Specifies the characteristics of a language.
Format
DEFINE LANGUAGE language-name
1 – Qualifiers
1.1 /CAPABILITIES
/CAPABILITIES=DIAGNOSTICS
/CAPABILITIES=NODIAGNOSTICS (D)
Specifies whether the compiler can generate diagnostic files.
1.2 /COMMENT
/COMMENT=(specifier, . . . )
Specifies the character sequences of comments in the language. The
specifiers are as follows:
o ASSOCIATED_IDENTIFIER=keyword
Indicates the preferred association of comments to identifier.
You can specify one of the following values:
- NEXT-Indicates that comments should be associated with the
next identifier.
- PREVIOUS-Indicates that comments should be associated with
the preceding identifier.
o BEGIN=list of quoted strings
END=list of quoted strings
Defines the character sequences that start and end bracketed
comments. A bracketed comment begins and ends with explicit
comment delimiters. (Note that the beginning and ending comment
delimiters can be the same, but need not be.) The list provided
with the specifiers BEGIN and END can be any of the following:
- A string that is the one open comment sequence for the
language. You must enclose this in quotes.
- A parenthesized list of strings, each one of which can be
an open comment sequence for the language. You must enclose
each one in quotes.
The list accompanying the BEGIN specifier must be consistent
with the list accompanying the END specifier. If the BEGIN
specifier lists a string, then the END specifier must also list
a string.
Bracketed comments are recognized by the formatting commands
(see the ALIGN and FILL commands) and placeholder operations
(see the ERASE PLACEHOLDER command and the /DUPLICATION
qualifier of the DEFINE PLACEHOLDER command).
o TRAILING=list of quoted strings
Defines the character sequence that introduces line-oriented
comments. A line-oriented comment begins with a special
character sequence (consisting of one or more characters)
and ends at the end of the line. The list provided with the
TRAILING specifier can be any of the following:
- A string that is the one-line comment sequence for the
language.
- A list of strings enclosed in parentheses; each string can
be a line-comment sequence for the language.
Line comments are recognized by the formatting commands and
placeholder operations, just as bracketed comments are.
o LINE=list of quoted strings
Requires that the comment delimiter be the first character that
is not blank on the line. The LINE specifier is particularly
useful with block comments, such as the following:
/*
** Here is the inside of a comment
** which has LINE="**" specified
*/
o FIXED=quoted string, column number
Used for languages that require that a specific comment
delimiter be placed in a specific column, such as FIXED=("*",1)
for COBOL.
Note that for the specifier you cannot use any character that you
used in the /PLACEHOLDER delimiter-specification.
1.3 /COMPILE_COMMAND
/COMPILE_COMMAND=string
Specifies the default command string for the COMPILE command. (See
the explanation of the command-string parameter in the COMPILE
command entry.)
1.4 /EXPAND_CASE
/EXPAND_CASE=AS_IS (D)
/EXPAND_CASE=LOWER
/EXPAND_CASE=UPPER
Specifies the case of the text of the inserted template. AS_
IS specifies that the inserted template be expanded according
to the case in the token or placeholder definition. LOWER and
UPPER specify that the inserted template be expanded lowercase or
uppercase, respectively.
1.5 /FILE_TYPES
/FILE_TYPES=(file-type[, . . . ])
Specifies a list of file types that are valid for the language
being defined. The file types must be enclosed in quoted strings.
When LSE reads a file into a buffer, it sets the language for that
buffer automatically if it recognizes the file type. For example,
a FORTRAN file type (.FOR) sets the language to FORTRAN. Note that
the period character must be included with the file type.
1.6 /FORTRAN
/FORTRAN=ANSI_FORMAT
/FORTRAN=NOANSI_FORMAT (D)
Specifies special processing for ANSI FORTRAN. Note that some
commands behave differently when you use the /FORTRAN qualifier.
Specifying NOANSI_FORMAT causes LSE to insert templates in non-
ANSI (tab) format.
1.7 /HELP_LIBRARY
/HELP_LIBRARY=file-spec
/NOHELP_LIBRARY (D)
Specifies the HELP library where you can find help text for
placeholders and tokens defined in this language. LSE applies
the default file specification SYS$HELP:HELPLIB.HLB. If you want
to access some HELP library other than SYS$HELP, you must supply
an explicit device name.
