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

Comparing CBOR-XS/XS.pm (file contents):
Revision 1.48 by root, Thu Feb 25 14:22:49 2016 UTC vs.
Revision 1.70 by root, Sat Nov 9 07:30:36 2019 UTC

…		…
38	with the added ability of supporting serialisation of Perl objects. (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	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	40	data later and speed is less important you might want to compare both
41	formats first).	41	formats first).
42		42
		43	The primary goal of this module is to be I<correct> and the secondary goal
		44	is to be I<fast>. To reach the latter goal it was written in C.
		45
43	To give you a general idea about speed, with texts in the megabyte range,	46	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	47	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	48	L<JSON::XS> and decodes about 15%-30% faster than those. The shorter the
46	data, the worse L<Storable> performs in comparison.	49	data, the worse L<Storable> performs in comparison.
47		50
…		…
52	In addition to the core CBOR data format, this module implements a	55	In addition to the core CBOR data format, this module implements a
53	number of extensions, to support cyclic and shared data structures	56	number of extensions, to support cyclic and shared data structures
54	(see C<allow_sharing> and C<allow_cycles>), string deduplication (see	57	(see C<allow_sharing> and C<allow_cycles>), string deduplication (see
55	C<pack_strings>) and scalar references (always enabled).	58	C<pack_strings>) and scalar references (always enabled).
56		59
57	The primary goal of this module is to be I<correct> and the secondary goal
58	is to be I<fast>. To reach the latter goal it was written in C.
59
60	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
61	vice versa.	61	vice versa.
62		62
63	=cut	63	=cut
64		64
65	package CBOR::XS;	65	package CBOR::XS;
66		66
67	use common::sense;	67	use common::sense;
68		68
69	our $VERSION = 1.41;	69	our $VERSION = 1.71;
70	our @ISA = qw(Exporter);	70	our @ISA = qw(Exporter);
71		71
72	our @EXPORT = qw(encode_cbor decode_cbor);	72	our @EXPORT = qw(encode_cbor decode_cbor);
73		73
74	use Exporter;	74	use Exporter;
…		…
112		112
113	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
114	be chained:	114	be chained:
115		115
116	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	}
117		142
118	=item $cbor = $cbor->max_depth ([$maximum_nesting_depth])	143	=item $cbor = $cbor->max_depth ([$maximum_nesting_depth])
119		144
120	=item $max_depth = $cbor->get_max_depth	145	=item $max_depth = $cbor->get_max_depth
121		146
…		…
137		162
138	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
139	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
140	crashing.	165	crashing.
141		166
142	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.
143		168
144	=item $cbor = $cbor->max_size ([$maximum_string_size])	169	=item $cbor = $cbor->max_size ([$maximum_string_size])
145		170
146	=item $max_size = $cbor->get_max_size	171	=item $max_size = $cbor->get_max_size
147		172
…		…
152	effect on C<encode> (yet).	177	effect on C<encode> (yet).
153		178
154	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
155	C<0> is specified).	180	C<0> is specified).
156		181
157	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.
158		183
159	=item $cbor = $cbor->allow_unknown ([$enable])	184	=item $cbor = $cbor->allow_unknown ([$enable])
160		185
161	=item $enabled = $cbor->get_allow_unknown	186	=item $enabled = $cbor->get_allow_unknown
162		187
…		…
180	reference to the earlier value.	205	reference to the earlier value.
181		206
182	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
183	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
184	sharing extension. This also makes it possible to encode cyclic data	209	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	210	structures (which need C<allow_cycles> to be enabled to be decoded by this
186	module).	211	module).
187		212
188	It is recommended to leave it off unless you know your	213	It is recommended to leave it off unless you know your
189	communication partner supports the value sharing extensions to CBOR	214	communication partner supports the value sharing extensions to CBOR
190	(L<http://cbor.schmorp.de/value-sharing>), as without decoder support, the	215	(L<http://cbor.schmorp.de/value-sharing>), as without decoder support, the
191	resulting data structure might be unusable.	216	resulting data structure might be unusable.
192		217
193	Detecting shared values incurs a runtime overhead when values are encoded	218	Detecting shared values incurs a runtime overhead when values are encoded
194	that have a reference counter large than one, and might unnecessarily	219	that have a reference counter large than one, and might unnecessarily
195	increase the encoded size, as potentially shared values are encode as	220	increase the encoded size, as potentially shared values are encoded as
196	shareable whether or not they are actually shared.	221	shareable whether or not they are actually shared.
