[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 1.76 by root, Tue Dec 1 01:49:47 2020 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 = 1.82;
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
…		…
286	strings as CBOR byte strings.	330	strings as CBOR byte strings.
287		331
288	This option does not affect C<decode> in any way.	332	This option does not affect C<decode> in any way.
289		333
290	This option has similar advantages and disadvantages as C<text_keys>. In	334	This option has similar advantages and disadvantages as C<text_keys>. In
291	addition, this option effectively removes the ability to encode byte	335	addition, this option effectively removes the ability to automatically
292	strings, which might break some C<FREEZE> and C<TO_CBOR> methods that rely	336	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	337	methods that rely on this.
294	simple data.	338
		339	A workaround is to use explicit type casts, which are unaffected by this option.
295		340
296	=item $cbor = $cbor->validate_utf8 ([$enable])	341	=item $cbor = $cbor->validate_utf8 ([$enable])
297		342
298	=item $enabled = $cbor->get_validate_utf8	343	=item $enabled = $cbor->get_validate_utf8
299		344
…		…
337	replace the tagged value in the decoded data structure, or no values,	382	replace the tagged value in the decoded data structure, or no values,
338	which will result in default handling, which currently means the decoder	383	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.	384	creates a C<CBOR::XS::Tagged> object to hold the tag and the value.
340		385
341	When the filter is cleared (the default state), the default filter	386	When the filter is cleared (the default state), the default filter
342	function, C<CBOR::XS::default_filter>, is used. This function simply looks	387	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	388	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	389	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.	390	responsible for decoding the value. If no entry exists, it returns no
		391	values. C<CBOR::XS> provides a number of default filter functions already,
		392	the the C<%CBOR::XS::FILTER> hash can be freely extended with more.
		393
		394	C<CBOR::XS> additionally provides an alternative filter function that is
		395	supposed to be safe to use with untrusted data (which the default filter
		396	might not), called C<CBOR::XS::safe_filter>, which works the same as
		397	the C<default_filter> but uses the C<%CBOR::XS::SAFE_FILTER> variable
		398	instead. It is prepopulated with the tag decoding functions that are
		399	deemed safe (basically the same as C<%CBOR::XS::FILTER> without all
		400	the bignum tags), and can be extended by user code as wlel, although,
		401	obviously, one should be very careful about adding decoding functions
		402	here, since the expectation is that they are safe to use on untrusted
		403	data, after all.
346		404
347	Example: decode all tags not handled internally into C<CBOR::XS::Tagged>	405	Example: decode all tags not handled internally into C<CBOR::XS::Tagged>
348	objects, with no other special handling (useful when working with	406	objects, with no other special handling (useful when working with
349	potentially "unsafe" CBOR data).	407	potentially "unsafe" CBOR data).
350		408
…		…
357	my ($tag, $value);	415	my ($tag, $value);
358		416
359	"tag 1347375694 value $value"	417	"tag 1347375694 value $value"
360	};	418	};
361		419
		420	Example: provide your own filter function that looks up tags in your own
		421	hash:
		422
		423	my %my_filter = (
		424	998347484 => sub {
		425	my ($tag, $value);
		426
		427	"tag 998347484 value $value"
		428	};
		429	);
		430
		431	my $coder = CBOR::XS->new->filter (sub {
		432	&{ $my_filter{$_[0]} or return }
		433	});
		434
		435
		436	Example: use the safe filter function (see L<SECURITY CONSIDERATIONS> for
		437	more considerations on security).
		438
		439	CBOR::XS->new->filter (\&CBOR::XS::safe_filter)->decode ($cbor_data);
		440
362	=item $cbor_data = $cbor->encode ($perl_scalar)	441	=item $cbor_data = $cbor->encode ($perl_scalar)
363		442
364	Converts the given Perl data structure (a scalar value) to its CBOR	443	Converts the given Perl data structure (a scalar value) to its CBOR
365	representation.	444	representation.
366		445
…		…
375	when there is trailing garbage after the CBOR string, it will silently	454	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.	455	stop parsing there and return the number of characters consumed so far.
