1 |
root |
1.1 |
=head1 NAME |
2 |
|
|
|
3 |
|
|
CBOR::XS - Concise Binary Object Representation (CBOR, RFC7049) |
4 |
|
|
|
5 |
|
|
=encoding utf-8 |
6 |
|
|
|
7 |
|
|
=head1 SYNOPSIS |
8 |
|
|
|
9 |
|
|
use CBOR::XS; |
10 |
|
|
|
11 |
|
|
$binary_cbor_data = encode_cbor $perl_value; |
12 |
|
|
$perl_value = decode_cbor $binary_cbor_data; |
13 |
|
|
|
14 |
|
|
# OO-interface |
15 |
|
|
|
16 |
|
|
$coder = CBOR::XS->new; |
17 |
root |
1.6 |
$binary_cbor_data = $coder->encode ($perl_value); |
18 |
|
|
$perl_value = $coder->decode ($binary_cbor_data); |
19 |
|
|
|
20 |
|
|
# prefix decoding |
21 |
|
|
|
22 |
|
|
my $many_cbor_strings = ...; |
23 |
|
|
while (length $many_cbor_strings) { |
24 |
|
|
my ($data, $length) = $cbor->decode_prefix ($many_cbor_strings); |
25 |
|
|
# data was decoded |
26 |
|
|
substr $many_cbor_strings, 0, $length, ""; # remove decoded cbor string |
27 |
|
|
} |
28 |
root |
1.1 |
|
29 |
|
|
=head1 DESCRIPTION |
30 |
|
|
|
31 |
root |
1.21 |
WARNING! This module is very new, and not very well tested (that's up |
32 |
|
|
to you to do). Furthermore, details of the implementation might change |
33 |
|
|
freely before version 1.0. And lastly, most extensions depend on an IANA |
34 |
|
|
assignment, and until that assignment is official, this implementation is |
35 |
|
|
not interoperable with other implementations (even future versions of this |
36 |
|
|
module) until the assignment is done. |
37 |
root |
1.9 |
|
38 |
|
|
You are still invited to try out CBOR, and this module. |
39 |
root |
1.5 |
|
40 |
|
|
This module converts Perl data structures to the Concise Binary Object |
41 |
|
|
Representation (CBOR) and vice versa. CBOR is a fast binary serialisation |
42 |
|
|
format that aims to use a superset of the JSON data model, i.e. when you |
43 |
|
|
can represent something in JSON, you should be able to represent it in |
44 |
|
|
CBOR. |
45 |
root |
1.1 |
|
46 |
root |
1.9 |
In short, CBOR is a faster and very compact binary alternative to JSON, |
47 |
root |
1.10 |
with the added ability of supporting serialisation of Perl objects. (JSON |
48 |
|
|
often compresses better than CBOR though, so if you plan to compress the |
49 |
|
|
data later you might want to compare both formats first). |
50 |
root |
1.5 |
|
51 |
root |
1.15 |
To give you a general idea about speed, with texts in the megabyte range, |
52 |
|
|
C<CBOR::XS> usually encodes roughly twice as fast as L<Storable> or |
53 |
|
|
L<JSON::XS> and decodes about 15%-30% faster than those. The shorter the |
54 |
|
|
data, the worse L<Storable> performs in comparison. |
55 |
|
|
|
56 |
|
|
As for compactness, C<CBOR::XS> encoded data structures are usually about |
57 |
|
|
20% smaller than the same data encoded as (compact) JSON or L<Storable>. |
58 |
root |
1.14 |
|
59 |
root |
1.21 |
In addition to the core CBOR data format, this module implements a number |
60 |
|
|
of extensions, to support cyclic and self-referencing data structures |
61 |
root |
1.25 |
(see C<allow_sharing>), string deduplication (see C<pack_strings>) and |
62 |
root |
1.21 |
scalar references (always enabled). |
63 |
|
|
|
64 |
root |
1.5 |
The primary goal of this module is to be I<correct> and the secondary goal |
65 |
|
|
is to be I<fast>. To reach the latter goal it was written in C. |
66 |
root |
1.1 |
|
67 |
|
|
See MAPPING, below, on how CBOR::XS maps perl values to CBOR values and |
