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

Comparing CBOR-XS/XS.pm (file contents):
Revision 1.5 by root, Sat Oct 26 23:02:55 2013 UTC vs.
Revision 1.52 by root, Mon Apr 25 18:17:17 2016 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
21	WARNING! THIS IS A PRE-ALPHA RELEASE! IT WILL CRASH, CORRUPT YOUR DATA
22	AND EAT YOUR CHILDREN! (Actually, apart from being untested and a bit
23	feature-limited, it might already be useful).
24		30
25	This module converts Perl data structures to the Concise Binary Object	31	This module converts Perl data structures to the Concise Binary Object
26	Representation (CBOR) and vice versa. CBOR is a fast binary serialisation	32	Representation (CBOR) and vice versa. CBOR is a fast binary serialisation
27	format that aims to use a superset of the JSON data model, i.e. when you	33	format that aims to use an (almost) superset of the JSON data model, i.e.
28	can represent something in JSON, you should be able to represent it in	34	when you can represent something useful in JSON, you should be able to
29	CBOR.	35	represent it in CBOR.
30		36
31	This makes it a faster and more compact binary alternative to JSON.	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).
32		56
33	The primary goal of this module is to be I<correct> and the secondary goal	57	The primary goal of this module is to be I<correct> and the secondary goal
34	is to be 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.
35		59
36	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
…		…
40		64
41	package CBOR::XS;	65	package CBOR::XS;
42		66
43	use common::sense;	67	use common::sense;
44		68
45	our $VERSION = 0.03;	69	our $VERSION = 1.41;
46	our @ISA = qw(Exporter);	70	our @ISA = qw(Exporter);
47		71
48	our @EXPORT = qw(encode_cbor decode_cbor);	72	our @EXPORT = qw(encode_cbor decode_cbor);
49		73
50	use Exporter;	74	use Exporter;
51	use XSLoader;	75	use XSLoader;
52		76
		77	use Types::Serialiser;
		78
53	our $MAGIC = "\xd9\xd9\xf7";	79	our $MAGIC = "\xd9\xd9\xf7";
54		80
55	=head1 FUNCTIONAL INTERFACE	81	=head1 FUNCTIONAL INTERFACE
56		82
57	The following convenience methods are provided by this module. They are	83	The following convenience methods are provided by this module. They are
…		…
85	strings. All boolean flags described below are by default I<disabled>.	111	strings. All boolean flags described below are by default I<disabled>.
86		112
87	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
88	be chained:	114	be chained:
89		115
90	#TODO
91	my $cbor = CBOR::XS->new->encode ({a => [1,2]});	116	my $cbor = CBOR::XS->new->encode ({a => [1,2]});
92		117
93	=item $cbor = $cbor->max_depth ([$maximum_nesting_depth])	118	=item $cbor = $cbor->max_depth ([$maximum_nesting_depth])
94		119
95	=item $max_depth = $cbor->get_max_depth	120	=item $max_depth = $cbor->get_max_depth
…		…
129	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
130	C<0> is specified).	155	C<0> is specified).
131		156
132	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.
133		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	FUTURE DIRECTION: the motivation behind this option is to avoid I<real>
		224	cycles - future versions of this module might chose to decode cyclic data
		225	structures using weak references when this option is off, instead of
		226	throwing an error.
		227
		228	This option does not affect C<encode> in any way - shared values and
		229	references will always be encoded properly if present.
		230
		231	=item $cbor = $cbor->pack_strings ([$enable])
		232
		233	=item $enabled = $cbor->get_pack_strings
		234
		235	If C<$enable> is true (or missing), then C<encode> will try not to encode
		236	the same string twice, but will instead encode a reference to the string
		237	instead. Depending on your data format, this can save a lot of space, but
		238	also results in a very large runtime overhead (expect encoding times to be
		239	2-4 times as high as without).
		240
		241	It is recommended to leave it off unless you know your
		242	communications partner supports the stringref extension to CBOR
		243	(L<http://cbor.schmorp.de/stringref>), as without decoder support, the
		244	resulting data structure might not be usable.
		245
		246	If C<$enable> is false (the default), then C<encode> will encode strings
		247	the standard CBOR way.
		248
		249	This option does not affect C<decode> in any way - string references will
		250	always be decoded properly if present.
		251
		252	=item $cbor = $cbor->text_keys ([$enable])
		253
		254	=item $enabled = $cbor->get_text_keys
		255
		256	If C<$enabled> is true (or missing), then C<encode> will encode all
		257	perl hash keys as CBOR text strings/UTF-8 string, upgrading them as needed.
		258
		259	If C<$enable> is false (the default), then C<encode> will encode hash keys
		260	normally - upgraded perl strings (strings internally encoded as UTF-8) as
		261	CBOR text strings, and downgraded perl strings as CBOR byte strings.
		262
		263	This option does not affect C<decode> in any way.
		264
		265	This option is useful for interoperability with CBOR decoders that don't
		266	treat byte strings as a form of text. It is especially useful as Perl
		267	gives very little control over hash keys.
		268
		269	Enabling this option can be slow, as all downgraded hash keys that are
		270	encoded need to be scanned and converted to UTF-8.
		271
		272	=item $cbor = $cbor->text_strings ([$enable])
		273
		274	=item $enabled = $cbor->get_text_strings
		275
		276	This option works similar to C<text_keys>, above, but works on all strings
		277	(including hash keys), so C<text_keys> has no further effect after
		278	enabling C<text_strings>.