377		456
378	This is useful if your CBOR texts are not delimited by an outer protocol	457	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	458	and you need to know where the first CBOR string ends amd the next one
380	starts.	459	starts - CBOR strings are self-delimited, so it is possible to concatenate
		460	CBOR strings without any delimiters or size fields and recover their data.
381		461
382	CBOR::XS->new->decode_prefix ("......")	462	CBOR::XS->new->decode_prefix ("......")
383	=> ("...", 3)	463	=> ("...", 3)
384		464
385	=back	465	=back
…		…
440		520
441	Resets the incremental decoder. This throws away any saved state, so that	521	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	522	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<$buffer> again.	523	a new CBOR value from the beginning of the C<$buffer> again.
444		524
445	This method can be caled at any time, but it I<must> be called if you want	525	This method can be called at any time, but it I<must> be called if you want
446	to change your C<$buffer> or there was a decoding error and you want to	526	to change your C<$buffer> or there was a decoding error and you want to
447	reuse the C<$cbor> object for future incremental parsings.	527	reuse the C<$cbor> object for future incremental parsings.
448		528
449	=back	529	=back
450		530
…		…
589	"$x"; # stringified	669	"$x"; # stringified
590	$x .= ""; # another, more awkward way to stringify	670	$x .= ""; # another, more awkward way to stringify
591	print $x; # perl does it for you, too, quite often	671	print $x; # perl does it for you, too, quite often
592		672
593	You can force whether a string is encoded as byte or text string by using	673	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):	674	C<utf8::upgrade> and C<utf8::downgrade> (if C<text_strings> is disabled).
595		675
596	utf8::upgrade $x; # encode $x as text string	676	utf8::upgrade $x; # encode $x as text string
597	utf8::downgrade $x; # encode $x as byte string	677	utf8::downgrade $x; # encode $x as byte string
		678
		679	More options are available, see L<TYPE CASTS>, below, and the C<text_keys>
		680	and C<text_strings> options.
598		681
599	Perl doesn't define what operations up- and downgrade strings, so if the	682	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	683	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	684	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>.	685	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	701	represent numerical values are supported, but might suffer loss of
619	precision.	702	precision.
620		703
621	=back	704	=back
622		705
		706	=head2 TYPE CASTS
		707
		708	B<EXPERIMENTAL>: As an experimental extension, C<CBOR::XS> allows you to
		709	force specific cbor types to be used when encoding. That allows you to
		710	encode types not normally accessible (e.g. half floats) as well as force
		711	string types even when C<text_strings> is in effect.
		712
		713	Type forcing is done by calling a special "cast" function which keeps a
		714	copy of the value and returns a new value that can be handed over to any
		715	CBOR encoder function.
		716
		717	The following casts are currently available (all of which are unary operators):
		718
		719	=over
		720
		721	=item CBOR::XS::as_int $value
		722
		723	Forces the value to be encoded as some form of (basic, not bignum) integer
		724	type.
		725
		726	=item CBOR::XS::as_text $value
		727
		728	Forces the value to be encoded as (UTF-8) text values.
		729
		730	=item CBOR::XS::as_bytes $value
		731
		732	Forces the value to be encoded as a (binary) string value.
		733
		734	=item CBOR::XS::as_bool $value
		735
		736	Converts a Perl boolean (which can be any kind of scalar) into a CBOR
		737	boolean. Strictly the same, but shorter to write, than:
		738
		739	$value ? Types::Serialiser::true : Types::Serialiser::false
		740
		741	=item CBOR::XS::as_float16 $value
		742
		743	Forces half-float (IEEE 754 binary16) encoding of the given value.
		744
		745	=item CBOR::XS::as_float32 $value
		746
		747	Forces single-float (IEEE 754 binary32) encoding of the given value.
		748
		749	=item CBOR::XS::as_float64 $value
		750
		751	Forces double-float (IEEE 754 binary64) encoding of the given value.
		752
		753	=item, CBOR::XS::as_cbor $cbor_text
		754
		755	Bot a type cast per-se, this type cast forces the argument to eb encoded
		756	as-is. This can be used to embed pre-encoded CBOR data.
		757
		758	Note that no checking on the validity of the C<$cbor_text> is done - it's
		759	the callers responsibility to correctly encode values.
		760
		761	=back
		762
		763	Example: encode a perl string as binary even though C<text_strings> is in
		764	effect.
