4. Operators

To make it easier to write Pike, and to make the code somewhat shorter, some functions can be called with just one or two characters in the code. These functions are called operators and we have already seen how they work in plenty of examples. In this chapter I will describe in detail what they do. The operators are divided into categories depending on their function, but beware that some operators have meanings that go way beyond the scope of the category they are in.

The arithmetic operators are the simplest ones, since they work just like you remember from math in school. The arithmetic operators are:

The third column, "Identifier" is the name of the function that actually evaluates the operation. For instance, a + b can also be written as `+(a, b). I will show you how useful this can be at the end of this chapter.

When applied to integers or floats these operators do exactly what they are supposed to do. The only operator in the list not known from basic math is the modulo operator. The modulo operator returns the remainder from an integer division. It is the same as calculating a - floor(a / b) * b. floor rounds the value down to closest lower integer value. Note that the call to floor isn't needed when operating on integers, since dividing two integers will return the result as an integer and it is always rounded down. For instance, 8 / 3 would return 2.

If all arguments to the operator are integers, the result will also be an integer. If one is a float and the other is an integer, the result will be a float. If both arguments are float, the result will of course be a float.

However, there are more types in Pike than integers and floats. Here is the complete list of combinations of types you can use with these operators:

int + int

float + int
int + float
float + float

string + string
int + string
float + string
string + int
string + float

array + array

mapping + mapping

multiset + multiset

int - int

float - int
int - float
float - float

string - string

array - array

mapping - mapping

multiset - multiset

- int

- float

int * int

float * int
int * float
float * float

array(string) * string

array(array) * array

string * int

string * float

array * int

array * float

int / int

float / int
int / float
float / float

string / string

string / int

string / float

array / int

array / float

int % int

float % float
int % float
float % int

string % int

array % int

Method `+

mixed `+(mixed arg)
mixed `+(object arg, mixed ... more)
int `+(int arg, int ... more)
float `+(float|int arg, float|int ... more)
string `+(string|float|int arg, string|float|int ... more)
array `+(array arg, array ... more)
mapping `+(mapping arg, mapping ... more)
multiset `+(multiset arg, multiset ... more)

Description

Addition/concatenation.

Every expression with the + operator becomes a call to this function, i.e. a+b is the same as predef::`+(a,b). Longer + expressions are normally optimized to one call, so e.g. a+b+c becomes predef::`+(a,b,c).

Returns

If there's a single argument, that argument is returned.

If arg is an object with only one reference and an lfun::`+=(), that function is called with the rest of the arguments, and its result is returned.

Otherwise, if arg is an object with an lfun::`+(), that function is called with the rest of the arguments, and its result is returned.

Otherwise, if any of the other arguments is an object that has an lfun::``+(), the first such function is called with the arguments leading up to it, and `+() is then called recursively with the result and the rest of the arguments.

Otherwise, if arg is UNDEFINED and the other arguments are either arrays, mappings or multisets, the first argument is ignored and the remaining are added together as described below. This is useful primarily when appending to mapping values since m[x] += ({foo}) will work even if m[x] doesn't exist yet.

Otherwise the result depends on the argument types:

int|float	The result is the sum of all the arguments. It's a float if any argument is a float.
string|int|float	If any argument is a string, all will be converted to strings and concatenated in order to form the result.
array	The array arguments are concatened in order to form the result.
mapping	The result is like arg but extended with the entries from the other arguments. If the same index (according to hash_value and `==) occur in several arguments, the value from the last one is used.
multiset	The result is like arg but extended with the entries from the other arguments. Subsequences with orderwise equal indices (i.e. where `< returns false) are concatenated into the result in argument order.

The function is not destructive on the arguments - the result is always a new instance.

Note

In Pike 7.0 and earlier the addition order was unspecified.

The treatment of UNDEFINED was new in Pike 7.0.

See also

`-(), lfun::`+(), lfun::``+()

Method `-

mixed `-(mixed arg1)
mixed `-(mixed arg1, mixed arg2, mixed ... extras)
mixed `-(object arg1, mixed arg2)
mixed `-(mixed arg1, object arg2)
int `-(int arg1, int arg2)
float `-(float arg1, int|float arg2)
float `-(int|float arg1, float arg2)
string `-(string arg1, string arg2)
array `-(array arg1, array arg2)
mapping `-(mapping arg1, array arg2)
mapping `-(mapping arg1, mapping arg2)
mapping `-(mapping arg1, multiset arg2)
multiset `-(multiset arg1, multiset arg2)

Description

Negation/subtraction/set difference.

Every expression with the - operator becomes a call to this function, i.e. -a is the same as predef::`-(a) and a-b is the same as predef::`-(a,b). Longer - expressions are normally optimized to one call, so e.g. a-b-c becomes predef::`-(a,b,c).

Returns

If there's a single argument, that argument is returned negated. If arg1 is an object with an lfun::`-(), that function is called without arguments, and its result is returned.

If there are more than two arguments the result is: `-(`-(arg1, arg2), @extras).

Otherwise, if arg1 is an object with an lfun::`-(), that function is called with arg2 as argument, and its result is returned.