1.8 /IDENTIFIER_CHARACTERS
/IDENTIFIER_CHARACTERS=string
Specifies the characters that may appear in token and alias names
in that language. This list of characters is used in various
contexts for the /INDICATED qualifier.
The list of identifier characters also determines what LSE
considers to be a word. A word is a sequence of identifier
characters, possibly followed by one or more blanks. All nonblank,
nonidentifier characters are considered to be distinct words.
If you do not specify the /IDENTIFIER_CHARACTERS qualifier, LSE
supplies the following values by default:
"abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ%$_0123456789"
1.9 /INITIAL_STRING
/INITIAL_STRING=string
Specifies the initial text that is to appear in a newly created
buffer.
1.10 /LEFT_MARGIN
/LEFT_MARGIN=n
/LEFT_MARGIN=1 (D)
/LEFT_MARGIN=CONTEXT_DEPENDENT
Specifies the left margin setting that is to be associated with
the language.
If you specify CONTEXT_DEPENDENT as the column number, then LSE
uses the indentation of the current line to determine the left
margin when you use the /WRAP qualifier. When you use the FILL
command, LSE uses the indentation of the first line of each
selected paragraph to determine the left margin.
1.11 /OVERVIEW_OPTIONS
/OVERVIEW_OPTIONS=(MINIMUM_LINES=m, TAB_RANGE=(t1,t2))
Specifies both the minimum number of lines an overview line must
hide and the range of acceptable tab increments.
The specifiers are as follows:
o MINIMUM_LINES=m
Specifies the minimum number of lines an overview line must
hide. The default is 1. For example, if the value of the
parameter on MINIMUM_LINES is 5, then a line hides other lines
only if there are at least five lines to hide. This specifier
helps the user to avoid having very small source-line groups,
and thus to avoid many expansion levels.
o TAB_RANGE=(t1,t2)
The TAB_RANGE specifier indicates the range of tab values for
which the adjustment definitions are valid. The default is
(4,8). The second value must be at least twice the first value;
both values must be positive. For example, if the tab range
is (4,8), then LSE assumes that the adjustment definitions
will work for any DEFINE LANGUAGE/TAB_INCREMENT value from 4
to 8 inclusive. If you specify a /TAB_INCREMENT value outside
the tab range, then LSE recomputes indentation to make the
adjustments work.
For best performance, it is recommended that you avoid
recomputation by choosing a range that covers reasonable
values. The numbers specified for the DEFINE ADJUSTMENT/CURRENT
and DEFINE ADJUSTMENT/SUBSEQUENT commands must work for any tab
increment value in the tab range.
1.12 /PLACEHOLDER_DELIMITERS
/PLACEHOLDER_DELIMITERS=(delimiter-specification[, . . . ])
Specifies starting and ending strings that delimit placeholders.
Placeholders can specify single constructs or lists of constructs.
The delimiters for each type of placeholder are specified as
a pair of quoted strings separated by commas and enclosed in
parentheses.
The format of a delimiter specification is as follows:
keyword=(starting-string,ending-string)
Possible keywords are REQUIRED, REQUIRED_LIST, OPTIONAL, OPTIONAL_
LIST, or PSEUDOCODE. If you do not use the PSEUDOCODE keyword, the
default is NOPSEUDOCODE. The maximum length of these strings is
seven characters.
