View Source beam (zigler v0.13.3)
This struct contains adapters designed to facilitate interfacing the BEAM's c-style helpers for NIFs with a more idiomatic Zig-style of programming, for example, the use of slices instead of null-terminated arrays as strings.
This struct derives from priv/beam/beam.zig
, and is provided to the
project as a package. You may import it into any project zig code
using the following code:
const beam = @import("beam")
If there's something in the BEAM nif API you need which is not provided,
you can also import erl_nif
package which provides direct access to the
equivalent calls from erl_nif.h
This can be done with the following code:
const e = @import("erl_nif");
Summary
Functions (Exceptions)
The equivalent of Kernel.raise/1
from elixir.
The equivalent of error
in erlang.
Raises with a special error datatype that contains an term-encoded stacktrace
datastructure. See also make_stacktrace
Functions (Concurrency)
periodic check-in function for long-running nifs.
Functions (Context)
convenience function to get the context's execution environment
Functions (Env)
Synonym for e.enif_alloc_env
Performs e.enif_clear_env
on the existing env in the context. This function is generally called by beam.send
,
but is provided if you need to manually trigger environment clearing. The function is
also passed a tuple which is a list of terms that should be persisted into the new
environment.
copies a term from one env to to another
Synonym for e.enif_free_env
creates an new independent
environment and attaches it to the existing
context. You may specify the allocator
option to set the default
allocator for the context. Defaults to beam.allocator
.
Functions (Term Management)
converts a e.ErlNifBinary
to []const u8
.
converts a []u8
to a term/0
. The binary must be encoded using erlang term format.
generic cleanup function that can be used to cleanup values that have
been created by get
.
compares two terms, following Elixir's compare interface.
converts BEAM term
dynamic types into static zig types
converts static zig types into BEAM term
dynamic types
returns the empty list term []
.
shortcut for make(.@"error", .{})
shortcut for make(env, .{.@"error", value}, options)
turns a []const u8
into a the corresponding atom/0
term.
performs a list cons operation for head
and tail
variables
causes the VM to generate a new reference term
equivalent to Kernel.make_ref/0
converts a zig std.builtin.StackTrace
into a special term
that is designed to be translated and concatenated onto a BEAM
stacktrace.
marks a e.ErlNifBinary
as qualified to be garbage
collected.
This is a thin wrapper over e.enif_release_binary
.
returns a pid
value that represents the current or
parent process.
sends data
(as a term) to a target process' mailbox.
converts a term/0
to a e.ErlNifBinary
using erlang term format serialization.
Types
a datastructure containing information that is local to the current call.
managed entrypoints into any given nif (any nif that isn't raw) will set
have access to the threadlocal context/0
variable.
a tag identifying the type of context in which the nif is running
A zig enum equivalent to e.ErlNifTermType
identical to ?*e.ErlNifEnv
identical to e.ErlNifEvent
. This is an event datatype that the BEAM
documentation does not describe.
boilerplate functions for nif module initialization. Contains
identical to e.ErlNifMonitor
. This is a monitor datatype that the BEAM
documentation does not describe.
identical to ?*e.ErlNifPid
identical to ?*e.ErlNifPort
wrapped term.
identical to e.ErlNifTid
. This is a thread id datatype that the BEAM.
documentation does not describe.
Constants
wraps e.enif_alloc
and e.enif_free
into the zig standard library allocator interface.
implements std.mem.Allocator
using the std.mem.GeneralPurposeAllocator
factory, backed by beam.wide_alignment_allocator
.
provides a BEAM allocator that can perform allocations with greater alignment than the machine word.
Variables
threadlocal variable that stores the execution context for the nif. Execution of any managed (non-raw) nif is guaranteed to have access to this variable, and the value of this variable will not be changed across any reentries into the function (for example, for yielding nifs)
Functions (Exceptions)
raise_elixir_exception(module: []u8, data: anytype, opts: anytype) term
View Source@spec raise_elixir_exception([]const u8, anytype, anytype) :: term
The equivalent of Kernel.raise/1
from elixir.
module
should be the name of the module that represents the exceptiondata
should be a struct (possibly anonymous) that represents the Elixir exception payload.
Example
~Z"""
pub fn raise_example() beam.term {
return beam.raise_elixir_exception(
"CompileError",
.{.file = "foo", .line = 42, .description = "something went wrong"},
.{});
}
"""
test "raise_elixir_execption" do
assert_raise CompileError, "foo:42: something went wrong", fn ->
raise_example()
end
end
Exception structs are not type checked
Unlike code in elixir, the validity of the exception struct is not checked when using this function.
@spec raise_exception(anytype, anytype) :: term
The equivalent of error
in erlang.
special-cased exception terms
N.B. certain erlang terms are caught and handled as special cases in elixir, see the following example:
Example
~Z"""
pub fn raise_exception_example() beam.term {
return beam.raise_exception(.badarg, .{});
}
"""
test "raise_exception" do
assert_raise ArgumentError, fn ->
raise_exception_example()
end
end
raise_with_error_return(err: anytype, maybe_return_trace: std.builtin.StackTrace, opts: anytype) term
View Source@spec raise_with_error_return(anytype, ?*std.builtin.StackTrace, anytype) :: term
Raises with a special error datatype that contains an term-encoded stacktrace
datastructure. See also make_stacktrace
This datastructure is designed to be concatenated onto the existing stacktrace. In order to concatenate this stacktrace onto your BEAM exception, the function that wraps the nif must be able to catch the error and append the zig error return trace to the existing stacktrace.
