manual/manual.of

@Ci{$Id: manual.of $}
@C{[(-------------------------------------------------------------------------}
@manual{

@sect1{@title{Introduction}

Lua is a powerful, efficient, lightweight, embeddable scripting language.
It supports procedural programming,
object-oriented programming, functional programming,
data-driven programming, and data description.

Lua combines simple procedural syntax with powerful data description
constructs based on associative arrays and extensible semantics.
Lua is dynamically typed,
runs by interpreting bytecode with a register-based
virtual machine,
and has automatic memory management with
a generational garbage collection,
making it ideal for configuration, scripting,
and rapid prototyping.

Lua is implemented as a library, written in @emphx{clean C},
the common subset of @N{standard C} and C++.
The Lua distribution includes a host program called @id{lua},
which uses the Lua library to offer a complete,
standalone Lua interpreter,
for interactive or batch use.
Lua is intended to be used both as a powerful, lightweight,
embeddable scripting language for any program that needs one,
and as a powerful but lightweight and efficient stand-alone language.

As an extension language, Lua has no notion of a @Q{main} program:
it works @emph{embedded} in a host client,
called the @emph{embedding program} or simply the @emphx{host}.
(Frequently, this host is the stand-alone @id{lua} program.)
The host program can invoke functions to execute a piece of Lua code,
can write and read Lua variables,
and can register @N{C functions} to be called by Lua code.
Through the use of @N{C functions}, Lua can be augmented to cope with
a wide range of different domains,
thus creating customized programming languages sharing a syntactical framework.

Lua is free software,
and is provided as usual with no guarantees,
as stated in its license.
The implementation described in this manual is available
at Lua's official web site, @id{www.lua.org}.

Like any other reference manual,
this document is dry in places.
For a discussion of the decisions behind the design of Lua,
see the technical papers available at Lua's web site.
For a detailed introduction to programming in Lua,
see Roberto's book, @emphx{Programming in Lua}.

}


@C{-------------------------------------------------------------------------}
@sect1{basic| @title{Basic Concepts}

@simplesect{

This section describes the basic concepts of the language.

}

@sect2{TypesSec| @title{Values and Types}

Lua is a dynamically typed language.
This means that
variables do not have types; only values do.
There are no type definitions in the language.
All values carry their own type.

All values in Lua are first-class values.
This means that all values can be stored in variables,
passed as arguments to other functions, and returned as results.

There are eight @x{basic types} in Lua:
@def{nil}, @def{boolean}, @def{number},
@def{string}, @def{function}, @def{userdata},
@def{thread}, and @def{table}.
The type @emph{nil} has one single value, @nil,
whose main property is to be different from any other value;
it often represents the absence of a useful value.
The type @emph{boolean} has two values, @false and @true.
Both @nil and @false make a condition false;
they are collectively called @def{false values}.
Any other value makes a condition true.
Despite its name,
@false is frequently used as an alternative to @nil,
with the key difference that @false behaves
like a regular value in a table,
while a @nil in a table represents an absent key.

The type @emph{number} represents both
integer numbers and real (floating-point) numbers,
using two @x{subtypes}: @def{integer} and @def{float}.
Standard Lua uses 64-bit integers and double-precision (64-bit) floats,
but you can also compile Lua so that it
uses 32-bit integers and/or single-precision (32-bit) floats.
The option with 32 bits for both integers and floats
is particularly attractive
for small machines and embedded systems.
(See macro @id{LUA_32BITS} in file @id{luaconf.h}.)

Unless stated otherwise,
any overflow when manipulating integer values @def{wrap around},
according to the usual rules of two-complement arithmetic.
(In other words,
the actual result is the unique representable integer
that is equal modulo @M{2@sp{n}} to the mathematical result,
where @M{n} is the number of bits of the integer type.)

Lua has explicit rules about when each subtype is used,
but it also converts between them automatically as needed @see{coercion}.
Therefore,
the programmer may choose to mostly ignore the difference
between integers and floats
or to assume complete control over the representation of each number.

The type @emph{string} represents immutable sequences of bytes.
@index{eight-bit clean}
Lua is 8-bit clean:
strings can contain any 8-bit value,
including @x{embedded zeros} (@Char{\0}).
Lua is also encoding-agnostic;
it makes no assumptions about the contents of a string.
The length of any string in Lua must fit in a Lua integer.

Lua can call (and manipulate) functions written in Lua and
functions written in C @see{functioncall}.
Both are represented by the type @emph{function}.

The type @emph{userdata} is provided to allow arbitrary @N{C data} to
be stored in Lua variables.
A userdata value represents a block of raw memory.
There are two kinds of userdata:
@emphx{full userdata},
which is an object with a block of memory managed by Lua,
and @emphx{light userdata},
which is simply a @N{C pointer} value.
Userdata has no predefined operations in Lua,
except assignment and identity test.
By using @emph{metatables},
the programmer can define operations for full userdata values
@see{metatable}.
Userdata values cannot be created or modified in Lua,
only through the @N{C API}.
This guarantees the integrity of data owned by
the host program and @N{C libraries}.

The type @def{thread} represents independent threads of execution
and it is used to implement coroutines @see{coroutine}.
Lua threads are not related to operating-system threads.
Lua supports coroutines on all systems,
even those that do not support threads natively.

The type @emph{table} implements @x{associative arrays},
that is, @x{arrays} that can have as indices not only numbers,
but any Lua value except @nil and @x{NaN}.
(@emphx{Not a Number} is a special floating-point value
used by the @x{IEEE 754} standard to represent
undefined numerical results, such as @T{0/0}.)
Tables can be @emph{heterogeneous};
that is, they can contain values of all types (except @nil).
Any key associated to the value @nil is not considered part of the table.
Conversely, any key that is not part of a table has
an associated value @nil.

Tables are the sole data-structuring mechanism in Lua;
they can be used to represent ordinary arrays, lists,
symbol tables, sets, records, graphs, trees, etc.
To represent @x{records}, Lua uses the field name as an index.
The language supports this representation by
providing @id{a.name} as syntactic sugar for @T{a["name"]}.
There are several convenient ways to create tables in Lua
@see{tableconstructor}.

Like indices,
the values of table fields can be of any type.
In particular,
because functions are first-class values,
table fields can contain functions.
Thus tables can also carry @emph{methods} @see{func-def}.

The indexing of tables follows
the definition of raw equality in the language.
The expressions @T{a[i]} and @T{a[j]}
denote the same table element
if and only if @id{i} and @id{j} are raw equal
(that is, equal without metamethods).
In particular, floats with integral values
are equal to their respective integers
(e.g., @T{1.0 == 1}).
To avoid ambiguities,
any float used as a key that is equal to an integer
is converted to that integer.
For instance, if you write @T{a[2.0] = true},
the actual key inserted into the table will be the integer @T{2}.


Tables, functions, threads, and (full) userdata values are @emph{objects}:
variables do not actually @emph{contain} these values,
only @emph{references} to them.
Assignment, parameter passing, and function returns
always manipulate references to such values;
these operations do not imply any kind of copy.

The library function @Lid{type} returns a string describing the type
of a given value @seeF{type}.

}

@sect2{globalenv| @title{Environments and the Global Environment}

As we will discuss further in @refsec{variables} and @refsec{assignment},
any reference to a free name
(that is, a name not bound to any declaration) @id{var}
is syntactically translated to @T{_ENV.var}.
Moreover, every chunk is compiled in the scope of
an external local variable named @id{_ENV} @see{chunks},
so @id{_ENV} itself is never a free name in a chunk.

Despite the existence of this external @id{_ENV} variable and
the translation of free names,
@id{_ENV} is a completely regular name.
In particular,
you can define new variables and parameters with that name.
Each reference to a free name uses the @id{_ENV} that is
visible at that point in the program,
following the usual visibility rules of Lua @see{visibility}.

Any table used as the value of @id{_ENV} is called an @def{environment}.

Lua keeps a distinguished environment called the @def{global environment}.
This value is kept at a special index in the C registry @see{registry}.
In Lua, the global variable @Lid{_G} is initialized with this same value.
(@Lid{_G} is never used internally,
so changing its value will affect only your own code.)

When Lua loads a chunk,
the default value for its @id{_ENV} variable
is the global environment @seeF{load}.
Therefore, by default,
free names in Lua code refer to entries in the global environment
and, therefore, they are also called @def{global variables}.
Moreover, all standard libraries are loaded in the global environment
and some functions there operate on that environment.
You can use @Lid{load} (or @Lid{loadfile})
to load a chunk with a different environment.
(In C, you have to load the chunk and then change the value
of its first upvalue; see @See{lua_setupvalue}.)

}

@sect2{error| @title{Error Handling}

Several operations in Lua can @emph{raise} an error.
An error interrupts the normal flow of the program,
which can continue by @emph{catching} the error.

Lua code can explicitly raise an error by calling the
@Lid{error} function.
(This function never returns.)

To catch errors in Lua,
you can do a @def{protected call},
using @Lid{pcall} (or @Lid{xpcall}).
The function @Lid{pcall} calls a given function in @def{protected mode}.
Any error while running the function stops its execution,
and control returns immediately to @id{pcall},
which returns a status code.

Because Lua is an embedded extension language,
Lua code starts running by a call
from @N{C code} in the host program.
(When you use Lua standalone,
the @id{lua} application is the host program.)
Usually, this call is protected;
so, when an otherwise unprotected error occurs during
the compilation or execution of a Lua chunk,
control returns to the host,
which can take appropriate measures,
such as printing an error message.

Whenever there is an error,
an @def{error object}
is propagated with information about the error.
Lua itself only generates errors whose error object is a string,
but programs can generate errors with
any value as the error object.
It is up to the Lua program or its host to handle such error objects.
For historical reasons,
an error object is often called an @def{error message},
even though it does not have to be a string.


When you use @Lid{xpcall} (or @Lid{lua_pcall}, in C)
you can give a @def{message handler}
to be called in case of errors.
This function is called with the original error object
and returns a new error object.
It is called before the error unwinds the stack,
so that it can gather more information about the error,
for instance by inspecting the stack and creating a stack traceback.
This message handler is still protected by the protected call;
so, an error inside the message handler
will call the message handler again.
If this loop goes on for too long,
Lua breaks it and returns an appropriate message.
The message handler is called only for regular runtime errors.
It is not called for memory-allocation errors
nor for errors while running finalizers or other message handlers.

Lua also offers a system of @emph{warnings} @seeF{warn}.
Unlike errors, warnings do not interfere
in any way with program execution.
They typically only generate a message to the user,
although this behavior can be adapted from C @seeC{lua_setwarnf}.

}

@sect2{metatable| @title{Metatables and Metamethods}

Every value in Lua can have a @emph{metatable}.
This @def{metatable} is an ordinary Lua table
that defines the behavior of the original value
under certain events.
You can change several aspects of the behavior
of a value by setting specific fields in its metatable.
For instance, when a non-numeric value is the operand of an addition,
Lua checks for a function in the field @idx{__add} of the value's metatable.
If it finds one,
Lua calls this function to perform the addition.

The key for each event in a metatable is a string
with the event name prefixed by two underscores;
the corresponding value is called a @def{metavalue}.
For most events, the metavalue must be a function,
which is then called a @def{metamethod}.
In the previous example, the key is the string @St{__add}
and the metamethod is the function that performs the addition.
Unless stated otherwise,
a metamethod can in fact be any @x{callable value},
which is either a function or a value with a @idx{__call} metamethod.

You can query the metatable of any value
using the @Lid{getmetatable} function.
Lua queries metamethods in metatables using a raw access @seeF{rawget}.

You can replace the metatable of tables
using the @Lid{setmetatable} function.
You cannot change the metatable of other types from Lua code,
except by using the @link{debuglib|debug library}.

Tables and full userdata have individual metatables,
although multiple tables and userdata can share their metatables.
Values of all other types share one single metatable per type;
that is, there is one single metatable for all numbers,
one for all strings, etc.
By default, a value has no metatable,
but the string library sets a metatable for the string type @see{strlib}.

A detailed list of operations controlled by metatables is given next.
Each event is identified by its corresponding key.
By convention, all metatable keys used by Lua are composed by
two underscores followed by lowercase Latin letters.

@description{

@item{@idx{__add}|
the addition (@T{+}) operation.
If any operand for an addition is not a number,
Lua will try to call a metamethod.
It starts by checking the first operand (even if it is a number);
if that operand does not define a metamethod for @idx{__add},
then Lua will check the second operand.
If Lua can find a metamethod,
it calls the metamethod with the two operands as arguments,
and the result of the call
(adjusted to one value)
is the result of the operation.
Otherwise, if no metamethod is found,
Lua raises an error.
}

@item{@idx{__sub}|
the subtraction (@T{-}) operation.
Behavior similar to the addition operation.
}

@item{@idx{__mul}|
the multiplication (@T{*}) operation.
Behavior similar to the addition operation.
}

@item{@idx{__div}|
the division (@T{/}) operation.
Behavior similar to the addition operation.
}

@item{@idx{__mod}|
the modulo (@T{%}) operation.
Behavior similar to the addition operation.
}

@item{@idx{__pow}|
the exponentiation (@T{^}) operation.
Behavior similar to the addition operation.
}

@item{@idx{__unm}|
the negation (unary @T{-}) operation.
Behavior similar to the addition operation.
}

@item{@idx{__idiv}|
the floor division (@T{//}) operation.
Behavior similar to the addition operation.
}

@item{@idx{__band}|
the bitwise AND (@T{&}) operation.
Behavior similar to the addition operation,
except that Lua will try a metamethod
if any operand is neither an integer
nor a float coercible to an integer @see{coercion}.
}

@item{@idx{__bor}|
the bitwise OR (@T{|}) operation.
Behavior similar to the bitwise AND operation.
}

@item{@idx{__bxor}|
the bitwise exclusive OR (binary @T{~}) operation.
Behavior similar to the bitwise AND operation.
}

@item{@idx{__bnot}|
the bitwise NOT (unary @T{~}) operation.
Behavior similar to the bitwise AND operation.
}

@item{@idx{__shl}|
the bitwise left shift (@T{<<}) operation.
Behavior similar to the bitwise AND operation.
}

@item{@idx{__shr}|
the bitwise right shift (@T{>>}) operation.
Behavior similar to the bitwise AND operation.
}

@item{@idx{__concat}|
the concatenation (@T{..}) operation.
Behavior similar to the addition operation,
except that Lua will try a metamethod
if any operand is neither a string nor a number
(which is always coercible to a string).
}

@item{@idx{__len}|
the length (@T{#}) operation.
If the object is not a string,
Lua will try its metamethod.
If there is a metamethod,
Lua calls it with the object as argument,
and the result of the call
(always adjusted to one value)
is the result of the operation.
If there is no metamethod but the object is a table,
then Lua uses the table length operation @see{len-op}.
Otherwise, Lua raises an error.
}

@item{@idx{__eq}|
the equal (@T{==}) operation.
Behavior similar to the addition operation,
except that Lua will try a metamethod only when the values
being compared are either both tables or both full userdata
and they are not primitively equal.
The result of the call is always converted to a boolean.
}

@item{@idx{__lt}|
the less than (@T{<}) operation.
Behavior similar to the addition operation,
except that Lua will try a metamethod only when the values
being compared are neither both numbers nor both strings.
Moreover, the result of the call is always converted to a boolean.
}

@item{@idx{__le}|
the less equal (@T{<=}) operation.
Behavior similar to the less than operation.
}

@item{@idx{__index}|
The indexing access operation @T{table[key]}.
This event happens when @id{table} is not a table or
when @id{key} is not present in @id{table}.
The metavalue is looked up in the metatable of @id{table}.

The metavalue for this event can be either a function, a table,
or any value with an @idx{__index} metavalue.
If it is a function,
it is called with @id{table} and @id{key} as arguments,
and the result of the call
(adjusted to one value)
is the result of the operation.
Otherwise,
the final result is the result of indexing this metavalue with @id{key}.
This indexing is regular, not raw,
and therefore can trigger another @idx{__index} metavalue.
}

@item{@idx{__newindex}|
The indexing assignment @T{table[key] = value}.
Like the index event,
this event happens when @id{table} is not a table or
when @id{key} is not present in @id{table}.
The metavalue is looked up in the metatable of @id{table}.

Like with indexing,
the metavalue for this event can be either a function, a table,
or any value with an @idx{__newindex} metavalue.
If it is a function,
it is called with @id{table}, @id{key}, and @id{value} as arguments.
Otherwise,
Lua repeats the indexing assignment over this metavalue
with the same key and value.
This assignment is regular, not raw,
and therefore can trigger another @idx{__newindex} metavalue.