197		222
198	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,
199	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
200	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
…		…
226	throwing an error.	251	throwing an error.
227		252
228	This option does not affect C<encode> in any way - shared values and	253	This option does not affect C<encode> in any way - shared values and
229	references will always be encoded properly if present.	254	references will always be encoded properly if present.
230		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
231	=item $cbor = $cbor->pack_strings ([$enable])	275	=item $cbor = $cbor->pack_strings ([$enable])
232		276
233	=item $enabled = $cbor->get_pack_strings	277	=item $enabled = $cbor->get_pack_strings
234		278
235	If C<$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
…		…
247	the standard CBOR way.	291	the standard CBOR way.
248		292
249	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
250	always be decoded properly if present.	294	always be decoded properly if present.
251		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
252	=item $cbor = $cbor->validate_utf8 ([$enable])	340	=item $cbor = $cbor->validate_utf8 ([$enable])
253		341
254	=item $enabled = $cbor->get_validate_utf8	342	=item $enabled = $cbor->get_validate_utf8
255		343
256	If C<$enable> is true (or missing), then C<decode> will validate that	344	If C<$enable> is true (or missing), then C<decode> will validate that
…		…
261	The concept of "valid UTF-8" used is perl's concept, which is a superset	349	The concept of "valid UTF-8" used is perl's concept, which is a superset
262	of the official UTF-8.	350	of the official UTF-8.
263		351
264	If C<$enable> is false (the default), then C<decode> will blindly accept	352	If C<$enable> is false (the default), then C<decode> will blindly accept
265	UTF-8 data, marking them as valid UTF-8 in the resulting data structure	353	UTF-8 data, marking them as valid UTF-8 in the resulting data structure
266	regardless of whether thats true or not.	354	regardless of whether that's true or not.
267		355
268	Perl isn't too happy about corrupted UTF-8 in strings, but should	356	Perl isn't too happy about corrupted UTF-8 in strings, but should
269	generally not crash or do similarly evil things. Extensions might be not	357	generally not crash or do similarly evil things. Extensions might be not
270	so forgiving, so it's recommended to turn on this setting if you receive	358	so forgiving, so it's recommended to turn on this setting if you receive
271	untrusted CBOR.	359	untrusted CBOR.
…		…
293	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,
294	which will result in default handling, which currently means the decoder	382	which will result in default handling, which currently means the decoder
295	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.
296		384
297	When the filter is cleared (the default state), the default filter	385	When the filter is cleared (the default state), the default filter
298	function, C<CBOR::XS::default_filter>, is used. This function simply looks	386	function, C<CBOR::XS::default_filter>, is used. This function simply
299	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
300	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
301	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.
		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.
302		403
303	Example: decode all tags not handled internally into C<CBOR::XS::Tagged>	404	Example: decode all tags not handled internally into C<CBOR::XS::Tagged>
304	objects, with no other special handling (useful when working with	405	objects, with no other special handling (useful when working with
305	potentially "unsafe" CBOR data).	406	potentially "unsafe" CBOR data).
306		407
…		…
313	my ($tag, $value);	414	my ($tag, $value);
314		415
315	"tag 1347375694 value $value"	416	"tag 1347375694 value $value"
316	};	417	};
317		418
		419	Example: provide your own filter function that looks up tags in your own
		420	hash:
		421
		422	my %my_filter = (
		423	998347484 => sub {
		424	my ($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);
		439
318	=item $cbor_data = $cbor->encode ($perl_scalar)	440	=item $cbor_data = $cbor->encode ($perl_scalar)
319		441
320	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
321	representation.	443	representation.
322		444
…		…
331	when there is trailing garbage after the CBOR string, it will silently	453	when there is trailing garbage after the CBOR string, it will silently
332	stop parsing there and return the number of characters consumed so far.	454	stop parsing there and return the number of characters consumed so far.
333		455
334	This is useful if your CBOR texts are not delimited by an outer protocol	456	This is useful if your CBOR texts are not delimited by an outer protocol
335	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
336	starts.	458	starts - CBOR strings are self-delimited, so it is possible to concatenate
		459	CBOR strings without any delimiters or size fields and recover their data.
337		460
338	CBOR::XS->new->decode_prefix ("......")	461	CBOR::XS->new->decode_prefix ("......")
339	=> ("...", 3)	462	=> ("...", 3)
340		463
341	=back	464	=back
…		…
396		519
397	Resets the incremental decoder. This throws away any saved state, so that	520	Resets the incremental decoder. This throws away any saved state, so that
398	subsequent calls to C<incr_parse> or C<incr_parse_multiple> start to parse	521	subsequent calls to C<incr_parse> or C<incr_parse_multiple> start to parse
399	a new CBOR value from the beginning of the C<$buffer> again.	522	a new CBOR value from the beginning of the C<$buffer> again.