Otherwise, if arg2 is an object with an lfun::``-(), that function is called with arg1 as argument, and its result is returned.

Otherwise the result depends on the argument types:

arg1 can have any of the following types:

int|float	The result is arg1 - arg2, and is a float if either arg1 or arg2 is a float.
string	The result is arg1 with all nonoverlapping occurrences of the substring arg2 removed. In cases with two overlapping occurrences, the leftmost is removed.
array|mapping|multiset	The result is like arg1 but without the elements/indices that match any in arg2 (according to `>, `<, `== and, in the case of mappings, hash_value).

The function is not destructive on the arguments - the result is always a new instance.

Note

In Pike 7.0 and earlier the subtraction order was unspecified.

Note

If this operator is used with arrays or multisets containing objects which implement lfun::`==() but not lfun::`>() and lfun::`<(), the result will be undefined.

See also

`+()

Method `*

mixed `*(mixed arg1)
mixed `*(object arg1, mixed arg2, mixed ... extras)
mixed `*(mixed arg1, object arg2)
array `*(array arg1, int arg2)
array `*(array arg1, float arg2)
string `*(string arg1, int arg2)
string `*(string arg1, float arg2)
string `*(array(string) arg1, string arg2)
array `*(array(array) arg1, array arg2)
float `*(float arg1, int|float arg2)
float `*(int arg1, float arg2)
int `*(int arg1, int arg2)
mixed `*(mixed arg1, mixed arg2, mixed ... extras)

Description

Multiplication/repetition/implosion.

Every expression with the * operator becomes a call to this function, i.e. a*b is the same as predef::`*(a,b). Longer * expressions are normally optimized to one call, so e.g. a*b*c becomes predef::`*(a,b,c).

Returns

If there's a single argument, that argument will be returned.

If the first argument is an object that implements lfun::`*(), that function will be called with the rest of the arguments.

If there are more than two arguments, the result will be `*(`*(arg1, arg2), @extras).

If arg2 is an object that implements lfun::``*(), that function will be called with arg1 as the single argument.

Otherwise the result will be as follows:

arg1 can have any of the following types:

array	arg2 can have any of the following types: int|float The result will be arg1 concatenated arg2 times. string|array The result will be the elements of arg1 concatenated with arg2 interspersed.
string	The result will be arg1 concatenated arg2 times.
int|float	The result will be arg1 * arg2, and will be a float if either arg1 or arg2 is a float.

Note

In Pike 7.0 and earlier the multiplication order was unspecified.

See also

`+(), `-(), `/(), lfun::`*(), lfun::``*()

Method `/

mixed `/(object arg1, mixed arg2)
mixed `/(mixed arg1, object arg2)
array(string) `/(string arg1, int arg2)
array(string) `/(string arg1, float arg2)
array(array) `/(array arg1, int arg2)
array(array) `/(array arg1, float arg2)
array(string) `/(string arg1, string arg2)
array(array) `/(array arg1, array arg2)
float `/(float arg1, int|float arg2)
float `/(int arg1, float arg2)
int `/(int arg1, int arg2)
mixed `/(mixed arg1, mixed arg2, mixed ... extras)

Description

Division/split.

Every expression with the / operator becomes a call to this function, i.e. a/b is the same as predef::`/(a,b).

Returns

If there are more than two arguments, the result will be `/(`/(arg1, arg2), @extras).

If arg1 is an object that implements lfun::`/(), that function will be called with arg2 as the single argument.

If arg2 is an object that implements lfun::``/(), that function will be called with arg1 as the single argument.

Otherwise the result will be as follows:

arg1 can have any of the following types:

string	arg2 can have any of the following types: int|float The result will be and array of arg1 split in segments of length arg2. If arg2 is negative the splitting will start from the end of arg1. string The result will be an array of arg1 split at each occurrence of arg2. Note that the segments that matched against arg2 will not be in the result.
array	arg2 can have any of the following types: int|float The result will be and array of arg1 split in segments of length arg2. If arg2 is negative the splitting will start from the end of arg1. array The result will be an array of arg1 split at each occurrence of arg2. Note that the elements that matched against arg2 will not be in the result.
float|int	The result will be arg1 / arg2. If both arguments are int, the result will be truncated to an int. Otherwise the result will be a float.

Note

Unlike in some languages, the function f(x) = x/n (x and n integers) behaves in a well-defined way and is always rounded down. When you increase x, f(x) will increase with one for each n:th increment. For all x, (x + n) / n = x/n + 1; crossing zero is not special. This also means that / and % are compatible, so that a = b*(a/b) + a%b for all a and b.

See also

`%

Method `%

mixed `%(object arg1, mixed arg2)
mixed `%(mixed arg1, object arg2)
string `%(string arg1, int arg2)
array `%(array arg1, int arg2)
float `%(float arg1, float|int arg2)
float `%(int arg1, float arg2)
int `%(int arg1, int arg2)

Description

Modulo.

Every expression with the % operator becomes a call to this function, i.e. a%b is the same as predef::`%(a,b).

Returns

If arg1 is an object that implements lfun::`%() then that function will be called with arg2 as the single argument.