The following is an example of a complete set of placeholder
delimiter specifications:
/PLACEHOLDER_DELIMITERS = ( -
REQUIRED =("{<",">}"), -
REQUIRED_LIST=("{<",">}..."), -
OPTIONAL =("[<",">]"), -
OPTIONAL_LIST=("[<",">]..."), -
PSEUDOCODE=("«" , "»"))
If any of the five keywords are not specified with the
/PLACEHOLDER_DELIMITERS qualifier, LSE applies the following
defaults:
/PLACEHOLDER_DELIMITERS = ( -
REQUIRED =("{","}"), -
REQUIRED_LIST=("{","}..."), -
OPTIONAL =("[","]"), -
OPTIONAL_LIST=("[","]..."), -
NOPSEUDOCODE)
The placeholder delimiters that are accepted by each compiler are
as follows:
ADA: { } , { }... , [ ] , [ ]... , « » , <| |>
BASIC: { } , { }... , [ ] , [ ]... , « » , << >>
BLISS: {~ ~} , {~ ~}... , [~ ~] , [~ ~]... , «» , <~ ~>
C: {@ @} , {@ @}... , [@ @] , [@ @]... , «» , <@ @>
COBOL: { } , { }... , [ ] , [ ]... , « » , << >>
FORTRAN: { } , { }... , [ ] , [ ]... , « » , << >>
PASCAL: %{ }% , %{ }%... , %[ ]% , %[ ]%... , « » , %< >%
PL1: { } , { }... , [ ] , [ ]... , « » , << >>
Note that for the specifier you cannot use any character that you
used in the /COMMENT specifier.
1.13 /PUNCTUATION_CHARACTERS
/PUNCTUATION_CHARACTERS=string
/PUNCTUATION_CHARACTERS=",;()" (D)
Specifies the characters that are considered punctuation marks,
or delimiters, in the language. When a placeholder name and its
enclosing brackets are deleted, preceding white space is also
deleted if there are punctuation characters to delimit the program
constructs.
1.14 /QUOTED_ITEM
/QUOTED_ITEM=(QUOTES=string [,ESCAPES=string])
/NOQUOTED_ITEM (D)
Describes the syntax of certain language elements, such as
strings, that require special handling for proper text formatting.
LSE uses the /QUOTED_ITEM qualifier to detect comments properly.
LSE does not acknowledge comment strings that occur within quoted
items, nor does it acknowledge quoted elements that occur within
comments.
The value of the /QUOTED_ITEM qualifier indicates the syntax of a
quoted item. This value must be a keyword list. The keywords are
as follows:
o QUOTES
This keyword is required and must have an explicit value.
The value must be a quoted string denoting all of the quote
characters in the language. LSE assumes that quoted items begin
and end with the same character.
o ESCAPES
This keyword is optional. If given, then the value is required
and must be a quoted string containing the escape characters
for quoted items. Some languages use escape characters to
insert quote characters into strings. For example, C uses
the backslash (\) as an escape character. If you omit this
keyword, then LSE assumes that the language inserts quote
characters into strings by doubling them.
1.15 /RIGHT_MARGIN
/RIGHT_MARGIN=n
/RIGHT_MARGIN=80 (D)
Specifies the right margin setting that is to be associated with
the language. By default, the right margin is set at column 80.
1.16 /TAB_INCREMENT
/TAB_INCREMENT=n
/TAB_INCREMENT=4 (D)
Specifies that tab stops be set every n columns, beginning with
column 1.
1.17 /TAG_TERMINATORS
/TAG_TERMINATORS=(string[, . . . ])
/TAG_TERMINATORS=(":") (D)
Specifies the character sequences that you an use to terminate a
tag. You can use each string in the list of strings to terminate
a tag. When you use the /DESIGN=COMMENTS qualifier to compile
programs, the compiler uses this information to detect tags inside
of comments.
You must not specify a string that ends in whitespace (space
characters or tabs). Compilers always allow whitespace between a
tag name and the tag terminator characters. If you want to require
whitespace between the tag and the tag terminator string, use
a space character as the first character of the tag terminator
string. HP does not recommend the use of embedded whitespace.
1.18 /TOPIC_STRING
/TOPIC_STRING=string
Specifies a prefix string to be concatenated to the /TOPIC_
STRING qualifier specified in a placeholder or token definition
before LSE looks up the help text for that placeholder or token.
(Typically, this is the name of the language in the HELP library.)
1.19 /VERSION
/VERSION=string
Specifies a string that represents the version number of the
tokens and placeholders associated with this language. You use
the SHOW LANGUAGE command to display this string.
1.20 /WRAP
/WRAP
/NOWRAP (D)
Specifies whether text should be wrapped to a new line when you
are typing beyond the right margin of the current line. The
/NOWRAP qualifier disables such text wrapping.
2 – Parameter
language-name
Specifies the name of the language whose characteristics are to be
defined.