proto(F)

Format

Description

When a job is submitted to at or batch, the job is constructed as a shell script. First, a prologue is constructed, consisting of:

• A header whether the job is an at job or a batch job (actually, at jobs submitted to all queues other than queue a, not just to the batch queue b, are listed as batch jobs); the header will be:

  : at job
  

  for an at job, and

  : batch job
  

  for a batch job.

• A set of Bourne shell commands to make the environment (see environ(M)) for the at job the same as the current environment;

• A command to run the user's shell (as specified by the SHELL environment variable) with the rest of the job file as input.

  at then reads a ``prototype file'' and constructs the rest of the job file from it.

Text from the prototype file is copied to the job file, except for special ``variables'' that are replaced by other text:

$d

is replaced by the current working directory

$l

is replaced by the current file size limit (see ulimit(S))

$m

is replaced by the current umask (see umask(S))

$t

is replaced by the time at which the job should be run, expressed as seconds since January 1, 1970, 00:00 Greenwich Mean Time, preceded by a colon

$<

is replaced by text read by at from the standard input (that is, the commands provided to at to be run in the job)

If the job is submitted in queue queue, at uses the file /usr/lib/cron/.proto.queue as the prototype file if it exists; otherwise it will use the file /usr/lib/cron/.proto.

Examples

   #ident	"@(#)adm:.proto	1.2"
   cd $d
   ulimit $l
   umask $m
   $<

Files

/usr/lib/cron/.proto

standard prototype job file

/usr/lib/cron/.proto.a

prototype at job file

/usr/lib/cron/.proto.b

prototype batch (see at(C)) job file

See also