If arg2 is an object that implements lfun::``%() then that function will be called with arg2 as the single argument.

Otherwise the result will be as follows:

arg1 can have any of the following types:

string|array	If arg2 is positive, the result will be the last `%(sizeof(arg1), arg2) elements of arg1. If arg2 is negative, the result will be the first `%(sizeof(arg1), -arg2) elements of arg1.
int|float	The result will be arg1 - arg2*floor(arg1/arg2). The result will be a float if either arg1 or arg2 is a float, and an int otherwise.

For numbers, this means that

1. a % b always has the same sign as b (typically b is positive; array size, rsa modulo, etc, and a varies a lot more than b).

2. The function f(x) = x % n behaves in a sane way; as x increases, f(x) cycles through the values 0,1, ..., n-1, 0, .... Nothing strange happens when you cross zero.

3. The % operator implements the binary "mod" operation, as defined by Donald Knuth (see the Art of Computer Programming, 1.2.4). It should be noted that Pike treats %-by-0 as an error rather than returning 0, though.

4. / and % are compatible, so that a == b*floor(a/b) + a%b for all a and b.

See also

`/, floor()

Method `**

object|int|float `**(object|int|float arg1, object|int|float arg2)

Description

Exponentiation. Raise arg1 to the power of arg2.

The arithmetic operators would be hard to use for anything interesting without the ability to compare the results to each other. For this purpose there are six comparison operators:

The == and != operators can be used on any type. For two values to be same they must be the same type. Therefore 1 and 1.0 are not same. Also, for two values of pointer types to be the same the two values must be pointers to the same object. It is not enough that the two objects are of the same size and contain the same data.

The other operators in the table above can only be used with integers, floats and strings. If you compare an integer with a float, the int will be promoted to a float before the comparison. When comparing strings, lexical order is used and the values of the environment variables LC_CTYPE and LC_LANG are respected.

Method `==

bool `==(mixed arg1, mixed arg2, mixed ... extras)

Description

Equality test.

Every expression with the == operator becomes a call to this function, i.e. a==b is the same as predef::`==(a,b).

If more than two arguments are given, each argument is compared with the following one as described below, and the test is successful iff all comparisons are successful.

If the first argument is an object with an lfun::`==(), that function is called with the second as argument, unless the second argument is the same as the first argument. The test is successful iff its result is nonzero (according to `!).

Otherwise, if the second argument is an object with an lfun::`==(), that function is called with the first as argument, and the test is successful iff its result is nonzero (according to `!).

Otherwise, if the arguments are of different types, the test is unsuccessful. Function pointers to programs are automatically converted to program pointers if necessary, though.

Otherwise the test depends on the type of the arguments:

int	Successful iff the two integers are numerically equal.
float	Successful iff the two floats are numerically equal and not NaN.
string	Successful iff the two strings are identical, character for character. (Since all strings are kept unique, this is actually a test whether the arguments point to the same string, and it therefore run in constant time.)
array|mapping|multiset|object|function(:void)|program|type	Successful iff the two arguments point to the same instance.

Returns

Returns 1 if the test is successful, 0 otherwise.

Note

Floats and integers are not automatically converted to test against each other, so e.g. 0==0.0 is false.

Note

Programs are not automatically converted to types to be compared type-wise.

See also

`!(), `!=()

Method `!=

bool `!=(mixed arg1, mixed arg2, mixed ... extras)

Description

Inequality test.

Every expression with the != operator becomes a call to this function, i.e. a!=b is the same as predef::`!=(a,b).

This is the inverse of `==(); see that function for further details.

Returns

Returns 1 if the test is successful, 0 otherwise.

See also

`==()

Method `>

bool `>(mixed arg1, mixed arg2, mixed ... extras)

Description

Greater than test.

Every expression with the > operator becomes a call to this function, i.e. a>b is the same as predef::`>(a,b).

Returns

Returns 1 if the arguments are strictly decreasing, and 0 (zero) otherwise.

See also

`<(), `<=(), `>=()

Method `>=

bool `>=(mixed arg1, mixed arg2, mixed ... extras)

Description

Greater than or equal test.

Every expression with the >= operator becomes a call to this function, i.e. a>=b is the same as predef::`>=(a,b).

Returns

Returns 1 if the test is successful, 0 otherwise.

Note

For total orders, e.g. integers, this is the inverse of `<().

See also

`<=(), `>(), `<()

Method `<

bool `<(mixed arg1, mixed arg2, mixed ... extras)

Description

Less than test.

Every expression with the < operator becomes a call to this function, i.e. a<b is the same as predef::`<(a,b).

Returns

Returns 1 if the test is successful, 0 otherwise.

See also

`<=(), `>(), `>=()

Method `<=

bool `<=(mixed arg1, mixed arg2, mixed ... extras)

Description

Less than or equal test.

Every expression with the <= operator becomes a call to this function, i.e. a<=b is the same as predef::`<=(a,b).

Returns

Returns 1 if the test is successful, 0 otherwise.

Note

For total orders, e.g. integers, this is the inverse of `>().