68 |
|
|
vice versa. |
69 |
|
|
|
70 |
|
|
=cut |
71 |
|
|
|
72 |
|
|
package CBOR::XS; |
73 |
|
|
|
74 |
|
|
use common::sense; |
75 |
|
|
|
76 |
root |
1.24 |
our $VERSION = 0.09; |
77 |
root |
1.1 |
our @ISA = qw(Exporter); |
78 |
|
|
|
79 |
|
|
our @EXPORT = qw(encode_cbor decode_cbor); |
80 |
|
|
|
81 |
|
|
use Exporter; |
82 |
|
|
use XSLoader; |
83 |
|
|
|
84 |
root |
1.6 |
use Types::Serialiser; |
85 |
|
|
|
86 |
root |
1.3 |
our $MAGIC = "\xd9\xd9\xf7"; |
87 |
|
|
|
88 |
root |
1.1 |
=head1 FUNCTIONAL INTERFACE |
89 |
|
|
|
90 |
|
|
The following convenience methods are provided by this module. They are |
91 |
|
|
exported by default: |
92 |
|
|
|
93 |
|
|
=over 4 |
94 |
|
|
|
95 |
|
|
=item $cbor_data = encode_cbor $perl_scalar |
96 |
|
|
|
97 |
|
|
Converts the given Perl data structure to CBOR representation. Croaks on |
98 |
|
|
error. |
99 |
|
|
|
100 |
|
|
=item $perl_scalar = decode_cbor $cbor_data |
101 |
|
|
|
102 |
|
|
The opposite of C<encode_cbor>: expects a valid CBOR string to parse, |
103 |
|
|
returning the resulting perl scalar. Croaks on error. |
104 |
|
|
|
105 |
|
|
=back |
106 |
|
|
|
107 |
|
|
|
108 |
|
|
=head1 OBJECT-ORIENTED INTERFACE |
109 |
|
|
|
110 |
|
|
The object oriented interface lets you configure your own encoding or |
111 |
|
|
decoding style, within the limits of supported formats. |
112 |
|
|
|
113 |
|
|
=over 4 |
114 |
|
|
|
115 |
|
|
=item $cbor = new CBOR::XS |
116 |
|
|
|
117 |
|
|
Creates a new CBOR::XS object that can be used to de/encode CBOR |
118 |
|
|
strings. All boolean flags described below are by default I<disabled>. |
119 |
|
|
|
120 |
|
|
The mutators for flags all return the CBOR object again and thus calls can |
121 |
|
|
be chained: |
122 |
|
|
|
123 |
|
|
my $cbor = CBOR::XS->new->encode ({a => [1,2]}); |
124 |
|
|
|
125 |
|
|
=item $cbor = $cbor->max_depth ([$maximum_nesting_depth]) |
126 |
|
|
|
127 |
|
|
=item $max_depth = $cbor->get_max_depth |
128 |
|
|
|
129 |
|
|
Sets the maximum nesting level (default C<512>) accepted while encoding |
130 |
|
|
or decoding. If a higher nesting level is detected in CBOR data or a Perl |
131 |
|
|
data structure, then the encoder and decoder will stop and croak at that |
132 |
|
|
point. |
133 |
|
|
|
134 |
|
|
Nesting level is defined by number of hash- or arrayrefs that the encoder |
135 |
|
|
needs to traverse to reach a given point or the number of C<{> or C<[> |
136 |
|
|
characters without their matching closing parenthesis crossed to reach a |
137 |
|
|
given character in a string. |
138 |
|
|
|
139 |
|
|
Setting the maximum depth to one disallows any nesting, so that ensures |
140 |
|
|
that the object is only a single hash/object or array. |
141 |
|
|
|
142 |
|
|
If no argument is given, the highest possible setting will be used, which |
143 |
|
|
is rarely useful. |
144 |
|
|
|
145 |
|
|
Note that nesting is implemented by recursion in C. The default value has |
146 |
|
|
been chosen to be as large as typical operating systems allow without |
147 |
|
|
crashing. |
148 |
|
|
|
149 |
|
|
See SECURITY CONSIDERATIONS, below, for more info on why this is useful. |
150 |
|
|
|
151 |
|
|
=item $cbor = $cbor->max_size ([$maximum_string_size]) |
152 |
|
|
|
153 |
|
|
=item $max_size = $cbor->get_max_size |
154 |
|
|
|
155 |
|
|
Set the maximum length a CBOR string may have (in bytes) where decoding |
156 |
|
|
