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

Comparing CBOR-XS/XS.pm (file contents):
Revision 1.56 by root, Mon Apr 25 21:44:13 2016 UTC vs.
Revision 1.88 by root, Thu Sep 7 23:52:24 2023 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 = 1.5;	69	our $VERSION = 1.86;
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	->validate_utf8
		127	->forbid_objects
		128	->filter (\&CBOR::XS::safe_filter)
		129	->max_size (1e8);
		130
		131	But is more future proof (it is better to crash because of a change than
		132	to be exploited in other ways).
		133
		134	=cut
		135
		136	sub new_safe {
		137	CBOR::XS
		138	->new
		139	->validate_utf8
		140	->forbid_objects
		141	->filter (\&CBOR::XS::safe_filter)
		142	->max_size (1e8)
		143	}
117		144
118	=item $cbor = $cbor->max_depth ([$maximum_nesting_depth])	145	=item $cbor = $cbor->max_depth ([$maximum_nesting_depth])
119		146
120	=item $max_depth = $cbor->get_max_depth	147	=item $max_depth = $cbor->get_max_depth
121		148
…		…
137		164
138	Note that nesting is implemented by recursion in C. The default value has	165	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	166	been chosen to be as large as typical operating systems allow without
140	crashing.	167	crashing.
141		168
142	See SECURITY CONSIDERATIONS, below, for more info on why this is useful.	169	See L<SECURITY CONSIDERATIONS>, below, for more info on why this is useful.
143		170
144	=item $cbor = $cbor->max_size ([$maximum_string_size])	171	=item $cbor = $cbor->max_size ([$maximum_string_size])
145		172
146	=item $max_size = $cbor->get_max_size	173	=item $max_size = $cbor->get_max_size
147		174
…		…
152	effect on C<encode> (yet).	179	effect on C<encode> (yet).
153		180
154	If no argument is given, the limit check will be deactivated (same as when	181	If no argument is given, the limit check will be deactivated (same as when
155	C<0> is specified).	182	C<0> is specified).
156		183
157	See SECURITY CONSIDERATIONS, below, for more info on why this is useful.	184	See L<SECURITY CONSIDERATIONS>, below, for more info on why this is useful.
158		185
159	=item $cbor = $cbor->allow_unknown ([$enable])	186	=item $cbor = $cbor->allow_unknown ([$enable])
160		187
161	=item $enabled = $cbor->get_allow_unknown	188	=item $enabled = $cbor->get_allow_unknown
162		189
…		…
180	reference to the earlier value.	207	reference to the earlier value.
181		208
182	This means that such values will only be encoded once, and will not result	209	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	210	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	211	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	212	structures (which need C<allow_cycles> to be enabled to be decoded by this
186	module).	213	module).
187		214
188	It is recommended to leave it off unless you know your	215	It is recommended to leave it off unless you know your
189	communication partner supports the value sharing extensions to CBOR	216	communication partner supports the value sharing extensions to CBOR
190	(L<http://cbor.schmorp.de/value-sharing>), as without decoder support, the	217	(L<http://cbor.schmorp.de/value-sharing>), as without decoder support, the
191	resulting data structure might be unusable.	218	resulting data structure might be unusable.
192		219
193	Detecting shared values incurs a runtime overhead when values are encoded	220	Detecting shared values incurs a runtime overhead when values are encoded
194	that have a reference counter large than one, and might unnecessarily	221	that have a reference counter larger than one, and might unnecessarily
195	increase the encoded size, as potentially shared values are encode as	222	increase the encoded size, as potentially shared values are encoded as
196	shareable whether or not they are actually shared.	223	shareable whether or not they are actually shared.
197		224
198	At the moment, only targets of references can be shared (e.g. scalars,	225	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	226	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	227	an array with multiple "copies" of the I<same> string, which are hard but
…		…
226	throwing an error.	253	throwing an error.
227		254
228	This option does not affect C<encode> in any way - shared values and	255	This option does not affect C<encode> in any way - shared values and
229	references will always be encoded properly if present.	256	references will always be encoded properly if present.
230		257
		258	=item $cbor = $cbor->forbid_objects ([$enable])
		259
		260	=item $enabled = $cbor->get_forbid_objects
		261
		262	Disables the use of the object serialiser protocol.
		263
		264	If C<$enable> is true (or missing), then C<encode> will will throw an
		265	exception when it encounters perl objects that would be encoded using the
		266	perl-object tag (26). When C<decode> encounters such tags, it will fall
		267	back to the general filter/tagged logic as if this were an unknown tag (by
		268	default resulting in a C<CBOR::XC::Tagged> object).
		269
		270	If C<$enable> is false (the default), then C<encode> will use the
		271	L<Types::Serialiser> object serialisation protocol to serialise objects
		272	into perl-object tags, and C<decode> will do the same to decode such tags.
		273
		274	See L<SECURITY CONSIDERATIONS>, below, for more info on why forbidding this
		275	protocol can be useful.