See also

`<(), `>(), `>=()

Logical operators are operators that operate with truth values. In Pike any value except zero is considered true. Logical operators are a very basic part of Pike. They can also decide which arguments to evaluate and which not to evaluate. Because of this most of the logical operators do not have any identifiers and can not be called as normal functions. There are four logical operators:

Method `!

bool `!(object|function(:void) arg)
int(1) `!(int(0) arg)
int(0) `!(mixed arg)

Description

Logical not.

Every expression with the ! operator becomes a call to this function, i.e. !a is the same as predef::`!(a).

It's also used when necessary to test truth on objects, i.e. in a statement if (o) ... where o is an object, the test becomes the equivalent of !!o so that any lfun::`!() the object might have gets called.

Returns

If arg is an object that implements lfun::`!(), that function will be called.

If arg is 0 (zero), a destructed object, or a function in a destructed object, 1 will be returned.

Otherwise 0 (zero) will be returned.

Note

No float is considered false, not even 0.0.

See also

`==(), `!=(), lfun::`!()

These operators are used to manipulate bits as members in sets. They can also manipulate arrays, multisets and mappings as sets.

The first three operators can only be used with integers and should be pretty obvious.

The other three, intersection, union and symmetric difference, can be used with integers, arrays, multisets and mappings. When used with integers, these operators considers each bit in the integer a separate element. If you do not know about how bits in integers work I suggest you go look it up in some other programming book or just don't use these operators on integers.

When intersection, union or symmetric difference is used on an array each element in the array is considered by itself. So intersecting two arrays will result in an array with all elements that are present in both arrays. Example: ({7,6,4,3,2,1}) & ({1, 23, 5, 4, 7}) will return ({7,4,1}). The order of the elements in the returned array will always be taken from the left array. Elements in multisets are treated the same as elements in arrays. When doing a set operation on a mapping however, only the indices are considered. The values are just copied with the indices. If a particular index is present in both the right and left argument to a set operator, the one from the right side will be used. Example: ([1:2]) | ([1:3]) will return ([1:3]).

Method `<<

int `<<(int arg1, int(0..) arg2)
mixed `<<(object arg1, int(0..)|object arg2)
mixed `<<(int arg1, object arg2)
mixed `<<(float arg1, int(0..) arg2)

Description

Left shift.

Every expression with the << operator becomes a call to this function, i.e. a<<b is the same as predef::`<<(a,b).

If arg1 is an object that implements lfun::`<<(), that function will be called with arg2 as the single argument.

If arg2 is an object that implements lfun::``<<(), that function will be called with arg1 as the single argument.

If arg1 is a float and arg2 is a non-negative integer, arg1 will be multiplied by 1<<arg2.

Otherwise arg1 will be shifted arg2 bits left.

See also

`>>()

Method `>>

int `>>(int arg1, int(0..) arg2)
mixed `>>(object arg1, int(0..)|object arg2)
mixed `>>(int arg1, object arg2)
float `>>(float arg1, int(0..) arg2)

Description

Right shift.

Every expression with the >> operator becomes a call to this function, i.e. a>>b is the same as predef::`>>(a,b).

If arg1 is an object that implements lfun::`>>(), that function will be called with arg2 as the single argument.

If arg2 is an object that implements lfun::``>>(), that function will be called with arg1 as the single argument.

If arg1 is a float and arg2 is a non-negative integer, arg1 will be divided by 1<<arg2.

Otherwise arg1 will be shifted arg2 bits right.

See also

`<<()

Method `~

mixed `~(object arg)
int `~(int arg)
float `~(float arg)
type `~(type|program arg)
string `~(string arg)

Description

Complement/inversion.

Every expression with the ~ operator becomes a call to this function, i.e. ~a is the same as predef::`~(a).

Returns

The result will be as follows:

arg can have any of the following types:

object	If arg implements lfun::`~(), that function will be called.
int	The bitwise inverse of arg will be returned.
float	The result will be -1.0 - arg.
type|program	The type inverse of arg will be returned.
string	If arg only contains characters in the range 0 - 255 (8-bit), a string containing the corresponding 8-bit inverses will be returned.

See also

`!(), lfun::`~()

Method `&

mixed `&(mixed arg1)
mixed `&(mixed arg1, mixed arg2, mixed ... extras)
mixed `&(object arg1, mixed arg2)
mixed `&(mixed arg1, object arg2)
int `&(int arg1, int arg2)
string `&(string arg1, string arg2)
array `&(array arg1, array arg2)
mapping `&(mapping arg1, mapping arg2)
mapping `&(mapping arg1, array arg2)
mapping `&(mapping arg1, multiset arg2)
multiset `&(multiset arg1, multiset arg2)
type `&(type|program arg1, type|program arg2)

Description

Bitwise and/intersection.

Every expression with the & operator becomes a call to this function, i.e. a&b is the same as predef::`&(a,b).

Returns

If there's a single argument, that argument is returned.

If there are more than two arguments the result is: `&(`&(arg1, arg2), @extras).

Otherwise, if arg1 is an object with an lfun::`&(), that function is called with arg2 as argument, and its result is returned.

Otherwise, if arg2 is an object with an lfun::``&(), that function is called with arg1 as argument, and its result is returned.

