<set/> or <set></set>

Provided by module: Tags: RXML tags

Sets a variable in any scope that isn't read-only.

Attributes

variable="string"

The name of the variable to set.

value="string"

The value the variable should have. This is an alternative to specifying the value in the content, and the content must be empty if this is used.

The difference is that the value always is parsed as text here. Even if a type is specified with the "type" attribute this is still parsed as text, and then converted to that type.

from="string"

Get the value from this variable. The content must be empty if this is used.

expr="string"

An expression that gets evaluated to produce the value for the variable. The content must be empty if this is used.

Expression values can take any of the following forms:

scope.var[.var...]	The value of the given RXML variable. This follows the usual RXML variable reference syntax, e.g. "." in a variable name is quoted as "..". The scope and variable names may only contain pike identifier chars, i.e. letters, numbers (except at the start of a scope name), and "_". An exception is that a var may be a positive or negative integer. To reference an RXML variable that doesn't obey these restrictions, e.g. when a variable name contains "-", use the var() or index() functions. Note that the variable is not written as an entity reference. I.e. it is written without the surrounding "&" and ";". Using e.g. "" instead of "form.x" works most of the time, but it is more susceptible to parse errors if form.x contains funny things, and it is slower since Roxen cannot cache the compiled expression very well.
49	A decimal integer.
0xFE, 0xfe	A hexadecimal integer is preceded by "0x".
040	An octal integer is preceded by "0".
0b10100110	A binary integer is preceded by "0b".
1.43, 1.6e-5	A floating point number contains "." and/or "E"/"e".
"hello"	A string inside double quotes. C-style backslash escapes can be used in the string, e.g. a double quote can be included using \".
(expr)	Parentheses can be used around an expression for grouping inside a larger expression.

Value conversion expressions:

(int) expr	Casts a numeric or string expression (containing a formatted number on one of the formats above) to an integer.
(float) expr	Casts a numeric or string expression (containing a formatted number on one of the formats above) to a floating point number.
(string) expr	Casts a numeric or string expression to a string. If it is an integer then it is formatted as a decimal number. If it is a floating point number then it is formatted on the form [-]XX[.XX][e[-]XX].
INT(expr), FLOAT(expr), STRING(expr)	These functions cast their arguments to integer, float, or string, respectively. They behave like the cast operators above except that they do not generate any error if expr cannot be cast successfully. Instead a zero number or an empty string is returned as appropriate. This is useful if expr is a value from the client that may be bogus.

Note that values in the RXML form scope often are strings even though they may appear as (formatted) numbers. It is therefore a good idea to use the INT() or FLOAT() functions on them before you do math.

Expressions for checking types:

arrayp(expr)	Returns 1 if the value of expr is an array, and 0 otherwise.
callablep(expr)	Returns 1 if the value of expr is a function or similar, and 0 otherwise.
floatp(expr)	Returns 1 if the value of expr is a floating point number, and 0 otherwise.
functionp(expr)	Returns 1 if the value of expr is a function, and 0 otherwise.
intp(expr)	Returns 1 if the value of expr is an integer, and 0 otherwise.
mappingp(expr)	Returns 1 if the value of expr is a mapping, and 0 otherwise.
multisetp(expr)	Returns 1 if the value of expr is a multiset, and 0 otherwise.
objectp(expr)	Returns 1 if the value of expr is an object, and 0 otherwise.
programp(expr)	Returns 1 if the value of expr is a program, and 0 otherwise.
stringp(expr)	Returns 1 if the value of expr is a string, and 0 otherwise.
undefinedp(expr)	Returns 1 if the value of expr is UNDEFINED, and 0 otherwise.

Expressions for checking contents:

has_index(haystack, index)	Returns 1 if index is in the index domain of haystack, and 0 otherwise.
has_prefix(string, prefix)	Returns 1 if string starts with prefix, and 0 otherwise.
has_suffix(string, suffix)	Returns 1 if string ends with suffix, and 0 otherwise.
has_value(haystack, value)	Returns 1 if value is in the value domain of haystack, and 0 otherwise.
indices(expr)	Returns an array with all indices present in expr.
values(expr)	Returns an array with all values present in expr.