Functions (Concurrency)
@spec yield() :: !void
periodic check-in function for long-running nifs.
always returns error.processterminated
if the calling
process has been terminated.
Note that the environment of the function may be invalid
so if you must send information back to the BEAM on a
process termination event, you should have a beam environment
allocated to communicate with the BEAM using send
For yielding nifs (not implemented yet):
- relinquishes control to the BEAM scheduler.
- If the thread's parent process has been killed, returns
error.processterminated
- when control is returned from the scheduler, resumes with no error.
- creates an async suspend point.
Example
~Z"""
pub fn yielding_example(pid: beam.pid) !void {
// for this to work, we must have an independent
// environment, as the process environment will have been
// destroyed when the beam.yield error return occurs.
const env = beam.alloc_env();
defer beam.free_env(env);
while (true) {
std.time.sleep(100000);
beam.yield() catch {
try beam.send(pid, .died, .{.env = env});
return;
};
}
}
"""
test "yield" do
test_pid = self()
spawn_pid = spawn(fn -> yielding_example(test_pid) end)
Process.sleep(100)
Process.exit(spawn_pid, :kill)
assert_receive :died, 1000
end
Functions (Context)
@spec get_env() :: env
convenience function to get the context's execution environment
Functions (Env)
@spec alloc_env() :: env
Synonym for e.enif_alloc_env
generally, zigler takes care of environments for you, but there may be cases where you have to manually allocate an environment. These are process independent environments which manage their own heaps.
use your own process independent environment, most beam
functions can take a
.env
option which lets you specify the environment; these functions generally
default to the context's environment (beam.context.env
).
pair with free_env
all calls to
alloc_env/0
should be balanced with a call tofree_env/1
.
~Z"""
pub fn alloc_env_example(pid: beam.pid) !void {
const env = beam.alloc_env();
defer beam.free_env(env);
try beam.send(pid, .{.ok, 47}, .{.env = env});
}
"""
test "alloc_env_example" do
alloc_env_example(self())
assert_receive {:ok, 47}
end
clear_env(env_: env, persist: anytype) ClearEnvReturn(@TypeOf(persist))
View Source@spec clear_env(env, anytype) :: anytype
Performs e.enif_clear_env
on the existing env in the context. This function is generally called by beam.send
,
but is provided if you need to manually trigger environment clearing. The function is
also passed a tuple which is a list of terms that should be persisted into the new
environment.
This function will panic if the context's environment is not one known to be created
by alloc_env
.
returns void
if the persist list is empty; otherwise returns a tuple of the same size
as the persist list, contaning the terms that are persisted, in the same order.
Example
~Z"""
pub fn clear_env_example(pid: beam.pid, term: beam.term) !void {
const env = beam.alloc_env();
defer beam.free_env(env);
const inner_term = beam.copy(env, term);
try beam.send(pid, .{.first, term}, .{.env = env, .clear = false});
// this function call is required since we did not clear our env
// after sending the term in the previous call.
const new_term = beam.clear_env(env, .{inner_term});
try beam.send(pid, .{.second, new_term}, .{.env = env, .clear = false});
return;
}
"""
test "clear_env" do
clear_env_example(self(), 47)
assert_receive {:first, 47}
assert_receive {:second, 47}
end
@spec copy(env, term) :: term
copies a term from one env to to another
@spec free_env(env) :: void
Synonym for e.enif_free_env
See alloc_env/0
for more information.
context env warning
unless you have created an independent context using
independent_context/1
, do not use this function to free thebeam.context.env
, as unexpected results may occur.
@spec independent_context(beam.InitEnvOpts) :: void
creates an new independent
environment and attaches it to the existing
context. You may specify the allocator
option to set the default
allocator for the context. Defaults to beam.allocator
.
~Z"""
pub fn independent_context_example(pid: beam.pid) !void {
beam.independent_context(.{});
try beam.send(pid, beam.context.mode, .{});
beam.free_env(beam.context.env);
}
"""
test "independent contexts" do
independent_context_example(self())
assert_receive :independent
end
free independent context envs
you must ALWAYS free
beam.context.env
if you have created an independent context.
Functions (Term Management)
@spec binary_to_slice(e.ErlNifBinary) :: []u8
converts a e.ErlNifBinary
to []const u8
.
Does not perform allocations or copies
Example
~Z"""
pub fn binary_to_slice_example(term: beam.term) u8 {
const e = @import("erl_nif");
var binary: e.ErlNifBinary = undefined;
_ = e.enif_inspect_binary(beam.context.env, term.v, &binary);
const slice = beam.binary_to_slice(binary);
return slice[2];
}
"""
test "binary_to_slice" do
assert ?o = binary_to_slice_example("foo")
end
binary data
This points to the data location of the binary, which might either be in the shared binary heap, or it might be in the process heap (for small binaries). This should be considered read-only, attempts to sneakily modify these data will have undefined effects, possibly including broken comparison operations.
@spec binary_to_term([]const u8, anytype) :: !term
converts a []u8
to a term/0
. The binary must be encoded using erlang term format.
This is a thin wrapper over e.enif_binary_to_term
.
Example
~Z"""
pub fn binary_to_term_example(encoded_number: []const u8) !u32 {
const term = try beam.binary_to_term(encoded_number, .{});
const number = try beam.get(u32, term, .{});
return number + 1;
}
"""
test "binary_to_term" do
assert 48 = binary_to_term_example(:erlang.term_to_binary(47))
end
@spec cleanup(anytype, anytype) :: void
generic cleanup function that can be used to cleanup values that have
been created by get
.