is being attempted. The default is C<0>, meaning no limit. When C<decode> |
157 |
|
|
is called on a string that is longer then this many bytes, it will not |
158 |
|
|
attempt to decode the string but throw an exception. This setting has no |
159 |
|
|
effect on C<encode> (yet). |
160 |
|
|
|
161 |
|
|
If no argument is given, the limit check will be deactivated (same as when |
162 |
|
|
C<0> is specified). |
163 |
|
|
|
164 |
|
|
See SECURITY CONSIDERATIONS, below, for more info on why this is useful. |
165 |
|
|
|
166 |
root |
1.19 |
=item $cbor = $cbor->allow_unknown ([$enable]) |
167 |
|
|
|
168 |
|
|
=item $enabled = $cbor->get_allow_unknown |
169 |
|
|
|
170 |
|
|
If C<$enable> is true (or missing), then C<encode> will I<not> throw an |
171 |
|
|
exception when it encounters values it cannot represent in CBOR (for |
172 |
|
|
example, filehandles) but instead will encode a CBOR C<error> value. |
173 |
|
|
|
174 |
|
|
If C<$enable> is false (the default), then C<encode> will throw an |
175 |
|
|
exception when it encounters anything it cannot encode as CBOR. |
176 |
|
|
|
177 |
|
|
This option does not affect C<decode> in any way, and it is recommended to |
178 |
|
|
leave it off unless you know your communications partner. |
179 |
|
|
|
180 |
root |
1.20 |
=item $cbor = $cbor->allow_sharing ([$enable]) |
181 |
root |
1.19 |
|
182 |
root |
1.20 |
=item $enabled = $cbor->get_allow_sharing |
183 |
root |
1.19 |
|
184 |
|
|
If C<$enable> is true (or missing), then C<encode> will not double-encode |
185 |
root |
1.20 |
values that have been referenced before (e.g. when the same object, such |
186 |
|
|
as an array, is referenced multiple times), but instead will emit a |
187 |
|
|
reference to the earlier value. |
188 |
root |
1.19 |
|
189 |
|
|
This means that such values will only be encoded once, and will not result |
190 |
|
|
in a deep cloning of the value on decode, in decoders supporting the value |
191 |
root |
1.25 |
sharing extension. This also makes it possible to encode cyclic data |
192 |
|
|
structures. |
193 |
root |
1.19 |
|
194 |
root |
1.21 |
It is recommended to leave it off unless you know your |
195 |
|
|
communication partner supports the value sharing extensions to CBOR |
196 |
root |
1.26 |
(L<http://cbor.schmorp.de/value-sharing>), as without decoder support, the |
197 |
root |
1.25 |
resulting data structure might be unusable. |
198 |
root |
1.21 |
|
199 |
root |
1.19 |
Detecting shared values incurs a runtime overhead when values are encoded |
200 |
|
|
that have a reference counter large than one, and might unnecessarily |
201 |
|
|
increase the encoded size, as potentially shared values are encode as |
202 |
|
|
sharable whether or not they are actually shared. |
203 |
|
|
|
204 |
root |
1.20 |
At the moment, only targets of references can be shared (e.g. scalars, |
205 |
|
|
arrays or hashes pointed to by a reference). Weirder constructs, such as |
206 |
|
|
an array with multiple "copies" of the I<same> string, which are hard but |
207 |
|
|
not impossible to create in Perl, are not supported (this is the same as |
208 |
root |
1.25 |
with L<Storable>). |
209 |
root |
1.19 |
|
210 |
root |
1.25 |
If C<$enable> is false (the default), then C<encode> will encode shared |
211 |
|
|
data structures repeatedly, unsharing them in the process. Cyclic data |
212 |
|
|
structures cannot be encoded in this mode. |
213 |
root |
1.19 |
|
214 |
|
|
