3.3.1. process.defaults
Program flow within sh_gamit and sh_glred is controlled by process.defaults
, shown below with annotations added for each section:
## LOCAL DIRECTORIES
# Directory for translation of raw data
set rawpth = "$procdir/raw"
# Directory path for raw archives (search all levels); e.g. /data18/simon
set rawfnd = ""
# Input files for RINEX translators
set mpth = "$procdir/mkrinex"
# RINEX files directory
set rpth = "$procdir/rinex"
# Directory path for RINEX archives (search all levels); e.g. /data18/simon
set rnxfnd = ""
# Broadcast orbit directory
set bpth = "$procdir/brdc"
# IGS files directory
set ipth = "$procdir/igs"
# G-files directory
set gpth = "$procdir/gfiles"
# GAMIT and GLOBK tables directory
set tpth = "$procdir/tables"
# Output directory for skyplots
set gifpth = "$procdir/figs"
# Globk solution directory
set glbpth = "$procdir/gsoln"
# Globk binary h-file directory
set glfpth = "$procdir/glbf"
# Directory path for other h-files (LA, LB, LC options; search all levels)
# e.g. "/raid1/tah/SIO_GLL"; ( /raid6/ftp/pub/MIT_GLL/H07 /raid2/simon/gps_analysis/cgps_hfiles )
set hfnd = ""
# Template files
set templatepth = "$procdir/templates"
# Place to store temporary control files
set cpth = "$procdir/control"
# Archive root directory (cannot be null)
set archivepth = "$procdir/archive"
This first section simply names the directories to be used for the processing.
Most of the directories will be created with these names whether or not you include the entry and whether or not you actually use the directory.
The only entry you are likely to change is glbpth
, for which you may want to assign successively different directory names (e.g., gsoln/
, hsoln/
, tsoln/
) if you perform parallel sh_glred solutions using, for example, different combinations of data.
##GAMIT
# Set sampling interval, number of epochs, and start time for processing
set sint = '30'
set nepc = '2880'
set stime = '0 0'
# Variables for updating tables
set stinf_unique = "-u"
set stinf_nosort = "-nosort"
set stinf_slthgt = "2.00"
# Set "Y" to use RINEX header coordinates not in lfile or apr file
set use_rxc = "N"
# Broadcast orbits
set brdc = 'brdc'
# Minimum x-file size to be processed (Def. 300 blocks; most OS use 1 Kb blocks)
set minxf = '300'
# Set search window for RINEX files which might contain data for day - default check the previous day
set rx_doy_plus = 0
set rx_doy_minus = 1
# Default globk .apr file
set aprf = igb14_comb.apr
# Set compress (copts), delete (dopts) and archive (aopts) options. (Don't forget to set the archivepth.)
# Possible d-, c-, and a- opts: D, H, ao, ac, as, b, c, d, e, g, h, i, j, k, l, m, o, p, q, t, x, ps, all"
set dopts = ( c )
set copts = ( x k ao )
set aopts = ''
The session variables are set here by default but can be overridden with the -sessinfo
option on the sh_gamit command line.
They specify, respectively, the sampling interval, number of epochs (2880 is 24 hours at 30 s intervals), and the start time (hr min).
The next four entries allow you to control how station.info
and the L-file are updated.
They work with sites.defaults
entries and are explained in the next section.
The rinex_ftpsites
option specifies what remote archives are to be searched for RINEX files;
you may add archives of your choosing by editing ~/gg/tables/ftp_info
.
The minimum x-file size option allows you to exclude from GAMIT processing data from sessions too short to be useful for your application.
The 300-block (300 Kb on most machines) default sets the limit at about 3 hours of tracking;
change this to a small number to process short sessions.
The search window options instruct sh_gamit to look for RINEX files named with day-of-year different from the one being processed.
This assures that you will not miss local data within a file named for a previous day (rx_doy_minus
), and for midnight-crossing sessions, that you will download and include IGS data from the part of the session on the following day (rx_doy_plus
).
The rx_doy_minus
option may need to be increased if you have RINEX files covering many days (the GAMIT maximum is 7).
Changing these to 0 will save a little time if you know that all of your data are within the expected 0–24h span.
Both of these options can be specified in the sh_gamit command line if you want to change them on a day-to-day basis.
## RESOURCES
# Minimum raw disk space in Mbytes
set minraw = '100'
# Minimum rinex disk space in Mbytes
set minrinex = '100'
# Minimum archive disk space in Mbytes
set minarchive = '100'
# Minimum working disk space in Mbytes
set minwork = '500'
The resource settings prevent processing from starting if there is inadequate disk space to complete it.
## SYSTEM-DEPENDENT SETTINGS
# UNIX df command must be set to return the correct form
# Most machines (
set udf = 'df -mk'
# but note that if you have free > 1 Tb, you will need to change this to Mb
# set udf = 'df -m'
# HP
# set udf = 'bdf'
# UNIX mail command
# Most machines
set umail = 'mail -s'
# HP
# set umail = 'mailx -s'
# Mail address for sending the processing report (if '' will default to `whoami` in sh_gamit)
set mailto = ''
# Host name for email and anonymous ftp password use (if '' will default to `hostname` in sh_gamit)
set machine = ''
# Ghostscript path
set gspath = '/usr/bin'
# ImageMagick path fir png conversion
# set impath = '/usr/bin/X11'
set impath = '/usr/bin'
The only system-dependent setting likely to need changing is the path for the convert program, used to convert the sky plots from PostScript to PNG.
If your UNIX installation does not have convert available, use -nopngs
in the sh_gamit command line so that the sky plots will be left as PostScript in the day directory.