20012014 Ericsson AB. All Rights Reserved. The contents of this file are subject to the Erlang Public License, Version 1.1, (the "License"); you may not use this file except in compliance with the License. You should have received a copy of the Erlang Public License along with this software. If not, it can be retrieved online at http://www.erlang.org/. Software distributed under the License is distributed on an "AS IS" basis, WITHOUT WARRANTY OF ANY KIND, either express or implied. See the License for the specific language governing rights and limitations under the License. ei Jakob Cederlund Kent Boortz 1 Kenneth Lundin 2000-11-27 PA1 ei.sgml
ei routines for handling the erlang binary term format

The library contains macros and functions to encode and decode the erlang binary term format.

With , you can convert atoms, lists, numbers and binaries to and from the binary format. This is useful when writing port programs and drivers. uses a given buffer, and no dynamic memory (with the exception of ), and is often quite fast.

It also handles C-nodes, C-programs that talks erlang distribution with erlang nodes (or other C-nodes) using the erlang distribution format. The difference between and is that uses the binary format directly when sending and receiving terms. It is also thread safe, and using threads, one process can handle multiple C-nodes. The library is built on top of , but of legacy reasons, it doesn't allow for multiple C-nodes. In general, is the preferred way of doing C-nodes.

The decode and encode functions use a buffer an index into the buffer, which points at the point where to encode and decode. The index is updated to point right after the term encoded/decoded. No checking is done whether the term fits in the buffer or not. If encoding goes outside the buffer, the program may crash.

All functions takes two parameter, is a pointer to the buffer where the binary data is / will be, is a pointer to an index into the buffer. This parameter will be incremented with the size of the term decoded / encoded. The data is thus at when an function is called.

The encode functions all assumes that the and parameters points to a buffer big enough for the data. To get the size of an encoded term, without encoding it, pass instead of a buffer pointer. The parameter will be incremented, but nothing will be encoded. This is the way in to "preflight" term encoding.

There are also encode-functions that uses a dynamic buffer. It is often more convenient to use these to encode data. All encode functions comes in two versions: those starting with , uses a dynamic buffer.

All functions return if successful, and if not. (For instance, if a term is not of the expected type, or the data to decode is not a valid erlang term.)

Some of the decode-functions needs a preallocated buffer. This buffer must be allocated big enough, and for non compound types the function returns the size required (note that for strings an extra byte is needed for the 0 string terminator).

DATA TYPES erlang_char_encoding

typedef enum { ERLANG_ASCII = 1, ERLANG_LATIN1 = 2, ERLANG_UTF8 = 4 }erlang_char_encoding;

The character encodings used for atoms. ERLANG_ASCII represents 7-bit ASCII. Latin1 and UTF8 are different extensions of 7-bit ASCII. All 7-bit ASCII characters are valid Latin1 and UTF8 characters. ASCII and Latin1 both represent each character by one byte. A UTF8 character can consist of one to four bytes. Note that these constants are bit-flags and can be combined with bitwise-or.

voidei_set_compat_rel(release_number) Set the ei library in compatibility mode unsigned release_number;

By default, the library is only guaranteed to be compatible with other Erlang/OTP components from the same release as the library itself. For example, from the OTP R10 release is not compatible with an Erlang emulator from the OTP R9 release by default.

A call to sets the library in compatibility mode of release . Valid range of is [7, current release]. This makes it possible to communicate with Erlang/OTP components from earlier releases.

If this function is called, it may only be called once and must be called before any other functions in the library is called.

You may run into trouble if this feature is used carelessly. Always make sure that all communicating components are either from the same Erlang/OTP release, or from release X and release Y where all components from release Y are in compatibility mode of release X.

intei_encode_version(char *buf, int *index) intei_x_encode_version(ei_x_buff* x) Encode version

Encodes a version magic number for the binary format. Must be the first token in a binary term.

intei_encode_long(char *buf, int *index, long p) intei_x_encode_long(ei_x_buff* x, long p) Encode integer

Encodes a long integer in the binary format. Note that if the code is 64 bits the function ei_encode_long() is exactly the same as ei_encode_longlong().

intei_encode_ulong(char *buf, int *index, unsigned long p) intei_x_encode_ulong(ei_x_buff* x, unsigned long p) Encode unsigned integer

Encodes an unsigned long integer in the binary format. Note that if the code is 64 bits the function ei_encode_ulong() is exactly the same as ei_encode_ulonglong().