The second parameter is an options
parameters, which should be passed a
struct (possibly anonymous) with the following fields:
allocator
: which allocator should be used to clean up allocations. optional, defaults to the threadlocalbeam.context.allocator
valuesize
: if the value does not have a defined size, for example aMany
pointer, you must provide this value.
noops on terms that do not require cleanup.
Example
~Z"""
const CleanupErrors = error{leaked};
pub fn cleanup_example(term: beam.term) !i32 {
var gpa = beam.make_general_purpose_allocator_instance();
const allocator = gpa.allocator();
const slice = try beam.get([]i32, term, .{.allocator = allocator});
const number = slice[0];
beam.cleanup(slice, .{.allocator = allocator});
if (gpa.detectLeaks()) return error.leaked;
return number + 1;
}
"""
test "cleanup" do
assert 48 = cleanup_example([47])
end
@spec compare(term, term) :: beam.Compared
compares two terms, following Elixir's compare interface.
Example
~Z"""
pub fn compare_example(lhs: beam.term, rhs: beam.term) beam.Compared {
return beam.compare(lhs, rhs);
}
"""
test "beam.compare" do
assert :gt == compare_example(:infinity, 47)
assert :eq == compare_example(self(), self())
assert :lt == compare_example(:atom, "string")
end
@spec get(T, beam.term, anytype) :: T
converts BEAM term
dynamic types into static zig types
The arguments are as follows:
- destination type
- term to convert
- struct (usually passed as anonymous) of keyword options for additional features. See supported options
See also make
for the reverse operation.
The following type classes (as passed as 1st argument) are supported by get
:
integer
- unsigned and signed integers are supported
- all integer sizes from 0..64 bits are supported (including non-power-of-2 sizes)
- for sizes bigger than 64, are supported, but the passed term must be a native-endian binary with size n*64 bits
Example
~Z"""
pub fn get_integer_example(term: beam.term) !i32 {
const x: i32 = try beam.get(i32, term, .{});
return x + 1;
}
// NOTE: the size of the integer in the following function
pub fn get_big_integer_example(term: beam.term) !u65 {
const x: u65 = try beam.get(u65, term, .{});
return x + 1;
}
"""
test "get small integer" do
assert 48 = get_integer_example(47)
end
test "get big integer" do
# note we must pass as a binary, with width 128 bits.
assert 0x1_0000_0000_0000_0000 =
get_big_integer_example(
<<0xffff_ffff_ffff_ffff :: native-unsigned-64, 0::64>>
)
end
enum
- may be passed the integer value of the enum.
- may be passed the atom representation of the enum.
- zero- and one- item enum types are not currently supported
Example
~Z"""
const EnumType = enum {foo, bar};
pub fn get_enum_example(term: beam.term) !EnumType {
return try beam.get(EnumType, term, .{});
}
"""
test "get integer enum" do
assert :foo = get_enum_example(0)
end
test "get atom enum" do
assert :foo = get_enum_example(:foo)
end
float
- supports
f16
,f32
, andf64
types. - may be passed a BEAM
float/0
term - atoms
:infinity
,:neg_infinity
,:NaN
are also supported
Example
~Z"""
const NanError = error {nanerror};
pub fn get_float_example(term: beam.term) !f32 {
const x: f32 = try beam.get(f32, term, .{});
if (!std.math.isNan(x)) {
return x + 1.0;
} else {
return error.nanerror;
}
}
"""
test "get real float" do
assert 48.0 = get_float_example(47.0)
end
test "get infinite float" do
assert :infinity = get_float_example(:infinity)
assert :neg_infinity = get_float_example(:neg_infinity)
end
test "get nan float" do
assert_raise ErlangError, fn -> get_float_example(:NaN) end
end
struct
- may be passed
map/0
withatom/0
keys and values of the appropriate type - may be passed a
keyword/0
list withatom/0
keys and values of the appropriate type. - inner values are recursively marshalled to to the appropriate type.
- if the struct is
packed
orextern
, supports binary data.
Example
~Z"""
const MathType = packed struct {
left: packed struct{ value: u32 },
right: packed struct{ value: u32 },
op: enum(u8) {add, sub}
};
pub fn get_struct_example(term: beam.term) !u32 {
const operation = try beam.get(MathType, term, .{});
switch (operation.op) {
.add => return operation.left.value + operation.right.value,
.sub => return operation.left.value - operation.right.value,
}
}
"""
test "get struct" do
assert 49 = get_struct_example(%{
op: :add,
left: %{value: 47},
right: %{value: 2}})
assert 42 = get_struct_example(%{
op: :sub,
left: %{value: 47},
right: %{value: 5}})
end
test "get struct as packed binary" do
assert 49 = get_struct_example(<<47::native-32, 2::native-32, 0::8, 0::56>>)
end
packed and extern structs
if you are doing binary encoding of these values, be mindful of native-endianness of the results and the need for padding to conform to alignment requirements.
note that the C ABI specification may reorder your values, whereas a packed struct may have wider than expected alignment.
bool
- supports
true
andfalse
boolean/0
terms only.
Example
~Z"""
pub fn get_bool_example(term: beam.term) !bool {
return try beam.get(bool, term, .{});
}
"""
test "get bool" do
assert get_bool_example(true)
refute get_bool_example(false)
end
array
- supports lists of terms that can be converted to the array's element type.
- note that arrays have compile-time known length.
- if the array's element is integers, floats, packed or extern structs, or arrays that support binaries, then the array can be passed binary data.