Otherwise the result depends on the argument types:

arg1 can have any of the following types:

int	Bitwise and of arg1 and arg2.
string	The result is a string where each character is the bitwise and of the characters in the same position in arg1 and arg2. The arguments must be strings of the same length.
array|mapping|multiset	The result is like arg1 but only with the elements/indices that match any in arg2 (according to `>, `<, `== and, in the case of mappings, hash_value).
type|program	Type intersection of arg1 and arg2.

The function is not destructive on the arguments - the result is always a new instance.

Note

If this operator is used with arrays or multisets containing objects which implement lfun::`==() but not lfun::`>() and lfun::`<(), the result will be undefined.

See also

`|(), lfun::`&(), lfun::``&()

Method `|

mixed `|(mixed arg1)
mixed `|(mixed arg1, mixed arg2, mixed ... extras)
mixed `|(object arg1, mixed arg2)
mixed `|(mixed arg1, object arg2)
int `|(int arg1, int arg2)
string `|(string arg1, string arg2)
array `|(array arg1, array arg2)
mapping `|(mapping arg1, mapping arg2)
multiset `|(multiset arg1, multiset arg2)
type `|(program|type arg1, program|type arg2)

Description

Bitwise or/union.

Every expression with the | operator becomes a call to this function, i.e. a|b is the same as predef::`|(a,b).

Returns

If there's a single argument, that argument is returned.

If there are more than two arguments, the result is: `|(`|(arg1, arg2), @extras).

Otherwise, if arg1 is an object with an lfun::`|(), that function is called with arg2 as argument, and its result is returned.

Otherwise, if arg2 is an object with an lfun::``|(), that function is called with arg1 as argument, and its result is returned.

Otherwise the result depends on the argument types:

arg1 can have any of the following types:

int	Bitwise or of arg1 and arg2.
zero	UNDEFINED may be or:ed with multisets, behaving as if it was an empty multiset.
string	The result is a string where each character is the bitwise or of the characters in the same position in arg1 and arg2. The arguments must be strings of the same length.
array	The result is an array with the elements in arg1 concatenated with those in arg2 that doesn't occur in arg1 (according to `>, `<, `==). The order between the elements that come from the same argument is kept. Every element in arg1 is only matched once against an element in arg2, so if arg2 contains several elements that are equal to each other and are more than their counterparts in arg1, the rightmost remaining elements in arg2 are kept.
mapping	The result is like arg1 but extended with the entries from arg2. If the same index (according to hash_value and `==) occur in both, the value from arg2 is used.
multiset	The result is like arg1 but extended with the entries in arg2 that don't already occur in arg1 (according to `>, `< and `==). Subsequences with orderwise equal entries (i.e. where `< returns false) are handled just like the array case above.
type|program	Type union of arg1 and arg2.

The function is not destructive on the arguments - the result is always a new instance.

Note

If this operator is used with arrays or multisets containing objects which implement lfun::`==() but not lfun::`>() and lfun::`<(), the result will be undefined.

The treatment of UNDEFINED with multisets was new in Pike 8.1.

See also

`&(), lfun::`|(), lfun::``|()

Method `^

mixed `^(mixed arg1)
mixed `^(mixed arg1, mixed arg2, mixed ... extras)
mixed `^(object arg1, mixed arg2)
mixed `^(mixed arg1, object arg2)
int `^(int arg1, int arg2)
string `^(string arg1, string arg2)
array `^(array arg1, array arg2)
mapping `^(mapping arg1, mapping arg2)
multiset `^(multiset arg1, multiset arg2)
type `^(program|type arg1, program|type arg2)

Description

Exclusive or.

Every expression with the ^ operator becomes a call to this function, i.e. a^b is the same as predef::`^(a,b).

Returns

If there's a single argument, that argument is returned.

If there are more than two arguments, the result is: `^(`^(arg1, arg2), @extras).

Otherwise, if arg1 is an object with an lfun::`^(), that function is called with arg2 as argument, and its result is returned.

Otherwise, if arg2 is an object with an lfun::``^(), that function is called with arg1 as argument, and its result is returned.

Otherwise the result depends on the argument types:

arg1 can have any of the following types:

int	Bitwise exclusive or of arg1 and arg2.
string	The result is a string where each character is the bitwise exclusive or of the characters in the same position in arg1 and arg2. The arguments must be strings of the same length.
array	The result is an array with the elements in arg1 that doesn't occur in arg2 concatenated with those in arg2 that doesn't occur in arg1 (according to `>, `< and `==). The order between the elements that come from the same argument is kept. Every element is only matched once against an element in the other array, so if one contains several elements that are equal to each other and are more than their counterparts in the other array, the rightmost remaining elements are kept.
mapping	The result is like arg1 but with the entries from arg1 and arg2 whose indices are different between them (according to hash_value and `==).
multiset	The result is like arg1 but with the entries from arg1 and arg2 that are different between them (according to `>, `< and `==). Subsequences with orderwise equal entries (i.e. where `< returns false) are handled just like the array case above.
type|program	The result is a type computed like this: (arg1&~arg2)|(~arg1&arg2).

