[ViewVC] Diff of: cvs/CBOR-XS/XS.pm

Comparing CBOR-XS/XS.pm (file contents):
Revision 1.4 by root, Sat Oct 26 22:25:47 2013 UTC vs.
Revision 1.40 by root, Sun Jan 5 14:24:54 2014 UTC

…		…
12	$perl_value = decode_cbor $binary_cbor_data;	12	$perl_value = decode_cbor $binary_cbor_data;
13		13
14	# OO-interface	14	# OO-interface
15		15
16	$coder = CBOR::XS->new;	16	$coder = CBOR::XS->new;
17	#TODO	17	$binary_cbor_data = $coder->encode ($perl_value);
		18	$perl_value = $coder->decode ($binary_cbor_data);
		19
		20	# prefix decoding
		21
		22	my $many_cbor_strings = ...;
		23	while (length $many_cbor_strings) {
		24	my ($data, $length) = $cbor->decode_prefix ($many_cbor_strings);
		25	# data was decoded
		26	substr $many_cbor_strings, 0, $length, ""; # remove decoded cbor string
		27	}
18		28
19	=head1 DESCRIPTION	29	=head1 DESCRIPTION
20		30
21	WARNING! THIS IS A PRE-ALPHA RELEASE! IT WILL CRASH, CORRUPT YOUR DATA AND	31	This module converts Perl data structures to the Concise Binary Object
22	EAT YOUR CHILDREN!	32	Representation (CBOR) and vice versa. CBOR is a fast binary serialisation
		33	format that aims to use an (almost) superset of the JSON data model, i.e.
		34	when you can represent something useful in JSON, you should be able to
		35	represent it in CBOR.
23		36
24	This module converts Perl data structures to CBOR and vice versa. Its	37	In short, CBOR is a faster and quite compact binary alternative to JSON,
		38	with the added ability of supporting serialisation of Perl objects. (JSON
		39	often compresses better than CBOR though, so if you plan to compress the
		40	data later and speed is less important you might want to compare both
		41	formats first).
		42
		43	To give you a general idea about speed, with texts in the megabyte range,
		44	C<CBOR::XS> usually encodes roughly twice as fast as L<Storable> or
		45	L<JSON::XS> and decodes about 15%-30% faster than those. The shorter the
		46	data, the worse L<Storable> performs in comparison.
		47
		48	Regarding compactness, C<CBOR::XS>-encoded data structures are usually
		49	about 20% smaller than the same data encoded as (compact) JSON or
		50	L<Storable>.
		51
		52	In addition to the core CBOR data format, this module implements a
		53	number of extensions, to support cyclic and shared data structures
		54	(see C<allow_sharing> and C<allow_cycles>), string deduplication (see
		55	C<pack_strings>) and scalar references (always enabled).
		56
25	primary goal is to be I<correct> and its secondary goal is to be	57	The primary goal of this module is to be I<correct> and the secondary goal
26	I<fast>. To reach the latter goal it was written in C.	58	is to be I<fast>. To reach the latter goal it was written in C.
27		59
28	See MAPPING, below, on how CBOR::XS maps perl values to CBOR values and	60	See MAPPING, below, on how CBOR::XS maps perl values to CBOR values and
29	vice versa.	61	vice versa.
30		62
31	=cut	63	=cut
32		64
33	package CBOR::XS;	65	package CBOR::XS;
34		66
35	use common::sense;	67	use common::sense;
36		68
37	our $VERSION = 0.02;	69	our $VERSION = 1.25;
38	our @ISA = qw(Exporter);	70	our @ISA = qw(Exporter);
39		71
40	our @EXPORT = qw(encode_cbor decode_cbor);	72	our @EXPORT = qw(encode_cbor decode_cbor);
41		73
42	use Exporter;	74	use Exporter;
43	use XSLoader;	75	use XSLoader;
44		76
		77	use Types::Serialiser;
		78
45	our $MAGIC = "\xd9\xd9\xf7";	79	our $MAGIC = "\xd9\xd9\xf7";
46		80
47	=head1 FUNCTIONAL INTERFACE	81	=head1 FUNCTIONAL INTERFACE
48		82
49	The following convenience methods are provided by this module. They are	83	The following convenience methods are provided by this module. They are
…		…
77	strings. All boolean flags described below are by default I<disabled>.	111	strings. All boolean flags described below are by default I<disabled>.
78		112
79	The mutators for flags all return the CBOR object again and thus calls can	113	The mutators for flags all return the CBOR object again and thus calls can
80	be chained:	114	be chained:
81		115
82	#TODO
83	my $cbor = CBOR::XS->new->encode ({a => [1,2]});	116	my $cbor = CBOR::XS->new->encode ({a => [1,2]});
84		117
85	=item $cbor = $cbor->max_depth ([$maximum_nesting_depth])	118	=item $cbor = $cbor->max_depth ([$maximum_nesting_depth])
86		119
87	=item $max_depth = $cbor->get_max_depth	120	=item $max_depth = $cbor->get_max_depth
…		…
121	If no argument is given, the limit check will be deactivated (same as when	154	If no argument is given, the limit check will be deactivated (same as when
122	C<0> is specified).	155	C<0> is specified).
123		156
124	See SECURITY CONSIDERATIONS, below, for more info on why this is useful.	157	See SECURITY CONSIDERATIONS, below, for more info on why this is useful.
125		158
		159	=item $cbor = $cbor->allow_unknown ([$enable])
		160
		161	=item $enabled = $cbor->get_allow_unknown
		162
		163	If C<$enable> is true (or missing), then C<encode> will I<not> throw an
		164	exception when it encounters values it cannot represent in CBOR (for
		165	example, filehandles) but instead will encode a CBOR C<error> value.
		166
		167	If C<$enable> is false (the default), then C<encode> will throw an
		168	exception when it encounters anything it cannot encode as CBOR.
		169
		170	This option does not affect C<decode> in any way, and it is recommended to
		171	leave it off unless you know your communications partner.
