6 Binary Instruction Codes

The following section contains a list of all DXB binary instruction codes.

6.1 Remarks

Instruction codes have a size of 1 byte. The codes from 0x60 to 0x9f are reserved for later use.

Number types:

• Uint8: Unsigned Integer, 8 Bits = 1 Byte
• Uint32: Unsigned Integer, 32 Bits = 4 Bytes
• Int8: Signed Integer, 8 Bits = 1 Byte, -2_147_483_648 to 2_147_483_647
• Int16: Signed Integer, 16 Bits = 2 Bytes, -2_147_483_648 to 2_147_483_647
• Int32: Signed Integer, 32 Bits = 4 Bytes, -2_147_483_648 to 2_147_483_647
• Int64: Signed Integer, 64 Bits = 8 Bytes, -9223372036854775808 to 9223372036854775807
• Float64: IEEE 754, Double-precision floating-point format, 64 Bits = 8 Bytes, 2.2e-308 to 1.7e+308

DATEX uses the little endian format for all numbers.

6.2 Stand-alone codes

Name	Hex Code	DX equivalent	Description
CLOSE_AND_STORE	a0	;	Close current statement and store result in the #scope_result variable. A DATEX Script must always end with CLOSE_AND_STORE.
SUBSCOPE_START	a1	(	Subscope starts, executed independently from parent scope
SUBSCOPE_END	a2	)	Subscope closed, #scope_result passed on to parent
END	00	end	The scope execution is immediately stopped. The #scope_result is returned as normal
RETURN	a4	return	Returns the current #scope_result, scope continues afterwards

6.3 Runtime Commands

Runtime commands can be standalone commands or commands with parameters. A command is normally expected to be followed by a specfic amount of values (parameters). Runtime commands are equivalent to operators (e.g. comparators) in DXB, but are viewed as a seperate group because they correspond to a string in DATEX Script (semantics similar to applying a value to a function).
Other operators always have a shorthand symbol in DATEX Script (e.g. !, ==).

Name	Hex Code	Structure	Expected Parameters	Return value type	DX equivalent	Description
JMP	a5	[INDEX]: Uint32	-	<Void>	jmp 42	Changes scope index to 'INDEX' and continues execution from there
JTR	a6	[INDEX]: Uint32	<boolean>	<Void>	jtr 42 ev1	Jumps if the parameter value is true-ish
JFA	66	[INDEX]: Uint32	<boolean>	<Void>	jfa 42 ev1	Jumps if the parameter value is false-ish
COUNT	ad	standalone	any	<integer>	count ev1	Counts the number of elements of the ev
ABOUT	ae	standalone	any	<Markdown>	about ev1	Returns a <Markdown> value describing the value
GET_TYPE	f5	standalone	any	<Type>	type ev1	Returns the <Type> of a value
RESOLVE_URL	52	standalone	<text>	<Url>	url "https://..."	Converts the string to a <Url>. The url is resolved to its content
TEMPLATE	53	standalone	<Type> <Object>	<Void>	template <Type> {}	Defines the template for a type
EXTENDS	54	standalone	any	<boolean>	av extends y	Returns true if the current active value (av) extends y
IMPLEMENTS	55	standalone	<Type>	<boolean>	av implements <Type>	Returns true if the current active value (av) extends the template of the type
DELETE_POINTER	ba	standalone	any (pointer)	<Void>	delete pointer	Permanently deletes the pointer on the current endpoint
SUBSCRIBE	bb	standalone	any (pointer)	any	subscribe pointer	Get future updates for this pointer. The current value of the pointer is returned
UNSUBSCRIBE	bc	standalone	any (pointer)	<Void>	unsubscribe pointer	Stop getting future updates for this pointer.
VALUE	bd	standalone	any (pointer)	any	value pointer	Get the value of a pointer
ORIGIN	be	standalone	any (pointer)	<Endpoint>	origin pointer	Get the origin endpoint of a pointer.
SUBSCRIBERS	bf	standalone	any (pointer)	<Filter>	subscribers pointer	Get all current subscribers for a pointer.

6.4 Operators

6.4.1 Comparators

A comparator operator compares the current active value with an effective value that follows after the comparator.

Name	Hex Code	DX equivalent	Description
EQUAL	a7	av == ev1	Active value is set to true if current active value is the same as ev1
NOT_EQUAL	a8	av ~= ev1	Active value is set to true if current active value is not the same as ev1
GREATER	a9	av > ev1	Active value is set to true if current active value is greater ev1
LESS	aa	av < ev1	Active value is set to true if current active value is less than ev1
GREATER_EQUAL	ab	av >= ev1	Active value is set to true if current active value is greater or equal to ev1
LESS_EQUAL	ac	av <= ev1	Active value is set to true if current active value is less or equal to ev1

6.4.2 Combinators

Combination operators combine two values and can be used as mathematical or logical operators, but can also have custom functions for different value types. For mathematical operations, Floats and Ints can be combined.