		765
		766	CBOR::XS->new->text_strings->encode ([4, "text", CBOR::XS::bytes "bytevalue"]);
		767
		768	=cut
		769
		770	sub CBOR::XS::as_cbor ($) { bless [$_[0], 0, undef], CBOR::XS::Tagged:: }
		771	sub CBOR::XS::as_int ($) { bless [$_[0], 1, undef], CBOR::XS::Tagged:: }
		772	sub CBOR::XS::as_bytes ($) { bless [$_[0], 2, undef], CBOR::XS::Tagged:: }
		773	sub CBOR::XS::as_text ($) { bless [$_[0], 3, undef], CBOR::XS::Tagged:: }
		774	sub CBOR::XS::as_float16 ($) { bless [$_[0], 4, undef], CBOR::XS::Tagged:: }
		775	sub CBOR::XS::as_float32 ($) { bless [$_[0], 5, undef], CBOR::XS::Tagged:: }
		776	sub CBOR::XS::as_float64 ($) { bless [$_[0], 6, undef], CBOR::XS::Tagged:: }
		777
		778	sub CBOR::XS::as_bool ($) { $_[0] ? $Types::Serialiser::true : $Types::Serialiser::false }
		779
623	=head2 OBJECT SERIALISATION	780	=head2 OBJECT SERIALISATION
624		781
625	This module implements both a CBOR-specific and the generic	782	This module implements both a CBOR-specific and the generic
626	L<Types::Serialier> object serialisation protocol. The following	783	L<Types::Serialier> object serialisation protocol. The following
627	subsections explain both methods.	784	subsections explain both methods.
…		…
935	objects. The corresponding C<Math::BigFloat::TO_CBOR> method I<always>	1092	objects. The corresponding C<Math::BigFloat::TO_CBOR> method I<always>
936	encodes into a decimal fraction (either tag 4 or 264).	1093	encodes into a decimal fraction (either tag 4 or 264).
937		1094
938	NaN and infinities are not encoded properly, as they cannot be represented	1095	NaN and infinities are not encoded properly, as they cannot be represented
939	in CBOR.	1096	in CBOR.
		1097
		1098	See L<BIGNUM SECURITY CONSIDERATIONS> for more info.
		1099
		1100	=item 30 (rational numbers)
		1101
		1102	These tags are decoded into L<Math::BigRat> objects. The corresponding
		1103	C<Math::BigRat::TO_CBOR> method encodes rational numbers with denominator
		1104	C<1> via their numerator only, i.e., they become normal integers or
		1105	C<bignums>.
940		1106
941	See L<BIGNUM SECURITY CONSIDERATIONS> for more info.	1107	See L<BIGNUM SECURITY CONSIDERATIONS> for more info.
942		1108
943	=item 21, 22, 23 (expected later JSON conversion)	1109	=item 21, 22, 23 (expected later JSON conversion)
944		1110
…		…
969	CBOR intact.	1135	CBOR intact.
970		1136
971		1137
972	=head1 SECURITY CONSIDERATIONS	1138	=head1 SECURITY CONSIDERATIONS
973		1139
974	When you are using CBOR in a protocol, talking to untrusted potentially	1140	Tl;dr... if you want to decode or encode CBOR from untrusted sources, you
975	hostile creatures requires relatively few measures.	1141	should start with a coder object created via C<new_safe> (which implements
		1142	the mitigations explained below):
976		1143
		1144	my $coder = CBOR::XS->new_safe;
		1145
		1146	my $data = $coder->decode ($cbor_text);
		1147	my $cbor = $coder->encode ($data);
		1148
		1149	Longer version: When you are using CBOR in a protocol, talking to
		1150	untrusted potentially hostile creatures requires some thought:
		1151
		1152	=over 4
		1153
		1154	=item Security of the CBOR decoder itself
		1155
977	First of all, your CBOR decoder should be secure, that is, should not have	1156	First and foremost, your CBOR decoder should be secure, that is, should
		1157	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	1158	exploited. Obviously, this module should ensure that and I am trying hard
979	trying hard on making that true, but you never know.	1159	on making that true, but you never know.
980		1160
		1161	=item CBOR::XS can invoke almost arbitrary callbacks during decoding
		1162
		1163	CBOR::XS supports object serialisation - decoding CBOR can cause calls
		1164	to I<any> C<THAW> method in I<any> package that exists in your process
		1165	(that is, CBOR::XS will not try to load modules, but any existing C<THAW>
		1166	method or function can be called, so they all have to be secure).
		1167
		1168	Less obviously, it will also invoke C<TO_CBOR> and C<FREEZE> methods -
		1169	even if all your C<THAW> methods are secure, encoding data structures from
		1170	untrusted sources can invoke those and trigger bugs in those.
