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

Comparing CBOR-XS/XS.pm (file contents):
Revision 1.25 by root, Thu Nov 28 12:08:07 2013 UTC vs.
Revision 1.66 by root, Tue Jun 27 02:03:24 2017 UTC

…		…
26	substr $many_cbor_strings, 0, $length, ""; # remove decoded cbor string	26	substr $many_cbor_strings, 0, $length, ""; # remove decoded cbor string
27	}	27	}
28		28
29	=head1 DESCRIPTION	29	=head1 DESCRIPTION
30		30
31	WARNING! This module is very new, and not very well tested (that's up
32	to you to do). Furthermore, details of the implementation might change
33	freely before version 1.0. And lastly, most extensions depend on an IANA
34	assignment, and until that assignment is official, this implementation is
35	not interoperable with other implementations (even future versions of this
36	module) until the assignment is done.
37
38	You are still invited to try out CBOR, and this module.
39
40	This module converts Perl data structures to the Concise Binary Object	31	This module converts Perl data structures to the Concise Binary Object
41	Representation (CBOR) and vice versa. CBOR is a fast binary serialisation	32	Representation (CBOR) and vice versa. CBOR is a fast binary serialisation
42	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.
43	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
44	CBOR.	35	represent it in CBOR.
45		36
46	In short, CBOR is a faster and very compact binary alternative to JSON,	37	In short, CBOR is a faster and quite compact binary alternative to JSON,
47	with the added ability of supporting serialisation of Perl objects. (JSON	38	with the added ability of supporting serialisation of Perl objects. (JSON
48	often compresses better than CBOR though, so if you plan to compress the	39	often compresses better than CBOR though, so if you plan to compress the
49	data later you might want to compare both formats first).	40	data later and speed is less important you might want to compare both
		41	formats first).
50		42
51	To give you a general idea about speed, with texts in the megabyte range,	43	To give you a general idea about speed, with texts in the megabyte range,
52	C<CBOR::XS> usually encodes roughly twice as fast as L<Storable> or	44	C<CBOR::XS> usually encodes roughly twice as fast as L<Storable> or
53	L<JSON::XS> and decodes about 15%-30% faster than those. The shorter the	45	L<JSON::XS> and decodes about 15%-30% faster than those. The shorter the
54	data, the worse L<Storable> performs in comparison.	46	data, the worse L<Storable> performs in comparison.
55		47
56	As for compactness, C<CBOR::XS> encoded data structures are usually about	48	Regarding compactness, C<CBOR::XS>-encoded data structures are usually
57	20% smaller than the same data encoded as (compact) JSON or L<Storable>.	49	about 20% smaller than the same data encoded as (compact) JSON or
		50	L<Storable>.
58		51
59	In addition to the core CBOR data format, this module implements a number	52	In addition to the core CBOR data format, this module implements a
60	of extensions, to support cyclic and self-referencing data structures	53	number of extensions, to support cyclic and shared data structures
61	(see C<allow_sharing>), string deduplication (see C<pack_strings>) and	54	(see C<allow_sharing> and C<allow_cycles>), string deduplication (see
62	scalar references (always enabled).	55	C<pack_strings>) and scalar references (always enabled).
63		56
64	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
65	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.
66		59
67	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
…		…
71		64
72	package CBOR::XS;	65	package CBOR::XS;
73		66
74	use common::sense;	67	use common::sense;
75		68
76	our $VERSION = 0.09;	69	our $VERSION = 1.7;
77	our @ISA = qw(Exporter);	70	our @ISA = qw(Exporter);
78		71
79	our @EXPORT = qw(encode_cbor decode_cbor);	72	our @EXPORT = qw(encode_cbor decode_cbor);
80		73
81	use Exporter;	74	use Exporter;
…		…
119		112
120	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
121	be chained:	114	be chained:
122		115
123	my $cbor = CBOR::XS->new->encode ({a => [1,2]});	116	my $cbor = CBOR::XS->new->encode ({a => [1,2]});
		117
		118	=item $cbor = new_safe CBOR::XS
		119
		120	Create a new, safe/secure CBOR::XS object. This is similar to C<new>,
		121	but configures the coder object to be safe to use with untrusted
		122	data. Currently, this is equivalent to:
		123
		124	my $cbor = CBOR::XS
		125	->new
		126	->forbid_objects
		127	->filter (\&CBOR::XS::safe_filter)
		128	->max_size (1e8);
		129
		130	But is more future proof (it is better to crash because of a change than
		131	to be exploited in other ways).
		132
		133	=cut
		134
		135	sub new_safe {
		136	CBOR::XS
		137	->new
		138	->forbid_objects
		139	->filter (\&CBOR::XS::safe_filter)
		140	->max_size (1e8)
		141	}
124		142
125	=item $cbor = $cbor->max_depth ([$maximum_nesting_depth])	143	=item $cbor = $cbor->max_depth ([$maximum_nesting_depth])
126		144
127	=item $max_depth = $cbor->get_max_depth	145	=item $max_depth = $cbor->get_max_depth
128		146
…		…
144		162
145	Note that nesting is implemented by recursion in C. The default value has	163	Note that nesting is implemented by recursion in C. The default value has
146	been chosen to be as large as typical operating systems allow without	164	been chosen to be as large as typical operating systems allow without
147	crashing.	165	crashing.
148		166
149	See SECURITY CONSIDERATIONS, below, for more info on why this is useful.	167	See L<SECURITY CONSIDERATIONS>, below, for more info on why this is useful.
150		168
151	=item $cbor = $cbor->max_size ([$maximum_string_size])	169	=item $cbor = $cbor->max_size ([$maximum_string_size])
152		170
153	=item $max_size = $cbor->get_max_size	171	=item $max_size = $cbor->get_max_size
154		172
…		…
159	effect on C<encode> (yet).	177	effect on C<encode> (yet).
160		178
161	If no argument is given, the limit check will be deactivated (same as when	179	If no argument is given, the limit check will be deactivated (same as when
162	C<0> is specified).	180	C<0> is specified).
163		181
164	See SECURITY CONSIDERATIONS, below, for more info on why this is useful.	182	See L<SECURITY CONSIDERATIONS>, below, for more info on why this is useful.
165		183
166	=item $cbor = $cbor->allow_unknown ([$enable])	184	=item $cbor = $cbor->allow_unknown ([$enable])
167		185
168	=item $enabled = $cbor->get_allow_unknown	186	=item $enabled = $cbor->get_allow_unknown
169		187
…		…
187	reference to the earlier value.	205	reference to the earlier value.
188		206
189	This means that such values will only be encoded once, and will not result	207	This means that such values will only be encoded once, and will not result
190	in a deep cloning of the value on decode, in decoders supporting the value	208	in a deep cloning of the value on decode, in decoders supporting the value
191	sharing extension. This also makes it possible to encode cyclic data	209	sharing extension. This also makes it possible to encode cyclic data
192	structures.	210	structures (which need C<allow_cycles> to be enabled to be decoded by this
		211	module).
193		212
194	It is recommended to leave it off unless you know your	213	It is recommended to leave it off unless you know your
195	communication partner supports the value sharing extensions to CBOR	214	communication partner supports the value sharing extensions to CBOR
196	(http://cbor.schmorp.de/value-sharing), as without decoder support, the	215	(L<http://cbor.schmorp.de/value-sharing>), as without decoder support, the
197	resulting data structure might be unusable.	216	resulting data structure might be unusable.