- does not perform allocation
Allocation warning
as allocation is not performed, getting could be a very expensive operation. these values will be on the stack.
Example
~Z"""
const PackedValue = packed struct { v: u32 };
pub fn get_array_example(term: beam.term) !u32 {
const array = try beam.get([3]PackedValue, term, .{});
var sum: u32 = 0;
for (array) |item| { sum += item.v; }
return sum;
}
"""
test "get array" do
assert 144 = get_array_example([%{v: 47}, %{v: 48}, %{v: 49}])
end
test "get array as binary" do
assert 144 = get_array_example(<<47::native-32, 48::native-32, 49::native-32>>)
end
single-item pointer
- used if an struct or array must be allocated on the heap.
- allocates memory based on allocator provided in the options, otherwise
defaults to
beam.allocator
- data may be passed as if they were not a pointer.
- returns an error if the allocation fails.
Allocation warning
it is the caller's responsibility to free the memory allocated by this function, via the
allocator.destroy
function, notallocator.free
Example
# see previous example cell for type definitions
~Z"""
pub fn get_struct_pointer_example(term: beam.term) !u32 {
const structptr = try beam.get(*PackedValue, term, .{});
defer beam.allocator.destroy(structptr);
return structptr.v;
}
pub fn get_array_pointer_example(term: beam.term) !u32 {
const arrayptr = try beam.get(*[3]PackedValue, term, .{});
defer beam.allocator.destroy(arrayptr);
var sum: u32 = 0;
for (arrayptr.*) |item| { sum += item.v; }
return sum;
}
"""
test "get struct pointer" do
assert 47 = get_struct_pointer_example(%{v: 47})
end
test "get array pointer" do
assert 144 = get_array_pointer_example(<<47::native-32, 48::native-32, 49::native-32>>)
end
slice
- allocates memory based on allocator provided in the options, otherwise
defaults to
beam.allocator
- note that slice carries a runtime length
- supports list of any type
- supports binary of any type that can be represented as a fixed size binary.
Allocation warning
it is the caller's responsibility to free the memory allocated by this function, via the
allocator.free
function, notallocator.destroy
Example
~Z"""
pub fn get_slice_example(term: beam.term) !u32 {
const slice = try beam.get([]packed struct{ v: u32 }, term, .{});
defer beam.allocator.free(slice);
var sum: u32 = 0;
for (slice) |item| { sum += item.v; }
return sum;
}
"""
test "get slice" do
assert 144 = get_slice_example([%{v: 47}, %{v: 48}, %{v: 49}])
end
test "get slice as binary" do
assert 144 = get_slice_example(<<47::native-32, 48::native-32, 49::native-32>>)
end
many-item-pointer
- allocates memory based on allocator provided in the options, otherwise
defaults to
beam.allocator
- supports list of any type
- supports binary of any type that can be represented as a fixed size binary.
- the runtime length is not a part of this datastructure, you are
expected to keep track of it using some other mechanism. Caller is responsible
for tracking the length.
Length warning
due to the fact that this datatype drops its length information, this datatype should be handled with extreme care.
- the allocated runtime length can be retrieved by passing a
*usize
parameter as thesize
parameter to the options.
Example
~Z"""
pub fn get_many_item_pointer_example(term: beam.term) !u32 {
var size: usize = undefined;
const pointer = try beam.get([*]u32, term, .{.size = &size});
defer beam.allocator.free(pointer[0..size]);
var sum: u32 = 0;
for (0..size) |index| { sum += pointer[index]; }
return sum;
}
"""
test "get many item pointer" do
assert 144 = get_many_item_pointer_example([47, 48, 49])
end
test "get many item pointer as binary" do
assert 144 = get_many_item_pointer_example(<<47::native-32, 48::native-32, 49::native-32>>)
end
cpointer
allocates memory based on allocator provided in the options, otherwise defaults to
beam.allocator
supports list of any type
supports binary of any type that can be represented as a fixed size binary.
the runtime length is not a part of this datastructure, you are expected to keep track of it using some other mechanism
Length warning
due to the fact that this datatype drops its length information, this datatype should only be used where c interop is needed
the allocated runtime length can be retrieved by passing a
*usize
parameter as thesize
parameter to the options.
Example
~Z"""
pub fn get_c_pointer_example(term: beam.term) !u32 {
var size: usize = undefined;
const pointer = try beam.get([*c]u32, term, .{.size = &size});
defer beam.allocator.free(pointer[0..size]);
var sum: u32 = 0;
for (0..size) |index| { sum += pointer[index]; }
return sum;
}
"""
test "get c pointer" do
assert 144 = get_c_pointer_example([47, 48, 49])
end
test "get c pointer as binary" do
assert 144 = get_c_pointer_example(<<47::native-32, 48::native-32, 49::native-32>>)
end
optional
- accepts
atom/0
nil
as well as whatever the child type is.
nil vs null
note that zig uses
null
as its nil value, but zigler only accepts atomnil
as its null value.