		172
		173	=item $cbor = $cbor->allow_sharing ([$enable])
		174
		175	=item $enabled = $cbor->get_allow_sharing
		176
		177	If C<$enable> is true (or missing), then C<encode> will not double-encode
		178	values that have been referenced before (e.g. when the same object, such
		179	as an array, is referenced multiple times), but instead will emit a
		180	reference to the earlier value.
		181
		182	This means that such values will only be encoded once, and will not result
		183	in a deep cloning of the value on decode, in decoders supporting the value
		184	sharing extension. This also makes it possible to encode cyclic data
		185	structures (which need C<allow_cycles> to ne enabled to be decoded by this
		186	module).
		187
		188	It is recommended to leave it off unless you know your
		189	communication partner supports the value sharing extensions to CBOR
		190	(L<http://cbor.schmorp.de/value-sharing>), as without decoder support, the
		191	resulting data structure might be unusable.
		192
		193	Detecting shared values incurs a runtime overhead when values are encoded
		194	that have a reference counter large than one, and might unnecessarily
		195	increase the encoded size, as potentially shared values are encode as
		196	shareable whether or not they are actually shared.
		197
		198	At the moment, only targets of references can be shared (e.g. scalars,
		199	arrays or hashes pointed to by a reference). Weirder constructs, such as
		200	an array with multiple "copies" of the I<same> string, which are hard but
		201	not impossible to create in Perl, are not supported (this is the same as
		202	with L<Storable>).
		203
		204	If C<$enable> is false (the default), then C<encode> will encode shared
		205	data structures repeatedly, unsharing them in the process. Cyclic data
		206	structures cannot be encoded in this mode.
		207
		208	This option does not affect C<decode> in any way - shared values and
		209	references will always be decoded properly if present.
		210
		211	=item $cbor = $cbor->allow_cycles ([$enable])
		212
		213	=item $enabled = $cbor->get_allow_cycles
		214
		215	If C<$enable> is true (or missing), then C<decode> will happily decode
		216	self-referential (cyclic) data structures. By default these will not be
		217	decoded, as they need manual cleanup to avoid memory leaks, so code that
		218	isn't prepared for this will not leak memory.
		219
		220	If C<$enable> is false (the default), then C<decode> will throw an error
		221	when it encounters a self-referential/cyclic data structure.
		222
		223	This option does not affect C<encode> in any way - shared values and
		224	references will always be decoded properly if present.
		225
		226	=item $cbor = $cbor->pack_strings ([$enable])
		227
		228	=item $enabled = $cbor->get_pack_strings
		229
		230	If C<$enable> is true (or missing), then C<encode> will try not to encode
		231	the same string twice, but will instead encode a reference to the string
		232	instead. Depending on your data format, this can save a lot of space, but
		233	also results in a very large runtime overhead (expect encoding times to be
		234	2-4 times as high as without).
		235
		236	It is recommended to leave it off unless you know your
		237	communications partner supports the stringref extension to CBOR
		238	(L<http://cbor.schmorp.de/stringref>), as without decoder support, the
		239	resulting data structure might not be usable.
		240
		241	If C<$enable> is false (the default), then C<encode> will encode strings
		242	the standard CBOR way.
		243
		244	This option does not affect C<decode> in any way - string references will
		245	always be decoded properly if present.
		246
		247	=item $cbor = $cbor->validate_utf8 ([$enable])
		248
		249	=item $enabled = $cbor->get_validate_utf8
		250
		251	If C<$enable> is true (or missing), then C<decode> will validate that
		252	elements (text strings) containing UTF-8 data in fact contain valid UTF-8
		253	data (instead of blindly accepting it). This validation obviously takes
		254	extra time during decoding.
		255
		256	The concept of "valid UTF-8" used is perl's concept, which is a superset
		257	of the official UTF-8.
		258
		259	If C<$enable> is false (the default), then C<decode> will blindly accept
		260	UTF-8 data, marking them as valid UTF-8 in the resulting data structure
		261	regardless of whether thats true or not.
		262
		263	Perl isn't too happy about corrupted UTF-8 in strings, but should
		264	generally not crash or do similarly evil things. Extensions might be not
		265	so forgiving, so it's recommended to turn on this setting if you receive
		266	untrusted CBOR.
		267
		268	This option does not affect C<encode> in any way - strings that are
		269	supposedly valid UTF-8 will simply be dumped into the resulting CBOR
		270	string without checking whether that is, in fact, true or not.
		271
		272	=item $cbor = $cbor->filter ([$cb->($tag, $value)])
		273
		274	=item $cb_or_undef = $cbor->get_filter
		275
		276	Sets or replaces the tagged value decoding filter (when C<$cb> is
		277	specified) or clears the filter (if no argument or C<undef> is provided).
		278
		279	The filter callback is called only during decoding, when a non-enforced
		280	tagged value has been decoded (see L<TAG HANDLING AND EXTENSIONS> for a
		281	list of enforced tags). For specific tags, it's often better to provide a
		282	default converter using the C<%CBOR::XS::FILTER> hash (see below).
		283
		284	The first argument is the numerical tag, the second is the (decoded) value
		285	that has been tagged.
		286
		287	The filter function should return either exactly one value, which will
		288	replace the tagged value in the decoded data structure, or no values,
		289	which will result in default handling, which currently means the decoder
		290	creates a C<CBOR::XS::Tagged> object to hold the tag and the value.
		291
		292	When the filter is cleared (the default state), the default filter
		293	function, C<CBOR::XS::default_filter>, is used. This function simply looks
		294	up the tag in the C<%CBOR::XS::FILTER> hash. If an entry exists it must be
		295	a code reference that is called with tag and value, and is responsible for
		296	decoding the value. If no entry exists, it returns no values.
		297
		298	Example: decode all tags not handled internally into C<CBOR::XS::Tagged>
		299	objects, with no other special handling (useful when working with
		300	potentially "unsafe" CBOR data).
		301
		302	CBOR::XS->new->filter (sub { })->decode ($cbor_data);
		303
		304	Example: provide a global filter for tag 1347375694, converting the value
		305	into some string form.