198		217
199	Detecting shared values incurs a runtime overhead when values are encoded	218	Detecting shared values incurs a runtime overhead when values are encoded
200	that have a reference counter large than one, and might unnecessarily	219	that have a reference counter large than one, and might unnecessarily
201	increase the encoded size, as potentially shared values are encode as	220	increase the encoded size, as potentially shared values are encode as
202	sharable whether or not they are actually shared.	221	shareable whether or not they are actually shared.
203		222
204	At the moment, only targets of references can be shared (e.g. scalars,	223	At the moment, only targets of references can be shared (e.g. scalars,
205	arrays or hashes pointed to by a reference). Weirder constructs, such as	224	arrays or hashes pointed to by a reference). Weirder constructs, such as
206	an array with multiple "copies" of the I<same> string, which are hard but	225	an array with multiple "copies" of the I<same> string, which are hard but
207	not impossible to create in Perl, are not supported (this is the same as	226	not impossible to create in Perl, are not supported (this is the same as
…		…
212	structures cannot be encoded in this mode.	231	structures cannot be encoded in this mode.
213		232
214	This option does not affect C<decode> in any way - shared values and	233	This option does not affect C<decode> in any way - shared values and
215	references will always be decoded properly if present.	234	references will always be decoded properly if present.
216		235
		236	=item $cbor = $cbor->allow_cycles ([$enable])
		237
		238	=item $enabled = $cbor->get_allow_cycles
		239
		240	If C<$enable> is true (or missing), then C<decode> will happily decode
		241	self-referential (cyclic) data structures. By default these will not be
		242	decoded, as they need manual cleanup to avoid memory leaks, so code that
		243	isn't prepared for this will not leak memory.
		244
		245	If C<$enable> is false (the default), then C<decode> will throw an error
		246	when it encounters a self-referential/cyclic data structure.
		247
		248	FUTURE DIRECTION: the motivation behind this option is to avoid I<real>
		249	cycles - future versions of this module might chose to decode cyclic data
		250	structures using weak references when this option is off, instead of
		251	throwing an error.
		252
		253	This option does not affect C<encode> in any way - shared values and
		254	references will always be encoded properly if present.
		255
		256	=item $cbor = $cbor->forbid_objects ([$enable])
		257
		258	=item $enabled = $cbor->get_forbid_objects
		259
		260	Disables the use of the object serialiser protocol.
		261
		262	If C<$enable> is true (or missing), then C<encode> will will throw an
		263	exception when it encounters perl objects that would be encoded using the
		264	perl-object tag (26). When C<decode> encounters such tags, it will fall
		265	back to the general filter/tagged logic as if this were an unknown tag (by
		266	default resulting in a C<CBOR::XC::Tagged> object).
		267
		268	If C<$enable> is false (the default), then C<encode> will use the
		269	L<Types::Serialiser> object serialisation protocol to serialise objects
		270	into perl-object tags, and C<decode> will do the same to decode such tags.
		271
		272	See L<SECURITY CONSIDERATIONS>, below, for more info on why forbidding this
		273	protocol can be useful.
		274
217	=item $cbor = $cbor->pack_strings ([$enable])	275	=item $cbor = $cbor->pack_strings ([$enable])
218		276
219	=item $enabled = $cbor->get_pack_strings	277	=item $enabled = $cbor->get_pack_strings
220		278
221	If C<$enable> is true (or missing), then C<encode> will try not to encode	279	If C<$enable> is true (or missing), then C<encode> will try not to encode
…		…
224	also results in a very large runtime overhead (expect encoding times to be	282	also results in a very large runtime overhead (expect encoding times to be
225	2-4 times as high as without).	283	2-4 times as high as without).
226		284
227	It is recommended to leave it off unless you know your	285	It is recommended to leave it off unless you know your
228	communications partner supports the stringref extension to CBOR	286	communications partner supports the stringref extension to CBOR
229	(http://cbor.schmorp.de/stringref), as without decoder support, the	287	(L<http://cbor.schmorp.de/stringref>), as without decoder support, the
230	resulting data structure might not be usable.	288	resulting data structure might not be usable.
231		289
232	If C<$enable> is false (the default), then C<encode> will encode strings	290	If C<$enable> is false (the default), then C<encode> will encode strings
233	the standard CBOR way.	291	the standard CBOR way.
234		292
235	This option does not affect C<decode> in any way - string references will	293	This option does not affect C<decode> in any way - string references will
236	always be decoded properly if present.	294	always be decoded properly if present.
		295
		296	=item $cbor = $cbor->text_keys ([$enable])
		297
		298	=item $enabled = $cbor->get_text_keys
		299
		300	If C<$enabled> is true (or missing), then C<encode> will encode all
		301	perl hash keys as CBOR text strings/UTF-8 string, upgrading them as needed.
		302
		303	If C<$enable> is false (the default), then C<encode> will encode hash keys
		304	normally - upgraded perl strings (strings internally encoded as UTF-8) as
		305	CBOR text strings, and downgraded perl strings as CBOR byte strings.
		306
		307	This option does not affect C<decode> in any way.
		308
		309	This option is useful for interoperability with CBOR decoders that don't
		310	treat byte strings as a form of text. It is especially useful as Perl
		311	gives very little control over hash keys.
		312
		313	Enabling this option can be slow, as all downgraded hash keys that are
		314	encoded need to be scanned and converted to UTF-8.
		315
		316	=item $cbor = $cbor->text_strings ([$enable])
		317
		318	=item $enabled = $cbor->get_text_strings
		319
		320	This option works similar to C<text_keys>, above, but works on all strings
		321	(including hash keys), so C<text_keys> has no further effect after
		322	enabling C<text_strings>.
		323
		324	If C<$enabled> is true (or missing), then C<encode> will encode all perl
		325	strings as CBOR text strings/UTF-8 strings, upgrading them as needed.
		326
		327	If C<$enable> is false (the default), then C<encode> will encode strings
		328	normally (but see C<text_keys>) - upgraded perl strings (strings
		329	internally encoded as UTF-8) as CBOR text strings, and downgraded perl
		330	strings as CBOR byte strings.
		331
		332	This option does not affect C<decode> in any way.
		333
		334	This option has similar advantages and disadvantages as C<text_keys>. In
		335	addition, this option effectively removes the ability to encode byte
		336	strings, which might break some C<FREEZE> and C<TO_CBOR> methods that rely
		337	on this, such as bignum encoding, so this option is mainly useful for very
		338	simple data.
		339
		340	=item $cbor = $cbor->validate_utf8 ([$enable])
		341
		342	=item $enabled = $cbor->get_validate_utf8
		343
		344	If C<$enable> is true (or missing), then C<decode> will validate that
		345	elements (text strings) containing UTF-8 data in fact contain valid UTF-8
		346	data (instead of blindly accepting it). This validation obviously takes
		347	extra time during decoding.
		348
		349	The concept of "valid UTF-8" used is perl's concept, which is a superset
		350	of the official UTF-8.
		351
		352	If C<$enable> is false (the default), then C<decode> will blindly accept
		353	UTF-8 data, marking them as valid UTF-8 in the resulting data structure
		354	regardless of whether that's true or not.
		355
		356	Perl isn't too happy about corrupted UTF-8 in strings, but should
		357	generally not crash or do similarly evil things. Extensions might be not
		358	so forgiving, so it's recommended to turn on this setting if you receive
		359	untrusted CBOR.
		360
		361	This option does not affect C<encode> in any way - strings that are
		362	supposedly valid UTF-8 will simply be dumped into the resulting CBOR
		363	string without checking whether that is, in fact, true or not.