Example
~Z"""
pub fn get_optional_example(term: beam.term) !u32 {
const x = try beam.get(?u32, term, .{});
if (x) |value| { return value + 1; } else { return 47; }
}
"""
test "get optional" do
assert 48 = get_optional_example(47)
assert 47 = get_optional_example(nil)
end
pid
- accepts
pid/0
Example
~Z"""
pub fn get_pid_example(term: beam.term) !void {
const pid = try beam.get(beam.pid, term, .{});
try beam.send(pid, .foo, .{});
}
"""
test "get pid" do
assert :ok = get_pid_example(self())
assert_receive :foo
end
Supported options
env
: required forraw
nifs. If not provided, defaults to the context's environment.allocator
: the allocator to use for allocations. If not provided, defaults to the context's environment.error_info
: pointer to aterm
that can be populated with error information that gets propagated on failure to convert. If omitted, the code to produce these errors will get optimized out.keep
(bool
, only applies to references): iftrue
(default) the refcount will be increased on the term as a result of performing theget
operation.size
(* usize
, only applies to manypointer or[*c]
):
@spec make(anytype, anytype) :: term
converts static zig types into BEAM term
dynamic types
The arguments are as follows:
- environment
value
to convert to termoptions
struct (usually passed as anonymous) of keyword options for additional features. See supported options. Note this struct must be comptime-known.
See also get
for the reverse operation.
The following zig types are supported:
term
- no conversion is performed
- this type is necessary for recursive make operations
Example
~Z"""
pub fn make_term_example(term: beam.term) beam.term {
return beam.make(term, .{});
}
"""
test "make term" do
assert 47 = make_term_example(47)
end
void
- returns atom
:ok
- supporting this type makes metaprogramming easier.
Example
~Z"""
pub fn make_void_example() beam.term {
const v: void = undefined;
return beam.make(v, .{});
}
"""
test "make void" do
assert :ok = make_void_example()
end
pid
~Z"""
pub fn make_pid_example(pid: beam.pid) beam.term {
return beam.make(pid, .{});
}
"""
test "make pid" do
assert self() == make_pid_example(self())
end
std.builtin.StackTrace
- special interface for returning stacktrace info to BEAM.
integers
- unsigned and signed integers supported
- all integer sizes from 0..64 bits supported (including non-power-of-2 sizes)
- returns a BEAM
integer/0
term - for sizes bigger than 64, supported, but the passed term will be converted into a binary term bearing the integer encoded with native endianness.
- comptime integers supported
Example
~Z"""
pub fn make_integer_example(integer: u32) beam.term {
return beam.make(integer + 1, .{});
}
pub fn make_big_integer_example(integer: u65) beam.term {
return beam.make(integer + 1, .{});
}
pub fn make_comptime_integer_example() beam.term {
return beam.make(47, .{});
}
"""
test "make integer" do
assert 48 = make_integer_example(47)
end
test "make big integer" do
assert <<0::64, 1::64-native>> = make_big_integer_example(0xFFFF_FFFF_FFFF_FFFF)
end
test "make comptime integer" do
assert 47 = make_comptime_integer_example()
end
floats
- supports
f16
,f32
, andf64
types. - supports comptime float type.
- returns a BEAM
float/0
term - may also return one of the
atom/0
type:infinity
,:neg_infinity
,:NaN
Example
~Z"""
pub fn make_float_example(float: f32) beam.term {
return beam.make(float + 1.0, .{});
}
pub fn make_comptime_float_example() beam.term {
return beam.make(47.0, .{});
}
"""
test "make float" do
assert 48.0 = make_float_example(47.0)
assert :infinity = make_float_example(:infinity)
assert :neg_infinity = make_float_example(:neg_infinity)
assert :NaN = make_float_example(:NaN)
end
test "make comptime float" do
assert 47.0 = make_comptime_float_example()
end
bool
- supports
bool
types. - returns a BEAM
boolean/0
term
Example
~Z"""
pub fn make_bool_example(value: bool) beam.term {
return beam.make(!value, .{});
}
"""
test "make bool" do
assert make_bool_example(false)
refute make_bool_example(true)
end
enum or error enum
- supports
enum
orerror
types. - doesn't support zero or one-item enums.
- returns a BEAM
atom/0
term - also supports enum literals.
Example
~Z"""
const MakeEnums = enum {foo, bar};
pub fn make_enum_example(value: MakeEnums) beam.term {
return beam.make(value, .{});
}
pub fn make_enum_as_int_example(value: MakeEnums) beam.term {
return beam.make(value, .{.as = .integer});
}
const MakeErrorSet = error { MakeEnumError };
pub fn make_error_example() beam.term {
return beam.make(error.MakeEnumError, .{});
}
pub fn make_enum_literal_example() beam.term {
return beam.make(.foobarbaz, .{});
}
"""
test "make enum" do
assert :foo = make_enum_example(:foo)
assert 0 = make_enum_as_int_example(:foo)
end
test "make error" do
assert :MakeEnumError = make_error_example()
end
test "make enum literal" do
assert :foobarbaz = make_enum_literal_example()
end
Enum literals
Enum literals are especially useful for returning atoms, such as
:ok
or:error
. Note thaterror
is a reserved word in zig, so you will need to use.@"error"
to generate the corresponding atom. See alsomake_error_atom
optionals or null
Example
~Z"""
pub fn make_null_example() beam.term {
return beam.make(null, .{});
}
pub fn make_optional_example(value: ?u32) beam.term {
return beam.make(value, .{});
}
"""
test "make null" do
assert is_nil(make_null_example())
end
test "make optional" do
assert is_nil(make_optional_example(nil))
assert 47 = make_optional_example(47)
end
arrays
- supports arrays of any term that can be encoded using
make
- note that arrays have compile-time known length.
- outputs as a list of the encoded terms
- arrays of
u8
default to outputting binary, this is the only exception to the above rule. - if the array's element is integers, floats, packed or extern structs,
or arrays that support binaries, then the array can be output as binary
data, by setting
as
option to.binary
- if the array's element is u8 and you would prefer outputting as a list,
setting
as
option to.list
will do this. - to specify the internal encoding of the array, pass to
as
a struct with the fieldlist
; the value of the list field should be the encoding of the internal terms.