		1171
		1172	So, if you are not sure about the security of all the modules you
		1173	have loaded (you shouldn't), you should disable this part using
		1174	C<forbid_objects> or using C<new_safe>.
		1175
		1176	=item CBOR can be extended with tags that call library code
		1177
		1178	CBOR can be extended with tags, and C<CBOR::XS> has a registry of
		1179	conversion functions for many existing tags that can be extended via
		1180	third-party modules (see the C<filter> method).
		1181
		1182	If you don't trust these, you should configure the "safe" filter function,
		1183	C<CBOR::XS::safe_filter> (C<new_safe> does this), which by default only
		1184	includes conversion functions that are considered "safe" by the author
		1185	(but again, they can be extended by third party modules).
		1186
		1187	Depending on your level of paranoia, you can use the "safe" filter:
		1188
		1189	$cbor->filter (\&CBOR::XS::safe_filter);
		1190
		1191	... your own filter...
		1192
		1193	$cbor->filter (sub { ... do your stuffs here ... });
		1194
		1195	... or even no filter at all, disabling all tag decoding:
		1196
		1197	$cbor->filter (sub { });
		1198
		1199	This is never a problem for encoding, as the tag mechanism only exists in
		1200	CBOR texts.
		1201
		1202	=item Resource-starving attacks: object memory usage
		1203
981	Second, you need to avoid resource-starving attacks. That means you should	1204	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	1205	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	1206	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	1207	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	1208	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	1209	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	1210	C<max_size> - done by C<new_safe>), it might be too late when you already
988	the size before you accept the string.	1211	have it in memory, so you might want to check the size before you accept
		1212	the string.
989		1213
		1214	As for encoding, it is possible to construct data structures that are
		1215	relatively small but result in large CBOR texts (for example by having an
		1216	array full of references to the same big data structure, which will all be
		1217	deep-cloned during encoding by default). This is rarely an actual issue
		1218	(and the worst case is still just running out of memory), but you can
		1219	reduce this risk by using C<allow_sharing>.
		1220
		1221	=item Resource-starving attacks: stack overflows
		1222
990	Third, CBOR::XS recurses using the C stack when decoding objects and	1223	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	1224	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	1225	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	1226	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	1227	temporary). If that is exceeded, the program crashes. To be conservative,
995	conservative, the default nesting limit is set to 512. If your process	1228	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	1229	stack, you should adjust this setting accordingly with the C<max_depth>
997	C<max_depth> method.	1230	method.
		1231
		1232	=item Resource-starving attacks: CPU en-/decoding complexity
		1233
		1234	CBOR::XS will use the L<Math::BigInt>, L<Math::BigFloat> and
		1235	L<Math::BigRat> libraries to represent encode/decode bignums. These can be
		1236	very slow (as in, centuries of CPU time) and can even crash your program
		1237	(and are generally not very trustworthy). See the next section on bignum
		1238	security for details.
		1239
		1240	=item Data breaches: leaking information in error messages
		1241
		1242	CBOR::XS might leak contents of your Perl data structures in its error
		1243	messages, so when you serialise sensitive information you might want to
		1244	make sure that exceptions thrown by CBOR::XS will not end up in front of
		1245	untrusted eyes.
		1246
		1247	=item Something else...
998		1248
999	Something else could bomb you, too, that I forgot to think of. In that	1249	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...	1250	case, you get to keep the pieces. I am always open for hints, though...
1001		1251
1002	Also keep in mind that CBOR::XS might leak contents of your Perl data	1252	=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		1253
1007		1254
1008	=head1 BIGNUM SECURITY CONSIDERATIONS	1255	=head1 BIGNUM SECURITY CONSIDERATIONS
1009		1256
1010	CBOR::XS provides a C<TO_CBOR> method for both L<Math::BigInt> and	1257	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	1258	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	1259	way, that is, either a CBOR integer, a CBOR bigint/decimal fraction (tag
1013	4) or an arbitrary-exponent decimal fraction (tag 264).	1260	4) or an arbitrary-exponent decimal fraction (tag 264). Rational numbers
		1261	(L<Math::BigRat>, tag 30) can also contain bignums as members.
1014		1262
1015	It will also understand base-2 bigfloat or arbitrary-exponent bigfloats	1263	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.	1264	bigfloats (tags 5 and 265), but it will never generate these on its own.