237		364
238	=item $cbor = $cbor->filter ([$cb->($tag, $value)])	365	=item $cbor = $cbor->filter ([$cb->($tag, $value)])
239		366
240	=item $cb_or_undef = $cbor->get_filter	367	=item $cb_or_undef = $cbor->get_filter
241		368
…		…
254	replace the tagged value in the decoded data structure, or no values,	381	replace the tagged value in the decoded data structure, or no values,
255	which will result in default handling, which currently means the decoder	382	which will result in default handling, which currently means the decoder
256	creates a C<CBOR::XS::Tagged> object to hold the tag and the value.	383	creates a C<CBOR::XS::Tagged> object to hold the tag and the value.
257		384
258	When the filter is cleared (the default state), the default filter	385	When the filter is cleared (the default state), the default filter
259	function, C<CBOR::XS::default_filter>, is used. This function simply looks	386	function, C<CBOR::XS::default_filter>, is used. This function simply
260	up the tag in the C<%CBOR::XS::FILTER> hash. If an entry exists it must be	387	looks up the tag in the C<%CBOR::XS::FILTER> hash. If an entry exists
261	a code reference that is called with tag and value, and is responsible for	388	it must be a code reference that is called with tag and value, and is
262	decoding the value. If no entry exists, it returns no values.	389	responsible for decoding the value. If no entry exists, it returns no
		390	values. C<CBOR::XS> provides a number of default filter functions already,
		391	the the C<%CBOR::XS::FILTER> hash can be freely extended with more.
263		392
		393	C<CBOR::XS> additionally provides an alternative filter function that is
		394	supposed to be safe to use with untrusted data (which the default filter
		395	might not), called C<CBOR::XS::safe_filter>, which works the same as
		396	the C<default_filter> but uses the C<%CBOR::XS::SAFE_FILTER> variable
		397	instead. It is prepopulated with the tag decoding functions that are
		398	deemed safe (basically the same as C<%CBOR::XS::FILTER> without all
		399	the bignum tags), and can be extended by user code as wlel, although,
		400	obviously, one should be very careful about adding decoding functions
		401	here, since the expectation is that they are safe to use on untrusted
		402	data, after all.
		403
264	Example: decode all tags not handled internally into CBOR::XS::Tagged	404	Example: decode all tags not handled internally into C<CBOR::XS::Tagged>
265	objects, with no other special handling (useful when working with	405	objects, with no other special handling (useful when working with
266	potentially "unsafe" CBOR data).	406	potentially "unsafe" CBOR data).
267		407
268	CBOR::XS->new->filter (sub { })->decode ($cbor_data);	408	CBOR::XS->new->filter (sub { })->decode ($cbor_data);
269		409
…		…
273	$CBOR::XS::FILTER{1347375694} = sub {	413	$CBOR::XS::FILTER{1347375694} = sub {
274	my ($tag, $value);	414	my ($tag, $value);
275		415
276	"tag 1347375694 value $value"	416	"tag 1347375694 value $value"
277	};	417	};
		418
		419	Example: provide your own filter function that looks up tags in your own
		420	hash:
		421
		422	my %my_filter = (
		423	998347484 => sub {
		424	my ($tag, $value);
		425
		426	"tag 998347484 value $value"
		427	};
		428	);
		429
		430	my $coder = CBOR::XS->new->filter (sub {
		431	&{ $my_filter{$_[0]} or return }
		432	});
		433
		434
		435	Example: use the safe filter function (see L<SECURITY CONSIDERATIONS> for
		436	more considerations on security).
		437
		438	CBOR::XS->new->filter (\&CBOR::XS::safe_filter)->decode ($cbor_data);
278		439
279	=item $cbor_data = $cbor->encode ($perl_scalar)	440	=item $cbor_data = $cbor->encode ($perl_scalar)
280		441
281	Converts the given Perl data structure (a scalar value) to its CBOR	442	Converts the given Perl data structure (a scalar value) to its CBOR
282	representation.	443	representation.
…		…
296	and you need to know where the first CBOR string ends amd the next one	457	and you need to know where the first CBOR string ends amd the next one
297	starts.	458	starts.
298		459
299	CBOR::XS->new->decode_prefix ("......")	460	CBOR::XS->new->decode_prefix ("......")
300	=> ("...", 3)	461	=> ("...", 3)
		462
		463	=back
		464
		465	=head2 INCREMENTAL PARSING
		466
		467	In some cases, there is the need for incremental parsing of JSON
		468	texts. While this module always has to keep both CBOR text and resulting
		469	Perl data structure in memory at one time, it does allow you to parse a
		470	CBOR stream incrementally, using a similar to using "decode_prefix" to see
		471	if a full CBOR object is available, but is much more efficient.
		472
		473	It basically works by parsing as much of a CBOR string as possible - if
		474	the CBOR data is not complete yet, the pasrer will remember where it was,
		475	to be able to restart when more data has been accumulated. Once enough
		476	data is available to either decode a complete CBOR value or raise an
		477	error, a real decode will be attempted.
		478
		479	A typical use case would be a network protocol that consists of sending
		480	and receiving CBOR-encoded messages. The solution that works with CBOR and
		481	about anything else is by prepending a length to every CBOR value, so the
		482	receiver knows how many octets to read. More compact (and slightly slower)
		483	would be to just send CBOR values back-to-back, as C<CBOR::XS> knows where
		484	a CBOR value ends, and doesn't need an explicit length.
		485
		486	The following methods help with this:
		487
		488	=over 4
		489
		490	=item @decoded = $cbor->incr_parse ($buffer)
		491
		492	This method attempts to decode exactly one CBOR value from the beginning
		493	of the given C<$buffer>. The value is removed from the C<$buffer> on
		494	success. When C<$buffer> doesn't contain a complete value yet, it returns
		495	nothing. Finally, when the C<$buffer> doesn't start with something
		496	that could ever be a valid CBOR value, it raises an exception, just as
		497	C<decode> would. In the latter case the decoder state is undefined and
		498	must be reset before being able to parse further.
		499
		500	This method modifies the C<$buffer> in place. When no CBOR value can be
		501	decoded, the decoder stores the current string offset. On the next call,
		502	continues decoding at the place where it stopped before. For this to make
		503	sense, the C<$buffer> must begin with the same octets as on previous
		504	unsuccessful calls.
		505
		506	You can call this method in scalar context, in which case it either
		507	returns a decoded value or C<undef>. This makes it impossible to
		508	distinguish between CBOR null values (which decode to C<undef>) and an
		509	unsuccessful decode, which is often acceptable.
		510
		511	=item @decoded = $cbor->incr_parse_multiple ($buffer)
		512
		513	Same as C<incr_parse>, but attempts to decode as many CBOR values as
		514	possible in one go, instead of at most one. Calls to C<incr_parse> and
		515	C<incr_parse_multiple> can be interleaved.
		516
		517	=item $cbor->incr_reset
		518
		519	Resets the incremental decoder. This throws away any saved state, so that
		520	subsequent calls to C<incr_parse> or C<incr_parse_multiple> start to parse
		521	a new CBOR value from the beginning of the C<$buffer> again.
		522
		523	This method can be called at any time, but it I<must> be called if you want
		524	to change your C<$buffer> or there was a decoding error and you want to
		525	reuse the C<$cbor> object for future incremental parsings.
301		526
302	=back	527	=back
303		528
304		529
305	=head1 MAPPING	530	=head1 MAPPING
…		…
323	CBOR integers become (numeric) perl scalars. On perls without 64 bit	548	CBOR integers become (numeric) perl scalars. On perls without 64 bit
324	support, 64 bit integers will be truncated or otherwise corrupted.	549	support, 64 bit integers will be truncated or otherwise corrupted.
325		550
326	=item byte strings	551	=item byte strings
327		552
328	Byte strings will become octet strings in Perl (the byte values 0..255	553	Byte strings will become octet strings in Perl (the Byte values 0..255
329	will simply become characters of the same value in Perl).	554	will simply become characters of the same value in Perl).
