My schedule, 2.3 schedules, and whatnot
rra at stanford.edu
Tue Dec 7 00:55:24 UTC 1999
Quick note to let everyone know that I got tons of work done on INN last
month largely due to the fact that I was on vacation most of last month,
and I'm definitely not going to be on vacation this month (the to-do list
for this month is intimidating to say the least). Since work is picking
up heavily here, don't expect to see too much INN-related activity from me
for the next month or so.
My expectation is that 2.3 should be mostly ready to go by early next
year, hopefully by the end of January although that may be somewhat
aggressive. I think all of the things I had mentally tagged as "must
have" are now in the tree.
On the list of things I've got floating around to work on are:
- Overhauling inn.conf parsing to make it easier to compile getconfig.c,
to clean up the maze of different places that have to be updated for
new options, to make sure that obsolete options are out of the tree
everywhere, and so forth. I need to write this up as a separate
message to cover all the details of what I'm thinking of, and may not
get to that for a bit.
- More configuration and Makefile cleanups. INN is currently linking
everything with libraries only used by one or two components; that
should be fixed. It should also use standard names between configure
and Makefile.global, the additional variables used by the Perl scripts
that are still configure targets should be moved into innshellvars,
and fixscript should make it possible for the signcontrol and pgpverify
in the source tree to be exact copies of the separately distributed
versions without even any @VAR@ magic.
- Remove support for Subject: cmsg and Also-Control.
- The [USEFOR] authenticated Path syntax, as well as whatever else they
come up with as a replacement for NNTP-Posting-Host/X-Trace.
- The history backend rewrite, which is definitely a 2.4 item.
- Additional work on slowly moving towards code that we can multithread.
The other pending things that I know are out there and are in various
stages of completion and that I'd like to see go into INN are:
- The controlchan rewrite, which includes better header parsing code that
the rest of INN should be switched over to use rather than the separate
implementations currently scattered all over the place. This will also
let us rip out the rest of the old control processing code.
- The universal configuration file parser.
Russ Allbery (rra at stanford.edu) <URL:http://www.eyrie.org/~eagle/>
More information about the inn-workers