1017		1265
1018	Using the built-in L<Math::BigInt::Calc> support, encoding and decoding	1266	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	1267	decimal fractions is generally fast. Decoding bigints can be slow for very
1020	big numbers, and decoding bigfloats or arbitrary-exponent bigfloats can be	1268	big numbers (tens of thousands of digits, something that could potentially
1021	extremely slow (minutes, decades) for large exponents.	1269	be caught by limiting the size of CBOR texts), and decoding bigfloats or
		1270	arbitrary-exponent bigfloats can be I<extremely> slow (minutes, decades)
		1271	for large exponents (roughly 40 bit and longer).
1022		1272
1023	Additionally, L<Math::BigInt> can take advantage of other bignum	1273	Additionally, L<Math::BigInt> can take advantage of other bignum
1024	libraries, such as L<Math::GMP>, which cannot handle big	1274	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,	1275	exponents, and might simply abort or crash your program, due to their code
1026	due to their code quality.	1276	quality.
1027		1277
1028	This can be a concern if you want to parse untrusted CBOR. If it is, you	1278	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,	1279	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	1280	types. You should also disable types 5 and 265, as these can be slow even
1031	types 4, 5, 264 and 265).	1281	without bigints.
		1282
		1283	Disabling bigints will also partially or fully disable types that rely on
		1284	them, e.g. rational numbers that use bignums.
1032		1285
1033		1286
1034	=head1 CBOR IMPLEMENTATION NOTES	1287	=head1 CBOR IMPLEMENTATION NOTES
1035		1288
1036	This section contains some random implementation notes. They do not	1289	This section contains some random implementation notes. They do not
…		…
1053	=head1 LIMITATIONS ON PERLS WITHOUT 64-BIT INTEGER SUPPORT	1306	=head1 LIMITATIONS ON PERLS WITHOUT 64-BIT INTEGER SUPPORT
1054		1307
1055	On perls that were built without 64 bit integer support (these are rare	1308	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	1309	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	1310	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	1311	value in CBOR is very limited - most likely, these 64 bit values will
1059	be truncated, corrupted, or otherwise not decoded correctly. This also	1312	be truncated, corrupted, or otherwise not decoded correctly. This also
1060	includes string, array and map sizes that are stored as 64 bit integers.	1313	includes string, float, array and map sizes that are stored as 64 bit
		1314	integers.
1061		1315
1062		1316
1063	=head1 THREADS	1317	=head1 THREADS
1064		1318
1065	This module is I<not> guaranteed to be thread safe and there are no	1319	This module is I<not> guaranteed to be thread safe and there are no
…		…
1078		1332
1079	Please refrain from using rt.cpan.org or any other bug reporting	1333	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.	1334	service. I put the contact address into my modules for a reason.
1081		1335
1082	=cut	1336	=cut
		1337
		1338	# clumsy and slow hv_store-in-hash helper function
		1339	sub _hv_store {
		1340	$_[0]{$_[1]} = $_[2];
		1341	}
1083		1342
1084	our %FILTER = (	1343	our %FILTER = (
1085	0 => sub { # rfc4287 datetime, utf-8	1344	0 => sub { # rfc4287 datetime, utf-8
1086	require Time::Piece;	1345	require Time::Piece;
1087	# Time::Piece::Strptime uses the "incredibly flexible date parsing routine"	1346	# Time::Piece::Strptime uses the "incredibly flexible date parsing routine"
…		…
1110	scalar Time::Piece::gmtime (pop)	1369	scalar Time::Piece::gmtime (pop)
1111	},	1370	},
1112		1371
1113	2 => sub { # pos bigint	1372	2 => sub { # pos bigint
1114	require Math::BigInt;	1373	require Math::BigInt;
1115	Math::BigInt->from_hex ("0x" . unpack "H*", pop)	1374	Math::BigInt->new ("0x" . unpack "H*", pop)