330		555
331	=item UTF-8 strings	556	=item UTF-8 strings
332		557
333	UTF-8 strings in CBOR will be decoded, i.e. the UTF-8 octets will be	558	UTF-8 strings in CBOR will be decoded, i.e. the UTF-8 octets will be
…		…
356	=item tagged values	581	=item tagged values
357		582
358	Tagged items consists of a numeric tag and another CBOR value.	583	Tagged items consists of a numeric tag and another CBOR value.
359		584
360	See L<TAG HANDLING AND EXTENSIONS> and the description of C<< ->filter >>	585	See L<TAG HANDLING AND EXTENSIONS> and the description of C<< ->filter >>
361	for details.	586	for details on which tags are handled how.
362		587
363	=item anything else	588	=item anything else
364		589
365	Anything else (e.g. unsupported simple values) will raise a decoding	590	Anything else (e.g. unsupported simple values) will raise a decoding
366	error.	591	error.
…		…
369		594
370		595
371	=head2 PERL -> CBOR	596	=head2 PERL -> CBOR
372		597
373	The mapping from Perl to CBOR is slightly more difficult, as Perl is a	598	The mapping from Perl to CBOR is slightly more difficult, as Perl is a
374	truly typeless language, so we can only guess which CBOR type is meant by	599	typeless language. That means this module can only guess which CBOR type
375	a Perl value.	600	is meant by a perl value.
376		601
377	=over 4	602	=over 4
378		603
379	=item hash references	604	=item hash references
380		605
381	Perl hash references become CBOR maps. As there is no inherent ordering in	606	Perl hash references become CBOR maps. As there is no inherent ordering in
382	hash keys (or CBOR maps), they will usually be encoded in a pseudo-random	607	hash keys (or CBOR maps), they will usually be encoded in a pseudo-random
383	order.	608	order. This order can be different each time a hash is encoded.
384		609
385	Currently, tied hashes will use the indefinite-length format, while normal	610	Currently, tied hashes will use the indefinite-length format, while normal
386	hashes will use the fixed-length format.	611	hashes will use the fixed-length format.
387		612
388	=item array references	613	=item array references
389		614
390	Perl array references become fixed-length CBOR arrays.	615	Perl array references become fixed-length CBOR arrays.
391		616
392	=item other references	617	=item other references
393		618
394	Other unblessed references are generally not allowed and will cause an	619	Other unblessed references will be represented using
395	exception to be thrown, except for references to the integers C<0> and	620	the indirection tag extension (tag value C<22098>,
396	C<1>, which get turned into false and true in CBOR.	621	L<http://cbor.schmorp.de/indirection>). CBOR decoders are guaranteed
		622	to be able to decode these values somehow, by either "doing the right
		623	thing", decoding into a generic tagged object, simply ignoring the tag, or
		624	something else.
397		625
398	=item CBOR::XS::Tagged objects	626	=item CBOR::XS::Tagged objects
399		627
400	Objects of this type must be arrays consisting of a single C<[tag, value]>	628	Objects of this type must be arrays consisting of a single C<[tag, value]>
401	pair. The (numerical) tag will be encoded as a CBOR tag, the value will	629	pair. The (numerical) tag will be encoded as a CBOR tag, the value will
402	be encoded as appropriate for the value. You cna use C<CBOR::XS::tag> to	630	be encoded as appropriate for the value. You must use C<CBOR::XS::tag> to
403	create such objects.	631	create such objects.
404		632
405	=item Types::Serialiser::true, Types::Serialiser::false, Types::Serialiser::error	633	=item Types::Serialiser::true, Types::Serialiser::false, Types::Serialiser::error
406		634
407	These special values become CBOR true, CBOR false and CBOR undefined	635	These special values become CBOR true, CBOR false and CBOR undefined
…		…
424	# dump as number	652	# dump as number
425	encode_cbor [2] # yields [2]	653	encode_cbor [2] # yields [2]
426	encode_cbor [-3.0e17] # yields [-3e+17]	654	encode_cbor [-3.0e17] # yields [-3e+17]
427	my $value = 5; encode_cbor [$value] # yields [5]	655	my $value = 5; encode_cbor [$value] # yields [5]
428		656
429	# used as string, so dump as string	657	# used as string, so dump as string (either byte or text)
430	print $value;	658	print $value;
431	encode_cbor [$value] # yields ["5"]	659	encode_cbor [$value] # yields ["5"]
432		660
433	# undef becomes null	661	# undef becomes null
434	encode_cbor [undef] # yields [null]	662	encode_cbor [undef] # yields [null]
…		…
437		665
438	my $x = 3.1; # some variable containing a number	666	my $x = 3.1; # some variable containing a number
439	"$x"; # stringified	667	"$x"; # stringified
440	$x .= ""; # another, more awkward way to stringify	668	$x .= ""; # another, more awkward way to stringify
441	print $x; # perl does it for you, too, quite often	669	print $x; # perl does it for you, too, quite often
		670
		671	You can force whether a string is encoded as byte or text string by using
		672	C<utf8::upgrade> and C<utf8::downgrade> (if C<text_strings> is disabled):
		673
		674	utf8::upgrade $x; # encode $x as text string
		675	utf8::downgrade $x; # encode $x as byte string
		676
		677	Perl doesn't define what operations up- and downgrade strings, so if the
		678	difference between byte and text is important, you should up- or downgrade
		679	your string as late as possible before encoding. You can also force the
		680	use of CBOR text strings by using C<text_keys> or C<text_strings>.
442		681
443	You can force the type to be a CBOR number by numifying it:	682	You can force the type to be a CBOR number by numifying it:
444		683
445	my $x = "3"; # some variable containing a string	684	my $x = "3"; # some variable containing a string
446	$x += 0; # numify it, ensuring it will be dumped as a number	685	$x += 0; # numify it, ensuring it will be dumped as a number
…		…
459		698
460	=back	699	=back
461		700
462	=head2 OBJECT SERIALISATION	701	=head2 OBJECT SERIALISATION
463		702
		703	This module implements both a CBOR-specific and the generic
		704	L<Types::Serialier> object serialisation protocol. The following
		705	subsections explain both methods.
		706
		707	=head3 ENCODING
		708
464	This module knows two way to serialise a Perl object: The CBOR-specific	709	This module knows two way to serialise a Perl object: The CBOR-specific
465	way, and the generic way.	710	way, and the generic way.
466		711
467	Whenever the encoder encounters a Perl object that it cnanot serialise	712	Whenever the encoder encounters a Perl object that it cannot serialise
468	directly (most of them), it will first look up the C<TO_CBOR> method on	713	directly (most of them), it will first look up the C<TO_CBOR> method on
469	it.	714	it.
470		715
471	If it has a C<TO_CBOR> method, it will call it with the object as only	716	If it has a C<TO_CBOR> method, it will call it with the object as only
472	argument, and expects exactly one return value, which it will then	717	argument, and expects exactly one return value, which it will then
…		…
478		723
479	The C<FREEZE> method can return any number of values (i.e. zero or	724	The C<FREEZE> method can return any number of values (i.e. zero or
480	more). These will be encoded as CBOR perl object, together with the	725	more). These will be encoded as CBOR perl object, together with the
481	classname.	726	classname.
482		727
		728	These methods I<MUST NOT> change the data structure that is being
		729	serialised. Failure to comply to this can result in memory corruption -
		730	and worse.
		731
483	If an object supports neither C<TO_CBOR> nor C<FREEZE>, encoding will fail	732	If an object supports neither C<TO_CBOR> nor C<FREEZE>, encoding will fail
484	with an error.	733	with an error.