		306
		307	$CBOR::XS::FILTER{1347375694} = sub {
		308	my ($tag, $value);
		309
		310	"tag 1347375694 value $value"
		311	};
		312
126	=item $cbor_data = $cbor->encode ($perl_scalar)	313	=item $cbor_data = $cbor->encode ($perl_scalar)
127		314
128	Converts the given Perl data structure (a scalar value) to its CBOR	315	Converts the given Perl data structure (a scalar value) to its CBOR
129	representation.	316	representation.
130		317
…		…
143	and you need to know where the first CBOR string ends amd the next one	330	and you need to know where the first CBOR string ends amd the next one
144	starts.	331	starts.
145		332
146	CBOR::XS->new->decode_prefix ("......")	333	CBOR::XS->new->decode_prefix ("......")
147	=> ("...", 3)	334	=> ("...", 3)
		335
		336	=back
		337
		338	=head2 INCREMENTAL PARSING
		339
		340	In some cases, there is the need for incremental parsing of JSON
		341	texts. While this module always has to keep both CBOR text and resulting
		342	Perl data structure in memory at one time, it does allow you to parse a
		343	CBOR stream incrementally, using a similar to using "decode_prefix" to see
		344	if a full CBOR object is available, but is much more efficient.
		345
		346	It basically works by parsing as much of a CBOR string as possible - if
		347	the CBOR data is not complete yet, the pasrer will remember where it was,
		348	to be able to restart when more data has been accumulated. Once enough
		349	data is available to either decode a complete CBOR value or raise an
		350	error, a real decode will be attempted.
		351
		352	A typical use case would be a network protocol that consists of sending
		353	and receiving CBOR-encoded messages. The solution that works with CBOR and
		354	about anything else is by prepending a length to every CBOR value, so the
		355	receiver knows how many octets to read. More compact (and slightly slower)
		356	would be to just send CBOR values back-to-back, as C<CBOR::XS> knows where
		357	a CBOR value ends, and doesn't need an explicit length.
		358
		359	The following methods help with this:
		360
		361	=over 4
		362
		363	=item @decoded = $cbor->incr_parse ($buffer)
		364
		365	This method attempts to decode exactly one CBOR value from the beginning
		366	of the given C<$buffer>. The value is removed from the C<$buffer> on
		367	success. When C<$buffer> doesn't contain a complete value yet, it returns
		368	nothing. Finally, when the C<$buffer> doesn't start with something
		369	that could ever be a valid CBOR value, it raises an exception, just as
		370	C<decode> would. In the latter case the decoder state is undefined and
		371	must be reset before being able to parse further.
		372
		373	This method modifies the C<$buffer> in place. When no CBOR value can be
		374	decoded, the decoder stores the current string offset. On the next call,
		375	continues decoding at the place where it stopped before. For this to make
		376	sense, the C<$buffer> must begin with the same octets as on previous
		377	unsuccessful calls.
		378
		379	You can call this method in scalar context, in which case it either
		380	returns a decoded value or C<undef>. This makes it impossible to
		381	distinguish between CBOR null values (which decode to C<undef>) and an
		382	unsuccessful decode, which is often acceptable.
		383
		384	=item @decoded = $cbor->incr_parse_multiple ($buffer)
		385
		386	Same as C<incr_parse>, but attempts to decode as many CBOR values as
		387	possible in one go, instead of at most one. Calls to C<incr_parse> and
		388	C<incr_parse_multiple> can be interleaved.
		389
		390	=item $cbor->incr_reset
		391
		392	Resets the incremental decoder. This throws away any saved state, so that
		393	subsequent calls to C<incr_parse> or C<incr_parse_multiple> start to parse
		394	a new CBOR value from the beginning of the C<$buffer> again.
		395
		396	This method can be caled at any time, but it I<must> be called if you want
		397	to change your C<$buffer> or there was a decoding error and you want to
		398	reuse the C<$cbor> object for future incremental parsings.
148		399
149	=back	400	=back
150		401
151		402
152	=head1 MAPPING	403	=head1 MAPPING
…		…
170	CBOR integers become (numeric) perl scalars. On perls without 64 bit	421	CBOR integers become (numeric) perl scalars. On perls without 64 bit
171	support, 64 bit integers will be truncated or otherwise corrupted.	422	support, 64 bit integers will be truncated or otherwise corrupted.
172		423
173	=item byte strings	424	=item byte strings
174		425
175	Byte strings will become octet strings in Perl (the byte values 0..255	426	Byte strings will become octet strings in Perl (the Byte values 0..255
176	will simply become characters of the same value in Perl).	427	will simply become characters of the same value in Perl).
177		428
178	=item UTF-8 strings	429	=item UTF-8 strings
179		430
180	UTF-8 strings in CBOR will be decoded, i.e. the UTF-8 octets will be	431	UTF-8 strings in CBOR will be decoded, i.e. the UTF-8 octets will be
…		…
186		437
187	CBOR arrays and CBOR maps will be converted into references to a Perl	438	CBOR arrays and CBOR maps will be converted into references to a Perl
188	array or hash, respectively. The keys of the map will be stringified	439	array or hash, respectively. The keys of the map will be stringified
189	during this process.	440	during this process.
190		441
		442	=item null
		443
		444	CBOR null becomes C<undef> in Perl.
		445
191	=item true, false	446	=item true, false, undefined
192		447
193	These CBOR values become C<CBOR::XS::true> and C<CBOR::XS::false>,	448	These CBOR values become C<Types:Serialiser::true>,
		449	C<Types:Serialiser::false> and C<Types::Serialiser::error>,
194	respectively. They are overloaded to act almost exactly like the numbers	450	respectively. They are overloaded to act almost exactly like the numbers
195	C<1> and C<0>. You can check whether a scalar is a CBOR boolean by using	451	C<1> and C<0> (for true and false) or to throw an exception on access (for
196	the C<CBOR::XS::is_bool> function.	452	error). See the L<Types::Serialiser> manpage for details.