		276
231	=item $cbor = $cbor->pack_strings ([$enable])	277	=item $cbor = $cbor->pack_strings ([$enable])
232		278
233	=item $enabled = $cbor->get_pack_strings	279	=item $enabled = $cbor->get_pack_strings
234		280
235	If C<$enable> is true (or missing), then C<encode> will try not to encode	281	If C<$enable> is true (or missing), then C<encode> will try not to encode
…		…
286	strings as CBOR byte strings.	332	strings as CBOR byte strings.
287		333
288	This option does not affect C<decode> in any way.	334	This option does not affect C<decode> in any way.
289		335
290	This option has similar advantages and disadvantages as C<text_keys>. In	336	This option has similar advantages and disadvantages as C<text_keys>. In
291	addition, this option effectively removes the ability to encode byte	337	addition, this option effectively removes the ability to automatically
292	strings, which might break some C<FREEZE> and C<TO_CBOR> methods that rely	338	encode byte strings, which might break some C<FREEZE> and C<TO_CBOR>
293	on this, such as bignum encoding, so this option is mainly useful for very	339	methods that rely on this.
294	simple data.	340
		341	A workaround is to use explicit type casts, which are unaffected by this option.
295		342
296	=item $cbor = $cbor->validate_utf8 ([$enable])	343	=item $cbor = $cbor->validate_utf8 ([$enable])
297		344
298	=item $enabled = $cbor->get_validate_utf8	345	=item $enabled = $cbor->get_validate_utf8
299		346
…		…
337	replace the tagged value in the decoded data structure, or no values,	384	replace the tagged value in the decoded data structure, or no values,
338	which will result in default handling, which currently means the decoder	385	which will result in default handling, which currently means the decoder
339	creates a C<CBOR::XS::Tagged> object to hold the tag and the value.	386	creates a C<CBOR::XS::Tagged> object to hold the tag and the value.
340		387
341	When the filter is cleared (the default state), the default filter	388	When the filter is cleared (the default state), the default filter
342	function, C<CBOR::XS::default_filter>, is used. This function simply looks	389	function, C<CBOR::XS::default_filter>, is used. This function simply
343	up the tag in the C<%CBOR::XS::FILTER> hash. If an entry exists it must be	390	looks up the tag in the C<%CBOR::XS::FILTER> hash. If an entry exists
344	a code reference that is called with tag and value, and is responsible for	391	it must be a code reference that is called with tag and value, and is
345	decoding the value. If no entry exists, it returns no values.	392	responsible for decoding the value. If no entry exists, it returns no
		393	values. C<CBOR::XS> provides a number of default filter functions already,
		394	the the C<%CBOR::XS::FILTER> hash can be freely extended with more.
		395
		396	C<CBOR::XS> additionally provides an alternative filter function that is
		397	supposed to be safe to use with untrusted data (which the default filter
		398	might not), called C<CBOR::XS::safe_filter>, which works the same as
		399	the C<default_filter> but uses the C<%CBOR::XS::SAFE_FILTER> variable
		400	instead. It is prepopulated with the tag decoding functions that are
		401	deemed safe (basically the same as C<%CBOR::XS::FILTER> without all
		402	the bignum tags), and can be extended by user code as wlel, although,
		403	obviously, one should be very careful about adding decoding functions
		404	here, since the expectation is that they are safe to use on untrusted
		405	data, after all.
346		406
347	Example: decode all tags not handled internally into C<CBOR::XS::Tagged>	407	Example: decode all tags not handled internally into C<CBOR::XS::Tagged>
348	objects, with no other special handling (useful when working with	408	objects, with no other special handling (useful when working with
349	potentially "unsafe" CBOR data).	409	potentially "unsafe" CBOR data).
350		410
…		…
357	my ($tag, $value);	417	my ($tag, $value);
358		418
359	"tag 1347375694 value $value"	419	"tag 1347375694 value $value"
360	};	420	};
361		421
		422	Example: provide your own filter function that looks up tags in your own
		423	hash:
		424
		425	my %my_filter = (
		426	998347484 => sub {
		427	my ($tag, $value);
		428
		429	"tag 998347484 value $value"
		430	};
		431	);
		432
		433	my $coder = CBOR::XS->new->filter (sub {
		434	&{ $my_filter{$_[0]} or return }
		435	});
		436
		437
		438	Example: use the safe filter function (see L<SECURITY CONSIDERATIONS> for
		439	more considerations on security).
		440
		441	CBOR::XS->new->filter (\&CBOR::XS::safe_filter)->decode ($cbor_data);
		442
362	=item $cbor_data = $cbor->encode ($perl_scalar)	443	=item $cbor_data = $cbor->encode ($perl_scalar)
363		444
364	Converts the given Perl data structure (a scalar value) to its CBOR	445	Converts the given Perl data structure (a scalar value) to its CBOR
365	representation.	446	representation.
366		447
…		…
375	when there is trailing garbage after the CBOR string, it will silently	456	when there is trailing garbage after the CBOR string, it will silently
376	stop parsing there and return the number of characters consumed so far.	457	stop parsing there and return the number of characters consumed so far.