intei_encode_longlong(char *buf, int *index, long long p) intei_x_encode_longlong(ei_x_buff* x, long long p) Encode integer

Encodes a GCC or Visual C++ (64 bit) integer in the binary format. Note that this function is missing in the VxWorks port.

intei_encode_ulonglong(char *buf, int *index, unsigned long long p) intei_x_encode_ulonglong(ei_x_buff* x, unsigned long long p) Encode unsigned integer

Encodes a GCC or Visual C++ (64 bit) integer in the binary format. Note that this function is missing in the VxWorks port.

intei_encode_bignum(char *buf, int *index, mpz_t obj) intei_x_encode_bignum(ei_x_buff *x, mpz_t obj) Encode an arbitrary precision integer

Encodes a GMP integer to binary format. To use this function the ei library needs to be configured and compiled to use the GMP library.

intei_encode_double(char *buf, int *index, double p) intei_x_encode_double(ei_x_buff* x, double p) Encode a double float

Encodes a double-precision (64 bit) floating point number in the binary format.

The function returns if the floating point number is not finite.

intei_encode_boolean(char *buf, int *index, int p) intei_x_encode_boolean(ei_x_buff* x, int p) Encode a boolean

Encodes a boolean value, as the atom if p is not zero or if p is zero.

intei_encode_char(char *buf, int *index, char p) intei_x_encode_char(ei_x_buff* x, char p) Encode an 8-bit integer between 0-255

Encodes a char (8-bit) as an integer between 0-255 in the binary format. Note that for historical reasons the integer argument is of type . Your C code should consider the given argument to be of type even if the C compilers and system may define to be signed.

intei_encode_string(char *buf, int *index, const char *p) intei_encode_string_len(char *buf, int *index, const char *p, int len) intei_x_encode_string(ei_x_buff* x, const char *p) intei_x_encode_string_len(ei_x_buff* x, const char* s, int len) Encode a string

Encodes a string in the binary format. (A string in erlang is a list, but is encoded as a character array in the binary format.) The string should be zero-terminated, except for the function.

intei_encode_atom(char *buf, int *index, const char *p) intei_encode_atom_len(char *buf, int *index, const char *p, int len) intei_x_encode_atom(ei_x_buff* x, const char *p) intei_x_encode_atom_len(ei_x_buff* x, const char *p, int len) Encode an atom

Encodes an atom in the binary format. The parameter is the name of the atom in latin1 encoding. Only upto MAXATOMLEN-1 bytes are encoded. The name should be zero-terminated, except for the function.

intei_encode_atom_as(char *buf, int *index, const char *p, erlang_char_encoding from_enc, erlang_char_encoding to_enc) intei_encode_atom_len_as(char *buf, int *index, const char *p, int len, erlang_char_encoding from_enc, erlang_char_encoding to_enc) intei_x_encode_atom_as(ei_x_buff* x, const char *p, erlang_char_encoding from_enc, erlang_char_encoding to_enc) intei_x_encode_atom_len_as(ei_x_buff* x, const char *p, int len, erlang_char_encoding from_enc, erlang_char_encoding to_enc) Encode an atom

Encodes an atom in the binary format with character encoding to_enc (latin1 or utf8). The p parameter is the name of the atom with character encoding from_enc (ascii, latin1 or utf8). The name must either be zero-terminated or a function variant with a len parameter must be used. If to_enc is set to the bitwise-or'd combination (ERLANG_LATIN1|ERLANG_UTF8), utf8 encoding is only used if the atom string can not be represented in latin1 encoding.

The encoding will fail if p is not a valid string in encoding from_enc, if the string is too long or if it can not be represented with character encoding to_enc.

These functions were introduced in R16 release of Erlang/OTP as part of a first step to support UTF8 atoms. Atoms encoded with ERLANG_UTF8 can not be decoded by earlier releases than R16.

intei_encode_binary(char *buf, int *index, const void *p, long len) intei_x_encode_binary(ei_x_buff* x, const void *p, long len) Encode a binary

Encodes a binary in the binary format. The data is at , of bytes length.

intei_encode_pid(char *buf, int *index, const erlang_pid *p) intei_x_encode_pid(ei_x_buff* x, const erlang_pid *p) Encode a pid

Encodes an erlang process identifier, pid, in the binary format. The parameter points to an structure (which should have been obtained earlier with ).

