[logon] updates to PET and [incr tsdb()] in the LOGON tree

[Stephan Oepen]

colleagues,

i did something moderately brave the past few days: revise the API that
[incr tsdb()] uses in communication with clients like PET.  the changes
are not dramatic (an additional argument to process_item()), and it may
even be possible for old client binaries to work with new [incr tsdb()]
code (but not vice versa).  nevertheless, timing of updates may be more
important for some of you than usually.

the LOGON tree should be internally consistent, so running `svn update'
should result in a functional environment (new [incr tsdb()] code; new
PET binaries, both from the `trunk' and the `cm' branch).  the changes
to PET sources are checked into the DFKI SVN (both branches).  but the
[incr tsdb()] API revisions are not yet checked into the LinGO CVS.  i
can do that in the next few days, but i thought we could test a little
within the LOGON tree first.  as a consequence of this asymmetry, one
should either not update anything for a few days, or avoid `mixing and
matching', i.e. make sure to run newer PET binaries only with the LOGON
version of [incr tsdb()].  and those compiling their own PET will need
to re-build those binaries, making sure to use [incr tsdb()] libraries
from the LOGON tree.

as i write this message, i realize that (i hope) very few of you would
have noticed these changes.  anyway, now everyone knows :-).

                                                      all best  -  oe

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
+++ Universitetet i Oslo (IFI); Boks 1080 Blindern; 0316 Oslo; (+47) 2284 0125
+++     CSLI Stanford; Ventura Hall; Stanford, CA 94305; (+1 650) 723 0515
+++       --- oe at ifi.uio.no; oe at csli.stanford.edu; stephan at oepen.net ---
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++