Ferret Version v6.04/v60.5 Release Notes

Version 6.04/6.05 is primarily a bug-fix release. V6.04 was released containing a bug, the first one listed in the "Bug Fixes" section below. This was corrected and v6.05 released one day later

New Features and Enhancements in Version 6.04/6.05

• New RETURN=xmod, RETURN=tmod to get the modulo character of a grid. These return the modulo length of the x or t axis of a variable.
  
• New transformation @NBD, parallel to @NGD. Lists number of data flagged as bad.
  
• New regridding transformation @MODNBD, parallel to @MODNGD. For a modulo regridding, lists the number of input data flagged as bad in each grid box.
  
• Long OPeNDAP URL's are now allowed. They may be up to 1024 characters.
  

Bug Fixes in Version 6.04/6.05

• Irregularly-spaced coordinate axes from OPeNDAP files detected as regular.
  
• Latest OPeNDAP libraries translate error messages.
  
• Plots of curvilinear data are drawn correctly across the modulo cut.
  
• Some new settings for axes for the point_spacing attribute does not get saved.
  
• For a user variable, accessing the list of attribute names gave an error
• In some viewport plots, the zero contour was not drawn as a dark line.
  
• When a time axis is redefined, the "since t0" portion of the units was missing.
  
• SHADE/LEV failed to repeat same levels, after SHADE/LINE
  
• Null operation produced an error.
  
• Under x86_64-linux, the DATE1900 function returned 0.
  

New Features and Enhancements in Version 6.04

• New RETURN=xmod, RETURN=tmod to get the modulo character of a grid. These return the modulo length of the axis for a varable, in the units of the axis.

     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
   

  
  
  
• New transformation @NBD, parallel to @NGD lists number of data flagged as bad.

     yes? USE coads
     yes? LIST sst[X=180,Y=20,L=100:500@NGD]
               VARIABLE : SEA SURFACE TEMPERATURE (# of points)
               DATA SET : COADS 2x2 Degree Monthly Average Surface Marine Observations
               FILENAME : coads.cdf
               FILEPATH : /home/shiva/data/
               LONGITUDE: 179E
               LATITUDE : 19N
               TIME     : 01-APR-1954 00:00 to 01-SEP-1987 00:00 (number of valid)
            365.0
  
     yes? LIST sst[X=180,Y=20,L=100:500@NBD]
               VARIABLE : SEA SURFACE TEMPERATURE (# of points)
               DATA SET : COADS 2x2 Degree Monthly Average Surface Marine Observations
               FILENAME : coads.cdf
               FILEPATH : /home/shiva/data/
               LONGITUDE: 179E
               LATITUDE : 19N
               TIME     : 01-APR-1954 00:00 to 01-SEP-1987 00:00 (number flagged bad)
            36.00
  

  
  
  
• New regridding transformation @MODNBD, parallel to @MODNGD lists for a modulo regridding, number of input data flagged as bad in each grid box.

  Example: Define a variable with no missing data and another with some missing data. See what MODNGD and MODNBD are, and check that the sum is the same as the number of pts in each month in the no-missing-data variable

     yes? use climatological_axes
      *** NOTE: regarding /home/porter/tmap/ferret/linux/go/climatological_axes.cdf ...
      *** NOTE: Climatological axes SEASONAL_REG, MONTH_REG, and MONTH_IRREG defined
     yes? can data climatological_axes
     yes? use coads
     
     yes? let var = sst[x=280,y=20]
     yes? let nomiss = missing(var,30)
     yes? let varngd = var[gt=month_reg@modngd]
     yes? let varnbd = var[gt=month_reg@modnbd]
     yes? let nomiss_ngd = nomiss[gt=month_reg@modngd]
     yes? let nbdPngd = varnbd + varngd
     