197		453
198	=item null, undefined	454	=item tagged values
199		455
200	CBOR null and undefined values becomes C<undef> in Perl (in the future,
201	Undefined may raise an exception or something else).
202
203	=item tags
204
205	Tagged items consists of a numeric tag and another CBOR value. The tag	456	Tagged items consists of a numeric tag and another CBOR value.
206	55799 is ignored (this tag implements the magic header).
207		457
208	All other tags are currently converted into a L<CBOR::XS::Tagged> object,	458	See L<TAG HANDLING AND EXTENSIONS> and the description of C<< ->filter >>
209	which is simply a blessed array reference consistsing of the numeric tag	459	for details on which tags are handled how.
210	value followed by the (decoded) BOR value.
211		460
212	=item anything else	461	=item anything else
213		462
214	Anything else (e.g. unsupported simple values) will raise a decoding	463	Anything else (e.g. unsupported simple values) will raise a decoding
215	error.	464	error.
…		…
218		467
219		468
220	=head2 PERL -> CBOR	469	=head2 PERL -> CBOR
221		470
222	The mapping from Perl to CBOR is slightly more difficult, as Perl is a	471	The mapping from Perl to CBOR is slightly more difficult, as Perl is a
223	truly typeless language, so we can only guess which CBOR type is meant by	472	typeless language. That means this module can only guess which CBOR type
224	a Perl value.	473	is meant by a perl value.
225		474
226	=over 4	475	=over 4
227		476
228	=item hash references	477	=item hash references
229		478
230	Perl hash references become CBOR maps. As there is no inherent ordering in	479	Perl hash references become CBOR maps. As there is no inherent ordering in
231	hash keys (or CBOR maps), they will usually be encoded in a pseudo-random	480	hash keys (or CBOR maps), they will usually be encoded in a pseudo-random
232	order.	481	order. This order can be different each time a hahs is encoded.
233		482
234	Currently, tied hashes will use the indefinite-length format, while normal	483	Currently, tied hashes will use the indefinite-length format, while normal
235	hashes will use the fixed-length format.	484	hashes will use the fixed-length format.
236		485
237	=item array references	486	=item array references
238		487
239	Perl array references become fixed-length CBOR arrays.	488	Perl array references become fixed-length CBOR arrays.
240		489
241	=item other references	490	=item other references
242		491
243	Other unblessed references are generally not allowed and will cause an	492	Other unblessed references will be represented using
244	exception to be thrown, except for references to the integers C<0> and	493	the indirection tag extension (tag value C<22098>,
245	C<1>, which get turned into false and true in CBOR.	494	L<http://cbor.schmorp.de/indirection>). CBOR decoders are guaranteed
		495	to be able to decode these values somehow, by either "doing the right
		496	thing", decoding into a generic tagged object, simply ignoring the tag, or
		497	something else.
246		498
247	=item CBOR::XS::Tagged objects	499	=item CBOR::XS::Tagged objects
248		500
249	Objects of this type must be arrays consisting of a single C<[tag, value]>	501	Objects of this type must be arrays consisting of a single C<[tag, value]>
250	pair. The (numerical) tag will be encoded as a CBOR tag, the value will be	502	pair. The (numerical) tag will be encoded as a CBOR tag, the value will
251	encoded as appropriate for the value.	503	be encoded as appropriate for the value. You must use C<CBOR::XS::tag> to
		504	create such objects.
252		505
253	=item CBOR::XS::true, CBOR::XS::false	506	=item Types::Serialiser::true, Types::Serialiser::false, Types::Serialiser::error
254		507
255	These special values become CBOR true and CBOR false values,	508	These special values become CBOR true, CBOR false and CBOR undefined
256	respectively. You can also use C<\1> and C<\0> directly if you want.	509	values, respectively. You can also use C<\1>, C<\0> and C<\undef> directly
		510	if you want.
257		511
258	=item blessed objects	512	=item other blessed objects
259		513
260	Other blessed objects currently need to have a C<TO_CBOR> method. It	514	Other blessed objects are serialised via C<TO_CBOR> or C<FREEZE>. See
261	will be called on every object that is being serialised, and must return	515	L<TAG HANDLING AND EXTENSIONS> for specific classes handled by this
262	something that can be encoded in CBOR.	516	module, and L<OBJECT SERIALISATION> for generic object serialisation.
263		517
264	=item simple scalars	518	=item simple scalars
265		519
266	TODO
267	Simple Perl scalars (any scalar that is not a reference) are the most	520	Simple Perl scalars (any scalar that is not a reference) are the most
268	difficult objects to encode: CBOR::XS will encode undefined scalars as	521	difficult objects to encode: CBOR::XS will encode undefined scalars as
269	CBOR null values, scalars that have last been used in a string context	522	CBOR null values, scalars that have last been used in a string context
270	before encoding as CBOR strings, and anything else as number value:	523	before encoding as CBOR strings, and anything else as number value:
271		524
272	# dump as number	525	# dump as number
273	encode_cbor [2] # yields [2]	526	encode_cbor [2] # yields [2]
274	encode_cbor [-3.0e17] # yields [-3e+17]	527	encode_cbor [-3.0e17] # yields [-3e+17]
275	my $value = 5; encode_cbor [$value] # yields [5]	528	my $value = 5; encode_cbor [$value] # yields [5]
276		529
277	# used as string, so dump as string	530	# used as string, so dump as string (either byte or text)
278	print $value;	531	print $value;
279	encode_cbor [$value] # yields ["5"]	532	encode_cbor [$value] # yields ["5"]
280		533
281	# undef becomes null	534	# undef becomes null
282	encode_cbor [undef] # yields [null]	535	encode_cbor [undef] # yields [null]
…		…
285		538
286	my $x = 3.1; # some variable containing a number	539	my $x = 3.1; # some variable containing a number
287	"$x"; # stringified	540	"$x"; # stringified
288	$x .= ""; # another, more awkward way to stringify	541	$x .= ""; # another, more awkward way to stringify
289	print $x; # perl does it for you, too, quite often	542	print $x; # perl does it for you, too, quite often
		543
		544	You can force whether a string ie encoded as byte or text string by using
		545	C<utf8::upgrade> and C<utf8::downgrade>):
		546
		547	utf8::upgrade $x; # encode $x as text string
		548	utf8::downgrade $x; # encode $x as byte string
		549
		550	Perl doesn't define what operations up- and downgrade strings, so if the
		551	difference between byte and text is important, you should up- or downgrade
		552	your string as late as possible before encoding.
