main answerscript is now bit more resistant to IM floods

diff --git a/purple/answerscripts.sh b/purple/answerscripts.sh
index bea26b3abad65449cf65d8d810f382aac58296ea..44ce2be292ec5952b1e4b072618a1edb98edd6ac 100755 (executable)
--- a/purple/answerscripts.sh
+++ b/purple/answerscripts.sh
@@ -12,7 +12,7 @@
 #              - AB?!_ scripts without numbers are NOT executed!
 #              - 00    executed immediately, zero or single line output (parallel async processing)
 #              - 01-48 executed immediately, multiline output (serial processing)
-#              - 49    delay script (adds random delay to emulate human factor)
+#              - 49    delay script (adds random delay to emulate human factor, no user scripts at this level!)
 #              - 50    executed after delay, zero or single line output (parallel async processing)
 #              - 51-79 executed after delay, multiline output (serial processing)
 #              - 80-99 reserved for future
@@ -31,7 +31,8 @@ if test -d "$dir"; then
 
                #sleep at 49 (this can be replaced by 49-delay.sh, but this should be faster)
                [ $i -eq 49 ] && {
-                       sleep $(( 2 + ($RANDOM % 8) )); #2-9 seconds of sleep
+                       find "$dir"/[5-9][0-9]-* -executable | grep . >/dev/null && #check if it's worth waiting
+                               sleep $(( 2 + ($RANDOM % 8) )); #2-9 seconds of sleep
                        continue;
                }