377		458
378	This is useful if your CBOR texts are not delimited by an outer protocol	459	This is useful if your CBOR texts are not delimited by an outer protocol
379	and you need to know where the first CBOR string ends amd the next one	460	and you need to know where the first CBOR string ends amd the next one
380	starts.	461	starts - CBOR strings are self-delimited, so it is possible to concatenate
		462	CBOR strings without any delimiters or size fields and recover their data.
381		463
382	CBOR::XS->new->decode_prefix ("......")	464	CBOR::XS->new->decode_prefix ("......")
383	=> ("...", 3)	465	=> ("...", 3)
384		466
385	=back	467	=back
…		…
391	Perl data structure in memory at one time, it does allow you to parse a	473	Perl data structure in memory at one time, it does allow you to parse a
392	CBOR stream incrementally, using a similar to using "decode_prefix" to see	474	CBOR stream incrementally, using a similar to using "decode_prefix" to see
393	if a full CBOR object is available, but is much more efficient.	475	if a full CBOR object is available, but is much more efficient.
394		476
395	It basically works by parsing as much of a CBOR string as possible - if	477	It basically works by parsing as much of a CBOR string as possible - if
396	the CBOR data is not complete yet, the pasrer will remember where it was,	478	the CBOR data is not complete yet, the parser will remember where it was,
397	to be able to restart when more data has been accumulated. Once enough	479	to be able to restart when more data has been accumulated. Once enough
398	data is available to either decode a complete CBOR value or raise an	480	data is available to either decode a complete CBOR value or raise an
399	error, a real decode will be attempted.	481	error, a real decode will be attempted.
400		482
401	A typical use case would be a network protocol that consists of sending	483	A typical use case would be a network protocol that consists of sending
…		…
440		522
441	Resets the incremental decoder. This throws away any saved state, so that	523	Resets the incremental decoder. This throws away any saved state, so that
442	subsequent calls to C<incr_parse> or C<incr_parse_multiple> start to parse	524	subsequent calls to C<incr_parse> or C<incr_parse_multiple> start to parse
443	a new CBOR value from the beginning of the C<$buffer> again.	525	a new CBOR value from the beginning of the C<$buffer> again.
444		526
445	This method can be caled at any time, but it I<must> be called if you want	527	This method can be called at any time, but it I<must> be called if you want
446	to change your C<$buffer> or there was a decoding error and you want to	528	to change your C<$buffer> or there was a decoding error and you want to
447	reuse the C<$cbor> object for future incremental parsings.	529	reuse the C<$cbor> object for future incremental parsings.
448		530
449	=back	531	=back
450		532
…		…
553	create such objects.	635	create such objects.
554		636
555	=item Types::Serialiser::true, Types::Serialiser::false, Types::Serialiser::error	637	=item Types::Serialiser::true, Types::Serialiser::false, Types::Serialiser::error
556		638
557	These special values become CBOR true, CBOR false and CBOR undefined	639	These special values become CBOR true, CBOR false and CBOR undefined
558	values, respectively. You can also use C<\1>, C<\0> and C<\undef> directly	640	values, respectively.
559	if you want.
560		641
561	=item other blessed objects	642	=item other blessed objects
562		643
563	Other blessed objects are serialised via C<TO_CBOR> or C<FREEZE>. See	644	Other blessed objects are serialised via C<TO_CBOR> or C<FREEZE>. See
564	L<TAG HANDLING AND EXTENSIONS> for specific classes handled by this	645	L<TAG HANDLING AND EXTENSIONS> for specific classes handled by this
…		…
589	"$x"; # stringified	670	"$x"; # stringified
590	$x .= ""; # another, more awkward way to stringify	671	$x .= ""; # another, more awkward way to stringify
591	print $x; # perl does it for you, too, quite often	672	print $x; # perl does it for you, too, quite often
592		673
593	You can force whether a string is encoded as byte or text string by using	674	You can force whether a string is encoded as byte or text string by using
594	C<utf8::upgrade> and C<utf8::downgrade> (if C<text_strings> is disabled):	675	C<utf8::upgrade> and C<utf8::downgrade> (if C<text_strings> is disabled).
595		676
596	utf8::upgrade $x; # encode $x as text string	677	utf8::upgrade $x; # encode $x as text string
597	utf8::downgrade $x; # encode $x as byte string	678	utf8::downgrade $x; # encode $x as byte string
		679
		680	More options are available, see L<TYPE CASTS>, below, and the C<text_keys>
		681	and C<text_strings> options.
598		682
599	Perl doesn't define what operations up- and downgrade strings, so if the	683	Perl doesn't define what operations up- and downgrade strings, so if the
600	difference between byte and text is important, you should up- or downgrade	684	difference between byte and text is important, you should up- or downgrade
601	your string as late as possible before encoding. You can also force the	685	your string as late as possible before encoding. You can also force the
602	use of CBOR text strings by using C<text_keys> or C<text_strings>.	686	use of CBOR text strings by using C<text_keys> or C<text_strings>.
…		…
618	represent numerical values are supported, but might suffer loss of	702	represent numerical values are supported, but might suffer loss of
619	precision.	703	precision.
