', tempdir;
+       $results = sql("INSERT INTO phpagiconf SET `debug`=$p_debug");
+       $results = sql("INSERT INTO phpagiconf SET `error_handler`=$p_error_handler");
+       $results = sql("INSERT INTO phpagiconf SET `err_email`='$p_err_email'");
+       $results = sql("INSERT INTO phpagiconf SET `hostname`='$p_hostname'");
+       $results = sql("INSERT INTO phpagiconf SET `tempdir`='$p_tempdir'");
+       $results = sql("INSERT INTO phpagiconf SET `festival_text2wave`='$p_festival_text2wave'");
+       $results = sql("INSERT INTO phpagiconf SET `asman_server`='$p_asman_server'");
+       $results = sql("INSERT INTO phpagiconf SET `asman_port`=$p_asman_port");
+       $results = sql("INSERT INTO phpagiconf SET `asman_user`='".$asmanager[0]."'");
+       $results = sql("INSERT INTO phpagiconf SET `asman_secret`='".$asmanager[1]."'");
+       $results = sql("INSERT INTO phpagiconf SET `cepstral_swift`='$p_cepstral_swift'");
+       $results = sql("INSERT INTO phpagiconf SET `cepstral_voice`='$p_cepstral_voice'");
+       $results = sql("INSERT INTO phpagiconf SET `setuid`=$p_setuid");
+       $results = sql("INSERT INTO phpagiconf SET `basedir`='$p_basedir'");
 }

 ?>

I am attaching a new patch - this one does the same work with an atomic "insert into". Both seem to work, but I like the first one better as it slides out of the code a lot less more.

Another fix for this module

(yes me again)

The new patch is the correct one, the older one just fails :(

moving to post 2.3 (unless someone wants to get these in soon)

deferring from lack of progress

r5887