Whenever a @idx{__newindex} metavalue is invoked,
Lua does not perform the primitive assignment.
If needed,
the metamethod itself can call @Lid{rawset}
to do the assignment.
}

@item{@idx{__call}|
The call operation @T{func(args)}.
This event happens when Lua tries to call a non-function value
(that is, @id{func} is not a function).
The metamethod is looked up in @id{func}.
If present,
the metamethod is called with @id{func} as its first argument,
followed by the arguments of the original call (@id{args}).
All results of the call
are the results of the operation.
This is the only metamethod that allows multiple results.
}

}

In addition to the previous list,
the interpreter also respects the following keys in metatables:
@idx{__gc} @see{finalizers},
@idx{__close} @see{to-be-closed},
@idx{__mode} @see{weak-table},
and @idx{__name}.
(The entry @idx{__name},
when it contains a string,
may be used by @Lid{tostring} and in error messages.)

For the unary operators (negation, length, and bitwise NOT),
the metamethod is computed and called with a dummy second operand,
equal to the first one.
This extra operand is only to simplify Lua's internals
(by making these operators behave like a binary operation)
and may be removed in future versions.
For most uses this extra operand is irrelevant.

Because metatables are regular tables,
they can contain arbitrary fields,
not only the event names defined above.
Some functions in the standard library
(e.g., @Lid{tostring})
use other fields in metatables for their own purposes.

It is a good practice to add all needed metamethods to a table
before setting it as a metatable of some object.
In particular, the @idx{__gc} metamethod works only when this order
is followed @see{finalizers}.
It is also a good practice to set the metatable of an object
right after its creation.

}

@sect2{GC| @title{Garbage Collection}

@simplesect{

Lua performs automatic memory management.
This means that
you do not have to worry about allocating memory for new objects
or freeing it when the objects are no longer needed.
Lua manages memory automatically by running
a @def{garbage collector} to collect all @emph{dead} objects.
All memory used by Lua is subject to automatic management:
strings, tables, userdata, functions, threads, internal structures, etc.

An object is considered @def{dead}
as soon as the collector can be sure the object
will not be accessed again in the normal execution of the program.
(@Q{Normal execution} here excludes finalizers,
which resurrect dead objects @see{finalizers},
and it excludes also some operations using the debug library.)
Note that the time when the collector can be sure that an object
is dead may not coincide with the programmer's expectations.
The only guarantees are that Lua will not collect an object
that may still be accessed in the normal execution of the program,
and it will eventually collect an object
that is inaccessible from Lua.
(Here,
@emph{inaccessible from Lua} means that neither a variable nor
another live object refer to the object.)
Because Lua has no knowledge about @N{C code},
it never collects objects accessible through the registry @see{registry},
which includes the global environment @see{globalenv} and
the main thread.


The garbage collector (GC) in Lua can work in two modes:
incremental and generational.

The default GC mode with the default parameters
are adequate for most uses.
However, programs that waste a large proportion of their time
allocating and freeing memory can benefit from other settings.
Keep in mind that the GC behavior is non-portable
both across platforms and across different Lua releases;
therefore, optimal settings are also non-portable.

You can change the GC mode and parameters by calling
@Lid{lua_gc} @N{in C}
or @Lid{collectgarbage} in Lua.
You can also use these functions to control the collector directly,
for instance to stop or restart it.

}

@sect3{incmode| @title{Incremental Garbage Collection}

In incremental mode,
each GC cycle performs a mark-and-sweep collection in small steps
interleaved with the program's execution.
In this mode,
the collector uses three numbers to control its garbage-collection cycles:
the @def{garbage-collector pause},
the @def{garbage-collector step multiplier},
and the @def{garbage-collector step size}.

The garbage-collector pause
controls how long the collector waits before starting a new cycle.
The collector starts a new cycle when the number of bytes
hits @M{n%} of the total after the previous collection.
Larger values make the collector less aggressive.
Values equal to or less than 100 mean the collector will not wait to
start a new cycle.
A value of 200 means that the collector waits for
the total number of bytes to double before starting a new cycle.

The garbage-collector step size controls the
size of each incremental step,
specifically how many bytes the interpreter allocates
before performing a step:
A value of @M{n} means the interpreter will allocate
approximately @M{n} bytes between steps.

The garbage-collector step multiplier
controls how much work each incremental step does.
A value of @M{n} means the interpreter will execute
@M{n%} @emphx{units of work} for each word allocated.
A unit of work corresponds roughly to traversing one slot
or sweeping one object.
Larger values make the collector more aggressive.
Beware that values too small can
make the collector too slow to ever finish a cycle.
As a special case, a zero value means unlimited work,
effectively producing a non-incremental, stop-the-world collector.

}

@sect3{genmode| @title{Generational Garbage Collection}

In generational mode,
the collector does frequent @emph{minor} collections,
which traverses only objects recently created.
If after a minor collection the number of bytes is above a limit,
the collector shifts to a @emph{major} collection,
which traverses all objects.
The collector will then stay doing major collections until
it detects that the program is generating enough garbage to justify
going back to minor collections.

The generational mode uses three parameters:
the @def{minor multiplier}, the @def{minor-major multiplier},
and the @def{major-minor multiplier}.

The minor multiplier controls the frequency of minor collections.
For a minor multiplier @M{x},
a new minor collection will be done when the number of bytes
grows @M{x%} larger than the number in use just
after the last major collection.
For instance, for a multiplier of 20,
the collector will do a minor collection when the number of bytes
gets 20% larger than the total after the last major collection.

The minor-major multiplier controls the shift to major collections.
For a multiplier @M{x},
the collector will shift to a major collection
when the number of bytes from old objects grows @M{x%} larger
than the total after the previous major collection.
For instance, for a multiplier of 100,
the collector will do a major collection when the number of old bytes
gets larger than twice the total after the previous major collection.
As a special case,
a value of 0 stops the collector from doing major collections.

The major-minor multiplier controls the shift back to minor collections.
For a multiplier @M{x},
the collector will shift back to minor collections
after a major collection collects at least @M{x%}
of the bytes allocated during the last cycle.
In particular, for a multiplier of 0,
the collector will immediately shift back to minor collections
after doing one major collection.

}

@sect3{finalizers| @title{Garbage-Collection Metamethods}

You can set garbage-collector metamethods for tables
and, using the @N{C API},
for full userdata @see{metatable}.
These metamethods, called @def{finalizers},
are called when the garbage collector detects that the
corresponding table or userdata is dead.
Finalizers allow you to coordinate Lua's garbage collection
with external resource management such as closing files,
network or database connections,
or freeing your own memory.

For an object (table or userdata) to be finalized when collected,
you must @emph{mark} it for finalization.
@index{mark (for finalization)}
You mark an object for finalization when you set its metatable
and the metatable has a @idx{__gc} metamethod.
Note that if you set a metatable without a @idx{__gc} field
and later create that field in the metatable,
the object will not be marked for finalization.

When a marked object becomes dead,
it is not collected immediately by the garbage collector.
Instead, Lua puts it in a list.
After the collection,
Lua goes through that list.
For each object in the list,
it checks the object's @idx{__gc} metamethod:
If it is present,
Lua calls it with the object as its single argument.

At the end of each garbage-collection cycle,
the finalizers are called in
the reverse order that the objects were marked for finalization,
among those collected in that cycle;
that is, the first finalizer to be called is the one associated
with the object marked last in the program.
The execution of each finalizer may occur at any point during
the execution of the regular code.

Because the object being collected must still be used by the finalizer,
that object (and other objects accessible only through it)
must be @emph{resurrected} by Lua.@index{resurrection}
Usually, this resurrection is transient,
and the object memory is freed in the next garbage-collection cycle.
However, if the finalizer stores the object in some global place
(e.g., a global variable),
then the resurrection is permanent.
Moreover, if the finalizer marks a finalizing object for finalization again,
its finalizer will be called again in the next cycle where the
object is dead.
In any case,
the object memory is freed only in a GC cycle where
the object is dead and not marked for finalization.

When you close a state @seeF{lua_close},
Lua calls the finalizers of all objects marked for finalization,
following the reverse order that they were marked.
If any finalizer marks objects for collection during that phase,
these marks have no effect.

Finalizers cannot yield nor run the garbage collector.
Because they can run in unpredictable times,
it is good practice to restrict each finalizer
to the minimum necessary to properly release
its associated resource.

Any error while running a finalizer generates a warning;
the error is not propagated.

}

@sect3{weak-table| @title{Weak Tables}

A @def{weak table} is a table whose elements are
@def{weak references}.
A weak reference is ignored by the garbage collector.
In other words,
if the only references to an object are weak references,
then the garbage collector will collect that object.

A weak table can have weak keys, weak values, or both.
A table with weak values allows the collection of its values,
but prevents the collection of its keys.
A table with both weak keys and weak values allows the collection of
both keys and values.
In any case, if either the key or the value is collected,
the whole pair is removed from the table.
The weakness of a table is controlled by the
@idx{__mode} field of its metatable.
This metavalue, if present, must be one of the following strings:
@St{k}, for a table with weak keys;
@St{v}, for a table with weak values;
or @St{kv}, for a table with both weak keys and values.

A table with weak keys and strong values
is also called an @def{ephemeron table}.
In an ephemeron table,
a value is considered reachable only if its key is reachable.
In particular,
if the only reference to a key comes through its value,
the pair is removed.

Any change in the weakness of a table may take effect only
at the next collect cycle.
In particular, if you change the weakness to a stronger mode,
Lua may still collect some items from that table
before the change takes effect.

Only objects that have an explicit construction
are removed from weak tables.
Values, such as numbers and @x{light @N{C functions}},
are not subject to garbage collection,
and therefore are not removed from weak tables
(unless their associated values are collected).
Although strings are subject to garbage collection,
they do not have an explicit construction and
their equality is by value;
they behave more like values than like objects.
Therefore, they are not removed from weak tables.

Resurrected objects
(that is, objects being finalized
and objects accessible only through objects being finalized)
have a special behavior in weak tables.
They are removed from weak values before running their finalizers,
but are removed from weak keys only in the next collection
after running their finalizers, when such objects are actually freed.
This behavior allows the finalizer to access properties
associated with the object through weak tables.

If a weak table is among the resurrected objects in a collection cycle,
it may not be properly cleared until the next cycle.

}

}

@sect2{coroutine| @title{Coroutines}

Lua supports coroutines,
also called @emphx{collaborative multithreading}.
A coroutine in Lua represents an independent thread of execution.
Unlike threads in multithread systems, however,
a coroutine only suspends its execution by explicitly calling
a yield function.

You create a coroutine by calling @Lid{coroutine.create}.
Its sole argument is a function
that is the main function of the coroutine.
The @id{create} function only creates a new coroutine and
returns a handle to it (an object of type @emph{thread});
it does not start the coroutine.

You execute a coroutine by calling @Lid{coroutine.resume}.
When you first call @Lid{coroutine.resume},
passing as its first argument
a thread returned by @Lid{coroutine.create},
the coroutine starts its execution by
calling its main function.
Extra arguments passed to @Lid{coroutine.resume} are passed
as arguments to that function.
After the coroutine starts running,
it runs until it terminates or @emph{yields}.

A coroutine can terminate its execution in two ways:
normally, when its main function returns
(explicitly or implicitly, after the last instruction);
and abnormally, if there is an unprotected error.
In case of normal termination,
@Lid{coroutine.resume} returns @true,
plus any values returned by the coroutine main function.
In case of errors, @Lid{coroutine.resume} returns @false
plus the error object.
In this case, the coroutine does not unwind its stack,
so that it is possible to inspect it after the error
with the debug API.

A coroutine yields by calling @Lid{coroutine.yield}.
When a coroutine yields,
the corresponding @Lid{coroutine.resume} returns immediately,
even if the yield happens inside nested function calls
(that is, not in the main function,
but in a function directly or indirectly called by the main function).
In the case of a yield, @Lid{coroutine.resume} also returns @true,
plus any values passed to @Lid{coroutine.yield}.
The next time you resume the same coroutine,
it continues its execution from the point where it yielded,
with the call to @Lid{coroutine.yield} returning any extra
arguments passed to @Lid{coroutine.resume}.

Like @Lid{coroutine.create},
the @Lid{coroutine.wrap} function also creates a coroutine,
but instead of returning the coroutine itself,
it returns a function that, when called, resumes the coroutine.
Any arguments passed to this function
go as extra arguments to @Lid{coroutine.resume}.
@Lid{coroutine.wrap} returns all the values returned by @Lid{coroutine.resume},
except the first one (the boolean error code).
Unlike @Lid{coroutine.resume},
the function created by @Lid{coroutine.wrap}
propagates any error to the caller.
In this case,
the function also closes the coroutine @seeF{coroutine.close}.

As an example of how coroutines work,
consider the following code:
@verbatim{
function foo (a)
  print("foo", a)
  return coroutine.yield(2*a)
end

co = coroutine.create(function (a,b)
      print("co-body", a, b)
      local r = foo(a+1)
      print("co-body", r)
      local r, s = coroutine.yield(a+b, a-b)
      print("co-body", r, s)
      return b, "end"
end)

print("main", coroutine.resume(co, 1, 10))
print("main", coroutine.resume(co, "r"))
print("main", coroutine.resume(co, "x", "y"))
print("main", coroutine.resume(co, "x", "y"))
}
When you run it, it produces the following output:
@verbatim{
co-body 1       10
foo     2
main    true    4
co-body r
main    true    11      -9
co-body x       y
main    true    10      end
main    false   cannot resume dead coroutine
}

You can also create and manipulate coroutines through the C API:
see functions @Lid{lua_newthread}, @Lid{lua_resume},
and @Lid{lua_yield}.

}

}


@C{-------------------------------------------------------------------------}
@sect1{language| @title{The Language}

@simplesect{

This section describes the lexis, the syntax, and the semantics of Lua.
In other words,
this section describes
which tokens are valid,
how they can be combined,
and what their combinations mean.

Language constructs will be explained using the usual extended BNF notation,
in which
@N{@bnfrep{@rep{a}} means 0} or more @rep{a}'s, and
@N{@bnfopt{@rep{a}} means} an optional @rep{a}.
Non-terminals are shown like @bnfNter{non-terminal},
keywords are shown like @rw{kword},
and other terminal symbols are shown like @bnfter{=}.
The complete syntax of Lua can be found in @refsec{BNF}
at the end of this manual.

}

@sect2{lexical| @title{Lexical Conventions}

Lua is a @x{free-form} language.
It ignores spaces and comments between lexical elements (@x{tokens}),
except as delimiters between two tokens.
In source code,
Lua recognizes as spaces the standard ASCII whitespace
characters space, form feed, newline,
carriage return, horizontal tab, and vertical tab.

@def{Names}
(also called @def{identifiers})
in Lua can be any string of Latin letters,
Arabic-Indic digits, and underscores,
not beginning with a digit and
not being a reserved word.
Identifiers are used to name variables, table fields, and labels.

The following @def{keywords} are reserved
and cannot be used as names:
@index{reserved words}
@verbatim{
and       break     do        else      elseif    end
false     for       function  goto      if        in
local     nil       not       or        repeat    return
then      true      until     while
}

Lua is a case-sensitive language:
@id{and} is a reserved word, but @id{And} and @id{AND}
are two different, valid names.
As a convention,
programs should avoid creating
names that start with an underscore followed by
one or more uppercase letters (such as @Lid{_VERSION}).

The following strings denote other @x{tokens}:
@verbatim{
+     -     *     /     %     ^     #
&     ~     |     <<    >>    //
==    ~=    <=    >=    <     >     =
(     )     {     }     [     ]     ::
;     :     ,     .     ..    ...
}

A @def{short literal string}
can be delimited by matching single or double quotes,
and can contain the following C-like escape sequences:
@Char{\a} (bell),
@Char{\b} (backspace),
@Char{\f} (form feed),
@Char{\n} (newline),
@Char{\r} (carriage return),
@Char{\t} (horizontal tab),
@Char{\v} (vertical tab),
@Char{\\} (backslash),
@Char{\"} (quotation mark [double quote]),
and @Char{\'} (apostrophe [single quote]).
A backslash followed by a line break
results in a newline in the string.
The escape sequence @Char{\z} skips the following span
of whitespace characters,
including line breaks;
it is particularly useful to break and indent a long literal string
into multiple lines without adding the newlines and spaces
into the string contents.
A short literal string cannot contain unescaped line breaks
nor escapes not forming a valid escape sequence.