485		734
		735	=head3 DECODING
		736
486	Objects encoded via C<TO_CBOR> cannot be automatically decoded, but	737	Objects encoded via C<TO_CBOR> cannot (normally) be automatically decoded,
487	objects encoded via C<FREEZE> can be decoded using the following protocol:	738	but objects encoded via C<FREEZE> can be decoded using the following
		739	protocol:
488		740
489	When an encoded CBOR perl object is encountered by the decoder, it will	741	When an encoded CBOR perl object is encountered by the decoder, it will
490	look up the C<THAW> method, by using the stored classname, and will fail	742	look up the C<THAW> method, by using the stored classname, and will fail
491	if the method cannot be found.	743	if the method cannot be found.
492		744
493	After the lookup it will call the C<THAW> method with the stored classname	745	After the lookup it will call the C<THAW> method with the stored classname
494	as first argument, the constant string C<CBOR> as second argument, and all	746	as first argument, the constant string C<CBOR> as second argument, and all
495	values returned by C<FREEZE> as remaining arguments.	747	values returned by C<FREEZE> as remaining arguments.
496		748
497	=head4 EXAMPLES	749	=head3 EXAMPLES
498		750
499	Here is an example C<TO_CBOR> method:	751	Here is an example C<TO_CBOR> method:
500		752
501	sub My::Object::TO_CBOR {	753	sub My::Object::TO_CBOR {
502	my ($obj) = @_;	754	my ($obj) = @_;
…		…
513		765
514	sub URI::TO_CBOR {	766	sub URI::TO_CBOR {
515	my ($self) = @_;	767	my ($self) = @_;
516	my $uri = "$self"; # stringify uri	768	my $uri = "$self"; # stringify uri
517	utf8::upgrade $uri; # make sure it will be encoded as UTF-8 string	769	utf8::upgrade $uri; # make sure it will be encoded as UTF-8 string
518	CBOR::XS::tagged 32, "$_[0]"	770	CBOR::XS::tag 32, "$_[0]"
519	}	771	}
520		772
521	This will encode URIs as a UTF-8 string with tag 32, which indicates an	773	This will encode URIs as a UTF-8 string with tag 32, which indicates an
522	URI.	774	URI.
523		775
…		…
534	"$self" # encode url string	786	"$self" # encode url string
535	}	787	}
536		788
537	sub URI::THAW {	789	sub URI::THAW {
538	my ($class, $serialiser, $uri) = @_;	790	my ($class, $serialiser, $uri) = @_;
539
540	$class->new ($uri)	791	$class->new ($uri)
541	}	792	}
542		793
543	Unlike C<TO_CBOR>, multiple values can be returned by C<FREEZE>. For	794	Unlike C<TO_CBOR>, multiple values can be returned by C<FREEZE>. For
544	example, a C<FREEZE> method that returns "type", "id" and "variant" values	795	example, a C<FREEZE> method that returns "type", "id" and "variant" values
…		…
675	additional tags (such as base64url).	926	additional tags (such as base64url).
676		927
677	=head2 ENFORCED TAGS	928	=head2 ENFORCED TAGS
678		929
679	These tags are always handled when decoding, and their handling cannot be	930	These tags are always handled when decoding, and their handling cannot be
680	overriden by the user.	931	overridden by the user.
681		932
682	=over 4	933	=over 4
683		934
684	=item <unassigned> (perl-object, L<http://cbor.schmorp.de/perl-object>)	935	=item 26 (perl-object, L<http://cbor.schmorp.de/perl-object>)
685		936
686	These tags are automatically created (and decoded) for serialisable	937	These tags are automatically created (and decoded) for serialisable
687	objects using the C<FREEZE/THAW> methods (the L<Types::Serialier> object	938	objects using the C<FREEZE/THAW> methods (the L<Types::Serialier> object
688	serialisation protocol). See L<OBJECT SERIALISATION> for details.	939	serialisation protocol). See L<OBJECT SERIALISATION> for details.
689		940
690	=item <unassigned>, <unassigned> (sharable, sharedref, L <http://cbor.schmorp.de/value-sharing>)	941	=item 28, 29 (shareable, sharedref, L<http://cbor.schmorp.de/value-sharing>)
691		942
692	These tags are automatically decoded when encountered, resulting in	943	These tags are automatically decoded when encountered (and they do not
		944	result in a cyclic data structure, see C<allow_cycles>), resulting in
693	shared values in the decoded object. They are only encoded, however, when	945	shared values in the decoded object. They are only encoded, however, when
694	C<allow_sharable> is enabled.	946	C<allow_sharing> is enabled.
695		947
		948	Not all shared values can be successfully decoded: values that reference
		949	themselves will I<currently> decode as C<undef> (this is not the same
		950	as a reference pointing to itself, which will be represented as a value
		951	that contains an indirect reference to itself - these will be decoded
		952	properly).
		953
		954	Note that considerably more shared value data structures can be decoded
		955	than will be encoded - currently, only values pointed to by references
		956	will be shared, others will not. While non-reference shared values can be
		957	generated in Perl with some effort, they were considered too unimportant
		958	to be supported in the encoder. The decoder, however, will decode these
		959	values as shared values.
		960
696	=item <unassigned>, <unassigned> (stringref-namespace, stringref, L <http://cbor.schmorp.de/stringref>)	961	=item 256, 25 (stringref-namespace, stringref, L<http://cbor.schmorp.de/stringref>)
697		962
698	These tags are automatically decoded when encountered. They are only	963	These tags are automatically decoded when encountered. They are only
699	encoded, however, when C<pack_strings> is enabled.	964	encoded, however, when C<pack_strings> is enabled.
700		965
701	=item 22098 (indirection, L<http://cbor.schmorp.de/indirection>)	966	=item 22098 (indirection, L<http://cbor.schmorp.de/indirection>)
702		967
703	This tag is automatically generated when a reference are encountered (with	968	This tag is automatically generated when a reference are encountered (with
704	the exception of hash and array refernces). It is converted to a reference	969	the exception of hash and array references). It is converted to a reference
705	when decoding.	970	when decoding.
706		971
707	=item 55799 (self-describe CBOR, RFC 7049)	972	=item 55799 (self-describe CBOR, RFC 7049)
708		973
709	This value is not generated on encoding (unless explicitly requested by	974	This value is not generated on encoding (unless explicitly requested by
…		…
712	=back	977	=back
713		978
714	=head2 NON-ENFORCED TAGS	979	=head2 NON-ENFORCED TAGS
715		980
716	These tags have default filters provided when decoding. Their handling can	981	These tags have default filters provided when decoding. Their handling can
717	be overriden by changing the C<%CBOR::XS::FILTER> entry for the tag, or by	982	be overridden by changing the C<%CBOR::XS::FILTER> entry for the tag, or by
718	providing a custom C<filter> callback when decoding.	983	providing a custom C<filter> callback when decoding.
719		984
720	When they result in decoding into a specific Perl class, the module	985	When they result in decoding into a specific Perl class, the module
721	usually provides a corresponding C<TO_CBOR> method as well.	986	usually provides a corresponding C<TO_CBOR> method as well.
722		987
…		…
725	provide these modules. The decoding usually fails with an exception if the	990	provide these modules. The decoding usually fails with an exception if the
726	required module cannot be loaded.	991	required module cannot be loaded.
727		992
728	=over 4	993	=over 4
729		994
		995	=item 0, 1 (date/time string, seconds since the epoch)
		996
		997	These tags are decoded into L<Time::Piece> objects. The corresponding
		998	C<Time::Piece::TO_CBOR> method always encodes into tag 1 values currently.
		999
		1000	The L<Time::Piece> API is generally surprisingly bad, and fractional
		1001	seconds are only accidentally kept intact, so watch out. On the plus side,
		1002	the module comes with perl since 5.10, which has to count for something.