		279
		280	If C<$enabled> is true (or missing), then C<encode> will encode all perl
		281	strings as CBOR text strings/UTF-8 strings, upgrading them as needed.
		282
		283	If C<$enable> is false (the default), then C<encode> will encode strings
		284	normally (but see C<text_keys>) - upgraded perl strings (strings
		285	internally encoded as UTF-8) as CBOR text strings, and downgraded perl
		286	strings as CBOR byte strings.
		287
		288	This option does not affect C<decode> in any way.
		289
		290	This option has similar advantages and disadvantages as C<text_keys>. In
		291	addition, this option effectively removes the ability to encode byte
		292	strings, which might break some C<FREEZE> and C<TO_CBOR> methods that rely
		293	on this, such as bignum encoding, so this option is mainly useful for very
		294	simple data.
		295
		296	=item $cbor = $cbor->validate_utf8 ([$enable])
		297
		298	=item $enabled = $cbor->get_validate_utf8
		299
		300	If C<$enable> is true (or missing), then C<decode> will validate that
		301	elements (text strings) containing UTF-8 data in fact contain valid UTF-8
		302	data (instead of blindly accepting it). This validation obviously takes
		303	extra time during decoding.
		304
		305	The concept of "valid UTF-8" used is perl's concept, which is a superset
		306	of the official UTF-8.
		307
		308	If C<$enable> is false (the default), then C<decode> will blindly accept
		309	UTF-8 data, marking them as valid UTF-8 in the resulting data structure
		310	regardless of whether that's true or not.
		311
		312	Perl isn't too happy about corrupted UTF-8 in strings, but should
		313	generally not crash or do similarly evil things. Extensions might be not
		314	so forgiving, so it's recommended to turn on this setting if you receive
		315	untrusted CBOR.
		316
		317	This option does not affect C<encode> in any way - strings that are
		318	supposedly valid UTF-8 will simply be dumped into the resulting CBOR
		319	string without checking whether that is, in fact, true or not.
		320
		321	=item $cbor = $cbor->filter ([$cb->($tag, $value)])
		322
		323	=item $cb_or_undef = $cbor->get_filter
		324
		325	Sets or replaces the tagged value decoding filter (when C<$cb> is
		326	specified) or clears the filter (if no argument or C<undef> is provided).
		327
		328	The filter callback is called only during decoding, when a non-enforced
		329	tagged value has been decoded (see L<TAG HANDLING AND EXTENSIONS> for a
		330	list of enforced tags). For specific tags, it's often better to provide a
		331	default converter using the C<%CBOR::XS::FILTER> hash (see below).
		332
		333	The first argument is the numerical tag, the second is the (decoded) value
		334	that has been tagged.
		335
		336	The filter function should return either exactly one value, which will
		337	replace the tagged value in the decoded data structure, or no values,
		338	which will result in default handling, which currently means the decoder
		339	creates a C<CBOR::XS::Tagged> object to hold the tag and the value.
		340
		341	When the filter is cleared (the default state), the default filter
		342	function, C<CBOR::XS::default_filter>, is used. This function simply looks
		343	up the tag in the C<%CBOR::XS::FILTER> hash. If an entry exists it must be
		344	a code reference that is called with tag and value, and is responsible for
		345	decoding the value. If no entry exists, it returns no values.
		346
		347	Example: decode all tags not handled internally into C<CBOR::XS::Tagged>
		348	objects, with no other special handling (useful when working with
		349	potentially "unsafe" CBOR data).
		350
		351	CBOR::XS->new->filter (sub { })->decode ($cbor_data);
		352
		353	Example: provide a global filter for tag 1347375694, converting the value
		354	into some string form.
		355
		356	$CBOR::XS::FILTER{1347375694} = sub {
		357	my ($tag, $value);
		358
		359	"tag 1347375694 value $value"
		360	};
		361
134	=item $cbor_data = $cbor->encode ($perl_scalar)	362	=item $cbor_data = $cbor->encode ($perl_scalar)
135		363
136	Converts the given Perl data structure (a scalar value) to its CBOR	364	Converts the given Perl data structure (a scalar value) to its CBOR
137	representation.	365	representation.
138		366
…		…
151	and you need to know where the first CBOR string ends amd the next one	379	and you need to know where the first CBOR string ends amd the next one
152	starts.	380	starts.
153		381
154	CBOR::XS->new->decode_prefix ("......")	382	CBOR::XS->new->decode_prefix ("......")
155	=> ("...", 3)	383	=> ("...", 3)
		384
		385	=back
		386
		387	=head2 INCREMENTAL PARSING
		388
		389	In some cases, there is the need for incremental parsing of JSON
		390	texts. While this module always has to keep both CBOR text and resulting
		391	Perl data structure in memory at one time, it does allow you to parse a
		392	CBOR stream incrementally, using a similar to using "decode_prefix" to see
		393	if a full CBOR object is available, but is much more efficient.
		394
		395	It basically works by parsing as much of a CBOR string as possible - if
		396	the CBOR data is not complete yet, the pasrer will remember where it was,
		397	to be able to restart when more data has been accumulated. Once enough
		398	data is available to either decode a complete CBOR value or raise an
		399	error, a real decode will be attempted.
		400
		401	A typical use case would be a network protocol that consists of sending
		402	and receiving CBOR-encoded messages. The solution that works with CBOR and
		403	about anything else is by prepending a length to every CBOR value, so the
		404	receiver knows how many octets to read. More compact (and slightly slower)
		405	would be to just send CBOR values back-to-back, as C<CBOR::XS> knows where
		406	a CBOR value ends, and doesn't need an explicit length.