1116	},	1375	},
1117		1376
1118	3 => sub { # neg bigint	1377	3 => sub { # neg bigint
1119	require Math::BigInt;	1378	require Math::BigInt;
1120	-Math::BigInt->from_hex ("0x" . unpack "H*", pop)	1379	-Math::BigInt->new ("0x" . unpack "H*", pop)
1121	},	1380	},
1122		1381
1123	4 => sub { # decimal fraction, array	1382	4 => sub { # decimal fraction, array
1124	require Math::BigFloat;	1383	require Math::BigFloat;
1125	Math::BigFloat->new ($_[1][1] . "E" . $_[1][0])	1384	Math::BigFloat->new ($_[1][1] . "E" . $_[1][0])
1126	},	1385	},
1127		1386
		1387	264 => sub { # decimal fraction with arbitrary exponent
		1388	require Math::BigFloat;
		1389	Math::BigFloat->new ($_[1][1] . "E" . $_[1][0])
		1390	},
		1391
1128	5 => sub { # bigfloat, array	1392	5 => sub { # bigfloat, array
1129	require Math::BigFloat;	1393	require Math::BigFloat;
1130	scalar Math::BigFloat->new ($_[1][1]) * Math::BigFloat->new (2)->bpow ($_[1][0])	1394	scalar Math::BigFloat->new ($_[1][1]) * Math::BigFloat->new (2)->bpow ($_[1][0])
1131	},	1395	},
1132		1396
		1397	265 => sub { # bigfloat with arbitrary exponent
		1398	require Math::BigFloat;
		1399	scalar Math::BigFloat->new ($_[1][1]) * Math::BigFloat->new (2)->bpow ($_[1][0])
		1400	},
		1401
		1402	30 => sub { # rational number
		1403	require Math::BigRat;
		1404	Math::BigRat->new ("$_[1][0]/$_[1][1]") # separate parameters only work in recent versons
		1405	},
		1406
1133	21 => sub { pop }, # expected conversion to base64url encoding	1407	21 => sub { pop }, # expected conversion to base64url encoding
1134	22 => sub { pop }, # expected conversion to base64 encoding	1408	22 => sub { pop }, # expected conversion to base64 encoding
1135	23 => sub { pop }, # expected conversion to base16 encoding	1409	23 => sub { pop }, # expected conversion to base16 encoding
1136		1410
1137	# 24 # embedded cbor, byte string	1411	# 24 # embedded cbor, byte string
…		…
1143		1417
1144	# 33 # base64url rfc4648, utf-8	1418	# 33 # base64url rfc4648, utf-8
1145	# 34 # base64 rfc46484, utf-8	1419	# 34 # base64 rfc46484, utf-8
1146	# 35 # regex pcre/ecma262, utf-8	1420	# 35 # regex pcre/ecma262, utf-8
1147	# 36 # mime message rfc2045, utf-8	1421	# 36 # mime message rfc2045, utf-8
1148
1149	264 => sub { # decimal fraction with arbitrary exponent
1150	require Math::BigFloat;
1151	Math::BigFloat->new ($_[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	);	1422	);
1159		1423
1160	sub CBOR::XS::default_filter {	1424	sub default_filter {
1161	&{ $FILTER{$_[0]} or return }	1425	&{ $FILTER{$_[0]} or return }
		1426	}
		1427
		1428	our %SAFE_FILTER = map { $_ => $FILTER{$_} } 0, 1, 21, 22, 23, 32;
		1429
		1430	sub safe_filter {
		1431	&{ $SAFE_FILTER{$_[0]} or return }
1162	}	1432	}
1163		1433
1164	sub URI::TO_CBOR {	1434	sub URI::TO_CBOR {
1165	my $uri = $_[0]->as_string;	1435	my $uri = $_[0]->as_string;
1166	utf8::upgrade $uri;	1436	utf8::upgrade $uri;
…		…
1183	-9223372036854775808 <= $e && $e <= 18446744073709551615	1453	-9223372036854775808 <= $e && $e <= 18446744073709551615
1184	? tag 4, [$e->numify, $m]	1454	? tag 4, [$e->numify, $m]
1185	: tag 264, [$e, $m]	1455	: tag 264, [$e, $m]
1186	}	1456	}
1187		1457
		1458	sub Math::BigRat::TO_CBOR {
		1459	my ($n, $d) = $_[0]->parts;
		1460
		1461	# older versions of BigRat need *1, as they not always return numbers
		1462
		1463	$d*1 == 1
		1464	? $n*1
		1465	: tag 30, [$n1, $d1]
		1466	}
		1467
1188	sub Time::Piece::TO_CBOR {	1468	sub Time::Piece::TO_CBOR {
1189	tag 1, 0 + $_[0]->epoch	1469	tag 1, 0 + $_[0]->epoch
1190	}	1470	}
1191		1471
1192	XSLoader::load "CBOR::XS", $VERSION;	1472	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 1.76 by root, Tue Dec 1 01:49:47 2020 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 1.76 by root, Tue Dec 1 01:49:47 2020 UTC