The function is not destructive on the arguments - the result is always a new instance.

Note

If this operator is used with arrays or multisets containing objects which implement lfun::`==() but not lfun::`>() and lfun::`<(), the result will be undefined.

See also

`&(), `|(), lfun::`^(), lfun::``^()

The index and range operators are used to retrieve information from a complex data type.

The index operator can be written in two different ways. It can be written as ob [ index ] or ob->identifier. However, the latter syntax is equal to ob [ "identifier" ].

You can only index strings, arrays, mapping, multisets and objects, and some of these can only be indexed on certain things as shown in this list:

string[int]

array[int]

array[int]=mixed

mapping[mixed]
mapping->identifier

mapping[mixed]=mixed
mapping->identifier=mixed

multiset[mixed]
multiset->identifier

multiset[mixed]=mixed
multiset->identifier=mixed

object[string]
object->identifier

object[string]=mixed
object->identifier=mixed

program[string]
program->identifier

string[int..int]

array[int..int]

When indexing an array or string it is sometimes convenient to access index from the end instead of from the beginning. This function can be performed by using a negative index. Thus  arr[-i]  is the same as arr[sizeof(arr)-i]. Note however that this behavior does not apply to the range operator. Instead the range operator clamps it's arguments to a suitable range. This means that a[b..c] will be treated as follows:

• If b is less than zero, the range will begin at the start of the array as if b had been zero.
• If b is greater or equal to sizeof(a) an empty array/string will be returned.
• If c is less than b, an empty array/string will be returned.
• If c is greater or equal to sizeof(a) the range will continue to the end of the array/string.
• No errors are generated in any of the above cases.

Method `[]

mixed `[](object arg, mixed index)
mixed `[](object arg, string index)
function(:void) `[](int arg, string index)
int `[](string arg, int index)
mixed `[](array arg, int index)
mixed `[](array arg, mixed index)
mixed `[](mapping arg, mixed index)
bool `[](multiset arg, mixed index)
mixed `[](program arg, string index)
mixed `[](object arg, mixed start, mixed end)
string `[](string arg, int start, int end)
array `[](array arg, int start, int end)

Description

Indexing.

This is the function form of expressions with the [] operator, i.e. a[i] is the same as predef::`[](a,i).

Returns

If arg is an object that implements lfun::`[](), that function is called with the index argument.

Otherwise, the action depends on the type of arg:

arg can have any of the following types:

object	The non-protected (i.e. public) symbol named index is looked up in arg.
int	The bignum function named index is looked up in arg. The bignum functions are the same as those in the Gmp.mpz class.
string	The character at index index in arg is returned as an integer. The first character in the string is at index 0 and the highest allowed index is therefore sizeof(arg)-1. A negative index number accesses the string from the end instead, from -1 for the last char back to -sizeof(arg) for the first.
array	If index is an int, index number index of arg is returned. Allowed index number are in the range [-sizeof(arg)..sizeof(arg)-1]; see the string case above for details. If index is not an int, an array of all elements in arg indexed with index are returned. I.e. it's the same as doing column(arg, index).
mapping	If index exists in arg the corresponding value is returned. Otherwise UNDEFINED is returned.
multiset	If index exists in arg, 1 is returned. Otherwise UNDEFINED is returned.
program	The non-protected (i.e. public) constant symbol index is looked up in arg.

As a compatibility measure, this function also performs range operations if it's called with three arguments. In that case it becomes equivalent to:

`[..] (arg, start, Pike.INDEX_FROM_BEG, end, Pike.INDEX_FROM_BEG)

See `[..] for further details.

Note

An indexing expression in an lvalue context, i.e. where the index is being assigned a new value, uses `[]= instead of this function.

See also

`->(), lfun::`[](), `[]=, `[..]

Method `->

mixed `->(object arg, string index)
mixed `->(int arg, string index)
mixed `->(array arg, string index)
mixed `->(mapping arg, string index)
bool `->(multiset arg, string index)
mixed `->(program arg, string index)

Description

Arrow indexing.

Every non-lvalue expression with the -> operator becomes a call to this function. a->b is the same as predef::`^(a,"b") where "b" is the symbol b in string form.

This function behaves like `[], except that the index is passed literally as a string instead of being evaluated.

Returns

If arg is an object that implements lfun::`->(), that function will be called with index as the single argument.

Otherwise the result will be as follows:

arg can have any of the following types:

object	The non-protected (ie public) symbol named index will be looked up in arg.
int	The bignum function named index will be looked up in arg.
array	An array of all elements in arg arrow indexed with index will be returned.
mapping	If index exists in arg the corresponding value will be returned. Otherwise UNDEFINED will be returned.
multiset	If index exists in arg, 1 will be returned. Otherwise UNDEFINED will be returned.
program	The non-protected (ie public) constant symbol index will be looked up in arg.

Note

In an expression a->b, the symbol b can be any token that matches the identifier syntax - keywords are disregarded in that context.

Note

An arrow indexing expression in an lvalue context, i.e. where the index is being assigned a new value, uses `->= instead of this function.