		407
		408	The following methods help with this:
		409
		410	=over 4
		411
		412	=item @decoded = $cbor->incr_parse ($buffer)
		413
		414	This method attempts to decode exactly one CBOR value from the beginning
		415	of the given C<$buffer>. The value is removed from the C<$buffer> on
		416	success. When C<$buffer> doesn't contain a complete value yet, it returns
		417	nothing. Finally, when the C<$buffer> doesn't start with something
		418	that could ever be a valid CBOR value, it raises an exception, just as
		419	C<decode> would. In the latter case the decoder state is undefined and
		420	must be reset before being able to parse further.
		421
		422	This method modifies the C<$buffer> in place. When no CBOR value can be
		423	decoded, the decoder stores the current string offset. On the next call,
		424	continues decoding at the place where it stopped before. For this to make
		425	sense, the C<$buffer> must begin with the same octets as on previous
		426	unsuccessful calls.
		427
		428	You can call this method in scalar context, in which case it either
		429	returns a decoded value or C<undef>. This makes it impossible to
		430	distinguish between CBOR null values (which decode to C<undef>) and an
		431	unsuccessful decode, which is often acceptable.
		432
		433	=item @decoded = $cbor->incr_parse_multiple ($buffer)
		434
		435	Same as C<incr_parse>, but attempts to decode as many CBOR values as
		436	possible in one go, instead of at most one. Calls to C<incr_parse> and
		437	C<incr_parse_multiple> can be interleaved.
		438
		439	=item $cbor->incr_reset
		440
		441	Resets the incremental decoder. This throws away any saved state, so that
		442	subsequent calls to C<incr_parse> or C<incr_parse_multiple> start to parse
		443	a new CBOR value from the beginning of the C<$buffer> again.
		444
		445	This method can be caled at any time, but it I<must> be called if you want
		446	to change your C<$buffer> or there was a decoding error and you want to
		447	reuse the C<$cbor> object for future incremental parsings.
156		448
157	=back	449	=back
158		450
159		451
160	=head1 MAPPING	452	=head1 MAPPING
…		…
178	CBOR integers become (numeric) perl scalars. On perls without 64 bit	470	CBOR integers become (numeric) perl scalars. On perls without 64 bit
179	support, 64 bit integers will be truncated or otherwise corrupted.	471	support, 64 bit integers will be truncated or otherwise corrupted.
180		472
181	=item byte strings	473	=item byte strings
182		474
183	Byte strings will become octet strings in Perl (the byte values 0..255	475	Byte strings will become octet strings in Perl (the Byte values 0..255
184	will simply become characters of the same value in Perl).	476	will simply become characters of the same value in Perl).
185		477
186	=item UTF-8 strings	478	=item UTF-8 strings
187		479
188	UTF-8 strings in CBOR will be decoded, i.e. the UTF-8 octets will be	480	UTF-8 strings in CBOR will be decoded, i.e. the UTF-8 octets will be
…		…
194		486
195	CBOR arrays and CBOR maps will be converted into references to a Perl	487	CBOR arrays and CBOR maps will be converted into references to a Perl
196	array or hash, respectively. The keys of the map will be stringified	488	array or hash, respectively. The keys of the map will be stringified
197	during this process.	489	during this process.
198		490
		491	=item null
		492
		493	CBOR null becomes C<undef> in Perl.
		494
199	=item true, false	495	=item true, false, undefined
200		496
201	These CBOR values become C<CBOR::XS::true> and C<CBOR::XS::false>,	497	These CBOR values become C<Types:Serialiser::true>,
		498	C<Types:Serialiser::false> and C<Types::Serialiser::error>,
202	respectively. They are overloaded to act almost exactly like the numbers	499	respectively. They are overloaded to act almost exactly like the numbers
203	C<1> and C<0>. You can check whether a scalar is a CBOR boolean by using	500	C<1> and C<0> (for true and false) or to throw an exception on access (for
204	the C<CBOR::XS::is_bool> function.	501	error). See the L<Types::Serialiser> manpage for details.
205		502
206	=item null, undefined	503	=item tagged values
207		504
208	CBOR null and undefined values becomes C<undef> in Perl (in the future,
209	Undefined may raise an exception or something else).
210
211	=item tags
212
213	Tagged items consists of a numeric tag and another CBOR value. The tag	505	Tagged items consists of a numeric tag and another CBOR value.
214	55799 is ignored (this tag implements the magic header).
215		506
216	All other tags are currently converted into a L<CBOR::XS::Tagged> object,	507	See L<TAG HANDLING AND EXTENSIONS> and the description of C<< ->filter >>
217	which is simply a blessed array reference consistsing of the numeric tag	508	for details on which tags are handled how.
218	value followed by the (decoded) BOR value.
219		509
220	=item anything else	510	=item anything else
221		511
222	Anything else (e.g. unsupported simple values) will raise a decoding	512	Anything else (e.g. unsupported simple values) will raise a decoding
223	error.	513	error.
…		…
226		516
227		517
228	=head2 PERL -> CBOR	518	=head2 PERL -> CBOR
229		519
230	The mapping from Perl to CBOR is slightly more difficult, as Perl is a	520	The mapping from Perl to CBOR is slightly more difficult, as Perl is a
231	truly typeless language, so we can only guess which CBOR type is meant by	521	typeless language. That means this module can only guess which CBOR type
232	a Perl value.	522	is meant by a perl value.