intei_encode_fun(char *buf, int *index, const erlang_fun *p) intei_x_encode_fun(ei_x_buff* x, const erlang_fun* fun) Encode a fun

Encodes a fun in the binary format. The parameter points to an structure. The is not freed automatically, the should be called if the fun is not needed after encoding.

intei_encode_port(char *buf, int *index, const erlang_port *p) intei_x_encode_port(ei_x_buff* x, const erlang_port *p) Encodes a port

Encodes an erlang port in the binary format. The parameter points to a structure (which should have been obtained earlier with .

intei_encode_ref(char *buf, int *index, const erlang_ref *p) intei_x_encode_ref(ei_x_buff* x, const erlang_ref *p) Encodes a ref

Encodes an erlang reference in the binary format. The parameter points to a structure (which should have been obtained earlier with .

intei_encode_term(char *buf, int *index, void *t) intei_x_encode_term(ei_x_buff* x, void *t) Encode an term

This function encodes an , as obtained from . The parameter is actually an pointer. This function doesn't free the .

intei_encode_trace(char *buf, int *index, const erlang_trace *p) intei_x_encode_trace(ei_x_buff* x, const erlang_trace *p) Encode a trace token

This function encodes an erlang trace token in the binary format. The parameter points to a structure (which should have been obtained earlier with .

intei_encode_tuple_header(char *buf, int *index, int arity) intei_x_encode_tuple_header(ei_x_buff* x, int arity) Encode a tuple

This function encodes a tuple header, with a specified arity. The next terms encoded will be the elements of the tuple. Tuples and lists are encoded recursively, so that a tuple may contain another tuple or list.

E.g. to encode the tuple :

ei_encode_tuple_header(buf, &i, 2);
ei_encode_atom(buf, &i, "a");
ei_encode_tuple_header(buf, &i, 2);
ei_encode_atom(buf, &i, "b");
ei_encode_tuple_header(buf, &i, 0);
        
intei_encode_list_header(char *buf, int *index, int arity) intei_x_encode_list_header(ei_x_buff* x, int arity) Encode a list

This function encodes a list header, with a specified arity. The next terms are the elements (actually its cons cells) and the tail of the list. Lists and tuples are encoded recursively, so that a list may contain another list or tuple.

E.g. to encode the list :

ei_encode_list_header(buf, &i, 3);
ei_encode_atom(buf, &i, "c");
ei_encode_atom(buf, &i, "d");
ei_encode_list_header(buf, &i, 1);
ei_encode_atom(buf, &i, "e");
ei_encode_atom(buf, &i, "f");
ei_encode_empty_list(buf, &i);
        

It may seem that there is no way to create a list without knowing the number of elements in advance. But indeed there is a way. Note that the list can be written as . Using this, a list can be written as conses.

To encode a list, without knowing the arity in advance:

while (something()) {
    ei_x_encode_list_header(&x, 1);
    ei_x_encode_ulong(&x, i); /* just an example */
}
ei_x_encode_empty_list(&x);
        
intei_encode_empty_list(char* buf, int* index) intei_x_encode_empty_list(ei_x_buff* x) Encode an empty list ()

This function encodes an empty list. It's often used at the tail of a list.

intei_encode_map_header(char *buf, int *index, int arity) intei_x_encode_map_header(ei_x_buff* x, int arity) Encode a map

This function encodes a map header, with a specified arity. The next arity*2 terms encoded will be the keys and values of the map encoded in the following order: K1, V1, K2, V2, ..., Kn, Vn.

E.g. to encode the map #{a => "Apple", b => "Banana"}:

ei_x_encode_map_header(&x, 2);
ei_x_encode_atom(&x, "a");
ei_x_encode_string(&x, "Apple");
ei_x_encode_atom(&x, "b");
ei_x_encode_string(&x, "Banana");
        

A correctly encoded map can not have duplicate keys.

intei_get_type(const char *buf, const int *index, int *type, int *size) Fetch the type and size of an encoded term

This function returns the type in and size in of the encoded term. For strings and atoms, size is the number of characters not including the terminating 0. For binaries, is the number of bytes. For lists and tuples, is the arity of the object. For other types, is 0. In all cases, is left unchanged.

intei_decode_version(const char *buf, int *index, int *version) Encode an empty list ()

This function decodes the version magic number for the erlang binary term format. It must be the first token in a binary term.