400		523
401	This method can be caled at any time, but it I<must> be called if you want	524	This method can be called at any time, but it I<must> be called if you want
402	to change your C<$buffer> or there was a decoding error and you want to	525	to change your C<$buffer> or there was a decoding error and you want to
403	reuse the C<$cbor> object for future incremental parsings.	526	reuse the C<$cbor> object for future incremental parsings.
404		527
405	=back	528	=back
406		529
…		…
481		604
482	=item hash references	605	=item hash references
483		606
484	Perl hash references become CBOR maps. As there is no inherent ordering in	607	Perl hash references become CBOR maps. As there is no inherent ordering in
485	hash keys (or CBOR maps), they will usually be encoded in a pseudo-random	608	hash keys (or CBOR maps), they will usually be encoded in a pseudo-random
486	order. This order can be different each time a hahs is encoded.	609	order. This order can be different each time a hash is encoded.
487		610
488	Currently, tied hashes will use the indefinite-length format, while normal	611	Currently, tied hashes will use the indefinite-length format, while normal
489	hashes will use the fixed-length format.	612	hashes will use the fixed-length format.
490		613
491	=item array references	614	=item array references
…		…
544	my $x = 3.1; # some variable containing a number	667	my $x = 3.1; # some variable containing a number
545	"$x"; # stringified	668	"$x"; # stringified
546	$x .= ""; # another, more awkward way to stringify	669	$x .= ""; # another, more awkward way to stringify
547	print $x; # perl does it for you, too, quite often	670	print $x; # perl does it for you, too, quite often
548		671
549	You can force whether a string ie encoded as byte or text string by using	672	You can force whether a string is encoded as byte or text string by using
550	C<utf8::upgrade> and C<utf8::downgrade>):	673	C<utf8::upgrade> and C<utf8::downgrade> (if C<text_strings> is disabled):
551		674
552	utf8::upgrade $x; # encode $x as text string	675	utf8::upgrade $x; # encode $x as text string
553	utf8::downgrade $x; # encode $x as byte string	676	utf8::downgrade $x; # encode $x as byte string
554		677
555	Perl doesn't define what operations up- and downgrade strings, so if the	678	Perl doesn't define what operations up- and downgrade strings, so if the
556	difference between byte and text is important, you should up- or downgrade	679	difference between byte and text is important, you should up- or downgrade
557	your string as late as possible before encoding.	680	your string as late as possible before encoding. You can also force the
		681	use of CBOR text strings by using C<text_keys> or C<text_strings>.
558		682
559	You can force the type to be a CBOR number by numifying it:	683	You can force the type to be a CBOR number by numifying it:
560		684
561	my $x = "3"; # some variable containing a string	685	my $x = "3"; # some variable containing a string
562	$x += 0; # numify it, ensuring it will be dumped as a number	686	$x += 0; # numify it, ensuring it will be dumped as a number
…		…
663	"$self" # encode url string	787	"$self" # encode url string
664	}	788	}
665		789
666	sub URI::THAW {	790	sub URI::THAW {
667	my ($class, $serialiser, $uri) = @_;	791	my ($class, $serialiser, $uri) = @_;
668
669	$class->new ($uri)	792	$class->new ($uri)
670	}	793	}
671		794
672	Unlike C<TO_CBOR>, multiple values can be returned by C<FREEZE>. For	795	Unlike C<TO_CBOR>, multiple values can be returned by C<FREEZE>. For
673	example, a C<FREEZE> method that returns "type", "id" and "variant" values	796	example, a C<FREEZE> method that returns "type", "id" and "variant" values
…		…
804	additional tags (such as base64url).	927	additional tags (such as base64url).
805		928
806	=head2 ENFORCED TAGS	929	=head2 ENFORCED TAGS
807		930
808	These tags are always handled when decoding, and their handling cannot be	931	These tags are always handled when decoding, and their handling cannot be
809	overriden by the user.	932	overridden by the user.