233		523
234	=over 4	524	=over 4
235		525
236	=item hash references	526	=item hash references
237		527
238	Perl hash references become CBOR maps. As there is no inherent ordering in	528	Perl hash references become CBOR maps. As there is no inherent ordering in
239	hash keys (or CBOR maps), they will usually be encoded in a pseudo-random	529	hash keys (or CBOR maps), they will usually be encoded in a pseudo-random
240	order.	530	order. This order can be different each time a hash is encoded.
241		531
242	Currently, tied hashes will use the indefinite-length format, while normal	532	Currently, tied hashes will use the indefinite-length format, while normal
243	hashes will use the fixed-length format.	533	hashes will use the fixed-length format.
244		534
245	=item array references	535	=item array references
246		536
247	Perl array references become fixed-length CBOR arrays.	537	Perl array references become fixed-length CBOR arrays.
248		538
249	=item other references	539	=item other references
250		540
251	Other unblessed references are generally not allowed and will cause an	541	Other unblessed references will be represented using
252	exception to be thrown, except for references to the integers C<0> and	542	the indirection tag extension (tag value C<22098>,
253	C<1>, which get turned into false and true in CBOR.	543	L<http://cbor.schmorp.de/indirection>). CBOR decoders are guaranteed
		544	to be able to decode these values somehow, by either "doing the right
		545	thing", decoding into a generic tagged object, simply ignoring the tag, or
		546	something else.
254		547
255	=item CBOR::XS::Tagged objects	548	=item CBOR::XS::Tagged objects
256		549
257	Objects of this type must be arrays consisting of a single C<[tag, value]>	550	Objects of this type must be arrays consisting of a single C<[tag, value]>
258	pair. The (numerical) tag will be encoded as a CBOR tag, the value will be	551	pair. The (numerical) tag will be encoded as a CBOR tag, the value will
259	encoded as appropriate for the value.	552	be encoded as appropriate for the value. You must use C<CBOR::XS::tag> to
		553	create such objects.
260		554
261	=item CBOR::XS::true, CBOR::XS::false	555	=item Types::Serialiser::true, Types::Serialiser::false, Types::Serialiser::error
262		556
263	These special values become CBOR true and CBOR false values,	557	These special values become CBOR true, CBOR false and CBOR undefined
264	respectively. You can also use C<\1> and C<\0> directly if you want.	558	values, respectively. You can also use C<\1>, C<\0> and C<\undef> directly
		559	if you want.
265		560
266	=item blessed objects	561	=item other blessed objects
267		562
268	Other blessed objects currently need to have a C<TO_CBOR> method. It	563	Other blessed objects are serialised via C<TO_CBOR> or C<FREEZE>. See
269	will be called on every object that is being serialised, and must return	564	L<TAG HANDLING AND EXTENSIONS> for specific classes handled by this
270	something that can be encoded in CBOR.	565	module, and L<OBJECT SERIALISATION> for generic object serialisation.
271		566
272	=item simple scalars	567	=item simple scalars
273		568
274	TODO
275	Simple Perl scalars (any scalar that is not a reference) are the most	569	Simple Perl scalars (any scalar that is not a reference) are the most
276	difficult objects to encode: CBOR::XS will encode undefined scalars as	570	difficult objects to encode: CBOR::XS will encode undefined scalars as
277	CBOR null values, scalars that have last been used in a string context	571	CBOR null values, scalars that have last been used in a string context
278	before encoding as CBOR strings, and anything else as number value:	572	before encoding as CBOR strings, and anything else as number value:
279		573
280	# dump as number	574	# dump as number
281	encode_cbor [2] # yields [2]	575	encode_cbor [2] # yields [2]
282	encode_cbor [-3.0e17] # yields [-3e+17]	576	encode_cbor [-3.0e17] # yields [-3e+17]
283	my $value = 5; encode_cbor [$value] # yields [5]	577	my $value = 5; encode_cbor [$value] # yields [5]
284		578
285	# used as string, so dump as string	579	# used as string, so dump as string (either byte or text)
286	print $value;	580	print $value;
287	encode_cbor [$value] # yields ["5"]	581	encode_cbor [$value] # yields ["5"]
288		582
289	# undef becomes null	583	# undef becomes null
290	encode_cbor [undef] # yields [null]	584	encode_cbor [undef] # yields [null]
…		…
293		587
294	my $x = 3.1; # some variable containing a number	588	my $x = 3.1; # some variable containing a number
295	"$x"; # stringified	589	"$x"; # stringified
296	$x .= ""; # another, more awkward way to stringify	590	$x .= ""; # another, more awkward way to stringify
297	print $x; # perl does it for you, too, quite often	591	print $x; # perl does it for you, too, quite often
		592
		593	You can force whether a string ie encoded as byte or text string by using
		594	C<utf8::upgrade> and C<utf8::downgrade>):
		595
		596	utf8::upgrade $x; # encode $x as text string
		597	utf8::downgrade $x; # encode $x as byte string
		598
		599	Perl doesn't define what operations up- and downgrade strings, so if the
		600	difference between byte and text is important, you should up- or downgrade
		601	your string as late as possible before encoding.
298		602
299	You can force the type to be a CBOR number by numifying it:	603	You can force the type to be a CBOR number by numifying it:
300		604
301	my $x = "3"; # some variable containing a string	605	my $x = "3"; # some variable containing a string
302	$x += 0; # numify it, ensuring it will be dumped as a number	606	$x += 0; # numify it, ensuring it will be dumped as a number
…		…
313	represent numerical values are supported, but might suffer loss of	617	represent numerical values are supported, but might suffer loss of
314	precision.	618	precision.