290		553
291	You can force the type to be a CBOR number by numifying it:	554	You can force the type to be a CBOR number by numifying it:
292		555
293	my $x = "3"; # some variable containing a string	556	my $x = "3"; # some variable containing a string
294	$x += 0; # numify it, ensuring it will be dumped as a number	557	$x += 0; # numify it, ensuring it will be dumped as a number
…		…
305	represent numerical values are supported, but might suffer loss of	568	represent numerical values are supported, but might suffer loss of
306	precision.	569	precision.
307		570
308	=back	571	=back
309		572
		573	=head2 OBJECT SERIALISATION
310		574
		575	This module implements both a CBOR-specific and the generic
		576	L<Types::Serialier> object serialisation protocol. The following
		577	subsections explain both methods.
		578
		579	=head3 ENCODING
		580
		581	This module knows two way to serialise a Perl object: The CBOR-specific
		582	way, and the generic way.
		583
		584	Whenever the encoder encounters a Perl object that it cannot serialise
		585	directly (most of them), it will first look up the C<TO_CBOR> method on
		586	it.
		587
		588	If it has a C<TO_CBOR> method, it will call it with the object as only
		589	argument, and expects exactly one return value, which it will then
		590	substitute and encode it in the place of the object.
		591
		592	Otherwise, it will look up the C<FREEZE> method. If it exists, it will
		593	call it with the object as first argument, and the constant string C<CBOR>
		594	as the second argument, to distinguish it from other serialisers.
		595
		596	The C<FREEZE> method can return any number of values (i.e. zero or
		597	more). These will be encoded as CBOR perl object, together with the
		598	classname.
		599
		600	These methods I<MUST NOT> change the data structure that is being
		601	serialised. Failure to comply to this can result in memory corruption -
		602	and worse.
		603
		604	If an object supports neither C<TO_CBOR> nor C<FREEZE>, encoding will fail
		605	with an error.
		606
		607	=head3 DECODING
		608
		609	Objects encoded via C<TO_CBOR> cannot (normally) be automatically decoded,
		610	but objects encoded via C<FREEZE> can be decoded using the following
		611	protocol:
		612
		613	When an encoded CBOR perl object is encountered by the decoder, it will
		614	look up the C<THAW> method, by using the stored classname, and will fail
		615	if the method cannot be found.
		616
		617	After the lookup it will call the C<THAW> method with the stored classname
		618	as first argument, the constant string C<CBOR> as second argument, and all
		619	values returned by C<FREEZE> as remaining arguments.
		620
		621	=head3 EXAMPLES
		622
		623	Here is an example C<TO_CBOR> method:
		624
		625	sub My::Object::TO_CBOR {
		626	my ($obj) = @_;
		627
		628	["this is a serialised My::Object object", $obj->{id}]
		629	}
		630
		631	When a C<My::Object> is encoded to CBOR, it will instead encode a simple
		632	array with two members: a string, and the "object id". Decoding this CBOR
		633	string will yield a normal perl array reference in place of the object.
		634
		635	A more useful and practical example would be a serialisation method for
		636	the URI module. CBOR has a custom tag value for URIs, namely 32:
		637
		638	sub URI::TO_CBOR {
		639	my ($self) = @_;
		640	my $uri = "$self"; # stringify uri
		641	utf8::upgrade $uri; # make sure it will be encoded as UTF-8 string
		642	CBOR::XS::tag 32, "$_[0]"
		643	}
		644
		645	This will encode URIs as a UTF-8 string with tag 32, which indicates an
		646	URI.
		647
		648	Decoding such an URI will not (currently) give you an URI object, but
		649	instead a CBOR::XS::Tagged object with tag number 32 and the string -
		650	exactly what was returned by C<TO_CBOR>.
		651
		652	To serialise an object so it can automatically be deserialised, you need
		653	to use C<FREEZE> and C<THAW>. To take the URI module as example, this
		654	would be a possible implementation:
		655
		656	sub URI::FREEZE {
		657	my ($self, $serialiser) = @_;
		658	"$self" # encode url string
		659	}
		660
		661	sub URI::THAW {
		662	my ($class, $serialiser, $uri) = @_;
		663
		664	$class->new ($uri)
		665	}
		666
		667	Unlike C<TO_CBOR>, multiple values can be returned by C<FREEZE>. For
		668	example, a C<FREEZE> method that returns "type", "id" and "variant" values
		669	would cause an invocation of C<THAW> with 5 arguments:
		670
		671	sub My::Object::FREEZE {
		672	my ($self, $serialiser) = @_;
		673
		674	($self->{type}, $self->{id}, $self->{variant})
		675	}
		676
		677	sub My::Object::THAW {
		678	my ($class, $serialiser, $type, $id, $variant) = @_;
		679
		680	$class-<new (type => $type, id => $id, variant => $variant)
		681	}
		682
		683
311	=head2 MAGIC HEADER	684	=head1 MAGIC HEADER
312		685
313	There is no way to distinguish CBOR from other formats	686	There is no way to distinguish CBOR from other formats
314	programmatically. To make it easier to distinguish CBOR from other	687	programmatically. To make it easier to distinguish CBOR from other
315	formats, the CBOR specification has a special "magic string" that can be	688	formats, the CBOR specification has a special "magic string" that can be
316	prepended to any CBOR string without changing it's meaning.	689	prepended to any CBOR string without changing its meaning.
317		690
318	This string is available as C<$CBOR::XS::MAGIC>. This module does not	691	This string is available as C<$CBOR::XS::MAGIC>. This module does not
319	prepend this string tot he CBOR data it generates, but it will ignroe it	692	prepend this string to the CBOR data it generates, but it will ignore it
320	if present, so users can prepend this string as a "file type" indicator as	693	if present, so users can prepend this string as a "file type" indicator as
321	required.	694	required.