intei_decode_long(const char *buf, int *index, long *p) Decode integer

This function decodes a long integer from the binary format. Note that if the code is 64 bits the function ei_decode_long() is exactly the same as ei_decode_longlong().

intei_decode_ulong(const char *buf, int *index, unsigned long *p) Decode unsigned integer

This function decodes an unsigned long integer from the binary format. Note that if the code is 64 bits the function ei_decode_ulong() is exactly the same as ei_decode_ulonglong().

intei_decode_longlong(const char *buf, int *index, long long *p) Decode integer

This function decodes a GCC or Visual C++ (64 bit) integer from the binary format. Note that this function is missing in the VxWorks port.

intei_decode_ulonglong(const char *buf, int *index, unsigned long long *p) Decode unsigned integer

This function decodes a GCC or Visual C++ (64 bit) integer from the binary format. Note that this function is missing in the VxWorks port.

intei_decode_bignum(const char *buf, int *index, mpz_t obj) Decode a GMP arbitrary precision integer

This function decodes an integer in the binary format to a GMP integer. To use this function the ei library needs to be configured and compiled to use the GMP library.

intei_decode_double(const char *buf, int *index, double *p) Decode a double

This function decodes an double-precision (64 bit) floating point number from the binary format.

intei_decode_boolean(const char *buf, int *index, int *p) Decode a boolean

This function decodes a boolean value from the binary format. A boolean is actually an atom, decodes 1 and decodes 0.

intei_decode_char(const char *buf, int *index, char *p) Decode an 8-bit integer between 0-255

This function decodes a char (8-bit) integer between 0-255 from the binary format. Note that for historical reasons the returned integer is of type . Your C code should consider the returned value to be of type even if the C compilers and system may define to be signed.

intei_decode_string(const char *buf, int *index, char *p) Decode a string

This function decodes a string from the binary format. A string in erlang is a list of integers between 0 and 255. Note that since the string is just a list, sometimes lists are encoded as strings by , even if it was not intended.

The string is copied to , and enough space must be allocated. The returned string is null terminated so you need to add an extra byte to the memory requirement.

intei_decode_atom(const char *buf, int *index, char *p) Decode an atom

This function decodes an atom from the binary format. The null terminated name of the atom is placed at . There can be at most bytes placed in the buffer.

intei_decode_atom_as(const char *buf, int *index, char *p, int plen, erlang_char_encoding want, erlang_char_encoding* was, erlang_char_encoding* result) Decode an atom

This function decodes an atom from the binary format. The null terminated name of the atom is placed in buffer at p of length plen bytes.

The wanted string encoding is specified by want. The original encoding used in the binary format (latin1 or utf8) can be obtained from *was. The actual encoding of the resulting string (7-bit ascii, latin1 or utf8) can be obtained from *result. Both was and result can be NULL. *result may differ from want if want is a bitwise-or'd combination like ERLANG_LATIN1|ERLANG_UTF8 or if *result turn out to be pure 7-bit ascii (compatible with both latin1 and utf8).

This function fails if the atom is too long for the buffer or if it can not be represented with encoding want.

This function was introduced in R16 release of Erlang/OTP as part of a first step to support UTF8 atoms.

intei_decode_binary(const char *buf, int *index, void *p, long *len) Decode a binary

This function decodes a binary from the binary format. The parameter is set to the actual size of the binary. Note that assumes that there are enough room for the binary. The size required can be fetched by .

intei_decode_fun(const char *buf, int *index, erlang_fun *p) voidfree_fun(erlang_fun* f) Decode a fun

This function decodes a fun from the binary format. The parameter should be NULL or point to an structure. This is the only decode function that allocates memory; when the is no longer needed, it should be freed with . (This has to do with the arbitrary size of the environment for a fun.)

intei_decode_pid(const char *buf, int *index, erlang_pid *p) Decode a

Decodes a pid, process identifier, from the binary format.

intei_decode_port(const char *buf, int *index, erlang_port *p) Decode a port

This function decodes a port identifier from the binary format.

intei_decode_ref(const char *buf, int *index, erlang_ref *p) Decode a reference

This function decodes a reference from the binary format.

intei_decode_trace(const char *buf, int *index, erlang_trace *p) Decode a trace token

Decodes an erlang trace token from the binary format.

intei_decode_tuple_header(const char *buf, int *index, int *arity) Decode a tuple

This function decodes a tuple header, the number of elements is returned in . The tuple elements follows in order in the buffer.

intei_decode_list_header(const char *buf, int *index, int *arity) Decode a list

This function decodes a list header from the binary format. The number of elements is returned in . The elements follows (the last one is the tail of the list, normally an empty list.) If is , it's an empty list.