620		704
621	=back	705	=back
622		706
		707	=head2 TYPE CASTS
		708
		709	B<EXPERIMENTAL>: As an experimental extension, C<CBOR::XS> allows you to
		710	force specific CBOR types to be used when encoding. That allows you to
		711	encode types not normally accessible (e.g. half floats) as well as force
		712	string types even when C<text_strings> is in effect.
		713
		714	Type forcing is done by calling a special "cast" function which keeps a
		715	copy of the value and returns a new value that can be handed over to any
		716	CBOR encoder function.
		717
		718	The following casts are currently available (all of which are unary
		719	operators, that is, have a prototype of C<$>):
		720
		721	=over
		722
		723	=item CBOR::XS::as_int $value
		724
		725	Forces the value to be encoded as some form of (basic, not bignum) integer
		726	type.
		727
		728	=item CBOR::XS::as_text $value
		729
		730	Forces the value to be encoded as (UTF-8) text values.
		731
		732	=item CBOR::XS::as_bytes $value
		733
		734	Forces the value to be encoded as a (binary) string value.
		735
		736	Example: encode a perl string as binary even though C<text_strings> is in
		737	effect.
		738
		739	CBOR::XS->new->text_strings->encode ([4, "text", CBOR::XS::bytes "bytevalue"]);
		740
		741	=item CBOR::XS::as_bool $value
		742
		743	Converts a Perl boolean (which can be any kind of scalar) into a CBOR
		744	boolean. Strictly the same, but shorter to write, than:
		745
		746	$value ? Types::Serialiser::true : Types::Serialiser::false
		747
		748	=item CBOR::XS::as_float16 $value
		749
		750	Forces half-float (IEEE 754 binary16) encoding of the given value.
		751
		752	=item CBOR::XS::as_float32 $value
		753
		754	Forces single-float (IEEE 754 binary32) encoding of the given value.
		755
		756	=item CBOR::XS::as_float64 $value
		757
		758	Forces double-float (IEEE 754 binary64) encoding of the given value.
		759
		760	=item CBOR::XS::as_cbor $cbor_text
		761
		762	Not a type cast per-se, this type cast forces the argument to be encoded
		763	as-is. This can be used to embed pre-encoded CBOR data.
		764
		765	Note that no checking on the validity of the C<$cbor_text> is done - it's
		766	the callers responsibility to correctly encode values.
		767
		768	=item CBOR::XS::as_map [key => value...]
		769
		770	Treat the array reference as key value pairs and output a CBOR map. This
		771	allows you to generate CBOR maps with arbitrary key types (or, if you
		772	don't care about semantics, duplicate keys or pairs in a custom order),
		773	which is otherwise hard to do with Perl.
		774
		775	The single argument must be an array reference with an even number of
		776	elements.
		777
		778	Note that only the reference to the array is copied, the array itself is
		779	not. Modifications done to the array before calling an encoding function
		780	will be reflected in the encoded output.
		781
		782	Example: encode a CBOR map with a string and an integer as keys.
		783
		784	encode_cbor CBOR::XS::as_map [string => "value", 5 => "value"]
		785
		786	=back
		787
		788	=cut
		789
		790	sub CBOR::XS::as_cbor ($) { bless [$_[0], 0, undef], CBOR::XS::Tagged:: }
		791	sub CBOR::XS::as_int ($) { bless [$_[0], 1, undef], CBOR::XS::Tagged:: }
		792	sub CBOR::XS::as_bytes ($) { bless [$_[0], 2, undef], CBOR::XS::Tagged:: }
		793	sub CBOR::XS::as_text ($) { bless [$_[0], 3, undef], CBOR::XS::Tagged:: }
		794	sub CBOR::XS::as_float16 ($) { bless [$_[0], 4, undef], CBOR::XS::Tagged:: }
		795	sub CBOR::XS::as_float32 ($) { bless [$_[0], 5, undef], CBOR::XS::Tagged:: }
		796	sub CBOR::XS::as_float64 ($) { bless [$_[0], 6, undef], CBOR::XS::Tagged:: }
		797
		798	sub CBOR::XS::as_bool ($) { $_[0] ? $Types::Serialiser::true : $Types::Serialiser::false }
		799
		800	sub CBOR::XS::as_map ($) {
		801	ARRAY:: eq ref $_[0]
		802	and $#{ $_[0] } & 1
		803	or do { require Carp; Carp::croak ("CBOR::XS::as_map only acepts array references with an even number of elements, caught") };
		804
		805	bless [$_[0], 7, undef], CBOR::XS::Tagged::
		806	}
		807
623	=head2 OBJECT SERIALISATION	808	=head2 OBJECT SERIALISATION
624		809
625	This module implements both a CBOR-specific and the generic	810	This module implements both a CBOR-specific and the generic
626	L<Types::Serialier> object serialisation protocol. The following	811	L<Types::Serialier> object serialisation protocol. The following
627	subsections explain both methods.	812	subsections explain both methods.