322		695
323		696
		697	=head1 THE CBOR::XS::Tagged CLASS
		698
		699	CBOR has the concept of tagged values - any CBOR value can be tagged with
		700	a numeric 64 bit number, which are centrally administered.
		701
		702	C<CBOR::XS> handles a few tags internally when en- or decoding. You can
		703	also create tags yourself by encoding C<CBOR::XS::Tagged> objects, and the
		704	decoder will create C<CBOR::XS::Tagged> objects itself when it hits an
		705	unknown tag.
		706
		707	These objects are simply blessed array references - the first member of
		708	the array being the numerical tag, the second being the value.
		709
		710	You can interact with C<CBOR::XS::Tagged> objects in the following ways:
		711
		712	=over 4
		713
		714	=item $tagged = CBOR::XS::tag $tag, $value
		715
		716	This function(!) creates a new C<CBOR::XS::Tagged> object using the given
		717	C<$tag> (0..2**64-1) to tag the given C<$value> (which can be any Perl
		718	value that can be encoded in CBOR, including serialisable Perl objects and
		719	C<CBOR::XS::Tagged> objects).
		720
		721	=item $tagged->[0]
		722
		723	=item $tagged->[0] = $new_tag
		724
		725	=item $tag = $tagged->tag
		726
		727	=item $new_tag = $tagged->tag ($new_tag)
		728
		729	Access/mutate the tag.
		730
		731	=item $tagged->[1]
		732
		733	=item $tagged->[1] = $new_value
		734
		735	=item $value = $tagged->value
		736
		737	=item $new_value = $tagged->value ($new_value)
		738
		739	Access/mutate the tagged value.
		740
		741	=back
		742
		743	=cut
		744
		745	sub tag($$) {
		746	bless [@_], CBOR::XS::Tagged::;
		747	}
		748
		749	sub CBOR::XS::Tagged::tag {
		750	$_[0][0] = $_[1] if $#_;
		751	$_[0][0]
		752	}
		753
		754	sub CBOR::XS::Tagged::value {
		755	$_[0][1] = $_[1] if $#_;
		756	$_[0][1]
		757	}
		758
		759	=head2 EXAMPLES
		760
		761	Here are some examples of C<CBOR::XS::Tagged> uses to tag objects.
		762
		763	You can look up CBOR tag value and emanings in the IANA registry at
		764	L<http://www.iana.org/assignments/cbor-tags/cbor-tags.xhtml>.
		765
		766	Prepend a magic header (C<$CBOR::XS::MAGIC>):
		767
		768	my $cbor = encode_cbor CBOR::XS::tag 55799, $value;
		769	# same as:
		770	my $cbor = $CBOR::XS::MAGIC . encode_cbor $value;
		771
		772	Serialise some URIs and a regex in an array:
		773
		774	my $cbor = encode_cbor [
		775	(CBOR::XS::tag 32, "http://www.nethype.de/"),
		776	(CBOR::XS::tag 32, "http://software.schmorp.de/"),
		777	(CBOR::XS::tag 35, "^[Pp][Ee][Rr][lL]\$"),
		778	];
		779
		780	Wrap CBOR data in CBOR:
		781
		782	my $cbor_cbor = encode_cbor
		783	CBOR::XS::tag 24,
		784	encode_cbor [1, 2, 3];
		785
		786	=head1 TAG HANDLING AND EXTENSIONS
		787
		788	This section describes how this module handles specific tagged values
		789	and extensions. If a tag is not mentioned here and no additional filters
		790	are provided for it, then the default handling applies (creating a
		791	CBOR::XS::Tagged object on decoding, and only encoding the tag when
		792	explicitly requested).
		793
		794	Tags not handled specifically are currently converted into a
		795	L<CBOR::XS::Tagged> object, which is simply a blessed array reference
		796	consisting of the numeric tag value followed by the (decoded) CBOR value.
		797
		798	Future versions of this module reserve the right to special case
		799	additional tags (such as base64url).
		800
		801	=head2 ENFORCED TAGS
		802
		803	These tags are always handled when decoding, and their handling cannot be
		804	overriden by the user.
		805
		806	=over 4
		807
		808	=item 26 (perl-object, L<http://cbor.schmorp.de/perl-object>)
		809
		810	These tags are automatically created (and decoded) for serialisable
		811	objects using the C<FREEZE/THAW> methods (the L<Types::Serialier> object
		812	serialisation protocol). See L<OBJECT SERIALISATION> for details.
		813
		814	=item 28, 29 (shareable, sharedref, L <http://cbor.schmorp.de/value-sharing>)
		815
		816	These tags are automatically decoded when encountered (and they do not
		817	result in a cyclic data structure, see C<allow_cycles>), resulting in
		818	shared values in the decoded object. They are only encoded, however, when
		819	C<allow_sharing> is enabled.
		820
		821	Not all shared values can be successfully decoded: values that reference
		822	themselves will I<currently> decode as C<undef> (this is not the same
		823	as a reference pointing to itself, which will be represented as a value
		824	that contains an indirect reference to itself - these will be decoded
		825	properly).
		826
		827	Note that considerably more shared value data structures can be decoded
		828	than will be encoded - currently, only values pointed to by references
		829	will be shared, others will not. While non-reference shared values can be
		830	generated in Perl with some effort, they were considered too unimportant
		831	to be supported in the encoder. The decoder, however, will decode these
		832	values as shared values.
		833
		834	=item 256, 25 (stringref-namespace, stringref, L <http://cbor.schmorp.de/stringref>)
		835
		836	These tags are automatically decoded when encountered. They are only
		837	encoded, however, when C<pack_strings> is enabled.
		838
		839	=item 22098 (indirection, L<http://cbor.schmorp.de/indirection>)
		840
		841	This tag is automatically generated when a reference are encountered (with
		842	the exception of hash and array refernces). It is converted to a reference
		843	when decoding.
		844
		845	=item 55799 (self-describe CBOR, RFC 7049)
		846
		847	This value is not generated on encoding (unless explicitly requested by
		848	the user), and is simply ignored when decoding.