315		619
316	=back	620	=back
317		621
		622	=head2 OBJECT SERIALISATION
318		623
		624	This module implements both a CBOR-specific and the generic
		625	L<Types::Serialier> object serialisation protocol. The following
		626	subsections explain both methods.
		627
		628	=head3 ENCODING
		629
		630	This module knows two way to serialise a Perl object: The CBOR-specific
		631	way, and the generic way.
		632
		633	Whenever the encoder encounters a Perl object that it cannot serialise
		634	directly (most of them), it will first look up the C<TO_CBOR> method on
		635	it.
		636
		637	If it has a C<TO_CBOR> method, it will call it with the object as only
		638	argument, and expects exactly one return value, which it will then
		639	substitute and encode it in the place of the object.
		640
		641	Otherwise, it will look up the C<FREEZE> method. If it exists, it will
		642	call it with the object as first argument, and the constant string C<CBOR>
		643	as the second argument, to distinguish it from other serialisers.
		644
		645	The C<FREEZE> method can return any number of values (i.e. zero or
		646	more). These will be encoded as CBOR perl object, together with the
		647	classname.
		648
		649	These methods I<MUST NOT> change the data structure that is being
		650	serialised. Failure to comply to this can result in memory corruption -
		651	and worse.
		652
		653	If an object supports neither C<TO_CBOR> nor C<FREEZE>, encoding will fail
		654	with an error.
		655
		656	=head3 DECODING
		657
		658	Objects encoded via C<TO_CBOR> cannot (normally) be automatically decoded,
		659	but objects encoded via C<FREEZE> can be decoded using the following
		660	protocol:
		661
		662	When an encoded CBOR perl object is encountered by the decoder, it will
		663	look up the C<THAW> method, by using the stored classname, and will fail
		664	if the method cannot be found.
		665
		666	After the lookup it will call the C<THAW> method with the stored classname
		667	as first argument, the constant string C<CBOR> as second argument, and all
		668	values returned by C<FREEZE> as remaining arguments.
		669
		670	=head3 EXAMPLES
		671
		672	Here is an example C<TO_CBOR> method:
		673
		674	sub My::Object::TO_CBOR {
		675	my ($obj) = @_;
		676
		677	["this is a serialised My::Object object", $obj->{id}]
		678	}
		679
		680	When a C<My::Object> is encoded to CBOR, it will instead encode a simple
		681	array with two members: a string, and the "object id". Decoding this CBOR
		682	string will yield a normal perl array reference in place of the object.
		683
		684	A more useful and practical example would be a serialisation method for
		685	the URI module. CBOR has a custom tag value for URIs, namely 32:
		686
		687	sub URI::TO_CBOR {
		688	my ($self) = @_;
		689	my $uri = "$self"; # stringify uri
		690	utf8::upgrade $uri; # make sure it will be encoded as UTF-8 string
		691	CBOR::XS::tag 32, "$_[0]"
		692	}
		693
		694	This will encode URIs as a UTF-8 string with tag 32, which indicates an
		695	URI.
		696
		697	Decoding such an URI will not (currently) give you an URI object, but
		698	instead a CBOR::XS::Tagged object with tag number 32 and the string -
		699	exactly what was returned by C<TO_CBOR>.
		700
		701	To serialise an object so it can automatically be deserialised, you need
		702	to use C<FREEZE> and C<THAW>. To take the URI module as example, this
		703	would be a possible implementation:
		704
		705	sub URI::FREEZE {
		706	my ($self, $serialiser) = @_;
		707	"$self" # encode url string
		708	}
		709
		710	sub URI::THAW {
		711	my ($class, $serialiser, $uri) = @_;
		712	$class->new ($uri)
		713	}
		714
		715	Unlike C<TO_CBOR>, multiple values can be returned by C<FREEZE>. For
		716	example, a C<FREEZE> method that returns "type", "id" and "variant" values
		717	would cause an invocation of C<THAW> with 5 arguments:
		718
		719	sub My::Object::FREEZE {
		720	my ($self, $serialiser) = @_;
		721
		722	($self->{type}, $self->{id}, $self->{variant})
		723	}
		724
		725	sub My::Object::THAW {
		726	my ($class, $serialiser, $type, $id, $variant) = @_;
		727
		728	$class-<new (type => $type, id => $id, variant => $variant)
		729	}
		730
		731
319	=head2 MAGIC HEADER	732	=head1 MAGIC HEADER
320		733
321	There is no way to distinguish CBOR from other formats	734	There is no way to distinguish CBOR from other formats
322	programmatically. To make it easier to distinguish CBOR from other	735	programmatically. To make it easier to distinguish CBOR from other
323	formats, the CBOR specification has a special "magic string" that can be	736	formats, the CBOR specification has a special "magic string" that can be
324	prepended to any CBOR string without changing it's meaning.	737	prepended to any CBOR string without changing its meaning.
325		738
326	This string is available as C<$CBOR::XS::MAGIC>. This module does not	739	This string is available as C<$CBOR::XS::MAGIC>. This module does not
327	prepend this string tot he CBOR data it generates, but it will ignroe it	740	prepend this string to the CBOR data it generates, but it will ignore it
328	if present, so users can prepend this string as a "file type" indicator as	741	if present, so users can prepend this string as a "file type" indicator as
329	required.	742	required.
330		743
331		744
		745	=head1 THE CBOR::XS::Tagged CLASS
		746
		747	CBOR has the concept of tagged values - any CBOR value can be tagged with
		748	a numeric 64 bit number, which are centrally administered.
