ucs-select - Select rows and/or columns from UCS data set


  ucs-select --count FROM data.ds.gz WHERE '%O11% < %E11%'

  ucs-select '*' 'am.%.pv' FROM data.ds.gz INTO new.ds.gz

  ucs-select '%' FROM data.ds.gz WHERE 'not defined %b.accept%'


This program is used to select rows and/or columns from a UCS data set file, very much like a SELECT statement in SQL. The general form of the ucs-select command is

  ucs-select [--verbose | -v] (<variables> | --count)
             [ FROM <input.ds> ] [ WHERE <condition> ] [ INTO <output.ds> ]

<variables> is a whitespace-separated list of variable names or wildcard patterns (see the ucsexp manpage), which are matched against the columns of the data set file <input.ds>. The list of variables may not be omitted: use '%' to select all columns, and --count to display the number of matching rows only. Note that wildcard patterns may need to be quoted individually (because they contain shell metacharacters).

<condition> is a UCS expression (see the ucsexp manpage) used to select rows from the data set for which it evaluates to a true value. When the WHERE clause is omitted, all rows are selected. Note that <condition> must be a single argument and will usually have to be quoted (single quotes are highly recommended).

The input data set file <input.ds> defaults to STDIN (when omitted). The resulting table is printed on STDOUT in UCS data set file format (see the ucsfile manpage), and can be written to a data set file <output.ds> with the optional INTO clause.

With the --verbose (or -v) option, some progress information is displayed while the program is running.


Copyright 2004 Stefan Evert.

This software is provided AS IS and the author makes no warranty as to its use and performance. You may use the software, redistribute and modify it under the same terms as Perl itself.