This option does not affect C<decode> in any way - shared values and |
215 |
root |
1.21 |
references will always be decoded properly if present. |
216 |
|
|
|
217 |
root |
1.25 |
=item $cbor = $cbor->pack_strings ([$enable]) |
218 |
root |
1.21 |
|
219 |
root |
1.25 |
=item $enabled = $cbor->get_pack_strings |
220 |
root |
1.21 |
|
221 |
|
|
If C<$enable> is true (or missing), then C<encode> will try not to encode |
222 |
|
|
the same string twice, but will instead encode a reference to the string |
223 |
root |
1.25 |
instead. Depending on your data format, this can save a lot of space, but |
224 |
root |
1.21 |
also results in a very large runtime overhead (expect encoding times to be |
225 |
|
|
2-4 times as high as without). |
226 |
|
|
|
227 |
|
|
It is recommended to leave it off unless you know your |
228 |
|
|
communications partner supports the stringref extension to CBOR |
229 |
root |
1.26 |
(L<http://cbor.schmorp.de/stringref>), as without decoder support, the |
230 |
root |
1.25 |
resulting data structure might not be usable. |
231 |
root |
1.21 |
|
232 |
root |
1.25 |
If C<$enable> is false (the default), then C<encode> will encode strings |
233 |
|
|
the standard CBOR way. |
234 |
root |
1.21 |
|
235 |
|
|
This option does not affect C<decode> in any way - string references will |
236 |
|
|
always be decoded properly if present. |
237 |
root |
1.19 |
|
238 |
root |
1.23 |
=item $cbor = $cbor->filter ([$cb->($tag, $value)]) |
239 |
|
|
|
240 |
|
|
=item $cb_or_undef = $cbor->get_filter |
241 |
|
|
|
242 |
root |
1.24 |
Sets or replaces the tagged value decoding filter (when C<$cb> is |
243 |
|
|
specified) or clears the filter (if no argument or C<undef> is provided). |
244 |
|
|
|
245 |
|
|
The filter callback is called only during decoding, when a non-enforced |
246 |
|
|
tagged value has been decoded (see L<TAG HANDLING AND EXTENSIONS> for a |
247 |
|
|
list of enforced tags). For specific tags, it's often better to provide a |
248 |
|
|
default converter using the C<%CBOR::XS::FILTER> hash (see below). |
249 |
|
|
|
250 |
|
|
The first argument is the numerical tag, the second is the (decoded) value |
251 |
|
|
that has been tagged. |
252 |
|
|
|
253 |
|
|
The filter function should return either exactly one value, which will |
254 |
|
|
replace the tagged value in the decoded data structure, or no values, |
255 |
|
|
which will result in default handling, which currently means the decoder |
256 |
|
|
creates a C<CBOR::XS::Tagged> object to hold the tag and the value. |
257 |
|
|
|
258 |
|
|
When the filter is cleared (the default state), the default filter |
259 |
|
|
function, C<CBOR::XS::default_filter>, is used. This function simply looks |
260 |
|
|
up the tag in the C<%CBOR::XS::FILTER> hash. If an entry exists it must be |
261 |
|
|
a code reference that is called with tag and value, and is responsible for |
262 |
|
|
decoding the value. If no entry exists, it returns no values. |
263 |
|
|
|
264 |
|
|
Example: decode all tags not handled internally into CBOR::XS::Tagged |
265 |
|
|
objects, with no other special handling (useful when working with |
266 |
|
|
potentially "unsafe" CBOR data). |
267 |
|
|
|
268 |
|
|
CBOR::XS->new->filter (sub { })->decode ($cbor_data); |
269 |
|
|
|
270 |
|
|
Example: provide a global filter for tag 1347375694, converting the value |
271 |
|
|
into some string form. |
272 |
|
|
|
273 |
|