		749
		750	C<CBOR::XS> handles a few tags internally when en- or decoding. You can
		751	also create tags yourself by encoding C<CBOR::XS::Tagged> objects, and the
		752	decoder will create C<CBOR::XS::Tagged> objects itself when it hits an
		753	unknown tag.
		754
		755	These objects are simply blessed array references - the first member of
		756	the array being the numerical tag, the second being the value.
		757
		758	You can interact with C<CBOR::XS::Tagged> objects in the following ways:
		759
		760	=over 4
		761
		762	=item $tagged = CBOR::XS::tag $tag, $value
		763
		764	This function(!) creates a new C<CBOR::XS::Tagged> object using the given
		765	C<$tag> (0..2**64-1) to tag the given C<$value> (which can be any Perl
		766	value that can be encoded in CBOR, including serialisable Perl objects and
		767	C<CBOR::XS::Tagged> objects).
		768
		769	=item $tagged->[0]
		770
		771	=item $tagged->[0] = $new_tag
		772
		773	=item $tag = $tagged->tag
		774
		775	=item $new_tag = $tagged->tag ($new_tag)
		776
		777	Access/mutate the tag.
		778
		779	=item $tagged->[1]
		780
		781	=item $tagged->[1] = $new_value
		782
		783	=item $value = $tagged->value
		784
		785	=item $new_value = $tagged->value ($new_value)
		786
		787	Access/mutate the tagged value.
		788
		789	=back
		790
		791	=cut
		792
		793	sub tag($$) {
		794	bless [@_], CBOR::XS::Tagged::;
		795	}
		796
		797	sub CBOR::XS::Tagged::tag {
		798	$_[0][0] = $_[1] if $#_;
		799	$_[0][0]
		800	}
		801
		802	sub CBOR::XS::Tagged::value {
		803	$_[0][1] = $_[1] if $#_;
		804	$_[0][1]
		805	}
		806
		807	=head2 EXAMPLES
		808
		809	Here are some examples of C<CBOR::XS::Tagged> uses to tag objects.
		810
		811	You can look up CBOR tag value and emanings in the IANA registry at
		812	L<http://www.iana.org/assignments/cbor-tags/cbor-tags.xhtml>.
		813
		814	Prepend a magic header (C<$CBOR::XS::MAGIC>):
		815
		816	my $cbor = encode_cbor CBOR::XS::tag 55799, $value;
		817	# same as:
		818	my $cbor = $CBOR::XS::MAGIC . encode_cbor $value;
		819
		820	Serialise some URIs and a regex in an array:
		821
		822	my $cbor = encode_cbor [
		823	(CBOR::XS::tag 32, "http://www.nethype.de/"),
		824	(CBOR::XS::tag 32, "http://software.schmorp.de/"),
		825	(CBOR::XS::tag 35, "^[Pp][Ee][Rr][lL]\$"),
		826	];
		827
		828	Wrap CBOR data in CBOR:
		829
		830	my $cbor_cbor = encode_cbor
		831	CBOR::XS::tag 24,
		832	encode_cbor [1, 2, 3];
		833
		834	=head1 TAG HANDLING AND EXTENSIONS
		835
		836	This section describes how this module handles specific tagged values
		837	and extensions. If a tag is not mentioned here and no additional filters
		838	are provided for it, then the default handling applies (creating a
		839	CBOR::XS::Tagged object on decoding, and only encoding the tag when
		840	explicitly requested).
		841
		842	Tags not handled specifically are currently converted into a
		843	L<CBOR::XS::Tagged> object, which is simply a blessed array reference
		844	consisting of the numeric tag value followed by the (decoded) CBOR value.
		845
		846	Future versions of this module reserve the right to special case
		847	additional tags (such as base64url).
		848
		849	=head2 ENFORCED TAGS
		850
		851	These tags are always handled when decoding, and their handling cannot be
		852	overridden by the user.
		853
		854	=over 4
		855
		856	=item 26 (perl-object, L<http://cbor.schmorp.de/perl-object>)
		857
		858	These tags are automatically created (and decoded) for serialisable
		859	objects using the C<FREEZE/THAW> methods (the L<Types::Serialier> object
		860	serialisation protocol). See L<OBJECT SERIALISATION> for details.
		861
		862	=item 28, 29 (shareable, sharedref, L<http://cbor.schmorp.de/value-sharing>)
		863
		864	These tags are automatically decoded when encountered (and they do not
		865	result in a cyclic data structure, see C<allow_cycles>), resulting in
		866	shared values in the decoded object. They are only encoded, however, when
		867	C<allow_sharing> is enabled.
		868
		869	Not all shared values can be successfully decoded: values that reference
		870	themselves will I<currently> decode as C<undef> (this is not the same
		871	as a reference pointing to itself, which will be represented as a value
		872	that contains an indirect reference to itself - these will be decoded
		873	properly).
		874
		875	Note that considerably more shared value data structures can be decoded
		876	than will be encoded - currently, only values pointed to by references
		877	will be shared, others will not. While non-reference shared values can be
		878	generated in Perl with some effort, they were considered too unimportant
		879	to be supported in the encoder. The decoder, however, will decode these
		880	values as shared values.
		881
		882	=item 256, 25 (stringref-namespace, stringref, L<http://cbor.schmorp.de/stringref>)
		883
		884	These tags are automatically decoded when encountered. They are only
		885	encoded, however, when C<pack_strings> is enabled.
