I imagine most glob patterns will look like this: *.foo.com
It would be cleaner and more powerful to use the existing
urlspace.c routines in the same manner that registered procs
work.
For example, you could specify a host pattern like
foo.bighost.com. The Host: header would be walked
backwards, piece by piece: com, bighost, foo, etc. until a
match was found, just as the URL is walked to match
registered procs.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Logged In: YES
user_id=87254
I imagine most glob patterns will look like this: *.foo.com
It would be cleaner and more powerful to use the existing
urlspace.c routines in the same manner that registered procs
work.
For example, you could specify a host pattern like
foo.bighost.com. The Host: header would be walked
backwards, piece by piece: com, bighost, foo, etc. until a
match was found, just as the URL is walked to match
registered procs.