See also

`[](), lfun::`->(), ::`->(), `->=

Method `[]=

mixed `[]=(object arg, mixed index, mixed val)
mixed `[]=(object arg, string index, mixed val)
mixed `[]=(array arg, int index, mixed val)
mixed `[]=(mapping arg, mixed index, mixed val)
bool `[]=(multiset arg, mixed index, bool val)

Description

Index assignment.

Every lvalue expression with the [] operator becomes a call to this function, i.e. a[b]=c is the same as predef::`[]=(a,b,c).

If arg is an object that implements lfun::`[]=(), that function will be called with index and val as the arguments.

arg can have any of the following types:

object	The non-protected (ie public) variable named index will be looked up in arg, and assigned val.
array|mapping	Index index in arg will be assigned val.
multiset	If val is 0 (zero), one occurrance of index in arg will be removed. Otherwise index will be added to arg if it is not already there.

Returns

val will be returned.

Note

An indexing expression in a non-lvalue context, i.e. where the index is being queried instead of assigned, uses `[] instead of this function.

See also

`->=(), lfun::`[]=(), `[]

Method `->=

mixed `->=(object arg, string index, mixed val)
mixed `->=(mapping arg, string index, mixed val)
bool `->=(multiset arg, string index, bool val)

Description

Arrow index assignment.

Every lvalue expression with the -> operator becomes a call to this function, i.e. a->b=c is the same as predef::`->=(a,"b",c) where "b" is the symbol b in string form.

This function behaves like `[]=, except that the index is passed literally as a string instead of being evaluated.

If arg is an object that implements lfun::`->=(), that function will be called with index and val as the arguments.

arg can have any of the following types:

object	The non-protected (ie public) variable named index will be looked up in arg, and assigned val.
array|mapping	Index index in arg will be assigned val.
multiset	If val is 0 (zero), one occurrance of index in arg will be removed. Otherwise index will be added to arg if it is not already there.

Returns

val will be returned.

Note

In an expression a->b=c, the symbol b can be any token that matches the identifier syntax - keywords are disregarded in that context.

Note

An arrow indexing expression in a non-lvalue context, i.e. where the index is being queried instead of assigned, uses `-> instead of this function.

See also

`[]=(), lfun::`->=(), `->

Method `[..]

mixed `[..](object arg, mixed start, int start_type, mixed end, int end_type)
string `[..](string arg, int start, int start_type, int end, int end_type)
array `[..](array arg, int start, int start_type, int end, int end_type)

Description

Extracts a subrange.

This is the function form of expressions with the [..] operator. arg is the thing from which the subrange is to be extracted. start is the lower bound of the subrange and end the upper bound.

start_type and end_type specifies how the start and end indices, respectively, are to be interpreted. The types are either Pike.INDEX_FROM_BEG, Pike.INDEX_FROM_END or Pike.OPEN_BOUND. In the last case, the index value is insignificant.

The relation between [..] expressions and this function is therefore as follows:

a[i..j]    <=>	`[..] (a, i, Pike.INDEX_FROM_BEG, j, Pike.INDEX_FROM_BEG)
    a[i..<j]   <=>	`[..] (a, i, Pike.INDEX_FROM_BEG, j, Pike.INDEX_FROM_END)
    a[i..]     <=>	`[..] (a, i, Pike.INDEX_FROM_BEG, 0, Pike.OPEN_BOUND)
    a[<i..j]   <=>	`[..] (a, i, Pike.INDEX_FROM_END, j, Pike.INDEX_FROM_BEG)
    a[<i..<j]  <=>	`[..] (a, i, Pike.INDEX_FROM_END, j, Pike.INDEX_FROM_END)
    a[<i..]    <=>	`[..] (a, i, Pike.INDEX_FROM_END, 0, Pike.OPEN_BOUND)
    a[..j]     <=>	`[..] (a, 0, Pike.OPEN_BOUND, j, Pike.INDEX_FROM_BEG)
    a[..<j]    <=>	`[..] (a, 0, Pike.OPEN_BOUND, j, Pike.INDEX_FROM_END)
    a[..]      <=>	`[..] (a, 0, Pike.OPEN_BOUND, 0, Pike.OPEN_BOUND)

The subrange is specified as follows by the two bounds:

• If the lower bound refers to an index before the lowest allowable (typically zero) then it's taken as an open bound which starts at the first index (without any error).

• Correspondingly, if the upper bound refers to an index past the last allowable then it's taken as an open bound which ends at the last index (without any error).

• If the lower bound is less than or equal to the upper bound, then the subrange is the inclusive range between them, i.e. from and including the element at the lower bound and up to and including the element at the upper bound.

• If the lower bound is greater than the upper bound then the result is an empty subrange (without any error).