Examples
~Z"""
pub fn make_array_example() beam.term {
const array = [_]u32{47, 48, 49};
return beam.make(array, .{});
}
pub fn make_array_u8_example() beam.term {
const array = "foo";
return beam.make(array, .{});
}
pub fn make_array_binary_example() beam.term {
const array = [_]u32{47, 48, 49};
return beam.make(array, .{.as = .binary});
}
pub fn make_array_u8_list_example() beam.term {
const array = "foo";
return beam.make(array, .{.as = .list});
}
pub fn make_array_internal_encoding_example() beam.term {
const aoa = [_][2]u32{[2]u32{47, 48}, [2]u32{49, 50}};
return beam.make(aoa, .{.as = .{.list = .binary}});
}
"""
test "make u8 array" do
assert "foo" = make_array_u8_example()
assert [102, 111, 111] = make_array_u8_list_example()
end
test "make u32 array" do
assert [47, 48, 49] = make_array_example()
assert <<47::native-32, 48::native-32, 49::native-32>> = make_array_binary_example()
end
test "make array with internal encoding" do
assert [
<<47::native-32, 48::native-32>>,
<<49::native-32, 50::native-32>>] = make_array_internal_encoding_example()
end
structs
- supports structs with fields of any term that can be encoded using
make
- outputs as a
map/0
with atom keys and the encoded terms as values - for
packed
orextern
structs, supports binary data by settingas
option to.binary
.extern
structs default to map encoding,packed
structs default to binary encoding. All structs can be forced tomap
encoding by passing setting.as = .map
. - encoding options can be specified by assigning
as
a struct with the fieldmap
; the value of the map field should be a struct with keys matching the struct fields and values being the encoding options for those fields. It's legal to omit a field from this specification, in which case the encoding will be.default
. - supports anonymous structs
- supports direct output as elixir structs using the
as
optionElixir structs
- non-elixir alias-style structs are not currently supported, but will be in a future release
- nested output formatting is not currently supported, but will be in a future release.
- if you need to output a struct that is a multilevel alias, use
@
syntax e.g..{.as = .@"Foo.Bar.Baz"}
the fields and types on the fields are not checked at compile time or runtime, so use with caution.
Examples
~Z"""
pub fn make_struct_example() beam.term {
return beam.make(.{.foo = 123, .bar = "bar", .baz = .baz}, .{});
}
pub fn make_elixir_struct_example() beam.term {
return beam.make(.{.first = 1, .last = 10, .step = 1}, .{.as = .Range});
}
const MakePacked = packed struct { value: u32 };
pub fn make_packed_struct() beam.term {
return beam.make(MakePacked{.value = 47}, .{});
}
pub fn make_struct_nested() beam.term {
const result = .{
.list = [2]u32{47, 48},
.binary = [2]u32{47, 48}
};
return beam.make(result, .{.as = .{.map = .{.binary = .binary}}});
}
"""
test "make general struct" do
assert %{foo: 123, bar: "bar", baz: :baz} = make_struct_example()
end
test "make elixir struct" do
assert 1..10 = make_elixir_struct_example()
end
test "make packed struct" do
assert <<47::native-32>> = make_packed_struct()
end
test "make struct with nested format info" do
assert %{
binary: <<47::native-32, 48::native-32>>,
list: [47, 48]
} = make_struct_nested()
end
tuples
- supports zig tuples with any term that can be encoded using
make
- outputs as a
tuple/0
. - encoding will always proceed using
.default
encoding. A scheme to specify encoding options is planned in the future. - note that the error atom should be encoded as
.@"error"
; you may also usebeam.make_error_atom(...)
Examples
~Z"""
pub fn make_tuple_example() beam.term {
return beam.make(.{.ok, "foo", 47}, .{});
}
"""
test "make tuple" do
assert {:ok, "foo", 47} = make_tuple_example()
end
single-item-pointer
- these pointers are only supported for arrays and structs
- these are only supported because they are assumed to be pointers to mutable data
- content will be dereferenced and encoded as if it were the child type
as
rules apply (see arrays and structs).
Examples
~Z"""
pub fn make_pointer_example() beam.term {
const array = [_]u32{47, 48, 49};
return beam.make(&array, .{});
}
"""
test "make pointer" do
assert [47, 48, 49] = make_pointer_example()
end
slice
- supports arrays of any term that can be encoded using
make
- note that arrays have compile-time known length.
- outputs as a list of the encoded terms
- slices of
u8
default to outputting binary, this is the only exception to the above rule. - if the slice's element is integers, floats, packed or extern structs,
or arrays that support binaries, then the slice can be output as binary
data, by setting
as
option to.binary
as
rules (see arrays) apply.
Examples
~Z"""
pub fn make_slice_example(slice: []u32) beam.term {
for (slice) |*item| { item.* += 1; }
return beam.make(slice, .{});
}
pub fn make_slice_binary_example(slice: []u32) beam.term {
return beam.make(slice, .{.as = .binary});
}
"""
test "make slice" do
assert [48, 49, 50] = make_slice_example([47, 48, 49])
end
test "make slice as binary" do
assert <<47::native-32, 48::native-32, 49::native-32>> = make_slice_binary_example([47, 48, 49])
end
many-item-pointer
- only supports [:0]u8 and [:null]?Pointer.
as
rules (see arrays) apply.