		1003
730	=item 2, 3 (positive/negative bignum)	1004	=item 2, 3 (positive/negative bignum)
731		1005
732	These tags are decoded into L<Math::BigInt> objects. The corresponding	1006	These tags are decoded into L<Math::BigInt> objects. The corresponding
733	C<Math::BigInt::TO_CBOR> method encodes "small" bigints into normal CBOR	1007	C<Math::BigInt::TO_CBOR> method encodes "small" bigints into normal CBOR
734	integers, and others into positive/negative CBOR bignums.	1008	integers, and others into positive/negative CBOR bignums.
735		1009
736	=item 4, 5 (decimal fraction/bigfloat)	1010	=item 4, 5, 264, 265 (decimal fraction/bigfloat)
737		1011
738	Both decimal fractions and bigfloats are decoded into L<Math::BigFloat>	1012	Both decimal fractions and bigfloats are decoded into L<Math::BigFloat>
739	objects. The corresponding C<Math::BigFloat::TO_CBOR> method I<always>	1013	objects. The corresponding C<Math::BigFloat::TO_CBOR> method I<always>
740	encodes into a decimal fraction.	1014	encodes into a decimal fraction (either tag 4 or 264).
741		1015
742	CBOR cannot represent bigfloats with I<very> large exponents - conversion	1016	NaN and infinities are not encoded properly, as they cannot be represented
743	of such big float objects is undefined.	1017	in CBOR.
744		1018
745	Also, NaN and infinities are not encoded properly.	1019	See L<BIGNUM SECURITY CONSIDERATIONS> for more info.
		1020
		1021	=item 30 (rational numbers)
		1022
		1023	These tags are decoded into L<Math::BigRat> objects. The corresponding
		1024	C<Math::BigRat::TO_CBOR> method encodes rational numbers with denominator
		1025	C<1> via their numerator only, i.e., they become normal integers or
		1026	C<bignums>.
		1027
		1028	See L<BIGNUM SECURITY CONSIDERATIONS> for more info.
746		1029
747	=item 21, 22, 23 (expected later JSON conversion)	1030	=item 21, 22, 23 (expected later JSON conversion)
748		1031
749	CBOR::XS is not a CBOR-to-JSON converter, and will simply ignore these	1032	CBOR::XS is not a CBOR-to-JSON converter, and will simply ignore these
750	tags.	1033	tags.
…		…
755	C<URI::TO_CBOR> method again results in a CBOR URI value.	1038	C<URI::TO_CBOR> method again results in a CBOR URI value.
756		1039
757	=back	1040	=back
758		1041
759	=cut	1042	=cut
760
761	our %FILTER = (
762	# 0 # rfc4287 datetime, utf-8
763	# 1 # unix timestamp, any
764
765	2 => sub { # pos bigint
766	require Math::BigInt;
767	Math::BigInt->new ("0x" . unpack "H*", pop)
768	},
769
770	3 => sub { # neg bigint
771	require Math::BigInt;
772	-Math::BigInt->new ("0x" . unpack "H*", pop)
773	},
774
775	4 => sub { # decimal fraction, array
776	require Math::BigFloat;
777	Math::BigFloat->new ($_[1][1] . "E" . $_[1][0])
778	},
779
780	5 => sub { # bigfloat, array
781	require Math::BigFloat;
782	scalar Math::BigFloat->new ($_[1][1])->blsft ($_[1][0], 2)
783	},
784
785	21 => sub { pop }, # expected conversion to base64url encoding
786	22 => sub { pop }, # expected conversion to base64 encoding
787	23 => sub { pop }, # expected conversion to base16 encoding
788
789	# 24 # embedded cbor, byte string
790
791	32 => sub {
792	require URI;
793	URI->new (pop)
794	},
795
796	# 33 # base64url rfc4648, utf-8
797	# 34 # base64 rfc46484, utf-8
798	# 35 # regex pcre/ecma262, utf-8
799	# 36 # mime message rfc2045, utf-8
800	);
801
802		1043
803	=head1 CBOR and JSON	1044	=head1 CBOR and JSON
804		1045
805	CBOR is supposed to implement a superset of the JSON data model, and is,	1046	CBOR is supposed to implement a superset of the JSON data model, and is,
806	with some coercion, able to represent all JSON texts (something that other	1047	with some coercion, able to represent all JSON texts (something that other
…		…
815	CBOR intact.	1056	CBOR intact.
816		1057
817		1058
818	=head1 SECURITY CONSIDERATIONS	1059	=head1 SECURITY CONSIDERATIONS
819		1060
820	When you are using CBOR in a protocol, talking to untrusted potentially	1061	Tl;dr... if you want to decode or encode CBOR from untrusted sources, you
821	hostile creatures requires relatively few measures.	1062	should start with a coder object created via C<new_safe>:
822		1063
		1064	my $coder = CBOR::XS->new_safe;
		1065
		1066	my $data = $coder->decode ($cbor_text);
		1067	my $cbor = $coder->encode ($data);
		1068
		1069	Longer version: When you are using CBOR in a protocol, talking to
		1070	untrusted potentially hostile creatures requires some thought:
		1071
		1072	=over 4
		1073
		1074	=item Security of the CBOR decoder itself
		1075
823	First of all, your CBOR decoder should be secure, that is, should not have	1076	First and foremost, your CBOR decoder should be secure, that is, should
		1077	not have any buffer overflows or similar bugs that could potentially be
824	any buffer overflows. Obviously, this module should ensure that and I am	1078	exploited. Obviously, this module should ensure that and I am trying hard
825	trying hard on making that true, but you never know.	1079	on making that true, but you never know.
826		1080
		1081	=item CBOR::XS can invoke almost arbitrary callbacks during decoding
		1082
		1083	CBOR::XS supports object serialisation - decoding CBOR can cause calls
		1084	to I<any> C<THAW> method in I<any> package that exists in your process
		1085	(that is, CBOR::XS will not try to load modules, but any existing C<THAW>
		1086	method or function can be called, so they all have to be secure).
		1087
		1088	Less obviously, it will also invoke C<TO_CBOR> and C<FREEZE> methods -
		1089	even if all your C<THAW> methods are secure, encoding data structures from
		1090	untrusted sources can invoke those and trigger bugs in those.
		1091
		1092	So, if you are not sure about the security of all the modules you
		1093	have loaded (you shouldn't), you should disable this part using
		1094	C<forbid_objects>.
		1095
		1096	=item CBOR can be extended with tags that call library code
		1097
		1098	CBOR can be extended with tags, and C<CBOR::XS> has a registry of
		1099	conversion functions for many existing tags that can be extended via
		1100	third-party modules (see the C<filter> method).
		1101
		1102	If you don't trust these, you should configure the "safe" filter function,
		1103	C<CBOR::XS::safe_filter>, which by default only includes conversion
		1104	functions that are considered "safe" by the author (but again, they can be
		1105	extended by third party modules).
		1106
		1107	Depending on your level of paranoia, you can use the "safe" filter:
		1108
		1109	$cbor->filter (\&CBOR::XS::safe_filter);
		1110
		1111	... your own filter...