We can specify any byte in a short literal string,
including @x{embedded zeros},
by its numeric value.
This can be done
with the escape sequence @T{\x@rep{XX}},
where @rep{XX} is a sequence of exactly two hexadecimal digits,
or with the escape sequence @T{\@rep{ddd}},
where @rep{ddd} is a sequence of up to three decimal digits.
(Note that if a decimal escape sequence is to be followed by a digit,
it must be expressed using exactly three digits.)

The @x{UTF-8} encoding of a @x{Unicode} character
can be inserted in a literal string with
the escape sequence @T{\u{@rep{XXX}}}
(with mandatory enclosing braces),
where @rep{XXX} is a sequence of one or more hexadecimal digits
representing the character code point.
This code point can be any value less than @M{2@sp{31}}.
(Lua uses the original UTF-8 specification here,
which is not restricted to valid Unicode code points.)

Literal strings can also be defined using a long format
enclosed by @def{long brackets}.
We define an @def{opening long bracket of level @rep{n}} as an opening
square bracket followed by @rep{n} equal signs followed by another
opening square bracket.
So, an opening long bracket of @N{level 0} is written as @T{[[}, @C{]]}
an opening long bracket of @N{level 1} is written as @T{[=[}, @C{]]}
and so on.
A @emph{closing long bracket} is defined similarly;
for instance,
a closing long bracket of @N{level 4} is written as @C{[[} @T{]====]}.
A @def{long literal} starts with an opening long bracket of any level and
ends at the first closing long bracket of the same level.
It can contain any text except a closing bracket of the same level.
Literals in this bracketed form can run for several lines,
do not interpret any escape sequences,
and ignore long brackets of any other level.
Any kind of end-of-line sequence
(carriage return, newline, carriage return followed by newline,
or newline followed by carriage return)
is converted to a simple newline.
When the opening long bracket is immediately followed by a newline,
the newline is not included in the string.

As an example, in a system using ASCII
(in which @Char{a} is coded @N{as 97},
newline is coded @N{as 10}, and @Char{1} is coded @N{as 49}),
the five literal strings below denote the same string:
@verbatim{
a = 'alo\n123"'
a = "alo\n123\""
a = '\97lo\10\04923"'
a = [[alo
123"]]
a = [==[
alo
123"]==]
}

Any byte in a literal string not
explicitly affected by the previous rules represents itself.
However, Lua opens files for parsing in text mode,
and the system's file functions may have problems with
some control characters.
So, it is safer to represent
binary data as a quoted literal with
explicit escape sequences for the non-text characters.

A @def{numeric constant} (or @def{numeral})
can be written with an optional fractional part
and an optional decimal exponent,
marked by a letter @Char{e} or @Char{E}.
Lua also accepts @x{hexadecimal constants},
which start with @T{0x} or @T{0X}.
Hexadecimal constants also accept an optional fractional part
plus an optional binary exponent,
marked by a letter @Char{p} or @Char{P} and written in decimal.
(For instance, @T{0x1.fp10} denotes 1984,
which is @M{0x1f / 16} multiplied by @M{2@sp{10}}.)

A numeric constant with a radix point or an exponent
denotes a float;
otherwise,
if its value fits in an integer or it is a hexadecimal constant,
it denotes an integer;
otherwise (that is, a decimal integer numeral that overflows),
it denotes a float.
Hexadecimal numerals with neither a radix point nor an exponent
always denote an integer value;
if the value overflows, it @emph{wraps around}
to fit into a valid integer.

Examples of valid integer constants are
@verbatim{
3   345   0xff   0xBEBADA
}
Examples of valid float constants are
@verbatim{
3.0     3.1416     314.16e-2     0.31416E1     34e1
0x0.1E  0xA23p-4   0X1.921FB54442D18P+1
}

A @def{comment} starts with a double hyphen (@T{--})
anywhere outside a string.
If the text immediately after @T{--} is not an opening long bracket,
the comment is a @def{short comment},
which runs until the end of the line.
Otherwise, it is a @def{long comment},
which runs until the corresponding closing long bracket.

}

@sect2{variables| @title{Variables}

Variables are places that store values.
There are three kinds of variables in Lua:
global variables, local variables, and table fields.

A single name can denote a global variable or a local variable
(or a function's formal parameter,
which is a particular kind of local variable):
@Produc{
@producname{var}@producbody{@bnfNter{Name}}
}
@bnfNter{Name} denotes identifiers @see{lexical}.

Any variable name is assumed to be global unless explicitly declared
as a local @see{localvar}.
@x{Local variables} are @emph{lexically scoped}:
local variables can be freely accessed by functions
defined inside their scope @see{visibility}.

Before the first assignment to a variable, its value is @nil.

Square brackets are used to index a table:
@Produc{
@producname{var}@producbody{prefixexp @bnfter{[} exp @bnfter{]}}
}
The meaning of accesses to table fields can be changed via metatables
@see{metatable}.

The syntax @id{var.Name} is just syntactic sugar for
@T{var["Name"]}:
@Produc{
@producname{var}@producbody{prefixexp @bnfter{.} @bnfNter{Name}}
}

An access to a global variable @id{x}
is equivalent to @id{_ENV.x}.
Due to the way that chunks are compiled,
the variable @id{_ENV} itself is never global @see{globalenv}.

}

@sect2{stats| @title{Statements}

@simplesect{

Lua supports an almost conventional set of @x{statements},
similar to those in other conventional languages.
This set includes
blocks, assignments, control structures, function calls,
and variable declarations.

}

@sect3{@title{Blocks}

A @x{block} is a list of statements,
which are executed sequentially:
@Produc{
@producname{block}@producbody{@bnfrep{stat}}
}
Lua has @def{empty statements}
that allow you to separate statements with semicolons,
start a block with a semicolon
or write two semicolons in sequence:
@Produc{
@producname{stat}@producbody{@bnfter{;}}
}

Both function calls and assignments
can start with an open parenthesis.
This possibility leads to an ambiguity in Lua's grammar.
Consider the following fragment:
@verbatim{
a = b + c
(print or io.write)('done')
}
The grammar could see this fragment in two ways:
@verbatim{
a = b + c(print or io.write)('done')

a = b + c; (print or io.write)('done')
}
The current parser always sees such constructions
in the first way,
interpreting the open parenthesis
as the start of the arguments to a call.
To avoid this ambiguity,
it is a good practice to always precede with a semicolon
statements that start with a parenthesis:
@verbatim{
;(print or io.write)('done')
}

A block can be explicitly delimited to produce a single statement:
@Produc{
@producname{stat}@producbody{@Rw{do} block @Rw{end}}
}
Explicit blocks are useful
to control the scope of variable declarations.
Explicit blocks are also sometimes used to
add a @Rw{return} statement in the middle
of another block @see{control}.

}

@sect3{chunks| @title{Chunks}

The unit of compilation of Lua is called a @def{chunk}.
Syntactically,
a chunk is simply a block:
@Produc{
@producname{chunk}@producbody{block}
}

Lua handles a chunk as the body of an anonymous function
with a variable number of arguments
@see{func-def}.
As such, chunks can define local variables,
receive arguments, and return values.
Moreover, such anonymous function is compiled as in the
scope of an external local variable called @id{_ENV} @see{globalenv}.
The resulting function always has @id{_ENV} as its only external variable,
even if it does not use that variable.

A chunk can be stored in a file or in a string inside the host program.
To execute a chunk,
Lua first @emph{loads} it,
precompiling the chunk's code into instructions for a virtual machine,
and then Lua executes the compiled code
with an interpreter for the virtual machine.

Chunks can also be precompiled into binary form;
see the program @idx{luac} and the function @Lid{string.dump} for details.
Programs in source and compiled forms are interchangeable;
Lua automatically detects the file type and acts accordingly @seeF{load}.

}

@sect3{assignment| @title{Assignment}

Lua allows @x{multiple assignments}.
Therefore, the syntax for assignment
defines a list of variables on the left side
and a list of expressions on the right side.
The elements in both lists are separated by commas:
@Produc{
@producname{stat}@producbody{varlist @bnfter{=} explist}
@producname{varlist}@producbody{var @bnfrep{@bnfter{,} var}}
@producname{explist}@producbody{exp @bnfrep{@bnfter{,} exp}}
}
Expressions are discussed in @See{expressions}.

Before the assignment,
the list of values is @emph{adjusted} to the length of
the list of variables @see{multires}.

If a variable is both assigned and read
inside a multiple assignment,
Lua ensures that all reads get the value of the variable
before the assignment.
Thus the code
@verbatim{
i = 3
i, a[i] = i+1, 20
}
sets @T{a[3]} to 20, without affecting @T{a[4]}
because the @id{i} in @T{a[i]} is evaluated (to 3)
before it is @N{assigned 4}.
Similarly, the line
@verbatim{
x, y = y, x
}
exchanges the values of @id{x} and @id{y},
and
@verbatim{
x, y, z = y, z, x
}
cyclically permutes the values of @id{x}, @id{y}, and @id{z}.

Note that this guarantee covers only accesses
syntactically inside the assignment statement.
If a function or a metamethod called during the assignment
changes the value of a variable,
Lua gives no guarantees about the order of that access.

An assignment to a global name @T{x = val}
is equivalent to the assignment
@T{_ENV.x = val} @see{globalenv}.

The meaning of assignments to table fields and
global variables (which are actually table fields, too)
can be changed via metatables @see{metatable}.

}

@sect3{control| @title{Control Structures}
The control structures
@Rw{if}, @Rw{while}, and @Rw{repeat} have the usual meaning and
familiar syntax:
@index{while-do statement}
@index{repeat-until statement}
@index{if-then-else statement}
@Produc{
@producname{stat}@producbody{@Rw{while} exp @Rw{do} block @Rw{end}}
@producname{stat}@producbody{@Rw{repeat} block @Rw{until} exp}
@producname{stat}@producbody{@Rw{if} exp @Rw{then} block
  @bnfrep{@Rw{elseif} exp @Rw{then} block}
   @bnfopt{@Rw{else} block} @Rw{end}}
}
Lua also has a @Rw{for} statement, in two flavors @see{for}.

The @x{condition expression} of a
control structure can return any value.
Both @false and @nil test false.
All values different from @nil and @false test true.
In particular, the number 0 and the empty string also test true.

In the @Rw{repeat}@En@Rw{until} loop,
the inner block does not end at the @Rw{until} keyword,
but only after the condition.
So, the condition can refer to local variables
declared inside the loop block.

The @Rw{goto} statement transfers the program control to a label.
For syntactical reasons,
labels in Lua are considered statements too:
@index{goto statement}
@index{label}
@Produc{
@producname{stat}@producbody{@Rw{goto} Name}
@producname{stat}@producbody{label}
@producname{label}@producbody{@bnfter{::} Name @bnfter{::}}
}

A label is visible in the entire block where it is defined,
except inside nested functions.
A goto can jump to any visible label as long as it does not
enter into the scope of a local variable.
A label should not be declared
where a label with the same name is visible,
even if this other label has been declared in an enclosing block.

The @Rw{break} statement terminates the execution of a
@Rw{while}, @Rw{repeat}, or @Rw{for} loop,
skipping to the next statement after the loop:
@index{break statement}
@Produc{
@producname{stat}@producbody{@Rw{break}}
}
A @Rw{break} ends the innermost enclosing loop.

The @Rw{return} statement is used to return values
from a function or a chunk
(which is handled as an anonymous function).
@index{return statement}
Functions can return more than one value,
so the syntax for the @Rw{return} statement is
@Produc{
@producname{stat}@producbody{@Rw{return} @bnfopt{explist} @bnfopt{@bnfter{;}}}
}

The @Rw{return} statement can only be written
as the last statement of a block.
If it is necessary to @Rw{return} in the middle of a block,
then an explicit inner block can be used,
as in the idiom @T{do return end},
because now @Rw{return} is the last statement in its (inner) block.

}

@sect3{for| @title{For Statement}

@index{for statement}
The @Rw{for} statement has two forms:
one numerical and one generic.

@sect4{@title{The numerical @Rw{for} loop}

The numerical @Rw{for} loop repeats a block of code while a
control variable goes through an arithmetic progression.
It has the following syntax:
@Produc{
@producname{stat}@producbody{@Rw{for} @bnfNter{Name} @bnfter{=}
  exp @bnfter{,} exp @bnfopt{@bnfter{,} exp} @Rw{do} block @Rw{end}}
}
The given identifier (@bnfNter{Name}) defines the control variable,
which is a new read-only variable local to the loop body (@emph{block}).

The loop starts by evaluating once the three control expressions.
Their values are called respectively
the @emph{initial value}, the @emph{limit}, and the @emph{step}.
If the step is absent, it defaults @N{to 1}.

If both the initial value and the step are integers,
the loop is done with integers;
note that the limit may not be an integer.
Otherwise, the three values are converted to
floats and the loop is done with floats.
Beware of floating-point accuracy in this case.

After that initialization,
the loop body is repeated with the value of the control variable
going through an arithmetic progression,
starting at the initial value,
with a common difference given by the step.
A negative step makes a decreasing sequence;
a step equal to zero raises an error.
The loop continues while the value is less than
or equal to the limit
(greater than or equal to for a negative step).
If the initial value is already greater than the limit
(or less than, if the step is negative),
the body is not executed.

For integer loops,
the control variable never wraps around;
instead, the loop ends in case of an overflow.

}

@sect4{@title{The generic @Rw{for} loop}


The generic @Rw{for} statement works over functions,
called @def{iterators}.
On each iteration, the iterator function is called to produce a new value,
stopping when this new value is @nil.
The generic @Rw{for} loop has the following syntax:
@Produc{
@producname{stat}@producbody{@Rw{for} namelist @Rw{in} explist
                    @Rw{do} block @Rw{end}}
@producname{namelist}@producbody{@bnfNter{Name} @bnfrep{@bnfter{,} @bnfNter{Name}}}
}
A @Rw{for} statement like
@verbatim{
for @rep{var_1}, @Cdots, @rep{var_n} in @rep{explist} do @rep{body} end
}
works as follows.

The names @rep{var_i} declare loop variables local to the loop body.
The first of these variables is the @emph{control variable},
which is a read-only variable.

The loop starts by evaluating @rep{explist}
to produce four values:
an @emph{iterator function},
a @emph{state},
an initial value for the control variable,
and a @emph{closing value}.

Then, at each iteration,
Lua calls the iterator function with two arguments:
the state and the control variable.
The results from this call are then assigned to the loop variables,
following the rules of multiple assignments @see{assignment}.
If the control variable becomes @nil,
the loop terminates.
Otherwise, the body is executed and the loop goes
to the next iteration.

The closing value behaves like a
to-be-closed variable @see{to-be-closed},
which can be used to release resources when the loop ends.
Otherwise, it does not interfere with the loop.

}

}

@sect3{funcstat| @title{Function Calls as Statements}
To allow possible side-effects,
function calls can be executed as statements:
@Produc{
@producname{stat}@producbody{functioncall}
}
In this case, all returned values are thrown away.
Function calls are explained in @See{functioncall}.

}

@sect3{localvar| @title{Local Declarations}
@x{Local variables} can be declared anywhere inside a block.
The declaration can include an initialization:
@Produc{
@producname{stat}@producbody{@Rw{local} attnamelist @bnfopt{@bnfter{=} explist}}
@producname{attnamelist}@producbody{
  @bnfNter{Name} attrib @bnfrep{@bnfter{,} @bnfNter{Name} attrib}}
}
If present, an initial assignment has the same semantics
of a multiple assignment @see{assignment}.
Otherwise, all variables are initialized with @nil.

Each variable name may be postfixed by an attribute
(a name between angle brackets):
@Produc{
@producname{attrib}@producbody{@bnfopt{@bnfter{<} @bnfNter{Name} @bnfter{>}}}
}
There are two possible attributes:
@id{const}, which declares a @emph{constant} or @emph{read-only} variable,
@index{constant variable}
that is, a variable that cannot be assigned to
after its initialization;
and @id{close}, which declares a to-be-closed variable @see{to-be-closed}.
A list of variables can contain at most one to-be-closed variable.

A chunk is also a block @see{chunks},
and so local variables can be declared in a chunk outside any explicit block.

The visibility rules for local variables are explained in @See{visibility}.