|
$CBOR::XS::FILTER{1347375694} = sub { |
274 |
|
|
my ($tag, $value); |
275 |
|
|
|
276 |
|
|
"tag 1347375694 value $value" |
277 |
|
|
}; |
278 |
root |
1.23 |
|
279 |
root |
1.1 |
=item $cbor_data = $cbor->encode ($perl_scalar) |
280 |
|
|
|
281 |
|
|
Converts the given Perl data structure (a scalar value) to its CBOR |
282 |
|
|
representation. |
283 |
|
|
|
284 |
|
|
=item $perl_scalar = $cbor->decode ($cbor_data) |
285 |
|
|
|
286 |
|
|
The opposite of C<encode>: expects CBOR data and tries to parse it, |
287 |
|
|
returning the resulting simple scalar or reference. Croaks on error. |
288 |
|
|
|
289 |
|
|
=item ($perl_scalar, $octets) = $cbor->decode_prefix ($cbor_data) |
290 |
|
|
|
291 |
|
|
This works like the C<decode> method, but instead of raising an exception |
292 |
|
|
when there is trailing garbage after the CBOR string, it will silently |
293 |
|
|
stop parsing there and return the number of characters consumed so far. |
294 |
|
|
|
295 |
|
|
This is useful if your CBOR texts are not delimited by an outer protocol |
296 |
|
|
and you need to know where the first CBOR string ends amd the next one |
297 |
|
|
starts. |
298 |
|
|
|
299 |
|
|
CBOR::XS->new->decode_prefix ("......") |
300 |
|
|
=> ("...", 3) |
301 |
|
|
|
302 |
|
|
=back |
303 |
|
|
|
304 |
|
|
|
305 |
|
|
=head1 MAPPING |
306 |
|
|
|
307 |
|
|
This section describes how CBOR::XS maps Perl values to CBOR values and |
308 |
|
|
vice versa. These mappings are designed to "do the right thing" in most |
309 |
|
|
circumstances automatically, preserving round-tripping characteristics |
310 |
|
|
(what you put in comes out as something equivalent). |
311 |
|
|
|
312 |
|
|
For the more enlightened: note that in the following descriptions, |
313 |
|
|
lowercase I<perl> refers to the Perl interpreter, while uppercase I<Perl> |
314 |
|
|
refers to the abstract Perl language itself. |
315 |
|
|
|
316 |
|
|
|
317 |
|
|
=head2 CBOR -> PERL |
318 |
|
|
|
319 |
|
|
=over 4 |
320 |
|
|
|
321 |
root |
1.4 |
=item integers |
322 |
|
|
|
323 |
|
|
CBOR integers become (numeric) perl scalars. On perls without 64 bit |
324 |
|
|
support, 64 bit integers will be truncated or otherwise corrupted. |
325 |
|
|
|
326 |
|
|
=item byte strings |
327 |
|
|
|
328 |
|
|
Byte strings will become octet strings in Perl (the byte values 0..255 |
329 |
|
|
will simply become characters of the same value in Perl). |
330 |
|
|
|
331 |
|
|
=item UTF-8 strings |
332 |
|
|
|
333 |
|
|
UTF-8 strings in CBOR will be decoded, i.e. the UTF-8 octets will be |
334 |
|
|
decoded into proper Unicode code points. At the moment, the validity of |
335 |
|
|
the UTF-8 octets will not be validated - corrupt input will result in |
336 |
|
|
corrupted Perl strings. |
337 |
|
|
|
338 |
|
|
=item arrays, maps |
339 |
|
|
|
340 |
|
|
CBOR arrays and CBOR maps will be converted into references to a Perl |
341 |
|
|
array or hash, respectively. The keys of the map will be stringified |
342 |
|
|
during this process. |
343 |
|
|
|
344 |
root |
1.6 |
=item null |
345 |
|
|
|
346 |
|
|
CBOR null becomes C<undef> in Perl. |
347 |
|
|
|
348 |
|
|
=item true, false, undefined |
349 |
root |
1.1 |
|
350 |
root |
1.6 |
These CBOR values become C<Types:Serialiser::true>, |
351 |
|
|
C<Types:Serialiser::false> and C<Types::Serialiser::error>, |
352 |
root |
1.1 |
respectively. They are overloaded to act almost exactly like the numbers |