810		933
811	=over 4	934	=over 4
812		935
813	=item 26 (perl-object, L<http://cbor.schmorp.de/perl-object>)	936	=item 26 (perl-object, L<http://cbor.schmorp.de/perl-object>)
814		937
…		…
842	encoded, however, when C<pack_strings> is enabled.	965	encoded, however, when C<pack_strings> is enabled.
843		966
844	=item 22098 (indirection, L<http://cbor.schmorp.de/indirection>)	967	=item 22098 (indirection, L<http://cbor.schmorp.de/indirection>)
845		968
846	This tag is automatically generated when a reference are encountered (with	969	This tag is automatically generated when a reference are encountered (with
847	the exception of hash and array refernces). It is converted to a reference	970	the exception of hash and array references). It is converted to a reference
848	when decoding.	971	when decoding.
849		972
850	=item 55799 (self-describe CBOR, RFC 7049)	973	=item 55799 (self-describe CBOR, RFC 7049)
851		974
852	This value is not generated on encoding (unless explicitly requested by	975	This value is not generated on encoding (unless explicitly requested by
…		…
855	=back	978	=back
856		979
857	=head2 NON-ENFORCED TAGS	980	=head2 NON-ENFORCED TAGS
858		981
859	These tags have default filters provided when decoding. Their handling can	982	These tags have default filters provided when decoding. Their handling can
860	be overriden by changing the C<%CBOR::XS::FILTER> entry for the tag, or by	983	be overridden by changing the C<%CBOR::XS::FILTER> entry for the tag, or by
861	providing a custom C<filter> callback when decoding.	984	providing a custom C<filter> callback when decoding.
862		985
863	When they result in decoding into a specific Perl class, the module	986	When they result in decoding into a specific Perl class, the module
864	usually provides a corresponding C<TO_CBOR> method as well.	987	usually provides a corresponding C<TO_CBOR> method as well.
865		988
…		…
883		1006
884	These tags are decoded into L<Math::BigInt> objects. The corresponding	1007	These tags are decoded into L<Math::BigInt> objects. The corresponding
885	C<Math::BigInt::TO_CBOR> method encodes "small" bigints into normal CBOR	1008	C<Math::BigInt::TO_CBOR> method encodes "small" bigints into normal CBOR
886	integers, and others into positive/negative CBOR bignums.	1009	integers, and others into positive/negative CBOR bignums.
887		1010
888	=item 4, 5 (decimal fraction/bigfloat)	1011	=item 4, 5, 264, 265 (decimal fraction/bigfloat)
889		1012
890	Both decimal fractions and bigfloats are decoded into L<Math::BigFloat>	1013	Both decimal fractions and bigfloats are decoded into L<Math::BigFloat>
891	objects. The corresponding C<Math::BigFloat::TO_CBOR> method I<always>	1014	objects. The corresponding C<Math::BigFloat::TO_CBOR> method I<always>
892	encodes into a decimal fraction.	1015	encodes into a decimal fraction (either tag 4 or 264).
893		1016
894	CBOR cannot represent bigfloats with I<very> large exponents - conversion	1017	NaN and infinities are not encoded properly, as they cannot be represented
895	of such big float objects is undefined.	1018	in CBOR.
896		1019
897	Also, NaN and infinities are not encoded properly.	1020	See L<BIGNUM SECURITY CONSIDERATIONS> for more info.
		1021
		1022	=item 30 (rational numbers)
		1023
		1024	These tags are decoded into L<Math::BigRat> objects. The corresponding
		1025	C<Math::BigRat::TO_CBOR> method encodes rational numbers with denominator
		1026	C<1> via their numerator only, i.e., they become normal integers or
		1027	C<bignums>.
		1028
		1029	See L<BIGNUM SECURITY CONSIDERATIONS> for more info.
898		1030
899	=item 21, 22, 23 (expected later JSON conversion)	1031	=item 21, 22, 23 (expected later JSON conversion)
900		1032
901	CBOR::XS is not a CBOR-to-JSON converter, and will simply ignore these	1033	CBOR::XS is not a CBOR-to-JSON converter, and will simply ignore these
902	tags.	1034	tags.
…		…
907	C<URI::TO_CBOR> method again results in a CBOR URI value.	1039	C<URI::TO_CBOR> method again results in a CBOR URI value.