…		…
935	objects. The corresponding C<Math::BigFloat::TO_CBOR> method I<always>	1120	objects. The corresponding C<Math::BigFloat::TO_CBOR> method I<always>
936	encodes into a decimal fraction (either tag 4 or 264).	1121	encodes into a decimal fraction (either tag 4 or 264).
937		1122
938	NaN and infinities are not encoded properly, as they cannot be represented	1123	NaN and infinities are not encoded properly, as they cannot be represented
939	in CBOR.	1124	in CBOR.
		1125
		1126	See L<BIGNUM SECURITY CONSIDERATIONS> for more info.
		1127
		1128	=item 30 (rational numbers)
		1129
		1130	These tags are decoded into L<Math::BigRat> objects. The corresponding
		1131	C<Math::BigRat::TO_CBOR> method encodes rational numbers with denominator
		1132	C<1> via their numerator only, i.e., they become normal integers or
		1133	C<bignums>.
940		1134
941	See L<BIGNUM SECURITY CONSIDERATIONS> for more info.	1135	See L<BIGNUM SECURITY CONSIDERATIONS> for more info.
942		1136
943	=item 21, 22, 23 (expected later JSON conversion)	1137	=item 21, 22, 23 (expected later JSON conversion)
944		1138
…		…
969	CBOR intact.	1163	CBOR intact.
970		1164
971		1165
972	=head1 SECURITY CONSIDERATIONS	1166	=head1 SECURITY CONSIDERATIONS
973		1167
974	When you are using CBOR in a protocol, talking to untrusted potentially	1168	Tl;dr... if you want to decode or encode CBOR from untrusted sources, you
975	hostile creatures requires relatively few measures.	1169	should start with a coder object created via C<new_safe> (which implements
		1170	the mitigations explained below):
976		1171
		1172	my $coder = CBOR::XS->new_safe;
		1173
		1174	my $data = $coder->decode ($cbor_text);
		1175	my $cbor = $coder->encode ($data);
		1176
		1177	Longer version: When you are using CBOR in a protocol, talking to
		1178	untrusted potentially hostile creatures requires some thought:
		1179
		1180	=over 4
		1181
		1182	=item Security of the CBOR decoder itself
		1183
977	First of all, your CBOR decoder should be secure, that is, should not have	1184	First and foremost, your CBOR decoder should be secure, that is, should
		1185	not have any buffer overflows or similar bugs that could potentially be
978	any buffer overflows. Obviously, this module should ensure that and I am	1186	exploited. Obviously, this module should ensure that and I am trying hard
979	trying hard on making that true, but you never know.	1187	on making that true, but you never know.
980		1188
		1189	=item CBOR::XS can invoke almost arbitrary callbacks during decoding
		1190
		1191	CBOR::XS supports object serialisation - decoding CBOR can cause calls
		1192	to I<any> C<THAW> method in I<any> package that exists in your process
		1193	(that is, CBOR::XS will not try to load modules, but any existing C<THAW>
		1194	method or function can be called, so they all have to be secure).
		1195
		1196	Less obviously, it will also invoke C<TO_CBOR> and C<FREEZE> methods -
		1197	even if all your C<THAW> methods are secure, encoding data structures from
		1198	untrusted sources can invoke those and trigger bugs in those.
		1199
		1200	So, if you are not sure about the security of all the modules you
		1201	have loaded (you shouldn't), you should disable this part using
		1202	C<forbid_objects> or using C<new_safe>.
		1203
		1204	=item CBOR can be extended with tags that call library code
		1205
		1206	CBOR can be extended with tags, and C<CBOR::XS> has a registry of
		1207	conversion functions for many existing tags that can be extended via
		1208	third-party modules (see the C<filter> method).
		1209
		1210	If you don't trust these, you should configure the "safe" filter function,
		1211	C<CBOR::XS::safe_filter> (C<new_safe> does this), which by default only
		1212	includes conversion functions that are considered "safe" by the author
		1213	(but again, they can be extended by third party modules).
		1214
		1215	Depending on your level of paranoia, you can use the "safe" filter:
		1216
		1217	$cbor->filter (\&CBOR::XS::safe_filter);
		1218
		1219	... your own filter...
		1220
		1221	$cbor->filter (sub { ... do your stuffs here ... });
		1222
		1223	... or even no filter at all, disabling all tag decoding:
		1224
		1225	$cbor->filter (sub { });
		1226
		1227	This is never a problem for encoding, as the tag mechanism only exists in
		1228	CBOR texts.
		1229
		1230	=item Resource-starving attacks: object memory usage
		1231
981	Second, you need to avoid resource-starving attacks. That means you should	1232	You need to avoid resource-starving attacks. That means you should limit
982	limit the size of CBOR data you accept, or make sure then when your	1233	the size of CBOR data you accept, or make sure then when your resources
983	resources run out, that's just fine (e.g. by using a separate process that	1234	run out, that's just fine (e.g. by using a separate process that can
984	can crash safely). The size of a CBOR string in octets is usually a good	1235	crash safely). The size of a CBOR string in octets is usually a good
985	indication of the size of the resources required to decode it into a Perl	1236	indication of the size of the resources required to decode it into a Perl
986	structure. While CBOR::XS can check the size of the CBOR text, it might be	1237	structure. While CBOR::XS can check the size of the CBOR text (using
987	too late when you already have it in memory, so you might want to check	1238	C<max_size> - done by C<new_safe>), it might be too late when you already
988	the size before you accept the string.	1239	have it in memory, so you might want to check the size before you accept
		1240	the string.
