#8852 - ldap sync output filter

edit.sh

Index: edit.sh
diff --git a/edit.sh b/edit.sh
new file mode 100755
index 0000000000000000000000000000000000000000..d4c1e239aa69f1e67453424557c2c8376276906b
--- /dev/null
+++ b/edit.sh
@@ -0,0 +1,35 @@
+#!/bin/sh
+
+# Filter out uninformative output lines from the ldap-sync cron job script
+
+last_action=""
+last_updated=""
+
+while read line; do
+  case "$line" in
+    *Updating*|*Creating*)

[mathias, 2018/06/11 14:10:56]

Please make the patterns a bit more specific, in order to minimize the risk of
filtering-out false-positives. Thank you! 

+      last_updated="$line"
+      ;;
+    *-\>*)
+      case "$last_action" in

[mathias, 2018/06/11 14:10:56]

It is not a good idea to base the decision whether to print `$last_updated` on a
replica of the former condition (`case "$last_action" in ...`). This is
repeating the same logic and therefore creating a 2nd point of possible failure
for the exact same condition.

In this case, for example, you could just set `last_updated=""` (an empty
string) after echoing, and `test -z "$last_updated"` to determine whether to
print the last line or not.

In general, in order to avoid conditional statements being evaluated a 2nd time
(which could have side-effects in each run), you can write the result to a
variable and just check on that one.

+        *-\>*)
+          echo "$line"
+          ;;
+        *)
+          echo "$last_updated"
+          echo "$line"
+          ;;
+      esac
+      ;;
+    *Found* | \
+    *\#* | \
+    *Synchronizing*)
+      continue
+      ;;
+    *)
+      echo "$line"
+      ;;
+  esac
+
+  last_action="$line"
+done