		886
		887	=item 22098 (indirection, L<http://cbor.schmorp.de/indirection>)
		888
		889	This tag is automatically generated when a reference are encountered (with
		890	the exception of hash and array references). It is converted to a reference
		891	when decoding.
		892
		893	=item 55799 (self-describe CBOR, RFC 7049)
		894
		895	This value is not generated on encoding (unless explicitly requested by
		896	the user), and is simply ignored when decoding.
		897
		898	=back
		899
		900	=head2 NON-ENFORCED TAGS
		901
		902	These tags have default filters provided when decoding. Their handling can
		903	be overridden by changing the C<%CBOR::XS::FILTER> entry for the tag, or by
		904	providing a custom C<filter> callback when decoding.
		905
		906	When they result in decoding into a specific Perl class, the module
		907	usually provides a corresponding C<TO_CBOR> method as well.
		908
		909	When any of these need to load additional modules that are not part of the
		910	perl core distribution (e.g. L<URI>), it is (currently) up to the user to
		911	provide these modules. The decoding usually fails with an exception if the
		912	required module cannot be loaded.
		913
		914	=over 4
		915
		916	=item 0, 1 (date/time string, seconds since the epoch)
		917
		918	These tags are decoded into L<Time::Piece> objects. The corresponding
		919	C<Time::Piece::TO_CBOR> method always encodes into tag 1 values currently.
		920
		921	The L<Time::Piece> API is generally surprisingly bad, and fractional
		922	seconds are only accidentally kept intact, so watch out. On the plus side,
		923	the module comes with perl since 5.10, which has to count for something.
		924
		925	=item 2, 3 (positive/negative bignum)
		926
		927	These tags are decoded into L<Math::BigInt> objects. The corresponding
		928	C<Math::BigInt::TO_CBOR> method encodes "small" bigints into normal CBOR
		929	integers, and others into positive/negative CBOR bignums.
		930
		931	=item 4, 5 (decimal fraction/bigfloat)
		932
		933	Both decimal fractions and bigfloats are decoded into L<Math::BigFloat>
		934	objects. The corresponding C<Math::BigFloat::TO_CBOR> method I<always>
		935	encodes into a decimal fraction.
		936
		937	CBOR cannot represent bigfloats with I<very> large exponents - conversion
		938	of such big float objects is undefined.
		939
		940	Also, NaN and infinities are not encoded properly.
		941
		942	=item 21, 22, 23 (expected later JSON conversion)
		943
		944	CBOR::XS is not a CBOR-to-JSON converter, and will simply ignore these
		945	tags.
		946
		947	=item 32 (URI)
		948
		949	These objects decode into L<URI> objects. The corresponding
		950	C<URI::TO_CBOR> method again results in a CBOR URI value.
		951
		952	=back
		953
		954	=cut
		955
332	=head2 CBOR and JSON	956	=head1 CBOR and JSON
333		957
334	CBOR is supposed to implement a superset of the JSON data model, and is,	958	CBOR is supposed to implement a superset of the JSON data model, and is,
335	with some coercion, able to represent all JSON texts (something that other	959	with some coercion, able to represent all JSON texts (something that other
336	"binary JSON" formats such as BSON generally do not support).	960	"binary JSON" formats such as BSON generally do not support).
337		961
…		…
396	properly. Half precision types are accepted, but not encoded.	1020	properly. Half precision types are accepted, but not encoded.
397		1021
398	Strict mode and canonical mode are not implemented.	1022	Strict mode and canonical mode are not implemented.
399		1023
400		1024
		1025	=head1 LIMITATIONS ON PERLS WITHOUT 64-BIT INTEGER SUPPORT
		1026
		1027	On perls that were built without 64 bit integer support (these are rare
		1028	nowadays, even on 32 bit architectures, as all major Perl distributions
		1029	are built with 64 bit integer support), support for any kind of 64 bit
		1030	integer in CBOR is very limited - most likely, these 64 bit values will
		1031	be truncated, corrupted, or otherwise not decoded correctly. This also
		1032	includes string, array and map sizes that are stored as 64 bit integers.
		1033
		1034
401	=head1 THREADS	1035	=head1 THREADS
402		1036
403	This module is I<not> guaranteed to be thread safe and there are no	1037	This module is I<not> guaranteed to be thread safe and there are no
404	plans to change this until Perl gets thread support (as opposed to the	1038	plans to change this until Perl gets thread support (as opposed to the
405	horribly slow so-called "threads" which are simply slow and bloated	1039	horribly slow so-called "threads" which are simply slow and bloated
…		…
417	Please refrain from using rt.cpan.org or any other bug reporting	1051	Please refrain from using rt.cpan.org or any other bug reporting
418	service. I put the contact address into my modules for a reason.	1052	service. I put the contact address into my modules for a reason.
419		1053
420	=cut	1054	=cut
421		1055
422	our $true = do { bless \(my $dummy = 1), "CBOR::XS::Boolean" };	1056	our %FILTER = (
423	our $false = do { bless \(my $dummy = 0), "CBOR::XS::Boolean" };	1057	0 => sub { # rfc4287 datetime, utf-8
		1058	require Time::Piece;
		1059	# Time::Piece::Strptime uses the "incredibly flexible date parsing routine"
		1060	# from FreeBSD, which can't parse ISO 8601, RFC3339, RFC4287 or much of anything
		1061	# else either. Whats incredibe over standard strptime totally escapes me.
		1062	# doesn't do fractional times, either. sigh.
		1063	# In fact, it's all a lie, it uses whatever strptime it wants, and of course,
		1064	# they are all incompatible. The openbsd one simply ignores %z (but according to the
		1065	# docs, it would be much more incredibly flexible indeed. If it worked, that is.).