}

@sect3{to-be-closed| @title{To-be-closed Variables}

A to-be-closed variable behaves like a constant local variable,
except that its value is @emph{closed} whenever the variable
goes out of scope, including normal block termination,
exiting its block by @Rw{break}/@Rw{goto}/@Rw{return},
or exiting by an error.

Here, to @emph{close} a value means
to call its @idx{__close} metamethod.
When calling the metamethod,
the value itself is passed as the first argument
and the error object that caused the exit (if any)
is passed as a second argument;
if there was no error, the second argument is @nil.

The value assigned to a to-be-closed variable
must have a @idx{__close} metamethod
or be a false value.
(@nil and @false are ignored as to-be-closed values.)

If several to-be-closed variables go out of scope at the same event,
they are closed in the reverse order that they were declared.

If there is any error while running a closing method,
that error is handled like an error in the regular code
where the variable was defined.
After an error,
the other pending closing methods will still be called.

If a coroutine yields and is never resumed again,
some variables may never go out of scope,
and therefore they will never be closed.
(These variables are the ones created inside the coroutine
and in scope at the point where the coroutine yielded.)
Similarly, if a coroutine ends with an error,
it does not unwind its stack,
so it does not close any variable.
In both cases,
you can either use finalizers
or call @Lid{coroutine.close} to close the variables.
However, if the coroutine was created
through @Lid{coroutine.wrap},
then its corresponding function will close the coroutine
in case of errors.

}

}

@sect2{expressions| @title{Expressions}

@simplesect{

The basic expressions in Lua are the following:
@Produc{
@producname{exp}@producbody{prefixexp}
@producname{exp}@producbody{@Rw{nil} @Or @Rw{false} @Or @Rw{true}}
@producname{exp}@producbody{@bnfNter{Numeral}}
@producname{exp}@producbody{@bnfNter{LiteralString}}
@producname{exp}@producbody{functiondef}
@producname{exp}@producbody{tableconstructor}
@producname{exp}@producbody{@bnfter{...}}
@producname{exp}@producbody{exp binop exp}
@producname{exp}@producbody{unop exp}
@producname{prefixexp}@producbody{var @Or functioncall @Or
                                  @bnfter{(} exp @bnfter{)}}
}

Numerals and literal strings are explained in @See{lexical};
variables are explained in @See{variables};
function definitions are explained in @See{func-def};
function calls are explained in @See{functioncall};
table constructors are explained in @See{tableconstructor}.
Vararg expressions,
denoted by three dots (@Char{...}), can only be used when
directly inside a variadic function;
they are explained in @See{func-def}.


Binary operators comprise arithmetic operators @see{arith},
bitwise operators @see{bitwise},
relational operators @see{rel-ops}, logical operators @see{logic},
and the concatenation operator @see{concat}.
Unary operators comprise the unary minus @see{arith},
the unary bitwise NOT @see{bitwise},
the unary logical @Rw{not} @see{logic},
and the unary @def{length operator} @see{len-op}.

}



@sect3{arith| @title{Arithmetic Operators}
Lua supports the following @x{arithmetic operators}:
@description{
@item{@T{+}|addition}
@item{@T{-}|subtraction}
@item{@T{*}|multiplication}
@item{@T{/}|float division}
@item{@T{//}|floor division}
@item{@T{%}|modulo}
@item{@T{^}|exponentiation}
@item{@T{-}|unary minus}
}

With the exception of exponentiation and float division,
the arithmetic operators work as follows:
If both operands are integers,
the operation is performed over integers and the result is an integer.
Otherwise, if both operands are numbers,
then they are converted to floats,
the operation is performed following the machine's rules
for floating-point arithmetic
(usually the @x{IEEE 754} standard),
and the result is a float.
(The string library coerces strings to numbers in
arithmetic operations; see @See{coercion} for details.)

Exponentiation and float division (@T{/})
always convert their operands to floats
and the result is always a float.
Exponentiation uses the @ANSI{pow},
so that it works for non-integer exponents too.

Floor division (@T{//}) is a division
that rounds the quotient towards minus infinity,
resulting in the floor of the division of its operands.

Modulo is defined as the remainder of a division
that rounds the quotient towards minus infinity (floor division).

In case of overflows in integer arithmetic,
all operations @emphx{wrap around}.
}

@sect3{bitwise| @title{Bitwise Operators}
Lua supports the following @x{bitwise operators}:
@description{
@item{@T{&}|bitwise AND}
@item{@T{@VerBar}|bitwise OR}
@item{@T{~}|bitwise exclusive OR}
@item{@T{>>}|right shift}
@item{@T{<<}|left shift}
@item{@T{~}|unary bitwise NOT}
}

All bitwise operations convert its operands to integers
@see{coercion},
operate on all bits of those integers,
and result in an integer.

Both right and left shifts fill the vacant bits with zeros.
Negative displacements shift to the other direction;
displacements with absolute values equal to or higher than
the number of bits in an integer
result in zero (as all bits are shifted out).

}

@sect3{coercion| @title{Coercions and Conversions}
Lua provides some automatic conversions between some
types and representations at run time.
Bitwise operators always convert float operands to integers.
Exponentiation and float division
always convert integer operands to floats.
All other arithmetic operations applied to mixed numbers
(integers and floats) convert the integer operand to a float.
The C API also converts both integers to floats and
floats to integers, as needed.
Moreover, string concatenation accepts numbers as arguments,
besides strings.

In a conversion from integer to float,
if the integer value has an exact representation as a float,
that is the result.
Otherwise,
the conversion gets the nearest higher or
the nearest lower representable value.
This kind of conversion never fails.

The conversion from float to integer
checks whether the float has an exact representation as an integer
(that is, the float has an integral value and
it is in the range of integer representation).
If it does, that representation is the result.
Otherwise, the conversion fails.

Several places in Lua coerce strings to numbers when necessary.
In particular,
the string library sets metamethods that try to coerce
strings to numbers in all arithmetic operations.
If the conversion fails,
the library calls the metamethod of the other operand
(if present) or it raises an error.
Note that bitwise operators do not do this coercion.

It is always a good practice not to rely on the
implicit coercions from strings to numbers,
as they are not always applied;
in particular, @T{"1"==1} is false and @T{"1"<1} raises an error
@see{rel-ops}.
These coercions exist mainly for compatibility and may be removed
in future versions of the language.

A string is converted to an integer or a float
following its syntax and the rules of the Lua lexer.
The string may have also leading and trailing whitespaces and a sign.
All conversions from strings to numbers
accept both a dot and the current locale mark
as the radix character.
(The Lua lexer, however, accepts only a dot.)
If the string is not a valid numeral,
the conversion fails.
If necessary, the result of this first step is then converted
to a specific number subtype following the previous rules
for conversions between floats and integers.

The conversion from numbers to strings uses a
non-specified human-readable format.
To convert numbers to strings in any specific way,
use the function @Lid{string.format}.

}

@sect3{rel-ops| @title{Relational Operators}
Lua supports the following @x{relational operators}:
@description{
@item{@T{==}|equality}
@item{@T{~=}|inequality}
@item{@T{<}|less than}
@item{@T{>}|greater than}
@item{@T{<=}|less or equal}
@item{@T{>=}|greater or equal}
}
These operators always result in @false or @true.

Equality (@T{==}) first compares the type of its operands.
If the types are different, then the result is @false.
Otherwise, the values of the operands are compared.
Strings are equal if they have the same byte content.
Numbers are equal if they denote the same mathematical value.

Tables, userdata, and threads
are compared by reference:
two objects are considered equal only if they are the same object.
Every time you create a new object
(a table, a userdata, or a thread),
this new object is different from any previously existing object.
A function is always equal to itself.
Functions with any detectable difference
(different behavior, different definition) are always different.
Functions created at different times but with no detectable differences
may be classified as equal or not
(depending on internal caching details).

You can change the way that Lua compares tables and userdata
by using the @idx{__eq} metamethod @see{metatable}.

Equality comparisons do not convert strings to numbers
or vice versa.
Thus, @T{"0"==0} evaluates to @false,
and @T{t[0]} and @T{t["0"]} denote different
entries in a table.

The operator @T{~=} is exactly the negation of equality (@T{==}).

The order operators work as follows.
If both arguments are numbers,
then they are compared according to their mathematical values,
regardless of their subtypes.
Otherwise, if both arguments are strings,
then their values are compared according to the current locale.
Otherwise, Lua tries to call the @idx{__lt} or the @idx{__le}
metamethod @see{metatable}.
A comparison @T{a > b} is translated to @T{b < a}
and @T{a >= b} is translated to @T{b <= a}.

Following the @x{IEEE 754} standard,
the special value @x{NaN} is considered neither less than,
nor equal to, nor greater than any value, including itself.

}

@sect3{logic| @title{Logical Operators}
The @x{logical operators} in Lua are
@Rw{and}, @Rw{or}, and @Rw{not}.
Like the control structures @see{control},
all logical operators consider both @false and @nil as false
and anything else as true.

The negation operator @Rw{not} always returns @false or @true.
The conjunction operator @Rw{and} returns its first argument
if this value is @false or @nil;
otherwise, @Rw{and} returns its second argument.
The disjunction operator @Rw{or} returns its first argument
if this value is different from @nil and @false;
otherwise, @Rw{or} returns its second argument.
Both @Rw{and} and @Rw{or} use @x{short-circuit evaluation};
that is,
the second operand is evaluated only if necessary.
Here are some examples:
@verbatim{
10 or 20            --> 10
10 or error()       --> 10
nil or "a"          --> "a"
nil and 10          --> nil
false and error()   --> false
false and nil       --> false
false or nil        --> nil
10 and 20           --> 20
}

}

@sect3{concat| @title{Concatenation}
The string @x{concatenation} operator in Lua is
denoted by two dots (@Char{..}).
If both operands are strings or numbers,
then the numbers are converted to strings
in a non-specified format @see{coercion}.
Otherwise, the @idx{__concat} metamethod is called @see{metatable}.

}

@sect3{len-op| @title{The Length Operator}

The length operator is denoted by the unary prefix operator @T{#}.

The length of a string is its number of bytes.
(That is the usual meaning of string length when each
character is one byte.)

The length operator applied on a table
returns a @x{border} in that table.
A @def{border} in a table @id{t} is any non-negative integer
that satisfies the following condition:
@verbatim{
(border == 0 or t[border] ~= nil) and
(t[border + 1] == nil or border == math.maxinteger)
}
In words,
a border is any positive integer index present in the table
that is followed by an absent index,
plus two limit cases:
zero, when index 1 is absent;
and the maximum value for an integer, when that index is present.
Note that keys that are not positive integers
do not interfere with borders.

A table with exactly one border is called a @def{sequence}.
For instance, the table @T{{10, 20, 30, 40, 50}} is a sequence,
as it has only one border (5).
The table @T{{10, 20, 30, nil, 50}} has two borders (3 and 5),
and therefore it is not a sequence.
(The @nil at index 4 is called a @emphx{hole}.)
The table @T{{nil, 20, 30, nil, nil, 60, nil}}
has three borders (0, 3, and 6),
so it is not a sequence, too.
The table @T{{}} is a sequence with border 0.

When @id{t} is a sequence,
@T{#t} returns its only border,
which corresponds to the intuitive notion of the length of the sequence.
When @id{t} is not a sequence,
@T{#t} can return any of its borders.
(The exact one depends on details of
the internal representation of the table,
which in turn can depend on how the table was populated and
the memory addresses of its non-numeric keys.)

The computation of the length of a table
has a guaranteed worst time of @M{O(log n)},
where @M{n} is the largest integer key in the table.

A program can modify the behavior of the length operator for
any value but strings through the @idx{__len} metamethod @see{metatable}.

}

@sect3{prec| @title{Precedence}
@x{Operator precedence} in Lua follows the table below,
from lower to higher priority:
@verbatim{
or
and
<     >     <=    >=    ~=    ==
|
~
&
<<    >>
..
+     -
*     /     //    %
unary operators (not   #     -     ~)
^
}
As usual,
you can use parentheses to change the precedences of an expression.
The concatenation (@Char{..}) and exponentiation (@Char{^})
operators are right associative.
All other binary operators are left associative.

}

@sect3{tableconstructor| @title{Table Constructors}
Table @x{constructors} are expressions that create tables.
Every time a constructor is evaluated, a new table is created.
A constructor can be used to create an empty table
or to create a table and initialize some of its fields.
The general syntax for constructors is
@Produc{
@producname{tableconstructor}@producbody{@bnfter{@Open} @bnfopt{fieldlist} @bnfter{@Close}}
@producname{fieldlist}@producbody{field @bnfrep{fieldsep field} @bnfopt{fieldsep}}
@producname{field}@producbody{@bnfter{[} exp @bnfter{]} @bnfter{=} exp @Or
               @bnfNter{Name} @bnfter{=} exp @Or exp}
@producname{fieldsep}@producbody{@bnfter{,} @Or @bnfter{;}}
}

Each field of the form @T{[exp1] = exp2} adds to the new table an entry
with key @id{exp1} and value @id{exp2}.
A field of the form @T{name = exp} is equivalent to
@T{["name"] = exp}.
Fields of the form @id{exp} are equivalent to
@T{[i] = exp}, where @id{i} are consecutive integers
starting with 1;
fields in the other formats do not affect this counting.
For example,
@verbatim{
a = { [f(1)] = g; "x", "y"; x = 1, f(x), [30] = 23; 45 }
}
is equivalent to
@verbatim{
do
  local t = {}
  t[f(1)] = g
  t[1] = "x"         -- 1st exp
  t[2] = "y"         -- 2nd exp
  t.x = 1            -- t["x"] = 1
  t[3] = f(x)        -- 3rd exp
  t[30] = 23
  t[4] = 45          -- 4th exp
  a = t
end
}

The order of the assignments in a constructor is undefined.
(This order would be relevant only when there are repeated keys.)

If the last field in the list has the form @id{exp}
and the expression is a multires expression,
then all values returned by this expression enter the list consecutively
@see{multires}.

The field list can have an optional trailing separator,
as a convenience for machine-generated code.

}

@sect3{functioncall| @title{Function Calls}
A @x{function call} in Lua has the following syntax:
@Produc{
@producname{functioncall}@producbody{prefixexp args}
}
In a function call,
first @bnfNter{prefixexp} and @bnfNter{args} are evaluated.
If the value of @bnfNter{prefixexp} has type @emph{function},
then this function is called
with the given arguments.
Otherwise, if present,
the @bnfNter{prefixexp} @idx{__call} metamethod is called:
its first argument is the value of @bnfNter{prefixexp},
followed by the original call arguments
@see{metatable}.

The form
@Produc{
@producname{functioncall}@producbody{prefixexp @bnfter{:} @bnfNter{Name} args}
}
can be used to emulate methods.
A call @T{v:name(@rep{args})}
is syntactic sugar for @T{v.name(v,@rep{args})},
except that @id{v} is evaluated only once.

Arguments have the following syntax:
@Produc{
@producname{args}@producbody{@bnfter{(} @bnfopt{explist} @bnfter{)}}
@producname{args}@producbody{tableconstructor}
@producname{args}@producbody{@bnfNter{LiteralString}}
}
All argument expressions are evaluated before the call.
A call of the form @T{f{@rep{fields}}} is
syntactic sugar for @T{f({@rep{fields}})};
that is, the argument list is a single new table.
A call of the form @T{f'@rep{string}'}
(or @T{f"@rep{string}"} or @T{f[[@rep{string}]]})
is syntactic sugar for @T{f('@rep{string}')};
that is, the argument list is a single literal string.