     yes? list varnbd, varngd, nbdPngd, nomiss_ngd
                  DATA SET: /home/shiva/data/coads.cdf
                  COADS 2x2 Degree Monthly Average Surface Marine Observations
                  TIME: 01-JAN 00:45 to 31-DEC 06:34
                  LONGITUDE: 81W
                  LATITUDE: 19N
      Column  1: VARNBD is VAR[GT=MONTH_REG@MODNBD]
      Column  2: VARNGD is VAR[GT=MONTH_REG@MODNGD]
      Column  3: NBDPNGD is VARNBD + VARNGD
      Column  4: NOMISS_NGD is NOMISS[GT=MONTH_REG@MODNGD]
                      VARNBD  VARNGD NBDPNGD NOMISS_NGD
     16-JAN      /  1:  8.000  46.00   54.00   54.00
     15-FEB      /  2:  8.000  46.00   54.00   54.00
     17-MAR      /  3:  9.000  45.00   54.00   54.00
     16-APR      /  4:  9.000  45.00   54.00   54.00
     16-MAY      /  5:  7.000  47.00   54.00   54.00
     16-JUN      /  6:  9.000  45.00   54.00   54.00
     16-JUL      /  7:  8.000  46.00   54.00   54.00
     16-AUG      /  8:  7.000  47.00   54.00   54.00
     15-SEP      /  9:  7.000  47.00   54.00   54.00
     16-OCT      / 10:  8.000  46.00   54.00   54.00
     15-NOV      / 11:  7.000  47.00   54.00   54.00
     16-DEC      / 12:  7.000  47.00   54.00   54.00
     
  

  
  
  
• Long OPeNDAP URL's are now allowed The URL length may be up to 1024 characters. File names for local files are limited to 256 characters by Fortran calls OPEN and INQUIRE. Stepfiles that are part of multi-file descriptor data sets are limited to 256 characters.
  

Bug fixes in Version 6.04/6.05

• Irregularly-spaced coordinate axes from OPeNDAP files detected as regular. These axes were then assigned regularly-spaced but incorrect coordinate values. This has been corrected in the v6.05 release.
  
  
• In Ferret versions 6.0 and for some cases in v6.02, we saw only a cryptic netCDF Error code 1001 when trying to open an unavailable dataset on an OPeNDAP server. Errors encountered when trying to open an OPeNDAP data set are now set apart by the Ferret error heading **Internet Data error whereas if the message starts with **netCDF error, that indicates that the netCDF libraries were able to open a file, but then encountered an error within that file.

  For example here is a valid OPeNDAP url, which we then corrupt in a couple of ways. Here is the output from Ferret v6.04, where the verbose message is coming directly from the OPeNDAP library.

          NOAA/PMEL TMAP
          FERRET v6.04  
          Linux(g77) 2.4.21-32 - 07/03/07
          10-Jul-07 11:16     
  
     yes? SET MODE ignore_errors
   
     yes? ! Here is a valid OPeNDAP address:
   
     yes? SET DATA "http://test.opendap.org:8080/opendap/wcs/DataFed/NetCDF/OCEXTTAU-INTEX.Organic.Carbon.Aerosol.wcs"
     yes? SHOW DATA
       currently SET data sets:
      1> http://test.opendap.org:8080/opendap/wcs/DataFed/NetCDF/OCEXTTAU-INTEX.Organic.Carbon.Aerosol.wcs  (default)
   name     title                             I         J         K         L
   OCEXTTAU                                  1:361     1:181     ...       ...
   
   
     yes? ! First mess up the dataset name
   
     yes? SET DATA "http://test.opendap.org:8080/opendap/wcs/DataFed/NetCDF/OCEXTTAU-INTEX.Organic.Carbon.Aerosol_mmmm.wcs"
   **Internet Data error
               OPeNDAP Error: Error while reading the URL: http://test.opendap.org:8080/opendap/wcs/DataFed/NetCDF/OCEXTTAU-INTEX.Organic.Carbon.Aerosol_mmmm.wcs.dds?.
  The OPeNDAP server returned the following message:
  Not Found: The data source or server could not be found.
          Often this means that the OPeNDAP server is missing or needs attention;
          Please contact the server administrator. (OPeNDAP/netCDF Error code 1001)
               Data set: http://test.opendap.org:8080/opendap/wcs/DataFed/NetCDF/OCEXTTAU-INTEX.Organic.Carbon.Aerosol_mmmm.wcs
  SET DATA "http://test.opendap.org:8080/opendap/wcs/DataFed/NetCDF/OCEXTTAU-INTEX.Organic.Carbon.Aerosol_mmmm.wcs"
   
  
     yes? ! Now corrupt the server name
     yes? SET DATA "http://test_MMM.opendap.org:8080/opendap/wcs/DataFed/NetCDF/OCEXTTAU-INTEX.Organic.Carbon.Aerosol.wcs"
   **Internet Data error
               OPeNDAP Error: Couldn't resolve host 'test_MMM.opendap.org' (OPeNDAP/netCDF Error code 1001)
               Data set: http://test_MMM.opendap.org:8080/opendap/wcs/DataFed/NetCDF/OCEXTTAU-INTEX.Organic.Carbon.Aerosol.wcs
  SET DATA "http://test_MMM.opendap.org:8080/opendap/wcs/DataFed/NetCDF/OCEXTTAU-INTEX.Organic.Carbon.Aerosol.wcs"
  

