{    The purpose of this command is to provide the user with a quick way of
{ examining the data.  Report formatting capabilities are limited.  This report
{ specification command cannot be used to report descriptive data strings.  The
{ output format is shown below.
{
{                              std         elap      sum  cnt
{               count sum mean dev min max time int /sec /sec
{ row label
{     .
{     .
{     .
{ row label
{
{    The elapsed_time_since_predecessor summary calculation is only valid for
{ fields taken from records with job or task predecessors specified.  When the
{ report exceeds the current display page of the output file, a page header is
{ written on the new page.  The first line of this header contains the
{ operating system level, the date, the time, and page number.  After
{ appropriate spacing, the current report column headers will be reprinted, if
{ they still apply.
{
{ PUT:  This optional parameter allows the user to name the report entry for
{       easy reference.
{
{ NUMBER:  The NUMBER parameter is an ordinal specifying where in the list of
{       output definition subcommands the current subcommand is to be placed.
{       If this parameter is not specified, the default is for the position to
{       be NEXT.  This means the new output specification subcommand will
{       follow all current output specification subcommands.  If an integer
{       value is specified, the new subcommand will occupy that ordinal
{       position.  The subcommands having ordinal position equal to or greater
{       than this position will all have their ordinal position increased by
{       one.  If an ordinal postion greater than the largest used ordinal is
{       specified, the effect will be the same as specifying NEXT.  The
{       specified ordinal position will not be used, but instead an ordinal
{       equal to one greater than the current last used ordinal is assigned to
{       this entry.
{
{ FIELD:  Valid values for this parameter are the names of previously defined
{       FIELDS.  The user may optionally associate a string to be used as the
{       row label with a field.  If no such string is specified, the row label
{       will be the field name.
{
{ SUMMARY_CALCULATION:  This parameter governs the data columns to be printed
{       as shown in the above output format template.  The columns contain
{       descriptive statistics specified by a list of keywords.  The keywords
{       may be specified as the only items in the list or format specifications
{       for the column the keyword will produce in the report can be included.
{       The format specification is determined by the start_column and the
{       column_width.
{
{    The calculations each keyword represents are specified below.
{
{     COUNT:  The number of occurrences of the field in the log.
{     SUM:  The sum of the values of all field occurrences.
{     MEAN:  The average as computed by 'sum/count'.
{     STANDARD_DEVIATION:  The normal standard deviation with a divisor of (n-1).
{     MINIMUM:  The minimum value of all field occurrences.
{     MAXIMUM:  The maximum value of all field occurrences.
{     INTERVAL:  The average time interval between data
{           item occurrences.  This is useful for periodic statistics such as
{           the OS statistics, where the interval is a fixed value.
{     COUNT_PER_SECOND:  The computed average 'count/interval'.  This is
{           useful for periodic statistics such as the OS statistics, where the
{           interval is a fixed value.
{     SUM_PER_SECOND:  The computed average 'sum/interval'.  This is useful for periodic
{           statistics such as the OS statistics, where the interval is a fixed value.
{     ELAPSED_TIME_SINCE_PREDECESSOR:  The average computed time
{           difference of the logged times between a record and its defined job
{           or task predecessor.  This is only valid for items where a
{           predecessor has been defined.
{
{    The column format fields are described below:
{
{     START_COLUMN:  This field allows the user to fix the column the display
{           of the corresponding named summary_calculation begins in.  If the
{           user does not specify a start_column for the summary calculation
{           list entry a start_column will be assigned to the column when the
{           report is generated.
{     COLUMN_WIDTH:  This field allows the user to
{           specify the width of this summary_calculation display.  If the user
{           does not specify a column_width for the summary calculation list
{           entry a column_width will be assigned to the column when the report
{           is generated.
{
{ DISPLAY_HEADERS:  This parameter selects whether or not column headers are
{       printed for the fields in this put_field_summary.  If a list of fields
{       is specified, new column headings will never be printed for the second
{       through last fields.  Thus, this parameter governs whether new column
{       headings are printed before the first row of data in a
{       put_field_summary command.  Having the ability to turn off column
{       header display allows the output generated by consecutive
{       put_field_summary commands to appear as though they were generated by a
{       single command.  If this parameter is not specified, column headings
{       are not printed.
{
{ ROW_LABEL_FORMAT:  This parameter allows the user to specify the format of
{       the row labels.  The start_column field fixes the column the row labels
{       will begin on.  The column width field limits the row labels to the
{       specified number of characters.
{
{ STATUS: This parameter specifies the request status.
{       CONDITIONS:
{             pte$error_processing_parameter
{             pte$field_out_of_line_limits
{             pte$field_overlap
{             pte$generate_log_mode
{             pte$redefined_put
{             pte$undefined_field_for_put
