NAME

tdiffer -- Create a new table which is the difference of two tables.

USAGE

tdiffer intable1 intable2 outtable colnam1 colnam2

DESCRIPTION

This task creates an output table containing all the rows of the first table which do not match the rows in the second table. The first, second, and output tables are given by the task parameters intable1, intable2, and outtable respectively. The match is done on the columns specified by the task parameters colnam1 and colnam2. Other columns are ignored. If the two tables are disjoint, the output table will be a copy of the first table, except the rows will be sorted. If the first table is a subset of the second, the output table will be empty.

PARAMETERS

intable[file name]

The name of the first input table.

intable[file name]

The name of the second input table.

outtable [file name]

The name of the output table. The output table has the same header parameters and column names as the first table.

colnam[string]

The column names in the first table used in the match. If more than one column is used, columns from the first and second table are associated with each other based on their position in the list and not on their names, i.e., the first column name in colnam1 is matched to the first column name passed to colnam2, regardless of whether the names match.

colnam[string]

The column names in the second table used in the match. The same number of column names must be passed to both the colnam1 and colnam2 parameters.

EXAMPLES

1. There are two tables, "targets.tab", containing a list of targets for observation, and "images.tab", containing a list of targets which have already been observed. Create a table named "new.tab" containing those targets which have not previously been observed:

tt> tdiffer targets.tab images.tab new.tab targetid targetid

BUGS

REFERENCES

This task was written by Bernie Simon.

SEE ALSO

tselect

Type "help tables opt=sys" for a higher-level description of the tables package.