Examples
~Z"""
pub fn make_many_item_pointer_example(pointer: []u8) beam.term {
pointer[5] = 0;
const truncated: [*:0]u8 = @ptrCast(pointer.ptr);
return beam.make(truncated, .{});
}
"""
test "make many item pointer" do
assert "hello" = make_many_item_pointer_example("hello world")
end
cpointer
- only supported if the pointer has child type
u8
or pointer. - in the case of
u8
interprets it as[*:0]u8
. - in the case of
Pointer
interprets it as[*:null]?Pointer
. - no other types are supported.
- note that the content will be interpreted as the pointer type, so rules on pointers (see single-item-pointers))
as
rules (see arrays) apply.
Examples
~Z"""
pub fn make_c_pointer_example(pointer: []u8) beam.term {
pointer[5] = 0;
const truncated: [*c]u8 = @ptrCast(pointer.ptr);
return beam.make(truncated, .{});
}
"""
test "make c pointer" do
assert "hello" = make_c_pointer_example("hello world")
end
@spec make_empty_list(anytype) :: term
returns the empty list term []
.
This is a thin wrapper over e.enif_make_empty_list
.
Example
~Z"""
pub fn make_empty_list_example() beam.term {
return beam.make_empty_list(.{});
}
"""
test "make empty list" do
assert [] = make_empty_list_example()
end
@spec make_error_atom(anytype) :: term
shortcut for make(.@"error", .{})
Example
~Z"""
pub fn make_error_atom_example() beam.term {
return beam.make_error_atom(.{});
}
"""
test "make error atom" do
assert :error = make_error_atom_example()
end
@spec make_error_pair(anytype, anytype) :: term
shortcut for make(env, .{.@"error", value}, options)
@spec make_into_atom([]const u8, anytype) :: term
turns a []const u8
into a the corresponding atom/0
term.
returns a raised ArgumentError
if the length of the string exceeds the
vm atom size limit (255 bytes)
This is a thin wrapper over e.enif_make_atom_len
.
Atom size limit
The BEAM VM has a limit of 255 bytes for atom size. This function does not protect against this and the API may change to accomodate this in the future.
Atom table exhaustion
The BEAM VM has a limit of 1_048_576 atoms. This function does not provide any protection against atom table exhaustion. A future version will protect against this case.
Example
~Z"""
pub fn make_into_atom_example() beam.term {
return beam.make_into_atom("foo", .{});
}
"""
test "make into atom" do
assert :foo = make_into_atom_example()
end
@spec make_list_cell(anytype, term, anytype) :: term
performs a list cons operation for head
and tail
variables
This is a thin wrapper over e.enif_make_list_cell
.
Example
~Z"""
pub fn make_list_cell_example() beam.term {
return beam.make_list_cell(47, beam.make_empty_list(.{}), .{});
}
"""
test "make list cell" do
assert [47] = make_list_cell_example()
end
@spec make_ref(anytype) :: term
causes the VM to generate a new reference term
equivalent to Kernel.make_ref/0
This is a thin wrapper over e.enif_make_ref
.
Example
~Z"""
pub fn make_ref_example() beam.term {
return beam.make_ref(.{});
}
"""
test "make_ref" do
assert is_reference(make_ref_example())
end
@spec make_stacktrace(*std.builtin.StackTrace, anytype) :: term
converts a zig std.builtin.StackTrace
into a special term
that is designed to be translated and concatenated onto a BEAM
stacktrace.
Example term:
[
%{
line_info: %{file_name: "/path/to/project/lib/my_app/.Elixir.MyApp.MyModule.zig", line: 15},
symbol_name: "my_fun",
compile_unit_name: "Elixir.MyApp.MyModule"
}
...
]
@spec release_binary(*e.ErlNifBinary) :: void
marks a e.ErlNifBinary
as qualified to be garbage
collected.
This is a thin wrapper over e.enif_release_binary
.
@spec self(anytype) :: !pid
returns a pid
value that represents the current or
parent process.
equivalent to Kernel.self/0
scope
This function succeeds in all contexts, except for callback contexts. For threaded processes, it will return the process that spawned the thread, whether or not that process is still alive.
Example:
~Z"""
pub fn self_example() !beam.pid {
return try beam.self(.{});
}
"""
test "self" do
assert self() == self_example()
end
@spec send(pid, anytype, anytype) :: anytype
sends data
(as a term) to a target process' mailbox.
equivalent to Kernel.send/2
This function is a context-aware wrapper over
e.enif_send
.
that also formats the message term using make
Note that send
is designed so that you can switch concurrency modes
without having to change your code.
Options
clear
(boolean): whether the environment should be cleared after sending the message. Defaults totrue
. See information below.persist
(tuple ofbeam.term
). Persists the terms into the new environment (seeclear_env
). It is not an error to passpersist
in a process-bound context, though that will no-op.as
: any formatting options to be passed when making the term to be sent.
Return value
If there are no persisted values, send
returns void
. Otherwise
it returns a tuple that corresponds to the tuple of persisted values,
with new beam.term
values.
send from raw nifs or non-process-bound threads
This function has undefined behaviour when called from
raw
nifs or calls that are not process-bound (e.g. if a new thread is started from a posix call that is not managed by zigler)in the case of raw nifs, use
e.enif_send
directly instead.in the case of non-process-bound threads or raw calls, if you use
independent_context
to initialize the environment, you can usesend
as normal.The safety characteristics of raw and non-process-bound sending may be subject to change in future releases of Zigler.
clearing your environment
normally when calling
e.enif_send
you would need to calle.enif_clear_env
to recycle the environment after sending. You do not need to do this for this function; it gets called automatically.If you are certain that the send operation is the last operation in your function call, you may call the function as
send(data, .{.clear = false})
and the clear_env function will not be called.