		1066	scalar eval {
		1067	my $s = $_[1];
424		1068
425	sub true() { $true }	1069	$s =~ s/Z$/+00:00/;
426	sub false() { $false }	1070	$s =~ s/(\.[0-9]+)?([+-][0-9][0-9]):([0-9][0-9])$//
		1071	or die;
427		1072
428	sub is_bool($) {	1073	my $b = $1 - ($2 * 60 + $3) * 60; # fractional part + offset. hopefully
429	UNIVERSAL::isa $_[0], "CBOR::XS::Boolean"	1074	my $d = Time::Piece->strptime ($s, "%Y-%m-%dT%H:%M:%S");
430	# or UNIVERSAL::isa $_[0], "CBOR::Literal"	1075
		1076	Time::Piece::gmtime ($d->epoch + $b)
		1077	} \|\| die "corrupted CBOR date/time string ($_[0])";
		1078	},
		1079
		1080	1 => sub { # seconds since the epoch, possibly fractional
		1081	require Time::Piece;
		1082	scalar Time::Piece::gmtime (pop)
		1083	},
		1084
		1085	2 => sub { # pos bigint
		1086	require Math::BigInt;
		1087	Math::BigInt->new ("0x" . unpack "H*", pop)
		1088	},
		1089
		1090	3 => sub { # neg bigint
		1091	require Math::BigInt;
		1092	-Math::BigInt->new ("0x" . unpack "H*", pop)
		1093	},
		1094
		1095	4 => sub { # decimal fraction, array
		1096	require Math::BigFloat;
		1097	Math::BigFloat->new ($_[1][1] . "E" . $_[1][0])
		1098	},
		1099
		1100	5 => sub { # bigfloat, array
		1101	require Math::BigFloat;
		1102	scalar Math::BigFloat->new ($_[1][1]) * Math::BigFloat->new (2)->bpow ($_[1][0])
		1103	},
		1104
		1105	21 => sub { pop }, # expected conversion to base64url encoding
		1106	22 => sub { pop }, # expected conversion to base64 encoding
		1107	23 => sub { pop }, # expected conversion to base16 encoding
		1108
		1109	# 24 # embedded cbor, byte string
		1110
		1111	32 => sub {
		1112	require URI;
		1113	URI->new (pop)
		1114	},
		1115
		1116	# 33 # base64url rfc4648, utf-8
		1117	# 34 # base64 rfc46484, utf-8
		1118	# 35 # regex pcre/ecma262, utf-8
		1119	# 36 # mime message rfc2045, utf-8
		1120	);
		1121
		1122	sub CBOR::XS::default_filter {
		1123	&{ $FILTER{$_[0]} or return }
431	}	1124	}
432		1125
		1126	sub URI::TO_CBOR {
		1127	my $uri = $_[0]->as_string;
		1128	utf8::upgrade $uri;
		1129	tag 32, $uri
		1130	}
		1131
		1132	sub Math::BigInt::TO_CBOR {
		1133	if ($_[0] >= -2147483648 && $_[0] <= 2147483647) {
		1134	$_[0]->numify
		1135	} else {
		1136	my $hex = substr $_[0]->as_hex, 2;
		1137	$hex = "0$hex" if 1 & length $hex; # sigh
		1138	tag $_[0] >= 0 ? 2 : 3, pack "H*", $hex
		1139	}
		1140	}
		1141
		1142	sub Math::BigFloat::TO_CBOR {
		1143	my ($m, $e) = $_[0]->parts;
		1144	tag 4, [$e->numify, $m]
		1145	}
		1146
		1147	sub Time::Piece::TO_CBOR {
		1148	tag 1, 0 + $_[0]->epoch
		1149	}
		1150
433	XSLoader::load "CBOR::XS", $VERSION;	1151	XSLoader::load "CBOR::XS", $VERSION;
434
435	package CBOR::XS::Boolean;
436
437	use overload
438	"0+" => sub { ${$_[0]} },
439	"++" => sub { $_[0] = ${$_[0]} + 1 },
440	"--" => sub { $_[0] = ${$_[0]} - 1 },
441	fallback => 1;
442
443	1;
444		1152
445	=head1 SEE ALSO	1153	=head1 SEE ALSO
446		1154
447	The L<JSON> and L<JSON::XS> modules that do similar, but human-readable,	1155	The L<JSON> and L<JSON::XS> modules that do similar, but human-readable,
448	serialisation.	1156	serialisation.
449		1157
		1158	The L<Types::Serialiser> module provides the data model for true, false
		1159	and error values.
		1160
450	=head1 AUTHOR	1161	=head1 AUTHOR
451		1162
452	Marc Lehmann <schmorp@schmorp.de>	1163	Marc Lehmann <schmorp@schmorp.de>
453	http://home.schmorp.de/	1164	http://home.schmorp.de/
454		1165
455	=cut	1166	=cut
456		1167
		1168	1
		1169

Diff Legend

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

Comparing CBOR-XS/XS.pm (file contents): Revision 1.5 by root, Sat Oct 26 23:02:55 2013 UTC vs. Revision 1.52 by root, Mon Apr 25 18:17:17 2016 UTC

Diff Legend

Comparing CBOR-XS/XS.pm (file contents):
Revision 1.5 by root, Sat Oct 26 23:02:55 2013 UTC vs.
Revision 1.52 by root, Mon Apr 25 18:17:17 2016 UTC