ansi VOL1
ansi HDR1
ansi HDR2
EOF
**** data set 1
EOF
ansi EOF1
ansi EOF2
EOF
ansi HDR1
ansi HDR2
EOF
**** data set 2
EOF
ansi EOF1
ansi EOF2
EOF
ansi HDR1
ansi HDR2
EOF
**** data set 3
EOF
ansi EOF1
ansi EOF2
EOF
ansi HDR1
ansi HDR2
EOF
**** data set 4
EOF
ansi EOF1
ansi EOF2
EOF
etc
Thus to get to data set # 3 it's best to skip to file
# 6 (trew /dev/nrst8 6 would do that) and then sort as
dmp_msu 100000000 3 2D 1- since that would allow you to see the two ansi header files, sort all the data in the 3'rd data set and see the two ansi trailor files! The dmp_msu utility does not try to be clever about counting files and data sets, that's up to the user.
The rule is: to get to the n'th data set, skip
(n'th-1)*3
files using the trew utility or mt utility. The first and last
block of the 'data set' are MSU header and trailor blocks with
some experiment information. These blocks are recognized and printed