The parsing correctly handles SMTPUTF8 Unicode in the string.
-.vitem &*${mask:*&<&'IP&~address'&>&*/*&<&'bit&~count'&>&*}*&
+.vitem &*${mask:*&<&'IP&~address'&>&*/*&<&'bit&~count'&>&*}*& &&&
+ &*${mask_n:*&<&'IP&~address'&>&*/*&<&'bit&~count'&>&*}*&
.cindex "masked IP address"
.cindex "IP address" "masking"
.cindex "CIDR notation"
.code
${mask:10.111.131.206/28}
.endd
-returns the string &"10.111.131.192/28"&. Since this operation is expected to
-be mostly used for looking up masked addresses in files, the result for an IPv6
+returns the string &"10.111.131.192/28"&.
+
+Since this operation is expected to
+be mostly used for looking up masked addresses in files, the
+.new
+normal
+.wen
+result for an IPv6
address uses dots to separate components instead of colons, because colon
terminates a key string in lsearch files. So, for example,
.code
.code
3ffe.ffff.836f.0a00.000a.0800.2000.0000/99
.endd
+.new
+If the optional form &*mask_n*& is used, IPv6 address result are instead
+returned in normailsed form, using colons and with zero-compression.
+.wen
Letters in IPv6 addresses are always output in lower case.
Creates and updates are marked with the current time.
Finally, a &"before"& test which succeeds, and for which the record
-is old enough, will be refreshed with a timstamp of the test time.
+is old enough, will be refreshed with a timestamp of the test time.
This can prevent tidying of the database from removing the entry.
The interval for this is, by default, 10 days.
An explicit interval can be set using a