908		1040
909	=back	1041	=back
910		1042
911	=cut	1043	=cut
912
913	our %FILTER = (
914	# 0 # rfc4287 datetime, utf-8
915	# 1 # unix timestamp, any
916
917	2 => sub { # pos bigint
918	require Math::BigInt;
919	Math::BigInt->new ("0x" . unpack "H*", pop)
920	},
921
922	3 => sub { # neg bigint
923	require Math::BigInt;
924	-Math::BigInt->new ("0x" . unpack "H*", pop)
925	},
926
927	4 => sub { # decimal fraction, array
928	require Math::BigFloat;
929	Math::BigFloat->new ($_[1][1] . "E" . $_[1][0])
930	},
931
932	5 => sub { # bigfloat, array
933	require Math::BigFloat;
934	scalar Math::BigFloat->new ($_[1][1])->blsft ($_[1][0], 2)
935	},
936
937	21 => sub { pop }, # expected conversion to base64url encoding
938	22 => sub { pop }, # expected conversion to base64 encoding
939	23 => sub { pop }, # expected conversion to base16 encoding
940
941	# 24 # embedded cbor, byte string
942
943	32 => sub {
944	require URI;
945	URI->new (pop)
946	},
947
948	# 33 # base64url rfc4648, utf-8
949	# 34 # base64 rfc46484, utf-8
950	# 35 # regex pcre/ecma262, utf-8
951	# 36 # mime message rfc2045, utf-8
952	);
953
954		1044
955	=head1 CBOR and JSON	1045	=head1 CBOR and JSON
956		1046
957	CBOR is supposed to implement a superset of the JSON data model, and is,	1047	CBOR is supposed to implement a superset of the JSON data model, and is,
958	with some coercion, able to represent all JSON texts (something that other	1048	with some coercion, able to represent all JSON texts (something that other
…		…
967	CBOR intact.	1057	CBOR intact.
968		1058
969		1059
970	=head1 SECURITY CONSIDERATIONS	1060	=head1 SECURITY CONSIDERATIONS
971		1061
972	When you are using CBOR in a protocol, talking to untrusted potentially	1062	Tl;dr... if you want to decode or encode CBOR from untrusted sources, you
973	hostile creatures requires relatively few measures.	1063	should start with a coder object created via C<new_safe> (which implements
		1064	the mitigations explained below):
974		1065
		1066	my $coder = CBOR::XS->new_safe;
		1067
		1068	my $data = $coder->decode ($cbor_text);
		1069	my $cbor = $coder->encode ($data);
		1070
		1071	Longer version: When you are using CBOR in a protocol, talking to
		1072	untrusted potentially hostile creatures requires some thought:
		1073
		1074	=over 4
		1075
		1076	=item Security of the CBOR decoder itself
		1077
975	First of all, your CBOR decoder should be secure, that is, should not have	1078	First and foremost, your CBOR decoder should be secure, that is, should
		1079	not have any buffer overflows or similar bugs that could potentially be
976	any buffer overflows. Obviously, this module should ensure that and I am	1080	exploited. Obviously, this module should ensure that and I am trying hard
977	trying hard on making that true, but you never know.	1081	on making that true, but you never know.
978		1082
		1083	=item CBOR::XS can invoke almost arbitrary callbacks during decoding
		1084
		1085	CBOR::XS supports object serialisation - decoding CBOR can cause calls
		1086	to I<any> C<THAW> method in I<any> package that exists in your process
		1087	(that is, CBOR::XS will not try to load modules, but any existing C<THAW>
		1088	method or function can be called, so they all have to be secure).
		1089
		1090	Less obviously, it will also invoke C<TO_CBOR> and C<FREEZE> methods -
		1091	even if all your C<THAW> methods are secure, encoding data structures from
		1092	untrusted sources can invoke those and trigger bugs in those.
		1093
		1094	So, if you are not sure about the security of all the modules you
		1095	have loaded (you shouldn't), you should disable this part using
		1096	C<forbid_objects> or using C<new_safe>.
		1097
		1098	=item CBOR can be extended with tags that call library code
		1099
		1100	CBOR can be extended with tags, and C<CBOR::XS> has a registry of
		1101	conversion functions for many existing tags that can be extended via
		1102	third-party modules (see the C<filter> method).
		1103
		1104	If you don't trust these, you should configure the "safe" filter function,
		1105	C<CBOR::XS::safe_filter> (C<new_safe> does this), which by default only
		1106	includes conversion functions that are considered "safe" by the author
		1107	(but again, they can be extended by third party modules).