Name	Hex Code	DX equivalent	Description
ADD	f8	1 + 2, 1.0 + 2, "a" + "b", [1,2] + [3,4] (whitespaces required)	Adds the active value to the next value
SUBTRACT	fa	10 - 2 (whitespaces required)	Subtracts next value from the active value
MULTIPLY	fb	10 * 2, 10 * 'a', (void * 4, 1, (1,2,3) * 2) (whitespaces required)	Multiplies the active value with the next value
DIVIDE	fb	22.22 / 2	Divides the active value from the next value
AND	ea	true & b, @user1 & #allowed	Logical AND
OR	eb	false | b, @user1 | @user2	Logical OR
RANGE	fd	a..b	Accepts only <integer> values, returns <Tuple> with ascending integers starting with a and excluding b

6.4.3 Other operators

Name	Hex Code	DX equivalent	Description
STREAM	ed	x << 'streamed data' 'more data'	Accepts an arbitrary number of immediately followed <text>, <Buffer>, and <Stream> values. x must be a <StreamSink>

6.5 Shortcuts for Standard Library types

The <Type> values for all standard library types are represented with a single byte:

Name	Hex Code	DX equivalent
STD_TYPE_STRING	10	<text>
STD_TYPE_INT	11	<integer>
STD_TYPE_FLOAT	12	<decimal>
STD_TYPE_BOOLEAN	13	<boolean>
STD_TYPE_NULL	14	<Null>
STD_TYPE_VOID	15	<Void>
STD_TYPE_BUFFER	16	<Buffer>
STD_TYPE_CODE_BLOCK	17	<DATEX>
STD_TYPE_UNIT	18	<Unit>
STD_TYPE_FILTER	19	<Filter>
STD_TYPE_ARRAY	1a	<Array>
STD_TYPE_OBJECT	1b	<Object>
STD_TYPE_SET	1c	<Set>
STD_TYPE_MAP	1d	<Map>
STD_TYPE_TUPLE	1e	<Tuple>
STD_TYPE_RECORD	1f	<Tuple>
STD_TYPE_FUNCTION	20	<Function>
STD_TYPE_STREAM	21	<Stream>

6.6 Values

Values instructions parse a value and then insert it into the scope.

6.6.1 Standalone values

Name	Hex Code	DX equivalent
TRUE	c4	true
FALSE	c5	false
NULL	c6	null
VOID	c7	void

6.6.2 Other primitive values

Name	Hex Code	Structure	DX example	Description
STRING	c0	LENGTH: Uint32 VALUE: Uint8[LENGTH]	"A String"	Maximum size 4.29 GB. DATEX Strings are not null-terminated, their length is stored
SHORT_STRING	ce	LENGTH: Uint8 VALUE: Uint8[LENGTH]	"A Short String"	Maximum size 255 bytes
INT_8	c1	VALUE: Int8	127	Range -128 -> 127
INT_16	c2	VALUE: Int16	32767	Range –32768 -> 32767
INT_32	c3	VALUE: Int32	127	Range –2147483648 -> 2147483647
INT_64	c4	VALUE: Int64	<integer>9e18	Range -9223372036854775808 -> 9223372036854775807
FLOAT_64	c5	VALUE: Float64	-12.34, infinity, nan	Range 2.2e-308 -> 1.7e+308
BUFFER	ca	LENGTH: Uint32 VALUE: Uint8[LENGTH]	`fafe334feaefe3`	A raw binary data buffer, maximum size 4.29 GB

6.6.3 Array, Objects, Tuples and Records

Name	Hex Code	DX equivalent	Description
ARRAY_START	e0	[	<Array> declaration start
ARRAY_END	e1	]	<Array> declaration end
OBJECT_START	e2	{	<Object> declaration start
OBJECT_END	e3	}	<Object> declaration end
TUPLE_START	e4	(	<Tuple> declaration start
TUPLE_END	e5	)	<Tuple> declaration end

Tuple, Record, and Subscope share the same symbol in DATEX Script.

6.7 Pointers and Variables

DATEX distinguishes between normal Variables, Internal Variables for specific purposes, and Pointers/Labels for persistant values.

Name	Hex Code	DX equivalent	Description
VAR	b0	myVar	Inserts the value of 'myVar' into the scope
SET_VAR	b1	myVar = x	Sets the value of 'myVar' to 'x'
VAR_ACTION	b2	myVar += x, myVar -= x, myVar &= x	Applies 'x' with the operator to 'myVar' internally (The value of 'myVar' is updated, but the variable is not overwritten)