353 |
root |
1.6 |
C<1> and C<0> (for true and false) or to throw an exception on access (for |
354 |
|
|
error). See the L<Types::Serialiser> manpage for details. |
355 |
|
|
|
356 |
root |
1.23 |
=item tagged values |
357 |
root |
1.1 |
|
358 |
root |
1.23 |
Tagged items consists of a numeric tag and another CBOR value. |
359 |
root |
1.4 |
|
360 |
root |
1.23 |
See L<TAG HANDLING AND EXTENSIONS> and the description of C<< ->filter >> |
361 |
|
|
for details. |
362 |
root |
1.4 |
|
363 |
|
|
=item anything else |
364 |
|
|
|
365 |
|
|
Anything else (e.g. unsupported simple values) will raise a decoding |
366 |
|
|
error. |
367 |
root |
1.1 |
|
368 |
|
|
=back |
369 |
|
|
|
370 |
|
|
|
371 |
|
|
=head2 PERL -> CBOR |
372 |
|
|
|
373 |
|
|
The mapping from Perl to CBOR is slightly more difficult, as Perl is a |
374 |
|
|
truly typeless language, so we can only guess which CBOR type is meant by |
375 |
|
|
a Perl value. |
376 |
|
|
|
377 |
|
|
=over 4 |
378 |
|
|
|
379 |
|
|
=item hash references |
380 |
|
|
|
381 |
root |
1.4 |
Perl hash references become CBOR maps. As there is no inherent ordering in |
382 |
|
|
hash keys (or CBOR maps), they will usually be encoded in a pseudo-random |
383 |
|
|
order. |
384 |
|
|
|
385 |
|
|
Currently, tied hashes will use the indefinite-length format, while normal |
386 |
|
|
hashes will use the fixed-length format. |
387 |
root |
1.1 |
|
388 |
|
|
=item array references |
389 |
|
|
|
390 |
root |
1.4 |
Perl array references become fixed-length CBOR arrays. |
391 |
root |
1.1 |
|
392 |
|
|
=item other references |
393 |
|
|
|
394 |
|
|
Other unblessed references are generally not allowed and will cause an |
395 |
|
|
exception to be thrown, except for references to the integers C<0> and |
396 |
root |
1.4 |
C<1>, which get turned into false and true in CBOR. |
397 |
|
|
|
398 |
|
|
=item CBOR::XS::Tagged objects |
399 |
|
|
|
400 |
|
|
Objects of this type must be arrays consisting of a single C<[tag, value]> |
401 |
root |
1.13 |
pair. The (numerical) tag will be encoded as a CBOR tag, the value will |
402 |
|
|
be encoded as appropriate for the value. You cna use C<CBOR::XS::tag> to |
403 |
|
|
create such objects. |
404 |
root |
1.1 |
|
405 |
root |
1.6 |
=item Types::Serialiser::true, Types::Serialiser::false, Types::Serialiser::error |
406 |
root |
1.1 |
|
407 |
root |
1.6 |
These special values become CBOR true, CBOR false and CBOR undefined |
408 |
|
|
values, respectively. You can also use C<\1>, C<\0> and C<\undef> directly |
409 |
|
|
if you want. |
410 |
root |
1.1 |
|
411 |
root |
1.7 |
=item other blessed objects |
412 |
root |
1.1 |
|
413 |
root |
1.7 |
Other blessed objects are serialised via C<TO_CBOR> or C<FREEZE>. See |
414 |
root |
1.23 |
L<TAG HANDLING AND EXTENSIONS> for specific classes handled by this |
415 |
|
|
module, and L<OBJECT SERIALISATION> for generic object serialisation. |
416 |
root |
1.1 |
|
417 |
|
|
=item simple scalars |
418 |
|
|
|
419 |
|
|
Simple Perl scalars (any scalar that is not a reference) are the most |
420 |
|
|
difficult objects to encode: CBOR::XS will encode undefined scalars as |
421 |
root |
1.4 |
CBOR null values, scalars that have last been used in a string context |
422 |
root |
1.1 |
before encoding as CBOR strings, and anything else as number value: |
423 |
|
|
|
424 |
|
|
# dump as number |
425 |
|
|
encode_cbor [2] # yields [2] |
426 |
|
|