		1108
		1109	Depending on your level of paranoia, you can use the "safe" filter:
		1110
		1111	$cbor->filter (\&CBOR::XS::safe_filter);
		1112
		1113	... your own filter...
		1114
		1115	$cbor->filter (sub { ... do your stuffs here ... });
		1116
		1117	... or even no filter at all, disabling all tag decoding:
		1118
		1119	$cbor->filter (sub { });
		1120
		1121	This is never a problem for encoding, as the tag mechanism only exists in
		1122	CBOR texts.
		1123
		1124	=item Resource-starving attacks: object memory usage
		1125
979	Second, you need to avoid resource-starving attacks. That means you should	1126	You need to avoid resource-starving attacks. That means you should limit
980	limit the size of CBOR data you accept, or make sure then when your	1127	the size of CBOR data you accept, or make sure then when your resources
981	resources run out, that's just fine (e.g. by using a separate process that	1128	run out, that's just fine (e.g. by using a separate process that can
982	can crash safely). The size of a CBOR string in octets is usually a good	1129	crash safely). The size of a CBOR string in octets is usually a good
983	indication of the size of the resources required to decode it into a Perl	1130	indication of the size of the resources required to decode it into a Perl
984	structure. While CBOR::XS can check the size of the CBOR text, it might be	1131	structure. While CBOR::XS can check the size of the CBOR text (using
985	too late when you already have it in memory, so you might want to check	1132	C<max_size> - done by C<new_safe>), it might be too late when you already
986	the size before you accept the string.	1133	have it in memory, so you might want to check the size before you accept
		1134	the string.
987		1135
		1136	As for encoding, it is possible to construct data structures that are
		1137	relatively small but result in large CBOR texts (for example by having an
		1138	array full of references to the same big data structure, which will all be
		1139	deep-cloned during encoding by default). This is rarely an actual issue
		1140	(and the worst case is still just running out of memory), but you can
		1141	reduce this risk by using C<allow_sharing>.
		1142
		1143	=item Resource-starving attacks: stack overflows
		1144
988	Third, CBOR::XS recurses using the C stack when decoding objects and	1145	CBOR::XS recurses using the C stack when decoding objects and arrays. The
989	arrays. The C stack is a limited resource: for instance, on my amd64	1146	C stack is a limited resource: for instance, on my amd64 machine with 8MB
990	machine with 8MB of stack size I can decode around 180k nested arrays but	1147	of stack size I can decode around 180k nested arrays but only 14k nested
991	only 14k nested CBOR objects (due to perl itself recursing deeply on croak	1148	CBOR objects (due to perl itself recursing deeply on croak to free the
992	to free the temporary). If that is exceeded, the program crashes. To be	1149	temporary). If that is exceeded, the program crashes. To be conservative,
993	conservative, the default nesting limit is set to 512. If your process	1150	the default nesting limit is set to 512. If your process has a smaller
994	has a smaller stack, you should adjust this setting accordingly with the	1151	stack, you should adjust this setting accordingly with the C<max_depth>
995	C<max_depth> method.	1152	method.
		1153
		1154	=item Resource-starving attacks: CPU en-/decoding complexity
		1155
		1156	CBOR::XS will use the L<Math::BigInt>, L<Math::BigFloat> and
		1157	L<Math::BigRat> libraries to represent encode/decode bignums. These can be
		1158	very slow (as in, centuries of CPU time) and can even crash your program
		1159	(and are generally not very trustworthy). See the next section on bignum
		1160	security for details.
		1161
		1162	=item Data breaches: leaking information in error messages
		1163
		1164	CBOR::XS might leak contents of your Perl data structures in its error
		1165	messages, so when you serialise sensitive information you might want to
		1166	make sure that exceptions thrown by CBOR::XS will not end up in front of
		1167	untrusted eyes.
		1168
		1169	=item Something else...
996		1170
997	Something else could bomb you, too, that I forgot to think of. In that	1171	Something else could bomb you, too, that I forgot to think of. In that
998	case, you get to keep the pieces. I am always open for hints, though...	1172	case, you get to keep the pieces. I am always open for hints, though...