A call of the form @T{return @rep{functioncall}} not in the
scope of a to-be-closed variable is called a @def{tail call}.
Lua implements @def{proper tail calls}
(or @def{proper tail recursion}):
In a tail call,
the called function reuses the stack entry of the calling function.
Therefore, there is no limit on the number of nested tail calls that
a program can execute.
However, a tail call erases any debug information about the
calling function.
Note that a tail call only happens with a particular syntax,
where the @Rw{return} has one single function call as argument,
and it is outside the scope of any to-be-closed variable.
This syntax makes the calling function return exactly
the returns of the called function,
without any intervening action.
So, none of the following examples are tail calls:
@verbatim{
return (f(x))        -- results adjusted to 1
return 2 * f(x)      -- result multiplied by 2
return x, f(x)       -- additional results
f(x); return         -- results discarded
return x or f(x)     -- results adjusted to 1
}

}

@sect3{func-def| @title{Function Definitions}

The syntax for function definition is
@Produc{
@producname{functiondef}@producbody{@Rw{function} funcbody}
@producname{funcbody}@producbody{@bnfter{(} @bnfopt{parlist} @bnfter{)} block @Rw{end}}
}

The following syntactic sugar simplifies function definitions:
@Produc{
@producname{stat}@producbody{@Rw{function} funcname funcbody}
@producname{stat}@producbody{@Rw{local} @Rw{function} @bnfNter{Name} funcbody}
@producname{funcname}@producbody{@bnfNter{Name} @bnfrep{@bnfter{.} @bnfNter{Name}} @bnfopt{@bnfter{:} @bnfNter{Name}}}
}
The statement
@verbatim{
function f () @rep{body} end
}
translates to
@verbatim{
f = function () @rep{body} end
}
The statement
@verbatim{
function t.a.b.c.f () @rep{body} end
}
translates to
@verbatim{
t.a.b.c.f = function () @rep{body} end
}
The statement
@verbatim{
local function f () @rep{body} end
}
translates to
@verbatim{
local f; f = function () @rep{body} end
}
not to
@verbatim{
local f = function () @rep{body} end
}
(This only makes a difference when the body of the function
contains references to @id{f}.)

A function definition is an executable expression,
whose value has type @emph{function}.
When Lua precompiles a chunk,
all its function bodies are precompiled too,
but they are not created yet.
Then, whenever Lua executes the function definition,
the function is @emph{instantiated} (or @emph{closed}).
This function instance, or @emphx{closure},
is the final value of the expression.

Parameters act as local variables that are
initialized with the argument values:
@Produc{
@producname{parlist}@producbody{namelist @bnfopt{@bnfter{,} @bnfter{...}} @Or
  @bnfter{...}}
}
When a Lua function is called,
it adjusts its list of @x{arguments} to
the length of its list of parameters @see{multires},
unless the function is a @def{variadic function},
which is indicated by three dots (@Char{...})
at the end of its parameter list.
A variadic function does not adjust its argument list;
instead, it collects all extra arguments and supplies them
to the function through a @def{vararg expression},
which is also written as three dots.
The value of this expression is a list of all actual extra arguments,
similar to a function with multiple results @see{multires}.


As an example, consider the following definitions:
@verbatim{
function f(a, b) end
function g(a, b, ...) end
function r() return 1,2,3 end
}
Then, we have the following mapping from arguments to parameters and
to the vararg expression:
@verbatim{
CALL             PARAMETERS

f(3)             a=3, b=nil
f(3, 4)          a=3, b=4
f(3, 4, 5)       a=3, b=4
f(r(), 10)       a=1, b=10
f(r())           a=1, b=2

g(3)             a=3, b=nil, ... -->  (nothing)
g(3, 4)          a=3, b=4,   ... -->  (nothing)
g(3, 4, 5, 8)    a=3, b=4,   ... -->  5  8
g(5, r())        a=5, b=1,   ... -->  2  3
}

Results are returned using the @Rw{return} statement @see{control}.
If control reaches the end of a function
without encountering a @Rw{return} statement,
then the function returns with no results.

@index{multiple return}
There is a system-dependent limit on the number of values
that a function may return.
This limit is guaranteed to be greater than 1000.

The @emphx{colon} syntax
is used to emulate @def{methods},
adding an implicit extra parameter @idx{self} to the function.
Thus, the statement
@verbatim{
function t.a.b.c:f (@rep{params}) @rep{body} end
}
is syntactic sugar for
@verbatim{
t.a.b.c.f = function (self, @rep{params}) @rep{body} end
}

}

@sect3{multires| @title{Lists of expressions, multiple results,
and adjustment}

Both function calls and vararg expressions can result in multiple values.
These expressions are called @def{multires expressions}.

When a multires expression is used as the last element
of a list of expressions,
all results from the expression are added to the
list of values produced by the list of expressions.
Note that a single expression
in a place that expects a list of expressions
is the last expression in that (singleton) list.

These are the places where Lua expects a list of expressions:
@description{

@item{A @rw{return} statement,
for instance @T{return e1, e2, e3} @see{control}.}

@item{A table constructor,
for instance @T{{e1, e2, e3}} @see{tableconstructor}.}

@item{The arguments of a function call,
for instance @T{foo(e1, e2, e3)} @see{functioncall}.}

@item{A multiple assignment,
for instance @T{a , b, c = e1, e2, e3} @see{assignment}.}

@item{A local declaration,
for instance @T{local a , b, c = e1, e2, e3} @see{localvar}.}

@item{The initial values in a generic @rw{for} loop,
for instance @T{for k in e1, e2, e3 do ... end} @see{for}.}

}
In the last four cases,
the list of values from the list of expressions
must be @emph{adjusted} to a specific length:
the number of parameters in a call to a non-variadic function
@see{func-def},
the number of variables in a multiple assignment or
a local declaration,
and exactly four values for a generic @rw{for} loop.
The @def{adjustment} follows these rules:
If there are more values than needed,
the extra values are thrown away;
if there are fewer values than needed,
the list is extended with @nil's.
When the list of expressions ends with a multires expression,
all results from that expression enter the list of values
before the adjustment.

When a multires expression is used
in a list of expressions without being the last element,
or in a place where the syntax expects a single expression,
Lua adjusts the result list of that expression to one element.
As a particular case,
the syntax expects a single expression inside a parenthesized expression;
therefore, adding parentheses around a multires expression
forces it to produce exactly one result.

We seldom need to use a vararg expression in a place
where the syntax expects a single expression.
(Usually it is simpler to add a regular parameter before
the variadic part and use that parameter.)
When there is such a need,
we recommend assigning the vararg expression
to a single variable and using that variable
in its place.

Here are some examples of uses of mutlres expressions.
In all cases, when the construction needs
@Q{the n-th result} and there is no such result,
it uses a @nil.
@verbatim{
print(x, f())      -- prints x and all results from f().
print(x, (f()))    -- prints x and the first result from f().
print(f(), x)      -- prints the first result from f() and x.
print(1 + f())     -- prints 1 added to the first result from f().
local x = ...      -- x gets the first vararg argument.
x,y = ...          -- x gets the first vararg argument,
                   -- y gets the second vararg argument.
x,y,z = w, f()     -- x gets w, y gets the first result from f(),
                   -- z gets the second result from f().
x,y,z = f()        -- x gets the first result from f(),
                   -- y gets the second result from f(),
                   -- z gets the third result from f().
x,y,z = f(), g()   -- x gets the first result from f(),
                   -- y gets the first result from g(),
                   -- z gets the second result from g().
x,y,z = (f())      -- x gets the first result from f(), y and z get nil.
return f()         -- returns all results from f().
return x, ...      -- returns x and all received vararg arguments.
return x,y,f()     -- returns x, y, and all results from f().
{f()}              -- creates a list with all results from f().
{...}              -- creates a list with all vararg arguments.
{f(), 5}           -- creates a list with the first result from f() and 5.
}

}

}

@sect2{visibility| @title{Visibility Rules}

@index{visibility}
Lua is a lexically scoped language.
The scope of a local variable begins at the first statement after
its declaration and lasts until the last non-void statement
of the innermost block that includes the declaration.
(@emph{Void statements} are labels and empty statements.)
Consider the following example:
@verbatim{
x = 10                -- global variable
do                    -- new block
  local x = x         -- new 'x', with value 10
  print(x)            --> 10
  x = x+1
  do                  -- another block
    local x = x+1     -- another 'x'
    print(x)          --> 12
  end
  print(x)            --> 11
end
print(x)              --> 10  (the global one)
}

Notice that, in a declaration like @T{local x = x},
the new @id{x} being declared is not in scope yet,
and so the second @id{x} refers to the outside variable.

Because of the @x{lexical scoping} rules,
local variables can be freely accessed by functions
defined inside their scope.
A local variable used by an inner function is called an @def{upvalue}
(or @emphx{external local variable}, or simply @emphx{external variable})
inside the inner function.

Notice that each execution of a @Rw{local} statement
defines new local variables.
Consider the following example:
@verbatim{
a = {}
local x = 20
for i = 1, 10 do
  local y = 0
  a[i] = function () y = y + 1; return x + y end
end
}
The loop creates ten closures
(that is, ten instances of the anonymous function).
Each of these closures uses a different @id{y} variable,
while all of them share the same @id{x}.

}

}


@C{-------------------------------------------------------------------------}
@sect1{API| @title{The Application Program Interface}

@simplesect{

@index{C API}
This section describes the @N{C API} for Lua, that is,
the set of @N{C functions} available to the host program to communicate
with Lua.
All API functions and related types and constants
are declared in the header file @defid{lua.h}.

Even when we use the term @Q{function},
any facility in the API may be provided as a macro instead.
Except where stated otherwise,
all such macros use each of their arguments exactly once
(except for the first argument, which is always a Lua state),
and so do not generate any hidden side-effects.

As in most @N{C libraries},
the Lua API functions do not check their arguments
for validity or consistency.
However, you can change this behavior by compiling Lua
with the macro @defid{LUA_USE_APICHECK} defined.

The Lua library is fully reentrant:
it has no global variables.
It keeps all information it needs in a dynamic structure,
called the @def{Lua state}.

Each Lua state has one or more threads,
which correspond to independent, cooperative lines of execution.
The type @Lid{lua_State} (despite its name) refers to a thread.
(Indirectly, through the thread, it also refers to the
Lua state associated to the thread.)

A pointer to a thread must be passed as the first argument to
every function in the library, except to @Lid{lua_newstate},
which creates a Lua state from scratch and returns a pointer
to the @emph{main thread} in the new state.

}


@sect2{@title{The Stack}

@simplesect{

Lua uses a @emph{virtual stack} to pass values to and from C.
Each element in this stack represents a Lua value
(@nil, number, string, etc.).
Functions in the API can access this stack through the
Lua state parameter that they receive.

Whenever Lua calls C, the called function gets a new stack,
which is independent of previous stacks and of stacks of
@N{C functions} that are still active.
This stack initially contains any arguments to the @N{C function}
and it is where the @N{C function} can store temporary
Lua values and must push its results
to be returned to the caller @seeC{lua_CFunction}.

For convenience,
most query operations in the API do not follow a strict stack discipline.
Instead, they can refer to any element in the stack
by using an @emph{index}:@index{index (API stack)}
A positive index represents an absolute stack position,
starting @N{at 1} as the bottom of the stack;
a negative index represents an offset relative to the top of the stack.
More specifically, if the stack has @rep{n} elements,
then @N{index 1} represents the first element
(that is, the element that was pushed onto the stack first)
and
@N{index @rep{n}} represents the last element;
@N{index @num{-1}} also represents the last element
(that is, the element at @N{the top})
and index @M{-n} represents the first element.

}

@sect3{stacksize| @title{Stack Size}

When you interact with the Lua API,
you are responsible for ensuring consistency.
In particular,
@emph{you are responsible for controlling stack overflow}.
When you call any API function,
you must ensure the stack has enough room to accommodate the results.

There is one exception to the above rule:
When you call a Lua function
without a fixed number of results @seeF{lua_call},
Lua ensures that the stack has enough space for all results.
However, it does not ensure any extra space.
So, before pushing anything on the stack after such a call
you should use @Lid{lua_checkstack}.

Whenever Lua calls C,
it ensures that the stack has space for
at least @defid{LUA_MINSTACK} extra elements;
that is, you can safely push up to @id{LUA_MINSTACK} values into it.
@id{LUA_MINSTACK} is defined as 20,
so that usually you do not have to worry about stack space
unless your code has loops pushing elements onto the stack.
Whenever necessary,
you can use the function @Lid{lua_checkstack}
to ensure that the stack has enough space for pushing new elements.

}

@sect3{@title{Valid and Acceptable Indices}

Any function in the API that receives stack indices
works only with @emphx{valid indices} or @emphx{acceptable indices}.

A @def{valid index} is an index that refers to a
position that stores a modifiable Lua value.
It comprises stack indices @N{between 1} and the stack top
(@T{1 @leq abs(index) @leq top})
@index{stack index}
plus @def{pseudo-indices},
which represent some positions that are accessible to @N{C code}
but that are not in the stack.
Pseudo-indices are used to access the registry @see{registry}
and the upvalues of a @N{C function} @see{c-closure}.

Functions that do not need a specific mutable position,
but only a value (e.g., query functions),
can be called with acceptable indices.
An @def{acceptable index} can be any valid index,
but it also can be any positive index after the stack top
within the space allocated for the stack,
that is, indices up to the stack size.
(Note that 0 is never an acceptable index.)
Indices to upvalues @see{c-closure} greater than the real number
of upvalues in the current @N{C function} are also acceptable (but invalid).
Except when noted otherwise,
functions in the API work with acceptable indices.

Acceptable indices serve to avoid extra tests
against the stack top when querying the stack.
For instance, a @N{C function} can query its third argument
without the need to check whether there is a third argument,
that is, without the need to check whether 3 is a valid index.

For functions that can be called with acceptable indices,
any non-valid index is treated as if it
contains a value of a virtual type @defid{LUA_TNONE},
which behaves like a nil value.

}

@sect3{constchar|@title{Pointers to strings}

Several functions in the API return pointers (@T{const char*})
to Lua strings in the stack.
(See @Lid{lua_pushfstring}, @Lid{lua_pushlstring},
@Lid{lua_pushstring}, and @Lid{lua_tolstring}.
See also @Lid{luaL_checklstring}, @Lid{luaL_checkstring},
and @Lid{luaL_tolstring} in the auxiliary library.)

In general,
Lua's garbage collection can free or move memory
and then invalidate pointers to strings handled by a Lua state.
To allow a safe use of these pointers,
the API guarantees that any pointer to a string in a stack index
is valid while the string value at that index is not removed from the stack.
(It can be moved to another index, though.)
When the index is a pseudo-index (referring to an upvalue),
the pointer is valid while the corresponding call is active and
the corresponding upvalue is not modified.

Some functions in the debug interface
also return pointers to strings,
namely @Lid{lua_getlocal}, @Lid{lua_getupvalue},
@Lid{lua_setlocal}, and @Lid{lua_setupvalue}.
For these functions, the pointer is guaranteed to
be valid while the caller function is active and
the given closure (if one was given) is in the stack.

Except for these guarantees,
the garbage collector is free to invalidate
any pointer to internal strings.

}

}

@sect2{c-closure| @title{C Closures}

When a @N{C function} is created,
it is possible to associate some values with it,
thus creating a @def{@N{C closure}}
@seeC{lua_pushcclosure};
these values are called @def{upvalues} and are
accessible to the function whenever it is called.

Whenever a @N{C function} is called,
its upvalues are located at specific pseudo-indices.
These pseudo-indices are produced by the macro
@Lid{lua_upvalueindex}.
The first upvalue associated with a function is at index
@T{lua_upvalueindex(1)}, and so on.
Any access to @T{lua_upvalueindex(@rep{n})},
where @rep{n} is greater than the number of upvalues of the
current function
(but not greater than 256,
which is one plus the maximum number of upvalues in a closure),
produces an acceptable but invalid index.

A @N{C closure} can also change the values
of its corresponding upvalues.

}

@sect2{registry| @title{Registry}

Lua provides a @def{registry},
a predefined table that can be used by any @N{C code} to
store whatever Lua values it needs to store.
The registry table is always accessible at pseudo-index
@defid{LUA_REGISTRYINDEX}.
Any @N{C library} can store data into this table,
but it must take care to choose keys
that are different from those used
by other libraries, to avoid collisions.
Typically, you should use as key a string containing your library name,
or a light userdata with the address of a @N{C object} in your code,
or any Lua object created by your code.
As with variable names,
string keys starting with an underscore followed by
uppercase letters are reserved for Lua.

The integer keys in the registry are used
by the reference mechanism @seeC{luaL_ref},
with some predefined values.
Therefore, integer keys in the registry
must not be used for other purposes.

When you create a new Lua state,
its registry comes with some predefined values.
These predefined values are indexed with integer keys
defined as constants in @id{lua.h}.
The following constants are defined:
@description{
@item{@defid{LUA_RIDX_MAINTHREAD}| At this index the registry has
the main thread of the state.
(The main thread is the one created together with the state.)
}

@item{@defid{LUA_RIDX_GLOBALS}| At this index the registry has
the @x{global environment}.
}
}

}

@sect2{C-error|@title{Error Handling in C}

@simplesect{

Internally, Lua uses the C @id{longjmp} facility to handle errors.
(Lua will use exceptions if you compile it as C++;
search for @id{LUAI_THROW} in the source code for details.)
When Lua faces any error,
such as a @x{memory allocation error} or a type error,
it @emph{raises} an error;
that is, it does a long jump.
A @emphx{protected environment} uses @id{setjmp}
to set a recovery point;
any error jumps to the most recent active recovery point.

