Version 6.02 is primarily a bug-fix release.
yes? SET AXIS/STRIDE=[/OFFSET=] axisname yes? CANCEL AXIS/STRIDE axisname
For example,
yes? SET DATA "http://iridl.ldeo.columbia.edu/SOURCES/.NOAA/.NGDC/.ETOPO5/dods"
yes? SHOW DATA
currently SET data sets:
1> http://iridl.ldeo.columbia.edu/SOURCES/.NOAA/.NGDC/.ETOPO5/dods (default)
name title I J K L
ELEV elevation/bathymetry 1:4320 1:2160 ... ...
yes? SET AXIS/STRIDE=5 `elev,RETURN=xaxis`
yes? sET AXIS/STRIDE=5 `elev,RETURN=yaxis`
yes? SHOW GRID elev
GRID GCR1
name axis # pts start end
(AX004) LONGITUDE 864mr 0E 0.41699W(359.58)
(AX005) LATITUDE 432 i 89.916S 89.666N
normal Z
normal T
This is equivalent to the older (but still valid) striding syntax, which required us to define a new variable
yes? SET DATA "http://iridl.ldeo.columbia.edu/SOURCES/.NOAA/.NGDC/.ETOPO5/dods" yes? LET strided_elev = elev[i=1:4320:5,j=1:2160:5]Now the data can be used as if it is on the smaller grid. If we want to resume use of the original grid,
yes? CANCEL AXIS/STRIDE `elev,RETURN=xaxis` yes? CANCEL AXIS/STRIDE `elev,RETURN=yaxis`
The offset if specified must be smaller than the stride. If a stride is not specified this qualifier is ignored. The OFFSET value refers to the first point to keep, i.e. these two command would define the same striding in X:
yes? SET AXIS/STRIDE=5/OFFSET=2 `elev,RETURN=xaxis` yes? LET strided_elev = elev[i=2:4320:5]For example,
yes? USE coads_climatology
yes? LIST sst[X=61:79,J=50,K=1,L=50]
VARIABLE : SEA SURFACE TEMPERATURE (Deg C)
FILENAME : coads_climatology.cdf
SUBSET : 10 points (LONGITUDE)
LATITUDE : 9N
TIME : 15-FEB-0012 14:19
9N
50
61E / 21: 26.86
63E / 22: 27.00
65E / 23: 27.29
67E / 24: 27.69
69E / 25: 27.93
71E / 26: 28.08
73E / 27: 28.23
75E / 28: 28.32
77E / 29: 28.23
79E / 30: 27.82
yes? SET AXIS/STRIDE=4/OFFSET=2 `sst, return=xaxis`
!-> SET AXIS/STRIDE=4/OFFSET=2 COADSX
yes? LIST sst[X=61:79,J=50,K=1,L=50]
VARIABLE : SEA SURFACE TEMPERATURE (Deg C)
FILENAME : coads_climatology.cdf
SUBSET : 3 points (LONGITUDE)
LATITUDE : 9N
TIME : 15-FEB-0012 14:19
9N
50
63E / 6: 27.00
71E / 7: 28.08
79E / 8: 27.82
This acts on the axis, so variables that share an axis, whether they come from a single dataset or several datasets that use the same exact axis, will all be affected by the strides.
Start Ferret with
> ferret -batch myfile.plt
When in this batch mode, xwindows are not used, and a new myfile.plt.~*~ file is started with each new plot "page" that we'd see when running interactively.
The following commands are ignored
New plots may be started with these commands
Note that there is still a choice of output graphics type when starting Ferret with -batch. Only one type may be chosen.
> ferret -batch metafiles.plt ! metafile > ferret -batch psfiles.ps ! postscript > ferret -gif ! gif outputIn -gif mode, recall that we must use use FRAME/FILE= commands to create each gif image files.
yes? USE coads_climatology
yes? SAY `sst,RETURN=xmod`
!-> MESSAGE/CONTINUE 360
360
yes? SAY `sst,RETURN=tmod`
!-> MESSAGE/CONTINUE 8765.8
8765.8
yes? SAY `sst,RETURN=tunits`
!-> MESSAGE/CONTINUE hour
hour
When initializing a dataset, if Ferret encounters an axis name that's already in use, it creates a new name by adding a digit, e.g. if TEMP is already in the list of axis names, then it uses TEMP1 for the new data for uniqueness. When this happens, we continue to new name with the dataset attribute information, and add an attribute orig_file_axname containing the original name. That original name is used on output to a netCDF file.