encode_cbor [-3.0e17] # yields [-3e+17] |
427 |
|
|
my $value = 5; encode_cbor [$value] # yields [5] |
428 |
|
|
|
429 |
|
|
# used as string, so dump as string |
430 |
|
|
print $value; |
431 |
|
|
encode_cbor [$value] # yields ["5"] |
432 |
|
|
|
433 |
|
|
# undef becomes null |
434 |
|
|
encode_cbor [undef] # yields [null] |
435 |
|
|
|
436 |
|
|
You can force the type to be a CBOR string by stringifying it: |
437 |
|
|
|
438 |
|
|
my $x = 3.1; # some variable containing a number |
439 |
|
|
"$x"; # stringified |
440 |
|
|
$x .= ""; # another, more awkward way to stringify |
441 |
|
|
print $x; # perl does it for you, too, quite often |
442 |
|
|
|
443 |
|
|
You can force the type to be a CBOR number by numifying it: |
444 |
|
|
|
445 |
|
|
my $x = "3"; # some variable containing a string |
446 |
|
|
$x += 0; # numify it, ensuring it will be dumped as a number |
447 |
|
|
$x *= 1; # same thing, the choice is yours. |
448 |
|
|
|
449 |
|
|
You can not currently force the type in other, less obscure, ways. Tell me |
450 |
|
|
if you need this capability (but don't forget to explain why it's needed |
451 |
|
|
:). |
452 |
|
|
|
453 |
root |
1.4 |
Perl values that seem to be integers generally use the shortest possible |
454 |
|
|
representation. Floating-point values will use either the IEEE single |
455 |
|
|
format if possible without loss of precision, otherwise the IEEE double |
456 |
|
|
format will be used. Perls that use formats other than IEEE double to |
457 |
|
|
represent numerical values are supported, but might suffer loss of |
458 |
|
|
precision. |
459 |
root |
1.1 |
|
460 |
|
|
=back |
461 |
|
|
|
462 |
root |
1.7 |
=head2 OBJECT SERIALISATION |
463 |
|
|
|
464 |
|
|
This module knows two way to serialise a Perl object: The CBOR-specific |
465 |
|
|
way, and the generic way. |
466 |
|
|
|
467 |
|
|
Whenever the encoder encounters a Perl object that it cnanot serialise |
468 |
|
|
directly (most of them), it will first look up the C<TO_CBOR> method on |
469 |
|
|
it. |
470 |
|
|
|
471 |
|
|
If it has a C<TO_CBOR> method, it will call it with the object as only |
472 |
|
|
argument, and expects exactly one return value, which it will then |
473 |
|
|
substitute and encode it in the place of the object. |
474 |
|
|
|
475 |
|
|
Otherwise, it will look up the C<FREEZE> method. If it exists, it will |
476 |
|
|
call it with the object as first argument, and the constant string C<CBOR> |
477 |
|
|
as the second argument, to distinguish it from other serialisers. |
478 |
|
|
|
479 |
|
|
The C<FREEZE> method can return any number of values (i.e. zero or |
480 |
|
|
more). These will be encoded as CBOR perl object, together with the |
481 |
|
|
classname. |
482 |
|
|
|
483 |
|
|
If an object supports neither C<TO_CBOR> nor C<FREEZE>, encoding will fail |
484 |
|
|
with an error. |
485 |
|
|
|
486 |
|
|
Objects encoded via C<TO_CBOR> cannot be automatically decoded, but |
487 |
|
|
objects encoded via C<FREEZE> can be decoded using the following protocol: |
488 |
|
|
|
489 |
|
|
When an encoded CBOR perl object is encountered by the decoder, it will |
490 |
|
|
look up the C<THAW> method, by using the stored classname, and will fail |
491 |
|
|
if the method cannot be found. |
492 |
|
|
|
493 |
|
|
After the lookup it will call the C<THAW> method with the stored classname |
494 |
|
|
as first argument, the constant string C<CBOR> as second argument, and all |
