Build a file where messages are grouped by similarity

[tecepe]

Taking one of the suggestions from @flatcap one step further, we could have a file where we group similar messages together. This would make easier to see what is the difference between them and if those differences are needed, or if there is an easy way to reuse the message. With that in mind, we could come up with "templates for messages".

[flatcap]

Here's one bunch you mentioned:

"%sgroup: warning: bad IDN '%s'."
"Bad IDN %s while preparing resent-from."
"Bad IDN in %s: '%s'"
"Bad IDN in \"%s\": '%s'"
"Bad IDN \"%s\"."
"Bad IDN: '%s'"
"Error: '%s' is a bad IDN."
"Warning: '%s' is a bad IDN."
"Warning: Bad IDN '%s' in alias '%s'."

If anyone can think of a way of automating finding similar messsages, I'd love to hear it.

[mgedmin]

No clues on automation, but I noticed a couple of messages that differ only in the type of the format parameters: %zu vs %d.

#: conn/ssl.c:999
#, c-format
msgid "SSL Certificate check (certificate %zu of %zu in chain)"

#: conn/ssl_gnutls.c:762
#, c-format
msgid "SSL Certificate check (certificate %d of %d in chain)"