Note that lists are encoded as strings, if they consist entirely of integers in the range 0..255. This function will not decode such strings, use instead.

intei_decode_map_header(const char *buf, int *index, int *arity) Decode a map

This function decodes a map header from the binary format. The number of key-value pairs is returned in *arity. Keys and values follow in the following order: K1, V1, K2, V2, ..., Kn, Vn. This makes a total of arity*2 terms. If arity is zero, it's an empty map. A correctly encoded map does not have duplicate keys.

intei_decode_ei_term(const char* buf, int* index, ei_term* term) Decode a term, without prior knowledge of type

This function decodes any term, or at least tries to. If the term pointed at by in fits in the union, it is decoded, and the appropriate field in value]]> is set, and is incremented by the term size.

The function returns 1 on successful decoding, -1 on error, and 0 if the term seems alright, but does not fit in the structure. If it returns 1, the will be incremented, and the contains the decoded term.

The structure will contain the arity for a tuple or list, size for a binary, string or atom. It will contains a term if it's any of the following: integer, float, atom, pid, port or ref.

intei_decode_term(const char *buf, int *index, void *t) Decode a

This function decodes a term from the binary format. The term is return in as a , so is actually an (see . The term should later be deallocated.

Note that this function is located in the erl_interface library.

intei_print_term(FILE* fp, const char* buf, int* index) intei_s_print_term(char** s, const char* buf, int* index) Print a term in clear text

This function prints a term, in clear text, to the file given by , or the buffer pointed to by . It tries to resemble the term printing in the erlang shell.

In , the parameter should point to a dynamically (malloc) allocated string of bytes or a NULL pointer. The string may be reallocated (and may be updated) by this function if the result is more than characters. The string returned is zero-terminated.

The return value is the number of characters written to the file or string, or -1 if doesn't contain a valid term. Unfortunately, I/O errors on is not checked.

The argument is updated, i.e. this function can be viewed as en decode function that decodes a term into a human readable format.

intei_x_format(ei_x_buff* x, const char* fmt, ...) intei_x_format_wo_ver(ei_x_buff* x, const char *fmt, ... ) Format a term from a format string and parameters.

Format a term, given as a string, to a buffer. This functions works like a sprintf for erlang terms. The contains a format string, with arguments like , to insert terms from variables. The following formats are supported (with the C types given):

~a - an atom, char*
~c - a character, char
~s - a string, char*
~i - an integer, int
~l - a long integer, long int
~u - a unsigned long integer, unsigned long int
~f - a float, float
~d - a double float, double float
~p - an Erlang PID, erlang_pid*
        

For instance, to encode a tuple with some stuff:

ei_x_format("{~a,~i,~d}", "numbers", 12, 3.14159)
encodes the tuple {numbers,12,3.14159}
        

The formats into a buffer, without the initial version byte.

intei_x_new(ei_x_buff* x) intei_x_new_with_version(ei_x_buff* x) Allocate a new buffer

This function allocates a new buffer. The fields of the structure pointed to by parameter is filled in, and a default buffer is allocated. The also puts an initial version byte, that is used in the binary format. (So that won't be needed.)

intei_x_free(ei_x_buff* x) Frees a buffer

This function frees an buffer. The memory used by the buffer is returned to the OS.

intei_x_append(ei_x_buff* x, const ei_x_buff* x2) intei_x_append_buf(ei_x_buff* x, const char* buf, int len) Appends a buffer at the end

These functions appends data at the end of the buffer .

intei_skip_term(const char* buf, int* index) skip a term

This function skips a term in the given buffer, it recursively skips elements of lists and tuples, so that a full term is skipped. This is a way to get the size of an erlang term.

is the buffer.

is updated to point right after the term in the buffer.

This can be useful when you want to hold arbitrary terms: just skip them and copy the binary term data to some buffer.

The function returns on success and on failure.

Debug Information

Some tips on what to check when the emulator doesn't seem to receive the terms that you send.

be careful with the version header, use when appropriate turn on distribution tracing on the erlang node check the result codes from ei_decode_-calls
See Also

erl_interface(3)