Example
~Z"""
pub fn send_example() !void {
const self = try beam.self(.{});
try beam.send(self, .{.foo, "bar", 47}, .{});
}
"""
test "send" do
send_example()
assert_receive {:foo, "bar", 47}
end
@spec term_to_binary(term, anytype) :: !e.ErlNifBinary
converts a term/0
to a e.ErlNifBinary
using erlang term format serialization.
This is a thin wrapper over e.enif_term_to_binary
.
returns error.OutOfMemory
if the allocation fails.
Functions
make_general_purpose_allocator_instance() std.heap.GeneralPurposeAllocator(...)
View Source@spec make_general_purpose_allocator_instance() :: std.heap.GeneralPurposeAllocator(...)
Types
@type Compared :: .lt | .gt | .eq
result type for compare
these atoms are used to conform to Elixir's Compare interface see: https://hexdocs.pm/elixir/1.13/Enum.html#sort/2-sorting-structs
@type Context :: {allocator :: mem.Allocator, env :: env, mode :: beam.ContextMode}
a datastructure containing information that is local to the current call.
managed entrypoints into any given nif (any nif that isn't raw) will set
have access to the threadlocal context/0
variable.
Fields
mode
: the concurrency mode of the nifenv
: the environment of the nifallocator
: the allocator to use for allocations
fields
mode
:ContextMode
env
:env
allocator
:std.mem.Allocator
@type ContextMode ::
.independent | .dirty_yield | .yielding | .threaded | .synchronous | .callback | .dirty
a tag identifying the type of context in which the nif is running
See nif documentation for more detailed information about the concurrency strategies.
.synchronous
: the execution context of a synchronous nif.threaded
: the execution context of a nif that runs in its own os thread.dirty
: the execution context of a nif that runs on a dirty scheduler.yielding
: the execution context of a nif that runs cooperatively with the BEAM scheduler.callback
: the execution context of module setup/teardown callbacks or a resource destruction callback.dirty_yield
: the execution context of a dirty nif whose parent process has been terminated but the nif is still running..independent
: the execution context of functions that are not associated with nifs. seeindependent_context
for how to set this as the context.
See context
for the threadlocal variable that stores this.
raw beam functions
nifs called in
raw
mode are not assigned an execution context.
@type TermType ::
.ref | .bitstring | .tuple | .fun | .map | .float | .port | .pid | .list | .integer | .atom
A zig enum equivalent to e.ErlNifTermType
retrievable from term
using the beam.term_type
method.
@type env :: env
identical to ?*e.ErlNifEnv
env
env
should be considered an opaque type that can be passed around without inspection of its contents.
@type event :: erl_nif_event
identical to e.ErlNifEvent
. This is an event datatype that the BEAM
documentation does not describe.
@type loader :: {}
boilerplate functions for nif module initialization. Contains:
blank_load
blank_upgrade
blank_unload
which are no-op versions of these functions.
@type monitor :: e.ErlNifMonitor
identical to e.ErlNifMonitor
. This is a monitor datatype that the BEAM
documentation does not describe.
@type pid :: pid
identical to ?*e.ErlNifPid
@type port :: e.ErlNifPort
identical to ?*e.ErlNifPort
@type term :: term
wrapped term.
e.ErlNifTerm
is, under the hood, an integer type. This is wrapped in a singleton struct
so that that semantic analysis can identify and distinguish between a
'plain' integer and a term.
The Zig compiler will optimize this away, so there is no runtime cost to
passing this around versus e.ErlNifTerm
, and the following conversion
operations are no-ops:
To convert to a raw
e.ErlNifTerm
, access the.v
field.To convert a raw
e.ErlNifTerm
to this term, use an anonymous struct:.{.v = erl_nif_term_value}
term_type
the struct function term_type
returns the TermType
of the
internal term.
const t = beam.term.make(env, 47, .{});
const term_type = t.term_type(env); // -> .integer
@type tid :: e.ErlNifTid
identical to e.ErlNifTid
. This is a thread id datatype that the BEAM.
documentation does not describe.
Constants
@spec allocator :: mem.Allocator
wraps e.enif_alloc
and e.enif_free
into the zig standard library allocator interface.
@spec general_purpose_allocator :: mem.Allocator
implements std.mem.Allocator
using the std.mem.GeneralPurposeAllocator
factory, backed by beam.wide_alignment_allocator
.
@spec wide_alignment_allocator :: mem.Allocator
provides a BEAM allocator that can perform allocations with greater alignment than the machine word.
Memory performance
This comes at the cost of some memory to store metadata
currently does not release memory that is resized. For this behaviour
use beam.general_purpose_allocator
.
not threadsafe. for a threadsafe allocator, use beam.general_purpose_allocator
Variables
@spec context :: beam.Context
threadlocal variable that stores the execution context for the nif. Execution of any managed (non-raw) nif is guaranteed to have access to this variable, and the value of this variable will not be changed across any reentries into the function (for example, for yielding nifs)
If you spawn a thread from your nif, you should copy anything you need out of this variable.
See Context
for the list of fields.
context starts undefined
This threadlocal is set to
undefined
because of architectural differences: we cannot trust loaded dynamic libraries to properly set this on thread creation.
raw
function calls do not setcontext