Folks;
Ok, I'll admit it. I can shell script, but arch_retrieve has me crossing
my eyes and praying for salvation. While I struggle to chart and understand
the complete flow of the script, I'm curious to know if anyone here has made
modifications to it (not extensions through the retrieve.local mech) they are
willing to share as learning tools. A few things that come to mind (in no
particular order) are:
1) Ignoring maxfiles; having a guy set maxfiles to 0 and then requesting a
few hundred files is probably not a good thing.
2) Disabling body commands, and limiting to a single subject command. (Ok,
this is a little silly, but might be useful for underpowered servers.)
3) Limiting ls/get on a per-directory basis. The latest/ directory in an
interactive list contains the archives...but for a digested list it contains
the parts for the _next_ issue, and I'm not sure it's a great idea to allow
subscribers to arbitrarily pull latest/digest.body. (A great reuse of an
existing directory that, in retrospect, was probably _not_ the way to go -
I'd argue the digest build should be outside the archive tree. But that's an
issue for the digest shell script instead, so nevermind.)
4) Dropping searches. Again, a little silly, but I could imagine some
situations where I'd want to restrict or remove searches.
I'm certain there are other possible modifications I haven't thought about
that someone might have found useful; I'd think almost anything would be
helpful in learning how to alter the behavior of arch_retrieve.
Charlie
