Experimental Physics and
Industrial Control System

Kay-Uwe Kasemir <[email protected]> · Fri, 17 Oct 2003 13:30:34 -0600

On 2003.10.17 08:17, Andy Foster wrote:

So, I'm assuming that, at the moment we cannot build the
ChannelArchiver against 3.14.4.
Is this correct?

Hi:
Sorta.
R3.14 is fundamentally different in every aspect that
pertains to the archiver:
Channel Access, Time Stamps, OSI support for threads.
(actually quite interesting:
The basic C tsStamp still works, but the C++ osiTime
class that was supposed to shield us from all the changes
in the intrinsics changed into a just-different-enough
epicsTime so that almost every time-stamp related line of
code in the archiver needed to be rewritten)
I did spend a little time to make it _compile_ again, but:
The result isn't fully tested, and then the different
releases of R3.14.x are again different enough that
so far I can barely keep up with getting them set up,
let alone have time to really check the archiver against them.
You can get a snapshot of this misery at
http://ics-web1.sns.ornl.gov/~kasemir/archiver/index.html
Hope is in sight, we are actually working on a better "index"
to replace the current "directory" files, one that will
also serve as a "master" index for multiple sub-archives.
The x in R3.14.x is also changing more slowly now, so
I hope we will get a better archiver which compiles under the
then current R3.14 within the next 6 month.
-Kay

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System