1 DAT_FETCH
  Syntax: DAT_FETCH filename

  Reads a file containing a series  of data points or an Ntuple which have
  been written in ASCII format. The file can also contain data vs. time.

  The first non-comment line of the file should contain the identifier
  for the plot.  If the identifier is omitted it will be set to 1 with
  the current secondary identifier.

  The following cards are allowed:

  !              denotes a comment card (must be in column 1 or 2 for the
                 datacards)
  ID    id [idb] to specify the identifier for the plot
  NTUPLE ndim name1 name2 ...
                 gives the number of dimensions and the names of the
                 variables for an Ntuple
  LIMIT xlo xhi  to specify the lower and upper limits for the plot
  TITLE title    to specify the title for the plot
  ORDER          to specify the order of the variables
  TIME           to specify the time mode to store and the reference time
  DATA           to flag that the cards following this contain the data
  END            to flag that this is the end of the data

  The default order for the data is: `x y dx dy' or if you want asymmetric
  error bars: `x y  dnx dny dpx dpy'.  You can change  the order either by
  having an `ORDER'  card or using  the `SET ORDER'  command
  (see HELP SET ORDER  for more  details). The  following  ways of  giving
  the time are recognized:

  DATE_TIM     YYMMDD HHMMSS
  DATE         YYMMDD
  TIME         HHMMSS
  DATE_MIN     YYMMDD HHMM
  TIME_MIN     HHMM
  VAXTIME      Char*23 Vaxtime DD-MMM-YYYY HH:MM:SS.SS

  `VAXTIME' can be abbreviated on VMS, but must be the full CHAR*23
  format on any other computer. These commands can be given on the
  `ORDER' card or with the `SET ORDER' command. The data are stored in
  the form given on the `TIME' card or with the `SET TIME' command and
  can be (`DAY', `HOUR', `MINUTE' or `SECOND', default is `DAY'). A
  reference time (T=0) can be given and this must be in the form
  `YYMMDD HHMMSS'. If this is omitted or not specified the first point
  will be used.

  If you want to ignore some columns in the data, give `DUMMY' as the
  name of the column. This works on both the `ORDER' and the `NTUPLE' card.

  A blank card, one with a number as the first element, or the `DATA' card
  will be taken to signal the start of data.

  You can also use `DAT_FETCH' to read in an Ntuple which you have written
  to a file. Use the `NTUPLE' card to  give the dimension and the names of
  the   variables  for  the   Ntuple  (there  must  be  as  many  names as
  dimensions). The  `LIMITS' and  `ORDER' cards will be  ignored if given.
  Again the data must be  given as 1 record per  line and should be in the
  same order as the names given in the `NTUPLE' card.

  Either an `END' card or an `ID' card signify the end of one dataset and the
  start of the next.

2 Examples
!\begin{enumerate}
!\item
!^
  Example 1:
    A simple plot:
      ID 1
      LIMIT  0.0 1.0
      TITLE This is a test file
      0.4 5.7 0.01 3
      0.2 1   0.02 0.5
      0.6,0.1 0.1  0.4

!\item
!^
  Example 2:
    A file containing dummy entries and an Ntuple:
      ID 2
      TITLE This file is more complicated
      ORDER X DX DUMMY Y DY
      DATA
      3.4 1.6,35.0,5 2
      2.2 1   45.0,6 1
      END
!^

      ID 3
      TITLE A second dataset in the same file
      0.4 0.1  35.0 3   1.0
      0.2 0.05 45.0 0.5 0.2
      0.6,0.1  55.0 0.4 0.1
!^

      ID 4
      NTUPLE 4 P MASS dummy THETA PHI
      TITLE Ntuple containing particle parameters
      0.35 0.498 Angle  0.6 3.6
      1.20 0.480 Angle -0.5, 1.6
      END

!\end{enumerate}