989		1241
		1242	As for encoding, it is possible to construct data structures that are
		1243	relatively small but result in large CBOR texts (for example by having an
		1244	array full of references to the same big data structure, which will all be
		1245	deep-cloned during encoding by default). This is rarely an actual issue
		1246	(and the worst case is still just running out of memory), but you can
		1247	reduce this risk by using C<allow_sharing>.
		1248
		1249	=item Resource-starving attacks: stack overflows
		1250
990	Third, CBOR::XS recurses using the C stack when decoding objects and	1251	CBOR::XS recurses using the C stack when decoding objects and arrays. The
991	arrays. The C stack is a limited resource: for instance, on my amd64	1252	C stack is a limited resource: for instance, on my amd64 machine with 8MB
992	machine with 8MB of stack size I can decode around 180k nested arrays but	1253	of stack size I can decode around 180k nested arrays but only 14k nested
993	only 14k nested CBOR objects (due to perl itself recursing deeply on croak	1254	CBOR objects (due to perl itself recursing deeply on croak to free the
994	to free the temporary). If that is exceeded, the program crashes. To be	1255	temporary). If that is exceeded, the program crashes. To be conservative,
995	conservative, the default nesting limit is set to 512. If your process	1256	the default nesting limit is set to 512. If your process has a smaller
996	has a smaller stack, you should adjust this setting accordingly with the	1257	stack, you should adjust this setting accordingly with the C<max_depth>
997	C<max_depth> method.	1258	method.
		1259
		1260	=item Resource-starving attacks: CPU en-/decoding complexity
		1261
		1262	CBOR::XS will use the L<Math::BigInt>, L<Math::BigFloat> and
		1263	L<Math::BigRat> libraries to represent encode/decode bignums. These can be
		1264	very slow (as in, centuries of CPU time) and can even crash your program
		1265	(and are generally not very trustworthy). See the next section on bignum
		1266	security for details.
		1267
		1268	=item Data breaches: leaking information in error messages
		1269
		1270	CBOR::XS might leak contents of your Perl data structures in its error
		1271	messages, so when you serialise sensitive information you might want to
		1272	make sure that exceptions thrown by CBOR::XS will not end up in front of
		1273	untrusted eyes.
		1274
		1275	=item Something else...
998		1276
999	Something else could bomb you, too, that I forgot to think of. In that	1277	Something else could bomb you, too, that I forgot to think of. In that
1000	case, you get to keep the pieces. I am always open for hints, though...	1278	case, you get to keep the pieces. I am always open for hints, though...
1001		1279
1002	Also keep in mind that CBOR::XS might leak contents of your Perl data	1280	=back
1003	structures in its error messages, so when you serialise sensitive
1004	information you might want to make sure that exceptions thrown by CBOR::XS
1005	will not end up in front of untrusted eyes.
1006		1281
1007		1282
1008	=head1 BIGNUM SECURITY CONSIDERATIONS	1283	=head1 BIGNUM SECURITY CONSIDERATIONS
1009		1284
1010	CBOR::XS provides a C<TO_CBOR> method for both L<Math::BigInt> and	1285	CBOR::XS provides a C<TO_CBOR> method for both L<Math::BigInt> and
1011	L<Math::BigFloat> that tries to encode the number in the simplest possible	1286	L<Math::BigFloat> that tries to encode the number in the simplest possible
1012	way, that is, either a CBOR integer, a CBOR bigint/decimal fraction (tag	1287	way, that is, either a CBOR integer, a CBOR bigint/decimal fraction (tag
1013	4) or an arbitrary-exponent decimal fraction (tag 264).	1288	4) or an arbitrary-exponent decimal fraction (tag 264). Rational numbers
		1289	(L<Math::BigRat>, tag 30) can also contain bignums as members.
1014		1290
1015	It will also understand base-2 bigfloat or arbitrary-exponent bigfloats	1291	CBOR::XS will also understand base-2 bigfloat or arbitrary-exponent
1016	(tags 5 and 265), but it will never generate these on its own.	1292	bigfloats (tags 5 and 265), but it will never generate these on its own.
1017		1293
1018	Using the built-in L<Math::BigInt::Calc> support, encoding and decoding	1294	Using the built-in L<Math::BigInt::Calc> support, encoding and decoding
1019	decimal fractions is generally fast. Decoding bigints can be slow for very	1295	decimal fractions is generally fast. Decoding bigints can be slow for very
1020	big numbers, and decoding bigfloats or arbitrary-exponent bigfloats can be	1296	big numbers (tens of thousands of digits, something that could potentially
1021	extremely slow (minutes, decades) for large exponents.	1297	be caught by limiting the size of CBOR texts), and decoding bigfloats or
		1298	arbitrary-exponent bigfloats can be I<extremely> slow (minutes, decades)
		1299	for large exponents (roughly 40 bit and longer).
