20020910: One last question.

Stonie,

GPWARN is capable of using file aliases. I have the following defined
in the $GEMTBL/config/datatype.tbl file:

! Text files for warnings
TORNADOWARN  $TEXT_WARN/torn_warn      YYYYMMDDHH.torn           CAT_NIL  etc.
TSTORMWARN   $TEXT_WARN/tstrm_warn     YYYYMMDDHH.tstrm          CAT_NIL  etc.
FLOODWARN    $TEXT_DATA/fflood/warn    YYYYMMDDHH.warn           CAT_NIL  etc.
WINTERWARN   $TEXT_WARN/winter         YYYYMMDDHH.winter         CAT_NIL  etc.
NONPCPWARN   $TEXT_WARN/noprcp         YYYYMMDDHH.noprcp         CAT_NIL  etc.


For my script which I use to plot the warnings, I have
WWFIL = nonpcpwarn;winterwarn;floodwarn;tstormwarn;tornadowarn

(you can define your own aliases...they aren't hardcoded into the program)

The program will go back through the files matching the aliases (all files
since things like Winter Storm Warnings get issued 24 to 72 hours in advance).


The colors and line types are currently hard coded into the routine, but I
will take it as a suggestion to make a $GEMTBL/unidata/gpwarn.tbl configuration
file.

Steve Chiswell
Unidata User Support










On Wed, 11 Sep 2002, Stonie R. Cooper wrote:

> Steve,
>
> You nailed it!
>
> Is there a way to get gpwarn to search through the $TEXT_DATA/nwx/watch_warn
> tree for all current data, or will it always have to have the files cat'ed to
> a large file (as mentioned in the phelp for gpwarn)?
>
> Also, is the color table compiled in, or a config file (for the fill and
> outline colors)?
>
> Thanks!
>
> Stonie
>
> On Wednesday 11 September 2002 18:25, Steve Chiswell wrote:
> > Stonie,
> > The problem you are having with gpwarn is apparently you had
> > set WWS=all. In reality WWS is either "current" or a gempak
> > format dattim. The segment fault was from trying to convert
> > the "all" string to a number.
> >
> > I'll try to make that more robust within the expected values.
> >
> > Steve
> --
> Stonie R. Cooper
> Planetary Data, Incorporated
> ph. (402) 782-6611
>