  
  
  
• Plots of curvilinear data are drawn correctly across the modulo cut. SHADE/MODULO and FILL/MODULO now work correctly for curvilinear 3-argument plot commands. Previously, SHADE plots had a gap acrosss the cut, and FILL plots showed a discontinuity.
  
  
• Some new settings for axes for the point_spacing attribute did not get saved.

     yes? use monthly_navy_winds
     yes? set att (`uwnd,return=xaxis`).point_spacing = "uneven"
  
     yes? DEFINE ATT/OUTPUT  (`uwnd,return=xaxis`).new_att = "somthing"
     yes? SAVE/file=new_uwnd.nc uwnd
  Previously, the new setting for point_spacing was not saved (it still showed up as "even"). And the new 
  attribute was not saved to the file.  Now these settings are saved for the axis.
  
  
  
  

• For a user variable, accessing the list of attribute names gave an error The syntax var.attnames refers to the list of names of all the attributes of a variable. The following commands previously resulted in an error message

     yes? USE coads_climatology
     yes? LET sst2 = 2*sst
     yes? SET ATT/LIKE=sst sst2
     yes? LET anames = sst2.attnames
     
     yes? LIST anames
                  VARIABLE : SST2.ATTNAMES
                  FILENAME : coads_climatology.cdf
                  FILEPATH : /home/porter/tmap/ferret/linux/fer_dsets/data/
                  SUBSET   : 5 points (X)
      1   / 1:"missing_value"
      2   / 2:"_FillValue"   
      3   / 3:"long_name"    
      4   / 4:"history"      
      5   / 5:"units"        
  

• In some viewport plots, the zero contour was not drawn as a dark line. For instance this plot, which now works correctly

     yes? use coads_climatology
     yes? set reg/l=1/x=120e:280e/y=20s:20n
     yes? set v ul; contour/lev=c uwnd
     yes? set v ur; contour/lev=(-10,10,2) uwnd
     yes? set v ll; contour/lev=(-10,10,1) uwnd
     yes? set v lr; contour/lev=(-10,10,2) uwnd
  

  
  
  
• When a time axis is redefined, the "since t0" portion of the units was missing.

          NOAA/PMEL TMAP
          FERRET v6.04  
          Linux(g77) 2.4.21-32 - 07/03/07
          10-Jul-07 11:40     
  
     yes? USE coads_climatology
     yes? SAY `T[GT=time],RETURN=T0`
      !-> MESSAGE/CONTINUE 01-JAN-0000 00:00:00
     01-JAN-0000 00:00:00
  
     yes? SHOW ATTRIBUTE (TIME).units
          attributes for dataset: ./coads_climatology.cdf
      (TIME).units = hour since 0000-01-01 00:00:00 
  
     yes? DEFINE AXIS/T=1:12:1/UNIT=months/T0=1-jan-2001 TIME
      *** NOTE: /UNIT=MONTHS is ambiguous ... using 1/12 of 365.2425 days
     Replacing definition of axis TIME
      *** NOTE: grid  used by data set coads_climatology
      *** NOTE: Redefinition may alter apparent contents of data set
  
     yes? SHOW ATTRIBUTE (TIME).units
          attributes for dataset: /home/porter/tmap/ferret/linux/fer_dsets/data/coads_climatology.cdf
      (TIME).units = MONTHS since 1-JAN-2001 
  

  The since 1-JAN-2001 after the redefinition was missing when running previous versions.
  
  
• SHADE/LEV failed to repeat same levels, after SHADE/LINE. Depending on the data values, in some cases, the levels were not matched. This is fixed.
  
  
• Null operation produced an error. Previously this sequence resulted in an error condition; now it does not.

     yes? def sym a " " 
     yes? ($a)
  

  
  
  
• Under x86_64-linux, the DATE1900 function returned 0. This has been fixed.

   
     yes? list date1900("23-jan-2007")
               VARIABLE : DATE1900("23-jan-2007")
            39103.