1022		1300
1023	Additionally, L<Math::BigInt> can take advantage of other bignum	1301	Additionally, L<Math::BigInt> can take advantage of other bignum
1024	libraries, such as L<Math::GMP>, which cannot handle big	1302	libraries, such as L<Math::GMP>, which cannot handle big floats with large
1025	floats with large exponents, and might simply abort or crash your program,	1303	exponents, and might simply abort or crash your program, due to their code
1026	due to their code quality.	1304	quality.
1027		1305
1028	This can be a concern if you want to parse untrusted CBOR. If it is, you	1306	This can be a concern if you want to parse untrusted CBOR. If it is, you
1029	need to disable decoding of tag 2 (bigint) and 3 (negative bigint) types,	1307	might want to disable decoding of tag 2 (bigint) and 3 (negative bigint)
1030	which will also disable bigfloat support (to be sure, you can also disable	1308	types. You should also disable types 5 and 265, as these can be slow even
1031	types 4, 5, 264 and 265).	1309	without bigints.
		1310
		1311	Disabling bigints will also partially or fully disable types that rely on
		1312	them, e.g. rational numbers that use bignums.
1032		1313
1033		1314
1034	=head1 CBOR IMPLEMENTATION NOTES	1315	=head1 CBOR IMPLEMENTATION NOTES
1035		1316
1036	This section contains some random implementation notes. They do not	1317	This section contains some random implementation notes. They do not
…		…
1053	=head1 LIMITATIONS ON PERLS WITHOUT 64-BIT INTEGER SUPPORT	1334	=head1 LIMITATIONS ON PERLS WITHOUT 64-BIT INTEGER SUPPORT
1054		1335
1055	On perls that were built without 64 bit integer support (these are rare	1336	On perls that were built without 64 bit integer support (these are rare
1056	nowadays, even on 32 bit architectures, as all major Perl distributions	1337	nowadays, even on 32 bit architectures, as all major Perl distributions
1057	are built with 64 bit integer support), support for any kind of 64 bit	1338	are built with 64 bit integer support), support for any kind of 64 bit
1058	integer in CBOR is very limited - most likely, these 64 bit values will	1339	value in CBOR is very limited - most likely, these 64 bit values will
1059	be truncated, corrupted, or otherwise not decoded correctly. This also	1340	be truncated, corrupted, or otherwise not decoded correctly. This also
1060	includes string, array and map sizes that are stored as 64 bit integers.	1341	includes string, float, array and map sizes that are stored as 64 bit
		1342	integers.
1061		1343
1062		1344
1063	=head1 THREADS	1345	=head1 THREADS
1064		1346
1065	This module is I<not> guaranteed to be thread safe and there are no	1347	This module is I<not> guaranteed to be thread safe and there are no
…		…
1078		1360
1079	Please refrain from using rt.cpan.org or any other bug reporting	1361	Please refrain from using rt.cpan.org or any other bug reporting
1080	service. I put the contact address into my modules for a reason.	1362	service. I put the contact address into my modules for a reason.
1081		1363
1082	=cut	1364	=cut
		1365
		1366	# clumsy and slow hv_store-in-hash helper function
		1367	sub _hv_store {
		1368	$_[0]{$_[1]} = $_[2];
		1369	}
1083		1370
1084	our %FILTER = (	1371	our %FILTER = (
1085	0 => sub { # rfc4287 datetime, utf-8	1372	0 => sub { # rfc4287 datetime, utf-8
1086	require Time::Piece;	1373	require Time::Piece;
1087	# Time::Piece::Strptime uses the "incredibly flexible date parsing routine"	1374	# Time::Piece::Strptime uses the "incredibly flexible date parsing routine"
…		…
1110	scalar Time::Piece::gmtime (pop)	1397	scalar Time::Piece::gmtime (pop)
1111	},	1398	},
1112		1399
1113	2 => sub { # pos bigint	1400	2 => sub { # pos bigint
1114	require Math::BigInt;	1401	require Math::BigInt;
1115	Math::BigInt->from_hex ("0x" . unpack "H*", pop)	1402	Math::BigInt->new ("0x" . unpack "H*", pop)
1116	},	1403	},
1117		1404
1118	3 => sub { # neg bigint	1405	3 => sub { # neg bigint
1119	require Math::BigInt;	1406	require Math::BigInt;