		849
		850	=back
		851
		852	=head2 NON-ENFORCED TAGS
		853
		854	These tags have default filters provided when decoding. Their handling can
		855	be overriden by changing the C<%CBOR::XS::FILTER> entry for the tag, or by
		856	providing a custom C<filter> callback when decoding.
		857
		858	When they result in decoding into a specific Perl class, the module
		859	usually provides a corresponding C<TO_CBOR> method as well.
		860
		861	When any of these need to load additional modules that are not part of the
		862	perl core distribution (e.g. L<URI>), it is (currently) up to the user to
		863	provide these modules. The decoding usually fails with an exception if the
		864	required module cannot be loaded.
		865
		866	=over 4
		867
		868	=item 0, 1 (date/time string, seconds since the epoch)
		869
		870	These tags are decoded into L<Time::Piece> objects. The corresponding
		871	C<Time::Piece::TO_CBOR> method always encodes into tag 1 values currently.
		872
		873	The L<Time::Piece> API is generally surprisingly bad, and fractional
		874	seconds are only accidentally kept intact, so watch out. On the plus side,
		875	the module comes with perl since 5.10, which has to count for something.
		876
		877	=item 2, 3 (positive/negative bignum)
		878
		879	These tags are decoded into L<Math::BigInt> objects. The corresponding
		880	C<Math::BigInt::TO_CBOR> method encodes "small" bigints into normal CBOR
		881	integers, and others into positive/negative CBOR bignums.
		882
		883	=item 4, 5 (decimal fraction/bigfloat)
		884
		885	Both decimal fractions and bigfloats are decoded into L<Math::BigFloat>
		886	objects. The corresponding C<Math::BigFloat::TO_CBOR> method I<always>
		887	encodes into a decimal fraction.
		888
		889	CBOR cannot represent bigfloats with I<very> large exponents - conversion
		890	of such big float objects is undefined.
		891
		892	Also, NaN and infinities are not encoded properly.
		893
		894	=item 21, 22, 23 (expected later JSON conversion)
		895
		896	CBOR::XS is not a CBOR-to-JSON converter, and will simply ignore these
		897	tags.
		898
		899	=item 32 (URI)
		900
		901	These objects decode into L<URI> objects. The corresponding
		902	C<URI::TO_CBOR> method again results in a CBOR URI value.
		903
		904	=back
		905
		906	=cut
		907
		908	our %FILTER = (
		909	# 0 # rfc4287 datetime, utf-8
		910	# 1 # unix timestamp, any
		911
		912	2 => sub { # pos bigint
		913	require Math::BigInt;
		914	Math::BigInt->new ("0x" . unpack "H*", pop)
		915	},
		916
		917	3 => sub { # neg bigint
		918	require Math::BigInt;
		919	-Math::BigInt->new ("0x" . unpack "H*", pop)
		920	},
		921
		922	4 => sub { # decimal fraction, array
		923	require Math::BigFloat;
		924	Math::BigFloat->new ($_[1][1] . "E" . $_[1][0])
		925	},
		926
		927	5 => sub { # bigfloat, array
		928	require Math::BigFloat;
		929	scalar Math::BigFloat->new ($_[1][1])->blsft ($_[1][0], 2)
		930	},
		931
		932	21 => sub { pop }, # expected conversion to base64url encoding
		933	22 => sub { pop }, # expected conversion to base64 encoding
		934	23 => sub { pop }, # expected conversion to base16 encoding
		935
		936	# 24 # embedded cbor, byte string
		937
		938	32 => sub {
		939	require URI;
		940	URI->new (pop)
		941	},
		942
		943	# 33 # base64url rfc4648, utf-8
		944	# 34 # base64 rfc46484, utf-8
		945	# 35 # regex pcre/ecma262, utf-8
		946	# 36 # mime message rfc2045, utf-8
		947	);
		948
		949
324	=head2 CBOR and JSON	950	=head1 CBOR and JSON
325		951
326	CBOR is supposed to implement a superset of the JSON data model, and is,	952	CBOR is supposed to implement a superset of the JSON data model, and is,
327	with some coercion, able to represent all JSON texts (something that other	953	with some coercion, able to represent all JSON texts (something that other
328	"binary JSON" formats such as BSON generally do not support).	954	"binary JSON" formats such as BSON generally do not support).
329		955
…		…
388	properly. Half precision types are accepted, but not encoded.	1014	properly. Half precision types are accepted, but not encoded.
389		1015
390	Strict mode and canonical mode are not implemented.	1016	Strict mode and canonical mode are not implemented.
391		1017
392		1018
		1019	=head1 LIMITATIONS ON PERLS WITHOUT 64-BIT INTEGER SUPPORT
		1020
		1021	On perls that were built without 64 bit integer support (these are rare
		1022	nowadays, even on 32 bit architectures), support for any kind of 64 bit
		1023	integer in CBOR is very limited - most likely, these 64 bit values will
		1024	be truncated, corrupted, or otherwise not decoded correctly. This also
		1025	includes string, array and map sizes that are stored as 64 bit integers.
		1026
		1027
393	=head1 THREADS	1028	=head1 THREADS
394		1029
395	This module is I<not> guaranteed to be thread safe and there are no	1030	This module is I<not> guaranteed to be thread safe and there are no
396	plans to change this until Perl gets thread support (as opposed to the	1031	plans to change this until Perl gets thread support (as opposed to the
397	horribly slow so-called "threads" which are simply slow and bloated	1032	horribly slow so-called "threads" which are simply slow and bloated
…		…
409	Please refrain from using rt.cpan.org or any other bug reporting	1044	Please refrain from using rt.cpan.org or any other bug reporting
410	service. I put the contact address into my modules for a reason.	1045	service. I put the contact address into my modules for a reason.
411		1046
412	=cut	1047	=cut
413		1048
414	our $true = do { bless \(my $dummy = 1), "CBOR::XS::Boolean" };	1049	our %FILTER = (
415	our $false = do { bless \(my $dummy = 0), "CBOR::XS::Boolean" };	1050	0 => sub { # rfc4287 datetime, utf-8
		1051	require Time::Piece;
		1052	# Time::Piece::Strptime uses the "incredibly flexible date parsing routine"
		1053	# from FreeBSD, which can't parse ISO 8601, RFC3339, RFC4287 or much of anything
		1054	# else either. Whats incredibe over standard strptime totally escapes me.