Inside a @N{C function} you can raise an error explicitly
by calling @Lid{lua_error}.

Most functions in the API can raise an error,
for instance due to a @x{memory allocation error}.
The documentation for each function indicates whether
it can raise errors.

If an error happens outside any protected environment,
Lua calls a @def{panic function} (see @Lid{lua_atpanic})
and then calls @T{abort},
thus exiting the host application.
Your panic function can avoid this exit by
never returning
(e.g., doing a long jump to your own recovery point outside Lua).

The panic function,
as its name implies,
is a mechanism of last resort.
Programs should avoid it.
As a general rule,
when a @N{C function} is called by Lua with a Lua state,
it can do whatever it wants on that Lua state,
as it should be already protected.
However,
when C code operates on other Lua states
(e.g., a Lua-state argument to the function,
a Lua state stored in the registry, or
the result of @Lid{lua_newthread}),
it should use them only in API calls that cannot raise errors.

The panic function runs as if it were a @x{message handler} @see{error};
in particular, the error object is on the top of the stack.
However, there is no guarantee about stack space.
To push anything on the stack,
the panic function must first check the available space @see{stacksize}.

}


@sect3{statuscodes|@title{Status Codes}

Several functions that report errors in the API use the following
status codes to indicate different kinds of errors or other conditions:
@description{

@item{@defid{LUA_OK} (0)| no errors.}

@item{@defid{LUA_ERRRUN}| a runtime error.}

@item{@defid{LUA_ERRMEM}|
@x{memory allocation error}.
For such errors, Lua does not call the @x{message handler}.
}

@item{@defid{LUA_ERRERR}| error while running the @x{message handler}.}

@item{@defid{LUA_ERRSYNTAX}| syntax error during precompilation
or format error in a binary chunk.}

@item{@defid{LUA_YIELD}| the thread (coroutine) yields.}

@item{@defid{LUA_ERRFILE}| a file-related error;
e.g., it cannot open or read the file.}

}
These constants are defined in the header file @id{lua.h}.

}

}

@sect2{continuations|@title{Handling Yields in C}

Internally, Lua uses the C @id{longjmp} facility to yield a coroutine.
Therefore, if a @N{C function} @id{foo} calls an API function
and this API function yields
(directly or indirectly by calling another function that yields),
Lua cannot return to @id{foo} any more,
because the @id{longjmp} removes its frame from the @N{C stack}.

To avoid this kind of problem,
Lua raises an error whenever it tries to yield across an API call,
except for three functions:
@Lid{lua_yieldk}, @Lid{lua_callk}, and @Lid{lua_pcallk}.
All those functions receive a @def{continuation function}
(as a parameter named @id{k}) to continue execution after a yield.

We need to set some terminology to explain continuations.
We have a @N{C function} called from Lua which we will call
the @emph{original function}.
This original function then calls one of those three functions in the C API,
which we will call the @emph{callee function},
that then yields the current thread.
This can happen when the callee function is @Lid{lua_yieldk},
or when the callee function is either @Lid{lua_callk} or @Lid{lua_pcallk}
and the function called by them yields.

Suppose the running thread yields while executing the callee function.
After the thread resumes,
it eventually will finish running the callee function.
However,
the callee function cannot return to the original function,
because its frame in the @N{C stack} was destroyed by the yield.
Instead, Lua calls a @def{continuation function},
which was given as an argument to the callee function.
As the name implies,
the continuation function should continue the task
of the original function.

As an illustration, consider the following function:
@verbatim{
int original_function (lua_State *L) {
  ...     /* code 1 */
  status = lua_pcall(L, n, m, h);  /* calls Lua */
  ...     /* code 2 */
}
}
Now we want to allow
the Lua code being run by @Lid{lua_pcall} to yield.
First, we can rewrite our function like here:
@verbatim{
int k (lua_State *L, int status, lua_KContext ctx) {
  ...  /* code 2 */
}

int original_function (lua_State *L) {
  ...     /* code 1 */
  return k(L, lua_pcall(L, n, m, h), ctx);
}
}
In the above code,
the new function @id{k} is a
@emph{continuation function} (with type @Lid{lua_KFunction}),
which should do all the work that the original function
was doing after calling @Lid{lua_pcall}.
Now, we must inform Lua that it must call @id{k} if the Lua code
being executed by @Lid{lua_pcall} gets interrupted in some way
(errors or yielding),
so we rewrite the code as here,
replacing @Lid{lua_pcall} by @Lid{lua_pcallk}:
@verbatim{
int original_function (lua_State *L) {
  ...     /* code 1 */
  return k(L, lua_pcallk(L, n, m, h, ctx2, k), ctx1);
}
}
Note the external, explicit call to the continuation:
Lua will call the continuation only if needed, that is,
in case of errors or resuming after a yield.
If the called function returns normally without ever yielding,
@Lid{lua_pcallk} (and @Lid{lua_callk}) will also return normally.
(Of course, instead of calling the continuation in that case,
you can do the equivalent work directly inside the original function.)

Besides the Lua state,
the continuation function has two other parameters:
the final status of the call and the context value (@id{ctx}) that
was passed originally to @Lid{lua_pcallk}.
Lua does not use this context value;
it only passes this value from the original function to the
continuation function.
For @Lid{lua_pcallk},
the status is the same value that would be returned by @Lid{lua_pcallk},
except that it is @Lid{LUA_YIELD} when being executed after a yield
(instead of @Lid{LUA_OK}).
For @Lid{lua_yieldk} and @Lid{lua_callk},
the status is always @Lid{LUA_YIELD} when Lua calls the continuation.
(For these two functions,
Lua will not call the continuation in case of errors,
because they do not handle errors.)
Similarly, when using @Lid{lua_callk},
you should call the continuation function
with @Lid{LUA_OK} as the status.
(For @Lid{lua_yieldk}, there is not much point in calling
directly the continuation function,
because @Lid{lua_yieldk} usually does not return.)

Lua treats the continuation function as if it were the original function.
The continuation function receives the same Lua stack
from the original function,
in the same state it would be if the callee function had returned.
(For instance,
after a @Lid{lua_callk} the function and its arguments are
removed from the stack and replaced by the results from the call.)
It also has the same upvalues.
Whatever it returns is handled by Lua as if it were the return
of the original function.

}

@sect2{@title{Functions and Types}

Here we list all functions and types from the @N{C API} in
alphabetical order.
Each function has an indicator like this:
@apii{o,p,x}

The first field, @T{o},
is how many elements the function pops from the stack.
The second field, @T{p},
is how many elements the function pushes onto the stack.
(Any function always pushes its results after popping its arguments.)
A field in the form @T{x|y} means the function can push (or pop)
@T{x} or @T{y} elements,
depending on the situation;
an interrogation mark @Char{?} means that
we cannot know how many elements the function pops/pushes
by looking only at its arguments.
(For instance, they may depend on what is in the stack.)
The third field, @T{x},
tells whether the function may raise errors:
@Char{-} means the function never raises any error;
@Char{m} means the function may raise only out-of-memory errors;
@Char{v} means the function may raise the errors explained in the text;
@Char{e} means the function can run arbitrary Lua code,
either directly or through metamethods,
and therefore may raise any errors.


@APIEntry{int lua_absindex (lua_State *L, int idx);|
@apii{0,0,-}

Converts the @x{acceptable index} @id{idx}
into an equivalent @x{absolute index}
(that is, one that does not depend on the stack size).

}


@APIEntry{
typedef void * (*lua_Alloc) (void *ud,
                             void *ptr,
                             size_t osize,
                             size_t nsize);|

The type of the @x{memory-allocation function} used by Lua states.
The allocator function must provide a
functionality similar to @id{realloc},
but not exactly the same.
Its arguments are
@id{ud}, an opaque pointer passed to @Lid{lua_newstate};
@id{ptr}, a pointer to the block being allocated/reallocated/freed;
@id{osize}, the original size of the block or some code about what
is being allocated;
and @id{nsize}, the new size of the block.

When @id{ptr} is not @id{NULL},
@id{osize} is the size of the block pointed by @id{ptr},
that is, the size given when it was allocated or reallocated.

When @id{ptr} is @id{NULL},
@id{osize} encodes the kind of object that Lua is allocating.
@id{osize} is any of
@Lid{LUA_TSTRING}, @Lid{LUA_TTABLE}, @Lid{LUA_TFUNCTION},
@Lid{LUA_TUSERDATA}, or @Lid{LUA_TTHREAD} when (and only when)
Lua is creating a new object of that type.
When @id{osize} is some other value,
Lua is allocating memory for something else.

Lua assumes the following behavior from the allocator function:

When @id{nsize} is zero,
the allocator must behave like @id{free}
and then return @id{NULL}.

When @id{nsize} is not zero,
the allocator must behave like @id{realloc}.
In particular, the allocator returns @id{NULL}
if and only if it cannot fulfill the request.

Here is a simple implementation for the @x{allocator function}.
It is used in the auxiliary library by @Lid{luaL_newstate}.
@verbatim{
static void *l_alloc (void *ud, void *ptr, size_t osize,
                                           size_t nsize) {
  (void)ud;  (void)osize;  /* not used */
  if (nsize == 0) {
    free(ptr);
    return NULL;
  }
  else
    return realloc(ptr, nsize);
}
}
Note that @N{ISO C} ensures
that @T{free(NULL)} has no effect and that
@T{realloc(NULL,size)} is equivalent to @T{malloc(size)}.

}

@APIEntry{void lua_arith (lua_State *L, int op);|
@apii{2|1,1,e}

Performs an arithmetic or bitwise operation over the two values
(or one, in the case of negations)
at the top of the stack,
with the value on the top being the second operand,
pops these values, and pushes the result of the operation.
The function follows the semantics of the corresponding Lua operator
(that is, it may call metamethods).

The value of @id{op} must be one of the following constants:
@description{

@item{@defid{LUA_OPADD}| performs addition (@T{+})}
@item{@defid{LUA_OPSUB}| performs subtraction (@T{-})}
@item{@defid{LUA_OPMUL}| performs multiplication (@T{*})}
@item{@defid{LUA_OPDIV}| performs float division (@T{/})}
@item{@defid{LUA_OPIDIV}| performs floor division (@T{//})}
@item{@defid{LUA_OPMOD}| performs modulo (@T{%})}
@item{@defid{LUA_OPPOW}| performs exponentiation (@T{^})}
@item{@defid{LUA_OPUNM}| performs mathematical negation (unary @T{-})}
@item{@defid{LUA_OPBNOT}| performs bitwise NOT (@T{~})}
@item{@defid{LUA_OPBAND}| performs bitwise AND (@T{&})}
@item{@defid{LUA_OPBOR}| performs bitwise OR (@T{|})}
@item{@defid{LUA_OPBXOR}| performs bitwise exclusive OR (@T{~})}
@item{@defid{LUA_OPSHL}| performs left shift (@T{<<})}
@item{@defid{LUA_OPSHR}| performs right shift (@T{>>})}

}

}

@APIEntry{lua_CFunction lua_atpanic (lua_State *L, lua_CFunction panicf);|
@apii{0,0,-}

Sets a new panic function and returns the old one @see{C-error}.

}

@APIEntry{void lua_call (lua_State *L, int nargs, int nresults);|
@apii{nargs+1,nresults,e}

Calls a function.
Like regular Lua calls,
@id{lua_call} respects the @idx{__call} metamethod.
So, here the word @Q{function}
means any callable value.

To do a call you must use the following protocol:
first, the function to be called is pushed onto the stack;
then, the arguments to the call are pushed
in direct order;
that is, the first argument is pushed first.
Finally you call @Lid{lua_call};
@id{nargs} is the number of arguments that you pushed onto the stack.
When the function returns,
all arguments and the function value are popped
and the call results are pushed onto the stack.
The number of results is adjusted to @id{nresults},
unless @id{nresults} is @defid{LUA_MULTRET},
which makes all results from the function to be pushed.
In the first case, an explicit number of results,
the caller must ensure that the stack has space for the
returned values.
In the second case, all results,
Lua takes care that the returned values fit into the stack space,
but it does not ensure any extra space in the stack.
The function results are pushed onto the stack in direct order
(the first result is pushed first),
so that after the call the last result is on the top of the stack.

The maximum value for @id{nresults} is 250.

Any error while calling and running the function is propagated upwards
(with a @id{longjmp}).

The following example shows how the host program can do the
equivalent to this Lua code:
@verbatim{
a = f("how", t.x, 14)
}
Here it is @N{in C}:
@verbatim{
lua_getglobal(L, "f");                  /* function to be called */
lua_pushliteral(L, "how");                       /* 1st argument */
lua_getglobal(L, "t");                    /* table to be indexed */
lua_getfield(L, -1, "x");        /* push result of t.x (2nd arg) */
lua_remove(L, -2);                  /* remove 't' from the stack */
lua_pushinteger(L, 14);                          /* 3rd argument */
lua_call(L, 3, 1);     /* call 'f' with 3 arguments and 1 result */
lua_setglobal(L, "a");                         /* set global 'a' */
}
Note that the code above is @emph{balanced}:
at its end, the stack is back to its original configuration.
This is considered good programming practice.

}

@APIEntry{
void lua_callk (lua_State *L,
                int nargs,
                int nresults,
                lua_KContext ctx,
                lua_KFunction k);|
@apii{nargs + 1,nresults,e}

This function behaves exactly like @Lid{lua_call},
but allows the called function to yield @see{continuations}.

}

@APIEntry{typedef int (*lua_CFunction) (lua_State *L);|

Type for @N{C functions}.

In order to communicate properly with Lua,
a @N{C function} must use the following protocol,
which defines the way parameters and results are passed:
a @N{C function} receives its arguments from Lua in its stack
in direct order (the first argument is pushed first).
So, when the function starts,
@T{lua_gettop(L)} returns the number of arguments received by the function.
The first argument (if any) is at index 1
and its last argument is at index @T{lua_gettop(L)}.
To return values to Lua, a @N{C function} just pushes them onto the stack,
in direct order (the first result is pushed first),
and returns in C the number of results.
Any other value in the stack below the results will be properly
discarded by Lua.
Like a Lua function, a @N{C function} called by Lua can also return
many results.

As an example, the following function receives a variable number
of numeric arguments and returns their average and their sum:
@verbatim{
static int foo (lua_State *L) {
  int n = lua_gettop(L);    /* number of arguments */
  lua_Number sum = 0.0;
  int i;
  for (i = 1; i <= n; i++) {
    if (!lua_isnumber(L, i)) {
      lua_pushliteral(L, "incorrect argument");
      lua_error(L);
    }
    sum += lua_tonumber(L, i);
  }
  lua_pushnumber(L, sum/n);        /* first result */
  lua_pushnumber(L, sum);         /* second result */
  return 2;                   /* number of results */
}
}



}


@APIEntry{int lua_checkstack (lua_State *L, int n);|
@apii{0,0,-}

Ensures that the stack has space for at least @id{n} extra elements,
that is, that you can safely push up to @id{n} values into it.
It returns false if it cannot fulfill the request,
either because it would cause the stack
to be greater than a fixed maximum size
(typically at least several thousand elements) or
because it cannot allocate memory for the extra space.
This function never shrinks the stack;
if the stack already has space for the extra elements,
it is left unchanged.

}

@APIEntry{void lua_close (lua_State *L);|
@apii{0,0,-}

Close all active to-be-closed variables in the main thread,
release all objects in the given Lua state
(calling the corresponding garbage-collection metamethods, if any),
and frees all dynamic memory used by this state.

On several platforms, you may not need to call this function,
because all resources are naturally released when the host program ends.
On the other hand, long-running programs that create multiple states,
such as daemons or web servers,
will probably need to close states as soon as they are not needed.

}

@APIEntry{void lua_closeslot (lua_State *L, int index);|
@apii{0,0,e}

Close the to-be-closed slot at the given index and set its value to @nil.
The index must be the last index previously marked to be closed
@see{lua_toclose} that is still active (that is, not closed yet).

A @idx{__close} metamethod cannot yield
when called through this function.

}

@APIEntry{int lua_closethread (lua_State *L, lua_State *from);|
@apii{0,?,-}

Resets a thread, cleaning its call stack and closing all pending
to-be-closed variables.
Returns a status code:
@Lid{LUA_OK} for no errors in the thread
(either the original error that stopped the thread or
errors in closing methods),
or an error status otherwise.
In case of error,
leaves the error object on the top of the stack.

