Documentation on uvsplit

Task: uvsplit
Purpose: Break a uv data-set into single source, single-band files.
Categories: uv analysis

        UVSPLIT breaks an input data-set into a number of output
        data-sets, each of which contains a single ``type'' of data.
        Generally this means single source and single frequency
        band, though this is controlled by some options.

        UVSPLIT chooses its own names for the output data-sets. These
        are in the form
            sourcename.frequency
        Where ``frequency'' is the center frequency of the band, in MHz,
        rounded to the nearest MHz.

Key: vis
        The name if an input uv data-set. No default.

Key: select
        The normal uv selection commands. The default is copy everything.

Key: options
        This gives extra processing options. Several options can be given,
        each separated by commas. They may be abbreivated to the minimum
        needed to avoid ambiguity.
          nocopy    By default, uvsplit copies any calibration tables
                    present in the input to the output. The nocopy
                    option suppresses this.
          mosaic    Typically this option is used when splitting datasets
                    observed in mosaic mode. UVSPLIT attempts to place all
                    pointing centres of the source of interest into the one
                    output dataset. This requires that the field names of
                    the different pointings be composed of two parts,
                    separated by an underscore, viz
                          a_b
                    Where "a" is common to all field names (typically
                    a source) and "b" is a field-specific name (typically
                    a field number). For example field 123 of a mosaic
                    experiment of the LMC might be called "lmc_123"
        The following three options determine which data-set characteristics
        result in UVSPLIT generating different output data-sets.
          nosource  Do not produce new data-sets based on source name. That
                    is each output data-set can contain multiple sources.
                    The default is for each output to contain only a single
                    source.
          nofreq    Do not produce new output data-sets based on a switch
                    in frequency. That is each output data-set can contain
                    frequency switches.
          nowindow  Do not generate a separate output data-set for each
                    spectral window. The default is to create a new
                    output for each spectral window.