 |
» |
|
|
|
Many commands in this chapter have the designation Native
Mode at the end of their definition. This means that
the command is parsed by the Native Mode Command Parser.
If Native Mode is not specified, the command
is parsed by the Compatibility Mode Command Parser.
(A command parser separates command parameters.)
There is no relationship between the parser a command uses and the
function(s) the command performs. Also, just because a command is
parsed by the Compatibility Mode parser does
not mean it functions in the same way it did in the Classic HP3000
environment. All new commands for MPE/iX use the NM
parser. Some commands used on MPE V/E which have been changed for
MPE/iX use the NM parser and some do not.
MPE V/E commands which have not been changed for MPE/iX generally
use the CM parser. The important thing to remember is that the Native Mode parser
accepts several different formats for commands that you enter at
the colon prompt (:).
You may enter these NM-parsed commands in one of the following ways: By using the formal
command specification shown in the syntax diagram for each command
in this chapter. By using positional parameter specifications to
enter keyword parameter values. By combining positional and keyword specification
Another difference between the NM parser and the CM parser
is that the CM parser restricts a single comand parameter value
to be <=255 characters. On the NM side, the value is limited
by the size of the CI's command buffer. Formal Command Specification | |
You may enter an NM-parsed command as shown in the syntax
diagram for each command, for example: COMMAND KEYWORD1=A;KEYWORD2=B;KEYWORD3=C |
Positional Parameter Specification | |
You may also enter an NM-parsed command by omitting the keyword
parameter name and only entering the values as positional parameters,
for example: If you omit the keyword specifications and enter the values
as positional parameters, the values must be treated as such, and
all rules for positional parameters must be followed. For example,
if you only specify A
and C, you must
use the positional place holder (,)
as shown in the following example: Combining Positional/Keyword Parameters | |
Another option is to enter NM-parsed commands by using a combination
of positional and keyword specifications, for example: There is one important rule to remember when you combine positional
and keyword parameters: once you specify a keyword parameter, you
may no use positional parameters. For example, entering the following
command would produce an error: An exception to the rule is that you may specify positional
parameters that are subparameters of
a keyword parameter. For example, in the BUILD
command shown below, REC
is a keyword but the next four parameters (which define records
as being 80 bytes long, blocked at 1 and in Fixed ASCII format)
are positional. This syntax is acceptable
because they are subparameters of the
key word REC. BUILD filename;REC=-80,1,F,ASCII |
The following example shows the correct way to combine positional
and keyword parameters where the keyword has no subparameters: COMMAND A;KEYWORD2=B;KEYWORD3=C |
Entering Numbers in Commands | |
You may enter numbers as parameters to NM-parsed commands
as follows: With or without leading
zeros. As positive or negative numbers. Preceded by the $
sign indicating hexadecimal or base 16. Precdeded by the %
sign indicating octal or base 8. Preceded by the #
sign indicating decimal or base 10 (if neither $, %
nor # is specified
base 10 is used). In the decimal range -2,147,483,648 to 2,147,483,647.
For example, suppose you wanted to suspend
spooling on LDEV
6, your system printer. You could enter : SPOOLER DEV=#0006;SUSPEND;SHOW |
Or, because decimal is the default
you could omit the #
sign and enter: SPOOLER DEV=0006;SUSPEND;SHOW |
Or, omitting the leading zeroes you could enter: SPOOLER DEV=6;SUSPEND;SHOW |
When entering numbers as command parameters, it is advisable
to omit leading zeros for some commands parsed by the compatibility
mode (CM) parser. Using Quotes and Strings | |
The NM parser optionally accepts any string
input in single or double quotes. For example, because the file
name parameter of the PRINT
command is a string parameter, you could
enter it as follows: or or General Rules for Using QuotesQuotes are required if the value
of any string parameter contains any of the following delimiters: - ,
comma - ;
semicolon blank (one or more spaces) - =
equal sign - (
left parentheses - )
right parentheses
For example, suppose you want to set a variable called MYVAR
to a value of ;(A).
Because this string contains both a semi-colon and parentheses,
you would enter SETVAR
as follows: As another example, suppose you wanted to use the INFO=
parameter of the RUN
command to pass the following string (which contains both commas
and spaces) BLUE RIGHT 24, SPLIT LEFT, 2.