The parameter @id{from} represents the coroutine that is resetting @id{L}.
If there is no such coroutine,
this parameter can be @id{NULL}.

}

@APIEntry{int lua_compare (lua_State *L, int index1, int index2, int op);|
@apii{0,0,e}

Compares two Lua values.
Returns 1 if the value at index @id{index1} satisfies @id{op}
when compared with the value at index @id{index2},
following the semantics of the corresponding Lua operator
(that is, it may call metamethods).
Otherwise @N{returns 0}.
Also @N{returns 0} if any of the indices is not valid.

The value of @id{op} must be one of the following constants:
@description{

@item{@defid{LUA_OPEQ}| compares for equality (@T{==})}
@item{@defid{LUA_OPLT}| compares for less than (@T{<})}
@item{@defid{LUA_OPLE}| compares for less or equal (@T{<=})}

}

}

@APIEntry{void lua_concat (lua_State *L, int n);|
@apii{n,1,e}

Concatenates the @id{n} values at the top of the stack,
pops them, and leaves the result on the top.
If @N{@T{n} is 1}, the result is the single value on the stack
(that is, the function does nothing);
if @id{n} is 0, the result is the empty string.
Concatenation is performed following the usual semantics of Lua
@see{concat}.

}

@APIEntry{void lua_copy (lua_State *L, int fromidx, int toidx);|
@apii{0,0,-}

Copies the element at index @id{fromidx}
into the valid index @id{toidx},
replacing the value at that position.
Values at other positions are not affected.

}

@APIEntry{void lua_createtable (lua_State *L, unsigned nseq, unsigned nrec);|
@apii{0,1,m}

Creates a new empty table and pushes it onto the stack.
Parameter @id{nseq} is a hint for how many elements the table
will have as a sequence;
parameter @id{nrec} is a hint for how many other elements
the table will have.
Lua may use these hints to preallocate memory for the new table.
This preallocation may help performance when you know in advance
how many elements the table will have.
Otherwise you can use the function @Lid{lua_newtable}.

}

@APIEntry{int lua_dump (lua_State *L,
                        lua_Writer writer,
                        void *data,
                        int strip);|
@apii{0,0,-}

Dumps a function as a binary chunk.
Receives a Lua function on the top of the stack
and produces a binary chunk that,
if loaded again,
results in a function equivalent to the one dumped.
As it produces parts of the chunk,
@Lid{lua_dump} calls function @id{writer} @seeC{lua_Writer}
with the given @id{data}
to write them.

The function @Lid{lua_dump} fully preserves the Lua stack
through the calls to the writer function,
except that it may push some values for internal use
before the first call,
and it restores the stack size to its original size
after the last call.

If @id{strip} is true,
the binary representation may not include all debug information
about the function,
to save space.

The value returned is the error code returned by the last
call to the writer;
@N{0 means} no errors.

}

@APIEntry{int lua_error (lua_State *L);|
@apii{1,0,v}

Raises a Lua error,
using the value on the top of the stack as the error object.
This function does a long jump,
and therefore never returns
@seeC{luaL_error}.

}

@APIEntry{int lua_gc (lua_State *L, int what, ...);|
@apii{0,0,-}

Controls the garbage collector.

This function performs several tasks,
according to the value of the parameter @id{what}.
For options that need extra arguments,
they are listed after the option.
@description{

@item{@defid{LUA_GCCOLLECT}|
Performs a full garbage-collection cycle.
}

@item{@defid{LUA_GCSTOP}|
Stops the garbage collector.
}

@item{@defid{LUA_GCRESTART}|
Restarts the garbage collector.
}

@item{@defid{LUA_GCCOUNT}|
Returns the current amount of memory (in Kbytes) in use by Lua.
}

@item{@defid{LUA_GCCOUNTB}|
Returns the remainder of dividing the current amount of bytes of
memory in use by Lua by 1024.
}

@item{@defid{LUA_GCSTEP} (size_t n)|
Performs a step of garbage collection.
}

@item{@defid{LUA_GCISRUNNING}|
Returns a boolean that tells whether the collector is running
(i.e., not stopped).
}

@item{@defid{LUA_GCINC}|
Changes the collector to incremental mode.
Returns the previous mode (@id{LUA_GCGEN} or @id{LUA_GCINC}).
}

@item{@defid{LUA_GCGEN}|
Changes the collector to generational mode.
Returns the previous mode (@id{LUA_GCGEN} or @id{LUA_GCINC}).
}

@item{@defid{LUA_GCPARAM} (int param, int val)|
Changes and/or returns the value of a parameter of the collector.
If @id{val} is negative, the call only returns the current value.
The argument @id{param} must have one of the following values:
@description{
@item{@defid{LUA_GCPMINORMUL}| The minor multiplier. }
@item{@defid{LUA_GCPMAJORMINOR}| The major-minor multiplier. }
@item{@defid{LUA_GCPMINORMAJOR}| The minor-major multiplier. }
@item{@defid{LUA_GCPPAUSE}| The garbage-collector pause. }
@item{@defid{LUA_GCPSTEPMUL}| The step multiplier. }
@item{@defid{LUA_GCPSTEPSIZE}| The step size. }
}
}

}

For more details about these options,
see @Lid{collectgarbage}.

This function should not be called by a finalizer.

}

@APIEntry{lua_Alloc lua_getallocf (lua_State *L, void **ud);|
@apii{0,0,-}

Returns the @x{memory-allocation function} of a given state.
If @id{ud} is not @id{NULL}, Lua stores in @T{*ud} the
opaque pointer given when the memory-allocator function was set.

}

@APIEntry{int lua_getfield (lua_State *L, int index, const char *k);|
@apii{0,1,e}

Pushes onto the stack the value @T{t[k]},
where @id{t} is the value at the given index.
As in Lua, this function may trigger a metamethod
for the @Q{index} event @see{metatable}.

Returns the type of the pushed value.

}

@APIEntry{void *lua_getextraspace (lua_State *L);|
@apii{0,0,-}

Returns a pointer to a raw memory area associated with the
given Lua state.
The application can use this area for any purpose;
Lua does not use it for anything.

Each new thread has this area initialized with a copy
of the area of the @x{main thread}.

By default, this area has the size of a pointer to void,
but you can recompile Lua with a different size for this area.
(See @id{LUA_EXTRASPACE} in @id{luaconf.h}.)

}

@APIEntry{int lua_getglobal (lua_State *L, const char *name);|
@apii{0,1,e}

Pushes onto the stack the value of the global @id{name}.
Returns the type of that value.

}

@APIEntry{int lua_geti (lua_State *L, int index, lua_Integer i);|
@apii{0,1,e}

Pushes onto the stack the value @T{t[i]},
where @id{t} is the value at the given index.
As in Lua, this function may trigger a metamethod
for the @Q{index} event @see{metatable}.

Returns the type of the pushed value.

}

@APIEntry{int lua_getmetatable (lua_State *L, int index);|
@apii{0,0|1,-}

If the value at the given index has a metatable,
the function pushes that metatable onto the stack and @N{returns 1}.
Otherwise,
the function @N{returns 0} and pushes nothing on the stack.

}

@APIEntry{int lua_gettable (lua_State *L, int index);|
@apii{1,1,e}

Pushes onto the stack the value @T{t[k]},
where @id{t} is the value at the given index
and @id{k} is the value on the top of the stack.

This function pops the key from the stack,
pushing the resulting value in its place.
As in Lua, this function may trigger a metamethod
for the @Q{index} event @see{metatable}.

Returns the type of the pushed value.

}

@APIEntry{int lua_gettop (lua_State *L);|
@apii{0,0,-}

Returns the index of the top element in the stack.
Because indices start @N{at 1},
this result is equal to the number of elements in the stack;
in particular, @N{0 means} an empty stack.

}

@APIEntry{int lua_getiuservalue (lua_State *L, int index, int n);|
@apii{0,1,-}

Pushes onto the stack the @id{n}-th user value associated with the
full userdata at the given index and
returns the type of the pushed value.

If the userdata does not have that value,
pushes @nil and returns @Lid{LUA_TNONE}.

}

@APIEntry{void lua_insert (lua_State *L, int index);|
@apii{1,1,-}

Moves the top element into the given valid index,
shifting up the elements above this index to open space.
This function cannot be called with a pseudo-index,
because a pseudo-index is not an actual stack position.

}

@APIEntry{typedef @ldots lua_Integer;|

The type of integers in Lua.

By default this type is @id{long long},
(usually a 64-bit two-complement integer),
but that can be changed to @id{long} or @id{int}
(usually a 32-bit two-complement integer).
(See @id{LUA_INT_TYPE} in @id{luaconf.h}.)

Lua also defines the constants
@defid{LUA_MININTEGER} and @defid{LUA_MAXINTEGER},
with the minimum and the maximum values that fit in this type.

}

@APIEntry{int lua_isboolean (lua_State *L, int index);|
@apii{0,0,-}

Returns 1 if the value at the given index is a boolean,
and @N{0 otherwise}.

}

@APIEntry{int lua_iscfunction (lua_State *L, int index);|
@apii{0,0,-}

Returns 1 if the value at the given index is a @N{C function},
and @N{0 otherwise}.

}

@APIEntry{int lua_isfunction (lua_State *L, int index);|
@apii{0,0,-}

Returns 1 if the value at the given index is a function
(either C or Lua), and @N{0 otherwise}.

}

@APIEntry{int lua_isinteger (lua_State *L, int index);|
@apii{0,0,-}

Returns 1 if the value at the given index is an integer
(that is, the value is a number and is represented as an integer),
and @N{0 otherwise}.

}

@APIEntry{int lua_islightuserdata (lua_State *L, int index);|
@apii{0,0,-}

Returns 1 if the value at the given index is a light userdata,
and @N{0 otherwise}.

}

@APIEntry{int lua_isnil (lua_State *L, int index);|
@apii{0,0,-}

Returns 1 if the value at the given index is @nil,
and @N{0 otherwise}.

}

@APIEntry{int lua_isnone (lua_State *L, int index);|
@apii{0,0,-}

Returns 1 if the given index is not valid,
and @N{0 otherwise}.

}

@APIEntry{int lua_isnoneornil (lua_State *L, int index);|
@apii{0,0,-}

Returns 1 if the given index is not valid
or if the value at this index is @nil,
and @N{0 otherwise}.

}

@APIEntry{int lua_isnumber (lua_State *L, int index);|
@apii{0,0,-}

Returns 1 if the value at the given index is a number
or a string convertible to a number,
and @N{0 otherwise}.

}

@APIEntry{int lua_isstring (lua_State *L, int index);|
@apii{0,0,-}

Returns 1 if the value at the given index is a string
or a number (which is always convertible to a string),
and @N{0 otherwise}.

}

@APIEntry{int lua_istable (lua_State *L, int index);|
@apii{0,0,-}

Returns 1 if the value at the given index is a table,
and @N{0 otherwise}.

}

@APIEntry{int lua_isthread (lua_State *L, int index);|
@apii{0,0,-}

Returns 1 if the value at the given index is a thread,
and @N{0 otherwise}.

}

@APIEntry{int lua_isuserdata (lua_State *L, int index);|
@apii{0,0,-}

Returns 1 if the value at the given index is a userdata
(either full or light), and @N{0 otherwise}.

}

@APIEntry{int lua_isyieldable (lua_State *L);|
@apii{0,0,-}

Returns 1 if the given coroutine can yield,
and @N{0 otherwise}.

}

@APIEntry{typedef @ldots lua_KContext;|

The type for continuation-function contexts.
It must be a numeric type.
This type is defined as @id{intptr_t}
when @id{intptr_t} is available,
so that it can store pointers too.
Otherwise, it is defined as @id{ptrdiff_t}.

}

@APIEntry{
typedef int (*lua_KFunction) (lua_State *L, int status, lua_KContext ctx);|

Type for continuation functions @see{continuations}.

}

@APIEntry{void lua_len (lua_State *L, int index);|
@apii{0,1,e}

Returns the length of the value at the given index.
It is equivalent to the @Char{#} operator in Lua @see{len-op} and
may trigger a metamethod for the @Q{length} event @see{metatable}.
The result is pushed on the stack.

}

@APIEntry{
int lua_load (lua_State *L,
              lua_Reader reader,
              void *data,
              const char *chunkname,
              const char *mode);|
@apii{0,1,-}

Loads a Lua chunk without running it.
If there are no errors,
@id{lua_load} pushes the compiled chunk as a Lua
function on top of the stack.
Otherwise, it pushes an error message.

The @id{lua_load} function uses a user-supplied @id{reader} function
to read the chunk @seeC{lua_Reader}.
The @id{data} argument is an opaque value passed to the reader function.

The @id{chunkname} argument gives a name to the chunk,
which is used for error messages and in debug information @see{debugI}.

@id{lua_load} automatically detects whether the chunk is text or binary
and loads it accordingly (see program @idx{luac}).
The string @id{mode} works as in function @Lid{load},
with the addition that
a @id{NULL} value is equivalent to the string @St{bt}.
Moreover, it may have a @Char{B} instead of a @Char{b},
meaning a @emphx{fixed buffer} with the binary dump.

A fixed buffer means that the address returned by the reader function
will contain the chunk until everything created by the chunk has
been collected;
therefore, Lua can avoid copying to internal structures
some parts of the chunk.
(In general, a fixed buffer would keep its contents
until the end of the program,
for instance with the chunk in ROM.)
Moreover, for a fixed buffer,
the reader function should return the entire chunk in the first read.
(As an example, @Lid{luaL_loadbufferx} does that,
which means that you can use it to load fixed buffers.)

The function @Lid{lua_load} fully preserves the Lua stack
through the calls to the reader function,
except that it may push some values for internal use
before the first call,
and it restores the stack size to its original size plus one
(for the pushed result) after the last call.

@id{lua_load} can return
@Lid{LUA_OK}, @Lid{LUA_ERRSYNTAX}, or @Lid{LUA_ERRMEM}.
The function may also return other values corresponding to
errors raised by the read function @see{statuscodes}.

If the resulting function has upvalues,
its first upvalue is set to the value of the @x{global environment}
stored at index @id{LUA_RIDX_GLOBALS} in the registry @see{registry}.
When loading main chunks,
this upvalue will be the @id{_ENV} variable @see{globalenv}.
Other upvalues are initialized with @nil.

}

@APIEntry{lua_State *lua_newstate (lua_Alloc f, void *ud,
                                   unsigned int seed);|
@apii{0,0,-}

Creates a new independent state and returns its main thread.
Returns @id{NULL} if it cannot create the state
(due to lack of memory).
The argument @id{f} is the @x{allocator function};
Lua will do all memory allocation for this state
through this function @seeF{lua_Alloc}.
The second argument, @id{ud}, is an opaque pointer that Lua
passes to the allocator in every call.
The third argument, @id{seed},
is a seed for the hashing of strings.

}

@APIEntry{void lua_newtable (lua_State *L);|
@apii{0,1,m}

Creates a new empty table and pushes it onto the stack.
It is equivalent to @T{lua_createtable(L, 0, 0)}.

}

@APIEntry{lua_State *lua_newthread (lua_State *L);|
@apii{0,1,m}

Creates a new thread, pushes it on the stack,
and returns a pointer to a @Lid{lua_State} that represents this new thread.
The new thread returned by this function shares with the original thread
its global environment,
but has an independent execution stack.

Threads are subject to garbage collection,
like any Lua object.

}

@APIEntry{void *lua_newuserdatauv (lua_State *L, size_t size, int nuvalue);|
@apii{0,1,m}

This function creates and pushes on the stack a new full userdata,
with @id{nuvalue} associated Lua values, called @id{user values},
plus an associated block of raw memory with @id{size} bytes.
(The user values can be set and read with the functions
@Lid{lua_setiuservalue} and @Lid{lua_getiuservalue}.)

The function returns the address of the block of memory.
Lua ensures that this address is valid as long as
the corresponding userdata is alive @see{GC}.
Moreover, if the userdata is marked for finalization @see{finalizers},
its address is valid at least until the call to its finalizer.

}

@APIEntry{int lua_next (lua_State *L, int index);|
@apii{1,2|0,v}

Pops a key from the stack,
and pushes a key@En{}value pair from the table at the given index,
the @Q{next} pair after the given key.
If there are no more elements in the table,
then @Lid{lua_next} @N{returns 0} and pushes nothing.