yes? DEFINE AXIS/T=1:5:1 time ! create an axis called TIME
yes? USE coads_climatology ! there is also a TIME axis in this dataset
yes? show grid sst ! The time axis is called TIME1
GRID GSQ1
name axis # pts start end
COADSX LONGITUDE 180mr 21E 19E(379)
COADSY LATITUDE 90 r 89S 89N
normal Z
TIME1 TIME 3mr 16-JAN 06:00 17-MAR 02:58
yes? ! Ferret automatically defines an attribute orig_file_axname
yes? SHOW ATTRIBUTE (TIME1).orig_file_axname
attributes for dataset: ./coads_climatology.cdf
(TIME1).orig_file_axname = TIME
yes? SAVE/FILE=a.nc/X=181/Y=1 sst ! This file has the time axis named TIME
When data contains missing values, LIST/FORMAT=comma or LIST/FORMAT=tab now writes a comma or tab as placeholder for the missing value. Previously these listings were ambiguous.
yes? LET a = {1,1,,1,1,,1,1}
yes? LET b = {4,4,4,4,4,4,4,4}
yes? LIST/FORM=comma a,b
X: 0.5 to 8.5
Column 1: A is {1,1,,1,1,,1,1}
Column 2: B is {4,4,4,4,4,4,4,4}
A,B
1 / 1:1,4
2 / 2:1,4
3 / 3:,4
4 / 4:1,4
5 / 5:1,4
6 / 6:,4
7 / 7:1,4
8 / 8:1,4
yes? USE mydat.nc
yes? LIST/Y=1/K=1/L=1 temp !See the title and units
VARIABLE : TEMPERATURE (deg. C)
FILENAME : mydat.cdf
SUBSET : 3 points (LONGITUDE)
LATITUDE : 0.8N
DEPTH (m): 5
TIME : 15-MAR-1983 10:00
0.83N
48
160.5W / 70: 31.20
159.5W / 71: 31.34
158.5W / 72: 31.45
yes? ! Make changes to variable with SET VARIABLE
yes? SET VAR/TITLE="Temp"/units="DegC" temp
yes? SAVE/FILE=a.nc/CLOBBER temp
yes? CANCEL DATA/ALL
yes? USE a.nc
yes? ! The file has the NEW var title and units
yes? lIST/Y=1/K=1/L=1 temp
VARIABLE : Temp (DegC)
FILENAME : a.nc
SUBSET : 3 points (LONGITUDE)
...
yes? USE mydat.nc ! a dataset whose time axis needs changes
yes? ! Use DEFINE AXIS to redefine it.
yes? SET AXIS/T0="1-jan-2004"/UNITS=days `temp,RETURN=taxis`
yes? ! We can make further chages with SET AXIS
yes? SET AXIS/MODULO/CALENDAR=noleap `temp,RETURN=taxis`
yes? SHOW ATT/ALL `temp,RETURN=taxis`
!-> SHOW ATT/ALL TIME
attributes for dataset: ./mydat.nc
(TIME).units = DAYS
(TIME).time_origin = 1-JAN-2004
(TIME).orig_file_axname = TIME
(TIME).calendar = NOLEAP
(TIME).modulo =
yes? ! CANCEL AXIS may also be used to change axis features
yes? CANCEL AXIS/MODULO `temp,RETURN=taxis`
yes? SHOW DATA/ATT ! see dataset with all the attributes
currently SET data sets:
1> ./mydat.nc (default)
VARIABLE ATTRIBUTE NAME TYPE SIZE OUTFLAG VALUE
--------------------------------------------------------------------------------
...
(TIME) units CHAR 4 T DAYS
time_origin CHAR 10 T 1-JAN-2004
orig_file_axname CHAR 4 F TIME
calendar CHAR 6 T NOLEAP
The error message was
" **TMAP ERR: attempt to redefine line
axis TIME doesnt match CDF file
This was a bug in matching the axis names, in the code that allows for CANCEL MODE UPCASE_OUTPUT. This is now fixed.
yes? LET a = {"a", "b", }
yes? LIST a
VARIABLE : {"a", "b", }
SUBSET : 3 points (X)
1 / 1:"a"
2 / 2:"b"
3 / 3:""