[btech-discussion] HUDINFO 0.7 and CVS

[Thomas W. <th...@xs...>]

I incorporated most of the feedback I got, from this list as well as private
channels, and updated the hudinfo spec and the implementation in CVS. The
new spec is uploaded to http://btech.sf.net/hudinfo.spec.txt, but here are
the major changes:

 - new 'bc' command for building contacts
 - added jump-target x/y to 'gs output
 - added heat to the weapon list
 - added optional map info items to the 't' output
 - required that all strings be non-empty

The latter change is really the biggest change, all the other changes are
100% backward compatible (for compliant implementations of the client side,
in any case.) The requirement for strings not being empty means that some of
the status/ability listings now turn up as ',-,' rather than ',,', possibly
breaking existing clients. The idea is that avoiding empty items eases
parsing and reduces the chance of errors. Do note that (currently, at least)
a string existing only of whitespace is still perfectly valid. Does anyone
expect this to be a problem ?

In addition to the above changes, the CVS implementation contains a few
bugfixes... Most notably the walkspeed and runspeed of 'sgi' were
transposed. 'oops'. If you were relying on their order in effect, rather
than standard, you should keep that in mind. I also fixed some other
glitches. The CVS implementation of hudinfo does support two of the three
map info items in 't' output, optionally, controlled by an @admin toggle.
See the CVS logs for more information.

3030MUX will probably update the binary (and thus switch to hudinfo 0.7) in
a couple of days. If it works out, I'll probably build a 1.4.4 release.

Aside from all that, I'm considering doing some serious work on the CVS HEAD
(but not the stable branch, btmux_1_4-branch) that would leave it in a very
unstable state for a while. It would also almost certainly screw up any
patches anyone has against the CVS HEAD, as the changes I have in mind touch
almost every aspect of every function in the source :) I also know Cord has
some (more) experimental code that he might want to check in.

If anyone is actually using the head (if you are using CVS and didn't check
out using '-r btmux_1_4-branch', or if, after you update now and install the
new binary, the version is reported as 1.5.0a0) let me know and I'll
consider making a seperate unstable branch. That, or you could refrain from
cvs updating for a while.

-- 
Thomas Wouters <th...@xs...>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!