999		1173
1000	Also keep in mind that CBOR::XS might leak contents of your Perl data	1174	=back
1001	structures in its error messages, so when you serialise sensitive	1175
1002	information you might want to make sure that exceptions thrown by CBOR::XS	1176
1003	will not end up in front of untrusted eyes.	1177	=head1 BIGNUM SECURITY CONSIDERATIONS
		1178
		1179	CBOR::XS provides a C<TO_CBOR> method for both L<Math::BigInt> and
		1180	L<Math::BigFloat> that tries to encode the number in the simplest possible
		1181	way, that is, either a CBOR integer, a CBOR bigint/decimal fraction (tag
		1182	4) or an arbitrary-exponent decimal fraction (tag 264). Rational numbers
		1183	(L<Math::BigRat>, tag 30) can also contain bignums as members.
		1184
		1185	CBOR::XS will also understand base-2 bigfloat or arbitrary-exponent
		1186	bigfloats (tags 5 and 265), but it will never generate these on its own.
		1187
		1188	Using the built-in L<Math::BigInt::Calc> support, encoding and decoding
		1189	decimal fractions is generally fast. Decoding bigints can be slow for very
		1190	big numbers (tens of thousands of digits, something that could potentially
		1191	be caught by limiting the size of CBOR texts), and decoding bigfloats or
		1192	arbitrary-exponent bigfloats can be I<extremely> slow (minutes, decades)
		1193	for large exponents (roughly 40 bit and longer).
		1194
		1195	Additionally, L<Math::BigInt> can take advantage of other bignum
		1196	libraries, such as L<Math::GMP>, which cannot handle big floats with large
		1197	exponents, and might simply abort or crash your program, due to their code
		1198	quality.
		1199
		1200	This can be a concern if you want to parse untrusted CBOR. If it is, you
		1201	might want to disable decoding of tag 2 (bigint) and 3 (negative bigint)
		1202	types. You should also disable types 5 and 265, as these can be slow even
		1203	without bigints.
		1204
		1205	Disabling bigints will also partially or fully disable types that rely on
		1206	them, e.g. rational numbers that use bignums.
		1207
1004		1208
1005	=head1 CBOR IMPLEMENTATION NOTES	1209	=head1 CBOR IMPLEMENTATION NOTES
1006		1210
1007	This section contains some random implementation notes. They do not	1211	This section contains some random implementation notes. They do not
1008	describe guaranteed behaviour, but merely behaviour as-is implemented	1212	describe guaranteed behaviour, but merely behaviour as-is implemented
…		…
1049		1253
1050	Please refrain from using rt.cpan.org or any other bug reporting	1254	Please refrain from using rt.cpan.org or any other bug reporting
1051	service. I put the contact address into my modules for a reason.	1255	service. I put the contact address into my modules for a reason.
1052		1256
1053	=cut	1257	=cut
		1258
		1259	# clumsy and slow hv_store-in-hash helper function
		1260	sub _hv_store {
		1261	$_[0]{$_[1]} = $_[2];
		1262	}
1054		1263
1055	our %FILTER = (	1264	our %FILTER = (
1056	0 => sub { # rfc4287 datetime, utf-8	1265	0 => sub { # rfc4287 datetime, utf-8
1057	require Time::Piece;	1266	require Time::Piece;
1058	# Time::Piece::Strptime uses the "incredibly flexible date parsing routine"	1267	# Time::Piece::Strptime uses the "incredibly flexible date parsing routine"
…		…
1094	4 => sub { # decimal fraction, array	1303	4 => sub { # decimal fraction, array
1095	require Math::BigFloat;	1304	require Math::BigFloat;
1096	Math::BigFloat->new ($_[1][1] . "E" . $_[1][0])	1305	Math::BigFloat->new ($_[1][1] . "E" . $_[1][0])
1097	},	1306	},
1098		1307
		1308	264 => sub { # decimal fraction with arbitrary exponent
