[fetchmail-devel] Patches continued... includes call for testing

[Matthias A. <ma...@dt...>]

...it's been silent for a while, and I've been looking at some more
patches.

I have committed  2003-10-23-6.2.5-mauermann-bounce.diff  with a minor
change (i. e. I dropped the hunk that changed logging in sink.c:361
(version 3939)) because it looked sane, but I do not have a setup that
allows me to test these bounces, can someone of you check if bounces
after SMTP delivery failed still work?


I have also moved Sunil's  2003-10-11-6.2.4-shetye-limitflush.diff
to the "post 6.2.6" list as it adds a new feature.


I don't know how to proceed WRT
2003-10-29-6.2.4-norman-fix_config_reload.diff, it doesn't apply.


The patch reads:
-    /* avoid parsing the config file if all we're doing is killing a daemon */ 
-    if (!quitmode && !pid)
+    /*
+     * Avoid parsing the config file if all we're doing is killing a daemon,
+     * unless this appears to be a self-restarted instance because we detected an 
+     * "under-our-noses" configuration file change, in which case we definitely want
+     * to reload the config file.
+     */
+    if ( ( !quitmode && !pid ) || pid == getpid() )

The code now reads:

    /* avoid parsing the config file if all we're doing is killing a daemon */ 
    if (!(quitmode && argc == 2))
        implicitmode = load_params(argc, argv, optind);

I wonder if the current shape is correct, and doubt it - assuming we're
just killing a daemon based on -q and just one argument is
bold. Opinions solicited.

-- 
Matthias Andree

Encrypted mail welcome: my GnuPG key ID is 0x052E7D95 (PGP/MIME preferred)