You would enter: RUN PROG;INFO="BLUE RIGHT 24, SPLIT LEFT, 2" |
MPE/iX string processing finds
the first double or single quote and pairs it with the last quote
of the same type to form a string. In
other words, single quotes pair only with other single quotes and
double quotes only with other double quotes. For this reason you
can use single quotes within double quotes, and double quotes within
single quotes. For example, all three of the following INFO
strings are correct: ...;INFO="THIS IS THE 'WRITE' WAY"
...;INFO='THIS IS THE "WRITE" WAY TOO'
...;INFO="YOU SIMPLY CAN'T GO WRONG" |
In all of the above cases, the quotes around the word WRITE
and in the word CAN'T are contained within the string and are treated
just like any other character. A technique called quote folding
enables you to embed single or double quotes in quoted strings.
For example, the following INFO=
string would pass the string shown below it: ... ;INFO="JUST SAY ""GATO""." |
Quote folding works as follows:
When the NM parser reads a quote (other than the very first quote
in an entire line), it checks the character to the immediate right
of the quote. If it is a quote of the same kind (single or double)
it is disregarded and the previous quote is treated like any other
non-quote character. For example, after being parsed, the following
quoted string becomes the string listed below it: ...;"PASS ""A"" TO ""X"" AND ""B"" TO ""Y""" |
PASS "A" TO "X" AND "B" TO "Y" |
Here is another example: "HERE ARE FOUR QUOTES "" "" "" """ |
HERE ARE FOUR QUOTES " " " " |
To delete spaces between the four quotes, you would enter
the string like this: "HERE ARE FOUR QUOTES """"""""" |
After being parsed, the string would look like this: HERE ARE FOUR QUOTES """" |
The NM parser processes quoted strings in the same way regardless
of the command or parameter with which they are used. For most CM commands, the CM parser
processes quoted strings in the same way as the NM parser. However,
the CM parser limits the length of quoted strings to 255 characters. Exceptions | |
There are four exceptions to the syntax governing MPE/iX commands: User command parameter lists (which
may affect string quoting rules).
These three exceptions allow the use of only specific delimiters
when specifying parameters, as defined below. Also the ECHO command accepts all delimiters and treats them
as part of the value to be echoed Invoking User Defined CommandsUser defined commands may be structured to accept the KEYWORD=parm
format, and you may mix keyword and positional parameters. User
command parameter lists allow you to use the following to delimit
parameters: - ,
comma - ;
semicolon blank (one or more spaces) - =
equal sign
For example, if the user defined command UDCA
is defined as UDCA parm1,parm2,parm3
you could invoke it as follows: or If the value of any parameter contains
any of the above delimiters you must
use quotes to delimit the parameter string. For example, if I;J;K
is a single string parameter value you
must delimit it with quotes (because it contains semi-colons) as
follows: The = sign
is used only to delimit a parameter name
from a parameter value. If the value
of a parameter contains an =
sign, then you must delimit the value with quotes. For example: Similarly, if a string value contains a quote, you must delimit
it by a quote. As an example, suppose you have a UDC which runs
a program with the INFO
string. The RUN
command within the UDC might look something like this: RUN PROGNAME;INFO="!PARM" |
If the value of the parameter were
something like this: THE "END" IS NEAR,
you would invoke the UDC like this: UDCA PARM="THE ""END"" IS NEAR" |
Or, you could enter this: UDCA "THE ""END"" IS NEAR" |
| | | |  | NOTE: If a parameter value begins with a quote it must
have a matching end quote. If it does not begin with a quote it
may contain embedded quotes which will be treated as any other character. | | | | |
For more information on the use of quotes, refer to the section
"Using Quotes and Strings" earlier in this chapter. The SETVAR
command allows you to use either spaces, semicolons, or commas to
delimit parameters, as follows: SETVAR NAME expression
SETVAR NAME,expression
SETVAR NAME;expression |
The rules for using quotes within strings containing delimiters
or quotes, previously discussed, apply to the SETVAR
command. For example, suppose you want to set a variable called BIGVAR
to a value of X,"Y";Z.
This expression contains two delimiters (comma and semicolon) as
well as quotes. The correct SETVAR
command would be: SETVAR BIGVAR "X,""Y"";Z" |
You could also delimit the expression from the variable name
using either a comma or semicolon as follows: SETVAR BIGVAR;"X,""Y"";Z" |
SETVAR BIGVAR,"X,""Y"";Z" |
For more information on the use of quotes, refer to the section
"Using Quotes and Strings" earlier in this chapter. |