		1309	require Math::BigFloat;
		1310	Math::BigFloat->new ($_[1][1] . "E" . $_[1][0])
		1311	},
		1312
1099	5 => sub { # bigfloat, array	1313	5 => sub { # bigfloat, array
1100	require Math::BigFloat;	1314	require Math::BigFloat;
1101	scalar Math::BigFloat->new ($_[1][1])->blsft ($_[1][0], 2)	1315	scalar Math::BigFloat->new ($_[1][1]) * Math::BigFloat->new (2)->bpow ($_[1][0])
		1316	},
		1317
		1318	265 => sub { # bigfloat with arbitrary exponent
		1319	require Math::BigFloat;
		1320	scalar Math::BigFloat->new ($_[1][1]) * Math::BigFloat->new (2)->bpow ($_[1][0])
		1321	},
		1322
		1323	30 => sub { # rational number
		1324	require Math::BigRat;
		1325	Math::BigRat->new ("$_[1][0]/$_[1][1]") # separate parameters only work in recent versons
1102	},	1326	},
1103		1327
1104	21 => sub { pop }, # expected conversion to base64url encoding	1328	21 => sub { pop }, # expected conversion to base64url encoding
1105	22 => sub { pop }, # expected conversion to base64 encoding	1329	22 => sub { pop }, # expected conversion to base64 encoding
1106	23 => sub { pop }, # expected conversion to base16 encoding	1330	23 => sub { pop }, # expected conversion to base16 encoding
…		…
1116	# 34 # base64 rfc46484, utf-8	1340	# 34 # base64 rfc46484, utf-8
1117	# 35 # regex pcre/ecma262, utf-8	1341	# 35 # regex pcre/ecma262, utf-8
1118	# 36 # mime message rfc2045, utf-8	1342	# 36 # mime message rfc2045, utf-8
1119	);	1343	);
1120		1344
1121	sub CBOR::XS::default_filter {	1345	sub default_filter {
1122	&{ $FILTER{$_[0]} or return }	1346	&{ $FILTER{$_[0]} or return }
		1347	}
		1348
		1349	our %SAFE_FILTER = map { $_ => $FILTER{$_} } 0, 1, 21, 22, 23, 32;
		1350
		1351	sub safe_filter {
		1352	&{ $SAFE_FILTER{$_[0]} or return }
1123	}	1353	}
1124		1354
1125	sub URI::TO_CBOR {	1355	sub URI::TO_CBOR {
1126	my $uri = $_[0]->as_string;	1356	my $uri = $_[0]->as_string;
1127	utf8::upgrade $uri;	1357	utf8::upgrade $uri;
1128	tag 32, $uri	1358	tag 32, $uri
1129	}	1359	}
1130		1360
1131	sub Math::BigInt::TO_CBOR {	1361	sub Math::BigInt::TO_CBOR {
1132	if ($_[0] >= -2147483648 && $_[0] <= 2147483647) {	1362	if (-2147483648 <= $_[0] && $_[0] <= 2147483647) {
1133	$_[0]->numify	1363	$_[0]->numify
1134	} else {	1364	} else {
1135	my $hex = substr $_[0]->as_hex, 2;	1365	my $hex = substr $_[0]->as_hex, 2;
1136	$hex = "0$hex" if 1 & length $hex; # sigh	1366	$hex = "0$hex" if 1 & length $hex; # sigh
1137	tag $_[0] >= 0 ? 2 : 3, pack "H*", $hex	1367	tag $_[0] >= 0 ? 2 : 3, pack "H*", $hex
1138	}	1368	}
1139	}	1369	}
1140		1370
1141	sub Math::BigFloat::TO_CBOR {	1371	sub Math::BigFloat::TO_CBOR {
1142	my ($m, $e) = $_[0]->parts;	1372	my ($m, $e) = $_[0]->parts;
		1373
		1374	-9223372036854775808 <= $e && $e <= 18446744073709551615
1143	tag 4, [$e->numify, $m]	1375	? tag 4, [$e->numify, $m]
		1376	: tag 264, [$e, $m]
		1377	}
		1378
		1379	sub Math::BigRat::TO_CBOR {
		1380	my ($n, $d) = $_[0]->parts;
		1381
		1382	# older versions of BigRat need *1, as they not always return numbers
		1383
		1384	$d*1 == 1
		1385	? $n*1
		1386	: tag 30, [$n1, $d1]
1144	}	1387	}
1145		1388
1146	sub Time::Piece::TO_CBOR {	1389	sub Time::Piece::TO_CBOR {
1147	tag 1, 0 + $_[0]->epoch	1390	tag 1, 0 + $_[0]->epoch
1148	}	1391	}

Diff Legend

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

Comparing CBOR-XS/XS.pm (file contents): Revision 1.48 by root, Thu Feb 25 14:22:49 2016 UTC vs. Revision 1.70 by root, Sat Nov 9 07:30:36 2019 UTC

Diff Legend

Comparing CBOR-XS/XS.pm (file contents):
Revision 1.48 by root, Thu Feb 25 14:22:49 2016 UTC vs.
Revision 1.70 by root, Sat Nov 9 07:30:36 2019 UTC