		1112
		1113	$cbor->filter (sub { ... do your stuffs here ... });
		1114
		1115	... or even no filter at all, disabling all tag decoding:
		1116
		1117	$cbor->filter (sub { });
		1118
		1119	This is never a problem for encoding, as the tag mechanism only exists in
		1120	CBOR texts.
		1121
		1122	=item Resource-starving attacks: object memory usage
		1123
827	Second, you need to avoid resource-starving attacks. That means you should	1124	You need to avoid resource-starving attacks. That means you should limit
828	limit the size of CBOR data you accept, or make sure then when your	1125	the size of CBOR data you accept, or make sure then when your resources
829	resources run out, that's just fine (e.g. by using a separate process that	1126	run out, that's just fine (e.g. by using a separate process that can
830	can crash safely). The size of a CBOR string in octets is usually a good	1127	crash safely). The size of a CBOR string in octets is usually a good
831	indication of the size of the resources required to decode it into a Perl	1128	indication of the size of the resources required to decode it into a Perl
832	structure. While CBOR::XS can check the size of the CBOR text, it might be	1129	structure. While CBOR::XS can check the size of the CBOR text (using
833	too late when you already have it in memory, so you might want to check	1130	C<max_size>), it might be too late when you already have it in memory, so
834	the size before you accept the string.	1131	you might want to check the size before you accept the string.
835		1132
		1133	As for encoding, it is possible to construct data structures that are
		1134	relatively small but result in large CBOR texts (for example by having an
		1135	array full of references to the same big data structure, which will all be
		1136	deep-cloned during encoding by default). This is rarely an actual issue
		1137	(and the worst case is still just running out of memory), but you can
		1138	reduce this risk by using C<allow_sharing>.
		1139
		1140	=item Resource-starving attacks: stack overflows
		1141
836	Third, CBOR::XS recurses using the C stack when decoding objects and	1142	CBOR::XS recurses using the C stack when decoding objects and arrays. The
837	arrays. The C stack is a limited resource: for instance, on my amd64	1143	C stack is a limited resource: for instance, on my amd64 machine with 8MB
838	machine with 8MB of stack size I can decode around 180k nested arrays but	1144	of stack size I can decode around 180k nested arrays but only 14k nested
839	only 14k nested CBOR objects (due to perl itself recursing deeply on croak	1145	CBOR objects (due to perl itself recursing deeply on croak to free the
840	to free the temporary). If that is exceeded, the program crashes. To be	1146	temporary). If that is exceeded, the program crashes. To be conservative,
841	conservative, the default nesting limit is set to 512. If your process	1147	the default nesting limit is set to 512. If your process has a smaller
842	has a smaller stack, you should adjust this setting accordingly with the	1148	stack, you should adjust this setting accordingly with the C<max_depth>
843	C<max_depth> method.	1149	method.
		1150
		1151	=item Resource-starving attacks: CPU en-/decoding complexity
		1152
		1153	CBOR::XS will use the L<Math::BigInt>, L<Math::BigFloat> and
		1154	L<Math::BigRat> libraries to represent encode/decode bignums. These can
		1155	be very slow (as in, centuries of CPU time) and can even crash your
		1156	program (and are generally not very trustworthy). See the next section for
		1157	details.
		1158
		1159	=item Data breaches: leaking information in error messages
		1160
		1161	CBOR::XS might leak contents of your Perl data structures in its error
		1162	messages, so when you serialise sensitive information you might want to
		1163	make sure that exceptions thrown by CBOR::XS will not end up in front of
		1164	untrusted eyes.
		1165
		1166	=item Something else...
844		1167
845	Something else could bomb you, too, that I forgot to think of. In that	1168	Something else could bomb you, too, that I forgot to think of. In that
846	case, you get to keep the pieces. I am always open for hints, though...	1169	case, you get to keep the pieces. I am always open for hints, though...
847		1170
848	Also keep in mind that CBOR::XS might leak contents of your Perl data	1171	=back
849	structures in its error messages, so when you serialise sensitive	1172
850	information you might want to make sure that exceptions thrown by CBOR::XS	1173
851	will not end up in front of untrusted eyes.	1174	=head1 BIGNUM SECURITY CONSIDERATIONS
		1175
		1176	CBOR::XS provides a C<TO_CBOR> method for both L<Math::BigInt> and
		1177	L<Math::BigFloat> that tries to encode the number in the simplest possible
		1178	way, that is, either a CBOR integer, a CBOR bigint/decimal fraction (tag
		1179	4) or an arbitrary-exponent decimal fraction (tag 264). Rational numbers
		1180	(L<Math::BigRat>, tag 30) can also contain bignums as members.
		1181
		1182	CBOR::XS will also understand base-2 bigfloat or arbitrary-exponent
		1183	bigfloats (tags 5 and 265), but it will never generate these on its own.
		1184
		1185	Using the built-in L<Math::BigInt::Calc> support, encoding and decoding
		1186	decimal fractions is generally fast. Decoding bigints can be slow for very
		1187	big numbers (tens of thousands of digits, something that could potentially
		1188	be caught by limiting the size of CBOR texts), and decoding bigfloats or
		1189	arbitrary-exponent bigfloats can be I<extremely> slow (minutes, decades)
		1190	for large exponents (roughly 40 bit and longer).
		1191
		1192	Additionally, L<Math::BigInt> can take advantage of other bignum
		1193	libraries, such as L<Math::GMP>, which cannot handle big floats with large
		1194	exponents, and might simply abort or crash your program, due to their code
		1195	quality.
		1196
		1197	This can be a concern if you want to parse untrusted CBOR. If it is, you
		1198	might want to disable decoding of tag 2 (bigint) and 3 (negative bigint)
		1199	types. You should also disable types 5 and 265, as these can be slow even
		1200	without bigints.
		1201
		1202	Disabling bigints will also partially or fully disable types that rely on
		1203	them, e.g. rational numbers that use bignums.
		1204
852		1205
853	=head1 CBOR IMPLEMENTATION NOTES	1206	=head1 CBOR IMPLEMENTATION NOTES
854		1207
855	This section contains some random implementation notes. They do not	1208	This section contains some random implementation notes. They do not
856	describe guaranteed behaviour, but merely behaviour as-is implemented	1209	describe guaranteed behaviour, but merely behaviour as-is implemented
…		…
865	Only the double data type is supported for NV data types - when Perl uses	1218	Only the double data type is supported for NV data types - when Perl uses
866	long double to represent floating point values, they might not be encoded	1219	long double to represent floating point values, they might not be encoded
867	properly. Half precision types are accepted, but not encoded.	1220	properly. Half precision types are accepted, but not encoded.
868		1221
869	Strict mode and canonical mode are not implemented.	1222	Strict mode and canonical mode are not implemented.
		1223
		1224
		1225	=head1 LIMITATIONS ON PERLS WITHOUT 64-BIT INTEGER SUPPORT
		1226
		1227	On perls that were built without 64 bit integer support (these are rare
		1228	nowadays, even on 32 bit architectures, as all major Perl distributions
		1229	are built with 64 bit integer support), support for any kind of 64 bit
		1230	integer in CBOR is very limited - most likely, these 64 bit values will
		1231	be truncated, corrupted, or otherwise not decoded correctly. This also
		1232	includes string, array and map sizes that are stored as 64 bit integers.
870		1233
871		1234
872	=head1 THREADS	1235	=head1 THREADS
873		1236
874	This module is I<not> guaranteed to be thread safe and there are no	1237	This module is I<not> guaranteed to be thread safe and there are no
…		…
888	Please refrain from using rt.cpan.org or any other bug reporting	1251	Please refrain from using rt.cpan.org or any other bug reporting
889	service. I put the contact address into my modules for a reason.	1252	service. I put the contact address into my modules for a reason.