Expressions for numeric operands:

expr1 * expr2	Multiplication.
expr1 / expr2	Division.
expr1 % expr2	Modulo.
expr1 + expr2	Addition.
expr1 - expr2	Subtraction.
expr1 < expr2	Less than.
expr1 > expr2	Greater than.
expr1 <= expr2	Less than or equal.
expr1 >= expr2	Greater than or equal.
expr1 & expr2	Bitwise AND (integer operands only).
expr1 ^ expr2	Bitwise XOR (integer operands only).
expr1 | expr2	Bitwise OR (integer operands only).
pow(expr1, expr2)	Returns the value expr1 raised to the power of expr2.
log(expr)	Returns the natural logarithm of the value expr. To get the logarithm in another base, divide the result with log(base). This is the inverse operation of exp().
exp(expr)	Returns the natural exponential of the value expr. This is the inverse operation of log().
abs(expr)	Returns the absolute value of expr.
floor(expr)	Returns the closest integer value less than or equal to the value expr.
ceil(expr)	Returns the closest integer value greater than or equal to the value expr.
round(expr)	Returns the closest integer value to the value expr.
max(expr, ...)	Returns the maximum value of the arguments.
min(expr, ...)	Returns the minimum value of the arguments.

Expressions for string operands:

expr * num	Returns expr repeated num times.
expr1 / expr2	Returns an array with expr1 split on expr2. E.g. the string "a,b,c" split on "," is an array with the three elements "a", "b", and "c".
expr1 + expr2	Returns expr1 concatenated with expr2.
expr1 - expr2	Returns expr1 without any occurrences of expr2.
var(expr)	Parses the string expr as an RXML variable reference and returns its value. Useful e.g. if the immediate scope.var form cannot be used due to strange characters in the scope or variable names.
sizeof(expr)	Returns the number of characters in expr.
strlen(expr)	Returns the number of characters in expr.
search(expr1, expr2)	Returns the starting position of the first occurrence of the substring expr2 inside expr1, counting from 1, or 0 if expr2 does not occur in expr1.
reverse(expr)	Returns the reverse of expr.
regexp_split(regexp, expr)	Matches regexp against the string expr. If it matches then an array is returned that has the full match in the first element, followed by what the corresponding submatches (if any) match. Returns &roxen.false; if the regexp doesn't match. The regexp follows PCRE syntax.
basename(expr)	Returns the basename of the path in expr.
dirname(expr)	Returns the dirname of the path in expr.
combine_path(base, relative_path, ...)	Returns the combined path base + "/" + relative_path, with any path-segments of '.' and '..' handled and removed.

Expressions for array operands:

expr * num	Returns expr repeated num times.
expr1 + expr2	Returns expr1 concatenated with expr2.
expr1 - expr2	Returns expr1 without any of the elements in expr2.
expr1 & expr2	Returns the elements that exist in both expr1 and expr2, ordered according to expr1.
expr1 ^ expr2	Returns the elements that exist in either expr1 or expr2 but not in both. The order is expr1 followed by expr2.
expr1 | expr2	Returns the elements that exist in either expr1 or expr2. The order is expr1 followed by expr2. (The difference from expr1 + expr2 is that elements in expr2 aren't repeated if they occur in expr1.)
sizeof(expr)	Returns the number of elements in expr.
search(arr, expr)	Returns the position of the first occurrence of the element expr inside the array arr, counting from 1, or 0 if expr does not exist in arr.
reverse(expr)	Returns the reverse of expr.
uniq(expr)	Returns expr with all duplicate elements removed. The order among the remaining elements is kept intact; it is always the first of several duplicate elements that is retained.

Expressions for all types of operands:

!expr	Logical negation: 1 (true) if expr is the integer 0 (zero), otherwise 0 (false).
expr1 == expr2	1 (true) if expr1 and expr2 are the same, 0 (false) otherwise. Note that arrays may be different even though they contain the same sequence of elements.
expr1 != expr2	1 (true) if expr1 and expr2 are different, 0 (false) otherwise. This is the inverse of the == operator.
equal(expr1, expr2)	1 (true) if expr1 and expr2 are structurally equal, 0 (false) otherwise. As opposed to the == operator above, this returns 1 if two arrays contain the same sequence of elements.
expr1 && expr2	Nonzero (true) if both expressions are nonzero (true), 0 (false) otherwise. A true return value is actually the value of expr2.
expr1 || expr2	Nonzero (true) if either expression is nonzero (true), 0 (false) otherwise. A true return value is actually the value of the last true expression.
index(scope, expr, ...)	Indexes the RXML scope specified by the string scope with expr, and then continue to index the result successively with the remaining arguments. The indexing is done according to RXML rules, so e.g. the first index in an array is 1.

type="type"

The type of the value. If the value is taken from the content then this is the context type while evaluating it. If "value", "from" or "expr" is used then the value is converted to this type. Defaults to "any".

Tip: The <value> tag is useful to construct nonstring values, such as arrays and mappings:

<set variable="var.x" type="array"> <value>alpha</value> <value>beta</value> </set> &var.x;

Array result: ({"alpha", "beta"})

split="string"

Split the value on this string to form an array which is set. The value gets converted to a string if it isn't one already.

<set variable="var.x" split=",">a,b,c</set> &var.x;

Array result: ({"a", "b", "c"})