20020423: Unidata-Wisconsin datastream generation on gold

>From: Jerrold Robaidek <address@hidden>
>Organization: University of Wisconsin SSEC
>Keywords: 200204221248.g3MCmda12543 Unidata-Wisconsin

Jerry,

I just wanted to touch base with you regarding the UW stream being
generated on gold.  In general, things are working, but they are not
quite like they were on unidata.ssec.

Today, I found that the Mollweide products were still being injected
into the datastream in delta-encoded format.  Janine helped me get in
touch with a system admin person (didn't catch his name) who would not
give me access onto oak (the access I had to oak and gold no longer
works; the only reason I am able to do things on gold is I kept a login
window open).  Eventually, we got him to copy all of the
~ldm/mcidas/unidata directory to /tmp/unidata on gold where I found the
revised code for unires.c that was needed to inject the Mollweide stuff
in PNG-compressed format.

What I still find to be different in the current stream is the default
AREA numbers being sent in the product headers no longer conform to
what they were from unidata.ssec.  After seeing this, I remembered that
the copy of pnginject.k that was running on unidata.ssec had been
heavily modified by me to provide AREA file numbers in specific ranges
for each product.  I really would like to find that copy of pnginject.k
if at all possible.  I can't imagine that the copy in use was anywhere
but in the ~ldm tree on unidata.ssec (i.e., it would not have been put
in /home/mcidas/bin, for instance).

Also, I need to root around on the restore on oak to find out what had
been put into place to get the CIMSS products.  I checked out systems
here, but there is no copy of the code that was implemented on
unidata.ssec.

Any help you can provide would be most appreciated.  I don't want to
have to reinvent the wheel to get things working like they were...

Hopefully, we can talk about these things tomorrow (Wednesday).  By
talk, I mostly mean email.

Tom