1120	-Math::BigInt->from_hex ("0x" . unpack "H*", pop)	1407	-Math::BigInt->new ("0x" . unpack "H*", pop)
1121	},	1408	},
1122		1409
1123	4 => sub { # decimal fraction, array	1410	4 => sub { # decimal fraction, array
1124	require Math::BigFloat;	1411	require Math::BigFloat;
1125	Math::BigFloat->new ($_[1][1] . "E" . $_[1][0])	1412	Math::BigFloat->new ($_[1][1] . "E" . $_[1][0])
1126	},	1413	},
1127		1414
		1415	264 => sub { # decimal fraction with arbitrary exponent
		1416	require Math::BigFloat;
		1417	Math::BigFloat->new ($_[1][1] . "E" . $_[1][0])
		1418	},
		1419
1128	5 => sub { # bigfloat, array	1420	5 => sub { # bigfloat, array
1129	require Math::BigFloat;	1421	require Math::BigFloat;
1130	scalar Math::BigFloat->new ($_[1][1]) * Math::BigFloat->new (2)->bpow ($_[1][0])	1422	scalar Math::BigFloat->new ($_[1][1]) * Math::BigFloat->new (2)->bpow ($_[1][0])
1131	},	1423	},
1132		1424
		1425	265 => sub { # bigfloat with arbitrary exponent
		1426	require Math::BigFloat;
		1427	scalar Math::BigFloat->new ($_[1][1]) * Math::BigFloat->new (2)->bpow ($_[1][0])
		1428	},
		1429
		1430	30 => sub { # rational number
		1431	require Math::BigRat;
		1432	Math::BigRat->new ("$_[1][0]/$_[1][1]") # separate parameters only work in recent versons
		1433	},
		1434
1133	21 => sub { pop }, # expected conversion to base64url encoding	1435	21 => sub { pop }, # expected conversion to base64url encoding
1134	22 => sub { pop }, # expected conversion to base64 encoding	1436	22 => sub { pop }, # expected conversion to base64 encoding
1135	23 => sub { pop }, # expected conversion to base16 encoding	1437	23 => sub { pop }, # expected conversion to base16 encoding
1136		1438
1137	# 24 # embedded cbor, byte string	1439	# 24 # embedded cbor, byte string
…		…
1143		1445
1144	# 33 # base64url rfc4648, utf-8	1446	# 33 # base64url rfc4648, utf-8
1145	# 34 # base64 rfc46484, utf-8	1447	# 34 # base64 rfc46484, utf-8
1146	# 35 # regex pcre/ecma262, utf-8	1448	# 35 # regex pcre/ecma262, utf-8
1147	# 36 # mime message rfc2045, utf-8	1449	# 36 # mime message rfc2045, utf-8
1148
1149	264 => sub { # decimal fraction with arbitrary exponent
1150	require Math::BigFloat;
1151	Math::BigFloat->new ($_[1][1] . "E" . $_[1][0])
1152	},
1153
1154	265 => sub { # bigfloat with arbitrary exponent
1155	require Math::BigFloat;
1156	scalar Math::BigFloat->new ($_[1][1]) * Math::BigFloat->new (2)->bpow ($_[1][0])
1157	},
1158	);	1450	);
1159		1451
1160	sub CBOR::XS::default_filter {	1452	sub default_filter {
1161	&{ $FILTER{$_[0]} or return }	1453	&{ $FILTER{$_[0]} or return }
		1454	}
		1455
		1456	our %SAFE_FILTER = map { $_ => $FILTER{$_} } 0, 1, 21, 22, 23, 32;
		1457
		1458	sub safe_filter {
		1459	&{ $SAFE_FILTER{$_[0]} or return }
1162	}	1460	}
1163		1461
1164	sub URI::TO_CBOR {	1462	sub URI::TO_CBOR {
1165	my $uri = $_[0]->as_string;	1463	my $uri = $_[0]->as_string;
1166	utf8::upgrade $uri;	1464	utf8::upgrade $uri;
…		…
1183	-9223372036854775808 <= $e && $e <= 18446744073709551615	1481	-9223372036854775808 <= $e && $e <= 18446744073709551615
1184	? tag 4, [$e->numify, $m]	1482	? tag 4, [$e->numify, $m]
1185	: tag 264, [$e, $m]	1483	: tag 264, [$e, $m]
1186	}	1484	}
1187		1485
		1486	sub Math::BigRat::TO_CBOR {
		1487	my ($n, $d) = $_[0]->parts;
		1488
		1489	# older versions of BigRat need *1, as they not always return numbers
		1490
		1491	$d*1 == 1
		1492	? $n*1
		1493	: tag 30, [$n1, $d1]
		1494	}
		1495
1188	sub Time::Piece::TO_CBOR {	1496	sub Time::Piece::TO_CBOR {
1189	tag 1, 0 + $_[0]->epoch	1497	tag 1, 0 + $_[0]->epoch
1190	}	1498	}
1191		1499
1192	XSLoader::load "CBOR::XS", $VERSION;	1500	XSLoader::load "CBOR::XS", $VERSION;

Diff Legend

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

Comparing CBOR-XS/XS.pm (file contents): Revision 1.56 by root, Mon Apr 25 21:44:13 2016 UTC vs. Revision 1.88 by root, Thu Sep 7 23:52:24 2023 UTC

Diff Legend

Comparing CBOR-XS/XS.pm (file contents):
Revision 1.56 by root, Mon Apr 25 21:44:13 2016 UTC vs.
Revision 1.88 by root, Thu Sep 7 23:52:24 2023 UTC