A typical table traversal looks like this:
@verbatim{
/* table is in the stack at index 't' */
lua_pushnil(L);  /* first key */
while (lua_next(L, t) != 0) {
  /* uses 'key' (at index -2) and 'value' (at index -1) */
  printf("%s - %s\n",
         lua_typename(L, lua_type(L, -2)),
         lua_typename(L, lua_type(L, -1)));
  /* removes 'value'; keeps 'key' for next iteration */
  lua_pop(L, 1);
}
}

While traversing a table,
avoid calling @Lid{lua_tolstring} directly on a key,
unless you know that the key is actually a string.
Recall that @Lid{lua_tolstring} may change
the value at the given index;
this confuses the next call to @Lid{lua_next}.

This function may raise an error if the given key
is neither @nil nor present in the table.
See function @Lid{next} for the caveats of modifying
the table during its traversal.

}

@APIEntry{typedef @ldots lua_Number;|

The type of floats in Lua.

By default this type is double,
but that can be changed to a single float or a long double.
(See @id{LUA_FLOAT_TYPE} in @id{luaconf.h}.)

}

@APIEntry{int lua_numbertointeger (lua_Number n, lua_Integer *p);|

Tries to convert a Lua float to a Lua integer;
the float @id{n} must have an integral value.
If that value is within the range of Lua integers,
it is converted to an integer and assigned to @T{*p}.
The macro results in a boolean indicating whether the
conversion was successful.
(Note that this range test can be tricky to do
correctly without this macro, due to rounding.)

This macro may evaluate its arguments more than once.

}

@APIEntry{unsigned (lua_numbertostrbuff) (lua_State *L, int idx,
                                          char *buff);|
@apii{0,0,-}

Converts the number at acceptable index @id{idx} to a string
and puts the result in @id{buff}.
The buffer must have a size of at least @Lid{LUA_N2SBUFFSZ} bytes.
The conversion follows a non-specified format @see{coercion}.
The function returns the number of bytes written to the buffer
(including the final zero),
or zero if the value at @id{idx} is not a number.

}

@APIEntry{int lua_pcall (lua_State *L, int nargs, int nresults, int msgh);|
@apii{nargs + 1,nresults|1,-}

Calls a function (or a callable object) in protected mode.


Both @id{nargs} and @id{nresults} have the same meaning as
in @Lid{lua_call}.
If there are no errors during the call,
@Lid{lua_pcall} behaves exactly like @Lid{lua_call}.
However, if there is any error,
@Lid{lua_pcall} catches it,
pushes a single value on the stack (the error object),
and returns an error code.
Like @Lid{lua_call},
@Lid{lua_pcall} always removes the function
and its arguments from the stack.

If @id{msgh} is 0,
then the error object returned on the stack
is exactly the original error object.
Otherwise, @id{msgh} is the stack index of a
@emph{message handler}.
(This index cannot be a pseudo-index.)
In case of runtime errors,
this handler will be called with the error object
and its return value will be the object
returned on the stack by @Lid{lua_pcall}.

Typically, the message handler is used to add more debug
information to the error object, such as a stack traceback.
Such information cannot be gathered after the return of @Lid{lua_pcall},
since by then the stack has unwound.

The @Lid{lua_pcall} function returns one of the following status codes:
@Lid{LUA_OK}, @Lid{LUA_ERRRUN}, @Lid{LUA_ERRMEM}, or @Lid{LUA_ERRERR}.

}

@APIEntry{
int lua_pcallk (lua_State *L,
                int nargs,
                int nresults,
                int msgh,
                lua_KContext ctx,
                lua_KFunction k);|
@apii{nargs + 1,nresults|1,-}

This function behaves exactly like @Lid{lua_pcall},
except that it allows the called function to yield @see{continuations}.

}

@APIEntry{void lua_pop (lua_State *L, int n);|
@apii{n,0,e}

Pops @id{n} elements from the stack.
It is implemented as a macro over @Lid{lua_settop}.

}

@APIEntry{void lua_pushboolean (lua_State *L, int b);|
@apii{0,1,-}

Pushes a boolean value with value @id{b} onto the stack.

}

@APIEntry{void lua_pushcclosure (lua_State *L, lua_CFunction fn, int n);|
@apii{n,1,m}

Pushes a new @N{C closure} onto the stack.
This function receives a pointer to a @N{C function}
and pushes onto the stack a Lua value of type @id{function} that,
when called, invokes the corresponding @N{C function}.
The parameter @id{n} tells how many upvalues this function will have
@see{c-closure}.

Any function to be callable by Lua must
follow the correct protocol to receive its parameters
and return its results @seeC{lua_CFunction}.

When a @N{C function} is created,
it is possible to associate some values with it,
the so called upvalues;
these upvalues are then accessible to the function whenever it is called.
This association is called a @x{@N{C closure}} @see{c-closure}.
To create a @N{C closure},
first the initial values for its upvalues must be pushed onto the stack.
(When there are multiple upvalues, the first value is pushed first.)
Then @Lid{lua_pushcclosure}
is called to create and push the @N{C function} onto the stack,
with the argument @id{n} telling how many values will be
associated with the function.
@Lid{lua_pushcclosure} also pops these values from the stack.

The maximum value for @id{n} is 255.

When @id{n} is zero,
this function creates a @def{light @N{C function}},
which is just a pointer to the @N{C function}.
In that case, it never raises a memory error.

}

@APIEntry{void lua_pushcfunction (lua_State *L, lua_CFunction f);|
@apii{0,1,-}

Pushes a @N{C function} onto the stack.
This function is equivalent to @Lid{lua_pushcclosure} with no upvalues.

}

@APIEntry{const char *(lua_pushextlstring) (lua_State *L,
                const char *s, size_t len, lua_Alloc falloc, void *ud);|
@apii{0,1,m}

Creates an @emphx{external string},
that is, a string that uses memory not managed by Lua.
The pointer @id{s} points to the external buffer
holding the string content,
and @id{len} is the length of the string.
The string should have a zero at its end,
that is, the condition @T{s[len] == '\0'} should hold.
As with any string in Lua,
the length must fit in a Lua integer.

If @id{falloc} is different from @id{NULL},
that function will be called by Lua
when the external buffer is no longer needed.
The contents of the buffer should not change before this call.
The function will be called with the given @id{ud},
the string @id{s} as the block,
the length plus one (to account for the ending zero) as the old size,
and 0 as the new size.

Lua always @x{internalizes} strings with lengths up to 40 characters.
So, for strings in that range,
this function will immediately internalize the string
and call @id{falloc} to free the buffer.

Even when using an external buffer,
Lua still has to allocate a header for the string.
In case of a memory-allocation error,
Lua will call @id{falloc} before raising the error.

}


@APIEntry{const char *lua_pushfstring (lua_State *L, const char *fmt, ...);|
@apii{0,1,m}

Pushes onto the stack a formatted string
and returns a pointer to this string @see{constchar}.
It is similar to the @ANSI{sprintf},
but has two important differences.
First,
you do not have to allocate space for the result;
the result is a Lua string and Lua takes care of memory allocation
(and deallocation, through garbage collection).
Second,
the conversion specifiers are quite restricted.
There are no flags, widths, or precisions.
The conversion specifiers can only be
@Char{%%} (inserts the character @Char{%}),
@Char{%s} (inserts a zero-terminated string, with no size restrictions),
@Char{%f} (inserts a @Lid{lua_Number}),
@Char{%I} (inserts a @Lid{lua_Integer}),
@Char{%p} (inserts a pointer),
@Char{%d} (inserts an @T{int}),
@Char{%c} (inserts an @T{int} as a one-byte character), and
@Char{%U} (inserts an @T{unsigned long} as a @x{UTF-8} byte sequence).

}

@APIEntry{void lua_pushglobaltable (lua_State *L);|
@apii{0,1,-}

Pushes the @x{global environment} onto the stack.

}

@APIEntry{void lua_pushinteger (lua_State *L, lua_Integer n);|
@apii{0,1,-}

Pushes an integer with value @id{n} onto the stack.

}

@APIEntry{void lua_pushlightuserdata (lua_State *L, void *p);|
@apii{0,1,-}

Pushes a light userdata onto the stack.

Userdata represent @N{C values} in Lua.
A @def{light userdata} represents a pointer, a @T{void*}.
It is a value (like a number):
you do not create it, it has no individual metatable,
and it is not collected (as it was never created).
A light userdata is equal to @Q{any}
light userdata with the same @N{C address}.

}

@APIEntry{const char *lua_pushliteral (lua_State *L, const char *s);|
@apii{0,1,m}

This macro is equivalent to @Lid{lua_pushstring},
but should be used only when @id{s} is a literal string.
(Lua may optimize this case.)

}

@APIEntry{const char *lua_pushlstring (lua_State *L, const char *s, size_t len);|
@apii{0,1,m}

Pushes the string pointed to by @id{s} with size @id{len}
onto the stack.
Lua will make or reuse an internal copy of the given string,
so the memory at @id{s} can be freed or reused immediately after
the function returns.
The string can contain any binary data,
including @x{embedded zeros}.

Returns a pointer to the internal copy of the string @see{constchar}.

}

@APIEntry{void lua_pushnil (lua_State *L);|
@apii{0,1,-}

Pushes a nil value onto the stack.

}

@APIEntry{void lua_pushnumber (lua_State *L, lua_Number n);|
@apii{0,1,-}

Pushes a float with value @id{n} onto the stack.

}

@APIEntry{const char *lua_pushstring (lua_State *L, const char *s);|
@apii{0,1,m}

Pushes the zero-terminated string pointed to by @id{s}
onto the stack.
Lua will make or reuse an internal copy of the given string,
so the memory at @id{s} can be freed or reused immediately after
the function returns.

Returns a pointer to the internal copy of the string @see{constchar}.

If @id{s} is @id{NULL}, pushes @nil and returns @id{NULL}.

}

@APIEntry{int lua_pushthread (lua_State *L);|
@apii{0,1,-}

Pushes the thread represented by @id{L} onto the stack.
Returns 1 if this thread is the @x{main thread} of its state.

}

@APIEntry{void lua_pushvalue (lua_State *L, int index);|
@apii{0,1,-}

Pushes a copy of the element at the given index
onto the stack.

}

@APIEntry{
const char *lua_pushvfstring (lua_State *L,
                              const char *fmt,
                              va_list argp);|
@apii{0,1,-}

Equivalent to @Lid{lua_pushfstring},
except that it receives a @id{va_list}
instead of a variable number of arguments,
and it does not raise errors.
Instead, in case of errors it pushes the error message
and returns @id{NULL}.

}

@APIEntry{int lua_rawequal (lua_State *L, int index1, int index2);|
@apii{0,0,-}

Returns 1 if the two values in indices @id{index1} and
@id{index2} are primitively equal
(that is, equal without calling the @idx{__eq} metamethod).
Otherwise @N{returns 0}.
Also @N{returns 0} if any of the indices are not valid.

}

@APIEntry{int lua_rawget (lua_State *L, int index);|
@apii{1,1,-}

Similar to @Lid{lua_gettable}, but does a raw access
(i.e., without metamethods).
The value at @id{index} must be a table.

}

@APIEntry{int lua_rawgeti (lua_State *L, int index, lua_Integer n);|
@apii{0,1,-}

Pushes onto the stack the value @T{t[n]},
where @id{t} is the table at the given index.
The access is raw,
that is, it does not use the @idx{__index} metavalue.

Returns the type of the pushed value.

}

@APIEntry{int lua_rawgetp (lua_State *L, int index, const void *p);|
@apii{0,1,-}

Pushes onto the stack the value @T{t[k]},
where @id{t} is the table at the given index and
@id{k} is the pointer @id{p} represented as a light userdata.
The access is raw;
that is, it does not use the @idx{__index} metavalue.

Returns the type of the pushed value.

}

@APIEntry{lua_Unsigned lua_rawlen (lua_State *L, int index);|
@apii{0,0,-}

Returns the raw @Q{length} of the value at the given index:
for strings, this is the string length;
for tables, this is the result of the length operator (@Char{#})
with no metamethods;
for userdata, this is the size of the block of memory allocated
for the userdata.
For other values, this call @N{returns 0}.

}

@APIEntry{void lua_rawset (lua_State *L, int index);|
@apii{2,0,m}

Similar to @Lid{lua_settable}, but does a raw assignment
(i.e., without metamethods).
The value at @id{index} must be a table.

}

@APIEntry{void lua_rawseti (lua_State *L, int index, lua_Integer i);|
@apii{1,0,m}

Does the equivalent of @T{t[i] = v},
where @id{t} is the table at the given index
and @id{v} is the value on the top of the stack.

This function pops the value from the stack.
The assignment is raw,
that is, it does not use the @idx{__newindex} metavalue.

}

@APIEntry{void lua_rawsetp (lua_State *L, int index, const void *p);|
@apii{1,0,m}

Does the equivalent of @T{t[p] = v},
where @id{t} is the table at the given index,
@id{p} is encoded as a light userdata,
and @id{v} is the value on the top of the stack.

This function pops the value from the stack.
The assignment is raw,
that is, it does not use the @idx{__newindex} metavalue.

}

@APIEntry{
typedef const char * (*lua_Reader) (lua_State *L,
                                    void *data,
                                    size_t *size);|

The reader function used by @Lid{lua_load}.
Every time @Lid{lua_load} needs another piece of the chunk,
it calls the reader,
passing along its @id{data} parameter.
The reader must return a pointer to a block of memory
with a new piece of the chunk
and set @id{size} to the block size.
The block must exist until the reader function is called again.
To signal the end of the chunk,
the reader must return @id{NULL} or set @id{size} to zero.
The reader function may return pieces of any size greater than zero.

}

@APIEntry{void lua_register (lua_State *L, const char *name, lua_CFunction f);|
@apii{0,0,e}

Sets the @N{C function} @id{f} as the new value of global @id{name}.
It is defined as a macro:
@verbatim{
#define lua_register(L,n,f) \
       (lua_pushcfunction(L, f), lua_setglobal(L, n))
}

}

@APIEntry{void lua_remove (lua_State *L, int index);|
@apii{1,0,-}

Removes the element at the given valid index,
shifting down the elements above this index to fill the gap.
This function cannot be called with a pseudo-index,
because a pseudo-index is not an actual stack position.

}

@APIEntry{void lua_replace (lua_State *L, int index);|
@apii{1,0,-}

Moves the top element into the given valid index
without shifting any element
(therefore replacing the value at that given index),
and then pops the top element.

}


@APIEntry{int lua_resume (lua_State *L, lua_State *from, int nargs,
                          int *nresults);|
@apii{?,?,-}

Starts and resumes a coroutine in the given thread @id{L}.

To start a coroutine,
you push the main function plus any arguments
onto the empty stack of the thread.
then you call @Lid{lua_resume},
with @id{nargs} being the number of arguments.
The function returns when the coroutine suspends,
finishes its execution, or raises an unprotected error.
When it returns without errors,
@id{*nresults} is updated and
the top of the stack contains
the @id{*nresults} values passed to @Lid{lua_yield}
or returned by the body function.
@Lid{lua_resume} returns
@Lid{LUA_YIELD} if the coroutine yields,
@Lid{LUA_OK} if the coroutine finishes its execution
without errors,
or an error code in case of errors @see{statuscodes}.
In case of errors,
the error object is pushed on the top of the stack.
(In that case, @id{nresults} is not updated,
as its value would have to be 1 for the sole error object.)

To resume a suspended coroutine,
you remove the @id{*nresults} yielded values from its stack,
push the values to be passed as results from @id{yield},
and then call @Lid{lua_resume}.

The parameter @id{from} represents the coroutine that is resuming @id{L}.
If there is no such coroutine,
this parameter can be @id{NULL}.

}

@APIEntry{void lua_rotate (lua_State *L, int idx, int n);|
@apii{0,0,-}

Rotates the stack elements between the valid index @id{idx}
and the top of the stack.
The elements are rotated @id{n} positions in the direction of the top,
for a positive @id{n},
or @T{-n} positions in the direction of the bottom,
for a negative @id{n}.
The absolute value of @id{n} must not be greater than the size
of the slice being rotated.
This function cannot be called with a pseudo-index,
because a pseudo-index is not an actual stack position.

}

@APIEntry{void lua_setallocf (lua_State *L, lua_Alloc f, void *ud);|
@apii{0,0,-}

Changes the @x{allocator function} of a given state to @id{f}
with user data @id{ud}.