		1055	# doesn't do fractional times, either. sigh.
		1056	# In fact, it's all a lie, it uses whatever strptime it wants, and of course,
		1057	# they are all incomptible. The openbsd one simply ignores %z (but according to the
		1058	# docs, it would be much more incredibly flexible indeed. If it worked, that is.).
		1059	scalar eval {
		1060	my $s = $_[1];
416		1061
417	sub true() { $true }	1062	$s =~ s/Z$/+00:00/;
418	sub false() { $false }	1063	$s =~ s/(\.[0-9]+)?([+-][0-9][0-9]):([0-9][0-9])$//
		1064	or die;
419		1065
420	sub is_bool($) {	1066	my $b = $1 - ($2 * 60 + $3) * 60; # fractional part + offset. hopefully
421	UNIVERSAL::isa $_[0], "CBOR::XS::Boolean"	1067	my $d = Time::Piece->strptime ($s, "%Y-%m-%dT%H:%M:%S");
422	# or UNIVERSAL::isa $_[0], "CBOR::Literal"	1068
		1069	Time::Piece::gmtime ($d->epoch + $b)
		1070	} \|\| die "corrupted CBOR date/time string ($_[0])";
		1071	},
		1072
		1073	1 => sub { # seconds since the epoch, possibly fractional
		1074	require Time::Piece;
		1075	scalar Time::Piece::gmtime (pop)
		1076	},
		1077
		1078	2 => sub { # pos bigint
		1079	require Math::BigInt;
		1080	Math::BigInt->new ("0x" . unpack "H*", pop)
		1081	},
		1082
		1083	3 => sub { # neg bigint
		1084	require Math::BigInt;
		1085	-Math::BigInt->new ("0x" . unpack "H*", pop)
		1086	},
		1087
		1088	4 => sub { # decimal fraction, array
		1089	require Math::BigFloat;
		1090	Math::BigFloat->new ($_[1][1] . "E" . $_[1][0])
		1091	},
		1092
		1093	5 => sub { # bigfloat, array
		1094	require Math::BigFloat;
		1095	scalar Math::BigFloat->new ($_[1][1])->blsft ($_[1][0], 2)
		1096	},
		1097
		1098	21 => sub { pop }, # expected conversion to base64url encoding
		1099	22 => sub { pop }, # expected conversion to base64 encoding
		1100	23 => sub { pop }, # expected conversion to base16 encoding
		1101
		1102	# 24 # embedded cbor, byte string
		1103
		1104	32 => sub {
		1105	require URI;
		1106	URI->new (pop)
		1107	},
		1108
		1109	# 33 # base64url rfc4648, utf-8
		1110	# 34 # base64 rfc46484, utf-8
		1111	# 35 # regex pcre/ecma262, utf-8
		1112	# 36 # mime message rfc2045, utf-8
		1113	);
		1114
		1115	sub CBOR::XS::default_filter {
		1116	&{ $FILTER{$_[0]} or return }
423	}	1117	}
424		1118
		1119	sub URI::TO_CBOR {
		1120	my $uri = $_[0]->as_string;
		1121	utf8::upgrade $uri;
		1122	tag 32, $uri
		1123	}
		1124
		1125	sub Math::BigInt::TO_CBOR {
		1126	if ($_[0] >= -2147483648 && $_[0] <= 2147483647) {
		1127	$_[0]->numify
		1128	} else {
		1129	my $hex = substr $_[0]->as_hex, 2;
		1130	$hex = "0$hex" if 1 & length $hex; # sigh
		1131	tag $_[0] >= 0 ? 2 : 3, pack "H*", $hex
		1132	}
		1133	}
		1134
		1135	sub Math::BigFloat::TO_CBOR {
		1136	my ($m, $e) = $_[0]->parts;
		1137	tag 4, [$e->numify, $m]
		1138	}
		1139
		1140	sub Time::Piece::TO_CBOR {
		1141	tag 1, 0 + $_[0]->epoch
		1142	}
		1143
425	XSLoader::load "CBOR::XS", $VERSION;	1144	XSLoader::load "CBOR::XS", $VERSION;
426
427	package CBOR::XS::Boolean;
428
429	use overload
430	"0+" => sub { ${$_[0]} },
431	"++" => sub { $_[0] = ${$_[0]} + 1 },
432	"--" => sub { $_[0] = ${$_[0]} - 1 },
433	fallback => 1;
434
435	1;
436		1145
437	=head1 SEE ALSO	1146	=head1 SEE ALSO
438		1147
439	The L<JSON> and L<JSON::XS> modules that do similar, but human-readable,	1148	The L<JSON> and L<JSON::XS> modules that do similar, but human-readable,
440	serialisation.	1149	serialisation.
441		1150
		1151	The L<Types::Serialiser> module provides the data model for true, false
		1152	and error values.
		1153
442	=head1 AUTHOR	1154	=head1 AUTHOR
443		1155
444	Marc Lehmann <schmorp@schmorp.de>	1156	Marc Lehmann <schmorp@schmorp.de>
445	http://home.schmorp.de/	1157	http://home.schmorp.de/
446		1158
447	=cut	1159	=cut
448		1160
		1161	1
		1162

Diff Legend

-–
+Removed lines
-+
+Added lines
-<
+Changed lines
->
+Changed lines

Comparing CBOR-XS/XS.pm (file contents): Revision 1.4 by root, Sat Oct 26 22:25:47 2013 UTC vs. Revision 1.40 by root, Sun Jan 5 14:24:54 2014 UTC

Diff Legend

Comparing CBOR-XS/XS.pm (file contents):
Revision 1.4 by root, Sat Oct 26 22:25:47 2013 UTC vs.
Revision 1.40 by root, Sun Jan 5 14:24:54 2014 UTC