CL File I/O from a Macro

The macro language provides the ability to reposition the input CL file pointer, read CL records from the input CL file, modify the last CL record read and process the last CL record read. The “last CL record read” is updated each time a macro TAPERD command is executed and each time that a CL record is read in the normal course of processing the input CL file.

Repositioning the CL File (SEARCH)

The SEARCH macro command requires a single numeric expression specifying a record number as its argument. It repositions the input CL file so that the next CL record read will be at the given position.

$\textbf{SEARCH}\;\textbf{/}\;\mathit{record\text{-}number}$

The SEARCH macro command is frequently used in conjunction with the $FCLREC() function, as illustrated in the following example:

$$ save the next CL record position in a local variable
%L01=$FCLREC()+1
$$ do some processing using TAPERD, SEARCH or TAPEWT
…
…
$$ reset the CL file record pointer when finished
SEARCH/%L01

Reading a CL Record (TAPERD)

The TAPERD macro command is used to read the next CL record from the input CL file.

$\textbf{TAPERD}$

The TAPERD command takes no arguments.

Functions to Examine the Last CL Record

The following functions provide information about the last CL record read:

Function

Value

$FEOF( )

$TRUE if CL record is FINI

$FCLREC( )

record number of CL record

$FCLASS( )

numeric class of CL record

$FSUBCL( )

numeric subclass of CL record

$FSIZE( )

number of arguments in CL record

$FCL(i)

ith argument of CL record

$FCLS(i,j)

sequence of arguments i through j of CL record

The expression $FEOF() has the same value as “$FCLASS().EQ.14000”. Note that the $FEOF function serves a dual purpose in that it can also be used to determine the end-of-file status of a file opened with the OPEN macro command.

The $FCLASS and $FSUBCL functions can be particularly useful when dealing with wildcard macros. They can indicate the original input record class and subclass of the record that matched the wildcard macro.

More detailed descriptions of these functions can be found in “CL Data Parsing Functions”.

Modifying the Last CL Record (CLPUT)

The CLPUT macro command is used to modify one or more elements of the last CL record read, although the modifications are not acted upon until a TAPEWT command is coded.

$\textbf{CLPUT}\;\textbf{/}$ $\icamhsp{0.707mm}\mathit{index}\;\icamrange{0.66}{1}\,\bigl(,\mathit{argument}\,\bigr)$ $\icamhsp{0.707mm}\bigl[,\textbf{{\char36}NULL}\,\bigr]$

The first CLPUT argument specifies the index of the first element of the input CL record to be modified, followed by one or more argument values to be put starting at that location. An index value of –1 refers to the ISO W2 (class) position. An index value of “0” refers to the ISO W3 (subclass) position. Index values of “1” or greater refer to ISO positions W4 and onwards. For example, if the last CL record read was SPINDL/300,IPM, the command:

CLPUT/1,200

would change the record to SPINDL/200,IPM the command:

CLPUT/1,200,RPM

would change the record to SPINDL/200,RPM.

A $NULL value can be used to delete CL elements from the given location onwards. This can only be done from element 1 (W4) onwards. For example, if the major word code of AIR is 1011 and the current CL record is SPINDL/300,RPM, the command:

CLPUT/-1,2000,1011,ON,$NULL

would change the record to AIR/ON. Note that the above command is equivalent to the following four commands given in any order:

CLPUT/-1,2000
CLPUT/0,1011
CLPUT/1,ON
CLPUT/2,$NULL

Processing the Last CL Record (TAPEWT)

The TAPEWT macro command is used to process the last CL record read from the input CL file, as modified by CLPUT commands if any.

$\textbf{TAPEWT}$

The TAPEWT command takes no arguments. If macro matching is not disabled, the processor will try to match the record against a user-defined macro.

Deleting CL File Records (TAPEOP)

The TAPEOP command allows the user to delete single records and ranges of records from the CL File. The syntax is as follows:

$\textbf{TAPEOP}\;\textbf{/}$ $\icamhsp{0.707mm}\textbf{DELETE}$ $\icamhsp{0.707mm}\raise0.05ex\hbox{$\icamrange{1.66}{1}\!\begin{pmatrix}\!,\mathit{cln}\begin{bmatrix},\textbf{THRU},\!\begin{pmatrix}\!\begin{array}{l}\mathit{cln}\\\textbf{ALL}\end{array}\!\end{pmatrix}\end{bmatrix}\end{pmatrix}$}$

One or more individual records can be deleted by specifying the CL record number cln. A range of records can be deleted by coding cln,THRU,cln. The keyword ALL can be specified instead of the ending CL record number to delete all records to the end of the CL file.

This command is useful if you have already read ahead in the CL file to get a specific record and have processed it using the TAPEWT function. The use of the TAPEOP command will prevent the “gotten” record from being reprocessed the next time it is read. The following example reads ahead in the CL file, gets the next spindle command, outputs it and then deletes it from the CL file so that it will not be processed again the next time it is read:

%L01=$FGET(SPINDL)
TAPEWT
TAPEOP/DELETE,%L01

Another available format of the TAPEOP command is:

$\textbf{TAPEOP}\;\textbf{/}\;\textbf{DELETE},\textbf{CLEAR}$

The CLEAR option clears the deletion table. This may be necessary if a large number of TAPEOP/DELETE commands are issued. The record numbers are stored in a buffer area. The CLEAR option can be used to remove any previous record numbers from that buffer area.

CL File Processing Examples

The following macro positions the CL file to the end-of-file.

WHILE/.NOT.$FEOF()
  TAPERD
ENDOF/WHILE

The following macro code counts the number of post-processor commands in the input CL file. The CL file pointer is reset at the end:

K=$FCLREC()
SEARCH/1
TAPERD
I=0
WHILE/.NOT.$FEOF()
  IF/$FCLASS().EQ.2000
    I=I+1
  ENDOF/IF
  TAPERD
ENDOF/WHILE
PPRINT/'!(s9) pp commands found ',I
SEARCH/K+1

The following macro code stores in %L23 the value coded after the minor word RANGE in the current CL record. If there is no RANGE minor word, %L23 will have $NULL value after the loop:

%L23=$NULL
DO/%L00=1,$FSIZE()
  IF/$FCL(%L00).EQ.RANGE
    %L23=$FCL(%L00+1)
    EXIT/1
  ENDOF/IF
ENDOF/DO

The exact same results could be achieved using the $FGETARG function, as follows:

%L23=$FGETARG(RANGE)