890		1253
891	=cut	1254	=cut
892		1255
		1256	# clumsy and slow hv_store-in-hash helper function
		1257	sub _hv_store {
		1258	$_[0]{$_[1]} = $_[2];
		1259	}
		1260
893	our %FILTER = (	1261	our %FILTER = (
894	# 0 # rfc4287 datetime, utf-8	1262	0 => sub { # rfc4287 datetime, utf-8
895	# 1 # unix timestamp, any	1263	require Time::Piece;
		1264	# Time::Piece::Strptime uses the "incredibly flexible date parsing routine"
		1265	# from FreeBSD, which can't parse ISO 8601, RFC3339, RFC4287 or much of anything
		1266	# else either. Whats incredibe over standard strptime totally escapes me.
		1267	# doesn't do fractional times, either. sigh.
		1268	# In fact, it's all a lie, it uses whatever strptime it wants, and of course,
		1269	# they are all incompatible. The openbsd one simply ignores %z (but according to the
		1270	# docs, it would be much more incredibly flexible indeed. If it worked, that is.).
		1271	scalar eval {
		1272	my $s = $_[1];
		1273
		1274	$s =~ s/Z$/+00:00/;
		1275	$s =~ s/(\.[0-9]+)?([+-][0-9][0-9]):([0-9][0-9])$//
		1276	or die;
		1277
		1278	my $b = $1 - ($2 * 60 + $3) * 60; # fractional part + offset. hopefully
		1279	my $d = Time::Piece->strptime ($s, "%Y-%m-%dT%H:%M:%S");
		1280
		1281	Time::Piece::gmtime ($d->epoch + $b)
		1282	} \|\| die "corrupted CBOR date/time string ($_[0])";
		1283	},
		1284
		1285	1 => sub { # seconds since the epoch, possibly fractional
		1286	require Time::Piece;
		1287	scalar Time::Piece::gmtime (pop)
		1288	},
896		1289
897	2 => sub { # pos bigint	1290	2 => sub { # pos bigint
898	require Math::BigInt;	1291	require Math::BigInt;
899	Math::BigInt->new ("0x" . unpack "H*", pop)	1292	Math::BigInt->new ("0x" . unpack "H*", pop)
900	},	1293	},
…		…
907	4 => sub { # decimal fraction, array	1300	4 => sub { # decimal fraction, array
908	require Math::BigFloat;	1301	require Math::BigFloat;
909	Math::BigFloat->new ($_[1][1] . "E" . $_[1][0])	1302	Math::BigFloat->new ($_[1][1] . "E" . $_[1][0])
910	},	1303	},
911		1304
		1305	264 => sub { # decimal fraction with arbitrary exponent
		1306	require Math::BigFloat;
		1307	Math::BigFloat->new ($_[1][1] . "E" . $_[1][0])
		1308	},
		1309
912	5 => sub { # bigfloat, array	1310	5 => sub { # bigfloat, array
913	require Math::BigFloat;	1311	require Math::BigFloat;
914	scalar Math::BigFloat->new ($_[1][1])->blsft ($_[1][0], 2)	1312	scalar Math::BigFloat->new ($_[1][1]) * Math::BigFloat->new (2)->bpow ($_[1][0])
		1313	},
		1314
		1315	265 => sub { # bigfloat with arbitrary exponent
		1316	require Math::BigFloat;
		1317	scalar Math::BigFloat->new ($_[1][1]) * Math::BigFloat->new (2)->bpow ($_[1][0])
		1318	},
		1319
		1320	30 => sub { # rational number
		1321	require Math::BigRat;
		1322	Math::BigRat->new ("$_[1][0]/$_[1][1]") # separate parameters only work in recent versons
915	},	1323	},
916		1324
917	21 => sub { pop }, # expected conversion to base64url encoding	1325	21 => sub { pop }, # expected conversion to base64url encoding
918	22 => sub { pop }, # expected conversion to base64 encoding	1326	22 => sub { pop }, # expected conversion to base64 encoding
919	23 => sub { pop }, # expected conversion to base16 encoding	1327	23 => sub { pop }, # expected conversion to base16 encoding
…		…
929	# 34 # base64 rfc46484, utf-8	1337	# 34 # base64 rfc46484, utf-8
930	# 35 # regex pcre/ecma262, utf-8	1338	# 35 # regex pcre/ecma262, utf-8
931	# 36 # mime message rfc2045, utf-8	1339	# 36 # mime message rfc2045, utf-8
932	);	1340	);
933		1341
934	sub CBOR::XS::default_filter {	1342	sub default_filter {
935	&{ $FILTER{$_[0]} or return }	1343	&{ $FILTER{$_[0]} or return }
		1344	}
		1345
		1346	our %SAFE_FILTER = map { $_ => $FILTER{$_} } 0, 1, 21, 22, 23, 32;
		1347
		1348	sub safe_filter {
		1349	&{ $SAFE_FILTER{$_[0]} or return }
936	}	1350	}
937		1351
938	sub URI::TO_CBOR {	1352	sub URI::TO_CBOR {
939	my $uri = $_[0]->as_string;	1353	my $uri = $_[0]->as_string;
940	utf8::upgrade $uri;	1354	utf8::upgrade $uri;
941	CBOR::XS::tag 32, $uri	1355	tag 32, $uri
942	}	1356	}
943		1357
944	sub Math::BigInt::TO_CBOR {	1358	sub Math::BigInt::TO_CBOR {
945	if ($_[0] >= -2147483648 && $_[0] <= 2147483647) {	1359	if (-2147483648 <= $_[0] && $_[0] <= 2147483647) {
946	$_[0]->numify	1360	$_[0]->numify
947	} else {	1361	} else {
948	my $hex = substr $_[0]->as_hex, 2;	1362	my $hex = substr $_[0]->as_hex, 2;
949	$hex = "0$hex" if 1 & length $hex; # sigh	1363	$hex = "0$hex" if 1 & length $hex; # sigh
950	CBOR::XS::tag $_[0] >= 0 ? 2 : 3, pack "H*", $hex	1364	tag $_[0] >= 0 ? 2 : 3, pack "H*", $hex
951	}	1365	}
952	}	1366	}
953		1367
954	sub Math::BigFloat::TO_CBOR {	1368	sub Math::BigFloat::TO_CBOR {
955	my ($m, $e) = $_[0]->parts;	1369	my ($m, $e) = $_[0]->parts;
		1370
		1371	-9223372036854775808 <= $e && $e <= 18446744073709551615
956	CBOR::XS::tag 4, [$e->numify, $m]	1372	? tag 4, [$e->numify, $m]
		1373	: tag 264, [$e, $m]
		1374	}
		1375
		1376	sub Math::BigRat::TO_CBOR {
		1377	my ($n, $d) = $_[0]->parts;
		1378
		1379	# older versions of BigRat need *1, as they not always return numbers
		1380
		1381	$d*1 == 1
		1382	? $n*1
		1383	: tag 30, [$n1, $d1]
		1384	}
		1385
		1386	sub Time::Piece::TO_CBOR {
		1387	tag 1, 0 + $_[0]->epoch
957	}	1388	}
958		1389
959	XSLoader::load "CBOR::XS", $VERSION;	1390	XSLoader::load "CBOR::XS", $VERSION;
960		1391
961	=head1 SEE ALSO	1392	=head1 SEE ALSO

Diff Legend

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

Comparing CBOR-XS/XS.pm (file contents): Revision 1.25 by root, Thu Nov 28 12:08:07 2013 UTC vs. Revision 1.66 by root, Tue Jun 27 02:03:24 2017 UTC

Diff Legend

Comparing CBOR-XS/XS.pm (file contents):
Revision 1.25 by root, Thu Nov 28 12:08:07 2013 UTC vs.
Revision 1.66 by root, Tue Jun 27 02:03:24 2017 UTC