Returns

The returned value depends on the type of arg:

arg can have any of the following types:

string	A string with the characters in the range is returned.
array	An array with the elements in the range is returned.
object	If the object implements lfun::`[..], that function is called with the four remaining arguments. As a compatibility measure, if the object does not implement lfun::`[..] but lfun::`[] then the latter is called with the bounds transformed to normal from-the-beginning indices in array-like fashion: `[..] (a, i, Pike.INDEX_FROM_BEG, j, Pike.INDEX_FROM_BEG) Calls a->`[] (i, j) `[..] (a, i, Pike.INDEX_FROM_BEG, j, Pike.INDEX_FROM_END) Calls a->`[] (i, a->_sizeof()-1-j) `[..] (a, i, Pike.INDEX_FROM_BEG, 0, Pike.OPEN_BOUND) Calls a->`[] (i, Int.NATIVE_MAX) `[..] (a, i, Pike.INDEX_FROM_END, j, Pike.INDEX_FROM_BEG) Calls a->`[] (a->_sizeof()-1-i, j) `[..] (a, i, Pike.INDEX_FROM_END, j, Pike.INDEX_FROM_END) Calls a->`[] (a->_sizeof()-1-i, a->_sizeof()-1-j), except that a->_sizeof() is called only once. `[..] (a, i, Pike.INDEX_FROM_END, 0, Pike.OPEN_BOUND) Calls a->`[] (a->_sizeof()-1-i, Int.NATIVE_MAX) `[..] (a, 0, Pike.OPEN_BOUND, j, Pike.INDEX_FROM_BEG) Calls a->`[] (0, j) `[..] (a, 0, Pike.OPEN_BOUND, j, Pike.INDEX_FROM_END) Calls a->`[] (0, a->_sizeof()-1-j) `[..] (a, 0, Pike.OPEN_BOUND, 0, Pike.OPEN_BOUND) Calls a->`[] (0, Int.NATIVE_MAX) Note that Int.NATIVE_MAX might be replaced with an even larger integer in the future.

See also

lfun::`[..], `[]

When indexing data-structures, it is common for some elements to be optional and thus evaluate to 0. To simplify deep indexing of such stuctures, the conditional (aka "safe") indexing operators exist (->?, [?] and (?)). They behave like the corresponding "unsafe" operator (ie ->, [] and ()), except for short-circuiting the remainder of the expression to 0 when the indexed value is 0.

Examples:

    mapping m = ([
      "a":([
        "x": ([]),
      ]),
      "b":0,
    ]);
  

There is really only one assignment operator, but it can be combined with lots of other operators to make the code shorter. An assignment looks like this:

	variable = expression;

	variable1 = variable2 = 1; // Assign 1 to both variables
	variable1 =(variable2 = 1); // Same as above

	// Write the value of the expression, if any
	if(variable = expression)
	  write(variable);

In all of the above expressions variable can actually be any of type of assignable values. Assignable values are also known as lvalues and here is a list of lvalues:

Now there are only a couple of operators left. I have grouped them together in this section, not because they are not important, but because they do not fit in any particular categories.

The most important of these operators is the calling operator. It is used to call functions. The operator itself is just a set of parenthesis placed after the expression that returns the function. Any arguments to the function should be placed between the parenthesis, separated by commas. We have already seen many examples of this operator, although you might not have realized it was an operator at the time. The function call operator can do more than just calling functions though; if the 'function' is in fact an array, the operator will loop over the array and call each element in the array and returns an array with the results.

If on the other hand, the 'function' is a program, the operator will clone an object from the program and call create() in the new object with the arguments given. In fact, the function clone is implemented like this:

object clone(mixed p, mixed ... args) { ( (program)p )(@args); }

Then there are the increment and decrement operators. The increment and decrement operators are somewhat limited: they can only be used on integers. They provide a short and fast way to add or subtract one to an integer. If the operator is written before the variable (++a) the returned value will be what the variable is after the operator has added/subtracted one to it. If the operator is after the variable (a++) it will instead return the value of the variable before it was incremented/decremented.

Casting is used to convert one type to another, not all casts are possible. Here is a list of all casts that actually _do_ anything:

You can also use the cast operator to tell the compiler things. If a is a variable of type mixed containing an int, then the expression (int)a can be used instead of a and that will tell the compiler that the type of that expression is int.

Last, and in some respect least, is the comma operator. It doesn't do much. In fact, it simply evaluates the two arguments and then returns the right hand one. This operator is mostly useful to produce smaller code, or to make defines that can be used in expressions.

Method `()
Method call_function

mixed `()(function(:void) fun, mixed ... args)
mixed call_function(function(:void) fun, mixed ... args)

Description

Call a function.

Calls the function fun with the arguments specified by args.

See also

lfun::`()()

When evaluating an expression, you can always use parenthesis to tell the compiler in which order to evaluate things. Normally, the compiler will evaluate things from left to right, but it will evaluate operators with higher priority before those with lower. The following table shows the relative priority of all the operators in descending order:

Examples:

As mentioned earlier a + b can just as well be written as `+(a, b). Together with the function map which calls a function for every index in an array and the splice operator this can be used to create some very very fast and compact code. Let's look at some examples:

map(arr, `-)

This will return an array with each element negated.

map(text/"\n",`/," ")

This will divide a text into lines, each line will then be mapped through `/ and divided into an array of words.

`+(0, @arr)

This will add all the integers in the array arr together.

int abs(int a) { return ( a>0 ? `+ : `-)(a); }

This is a rather absurd but working function which will return the absolute value of a.