495 |
|
|
values returned by C<FREEZE> as remaining arguments. |
496 |
|
|
|
497 |
|
|
=head4 EXAMPLES |
498 |
|
|
|
499 |
|
|
Here is an example C<TO_CBOR> method: |
500 |
|
|
|
501 |
|
|
sub My::Object::TO_CBOR { |
502 |
|
|
my ($obj) = @_; |
503 |
|
|
|
504 |
|
|
["this is a serialised My::Object object", $obj->{id}] |
505 |
|
|
} |
506 |
|
|
|
507 |
|
|
When a C<My::Object> is encoded to CBOR, it will instead encode a simple |
508 |
|
|
array with two members: a string, and the "object id". Decoding this CBOR |
509 |
|
|
string will yield a normal perl array reference in place of the object. |
510 |
|
|
|
511 |
|
|
A more useful and practical example would be a serialisation method for |
512 |
|
|
the URI module. CBOR has a custom tag value for URIs, namely 32: |
513 |
|
|
|
514 |
|
|
sub URI::TO_CBOR { |
515 |
|
|
my ($self) = @_; |
516 |
|
|
my $uri = "$self"; # stringify uri |
517 |
|
|
utf8::upgrade $uri; # make sure it will be encoded as UTF-8 string |
518 |
|
|
CBOR::XS::tagged 32, "$_[0]" |
519 |
|
|
} |
520 |
|
|
|
521 |
|
|
This will encode URIs as a UTF-8 string with tag 32, which indicates an |
522 |
|
|
URI. |
523 |
|
|
|
524 |
|
|
Decoding such an URI will not (currently) give you an URI object, but |
525 |
|
|
instead a CBOR::XS::Tagged object with tag number 32 and the string - |
526 |
|
|
exactly what was returned by C<TO_CBOR>. |
527 |
|
|
|
528 |
|
|
To serialise an object so it can automatically be deserialised, you need |
529 |
|
|
to use C<FREEZE> and C<THAW>. To take the URI module as example, this |
530 |
|
|
would be a possible implementation: |
531 |
|
|
|
532 |
|
|
sub URI::FREEZE { |
533 |
|
|
my ($self, $serialiser) = @_; |
534 |
|
|
"$self" # encode url string |
535 |
|
|
} |
536 |
|
|
|
537 |
|
|
sub URI::THAW { |
538 |
|
|
my ($class, $serialiser, $uri) = @_; |
539 |
|
|
|
540 |
|
|
$class->new ($uri) |
541 |
|
|
} |
542 |
|
|
|
543 |
|
|
Unlike C<TO_CBOR>, multiple values can be returned by C<FREEZE>. For |
544 |
|
|
example, a C<FREEZE> method that returns "type", "id" and "variant" values |
545 |
|
|
would cause an invocation of C<THAW> with 5 arguments: |
546 |
|
|
|
547 |
|
|
sub My::Object::FREEZE { |
548 |
|
|
my ($self, $serialiser) = @_; |
549 |
|
|
|
550 |
|
|
($self->{type}, $self->{id}, $self->{variant}) |
551 |
|
|
} |
552 |
|
|
|
553 |
|
|
sub My::Object::THAW { |
554 |
|
|
my ($class, $serialiser, $type, $id, $variant) = @_; |
555 |
|
|
|
556 |
|
|
$class-<new (type => $type, id => $id, variant => $variant) |
557 |
|
|
} |
558 |
|
|
|
559 |
root |
1.1 |
|
560 |
root |
1.7 |
=head1 MAGIC HEADER |
561 |
root |
1.3 |
|
562 |
|
|
There is no way to distinguish CBOR from other formats |
563 |
|
|
programmatically. To make it easier to distinguish CBOR from other |
564 |
|
|
formats, the CBOR specification has a special "magic string" that can be |
565 |
root |
1.18 |
prepended to any CBOR string without changing its meaning. |
566 |
root |
1.3 |
|
567 |
|
|
This string is available as C<$CBOR::XS::MAGIC>. This module does not |
568 |
root |
1.18 |
prepend this string to the CBOR data it generates, but it will ignore it |
569 |
root |
1.3 |
if present, so users can prepend this string as a "file type" indicator as |
570 |
|
|
required. |
571 |
|
|
|
572 |
|
|
|
573 |
root |
1.12 |
=head1 THE CBOR::XS::Tagged CLASS |
574 |
|
|
|
575 |
|
|
CBOR has the concept of tagged values - any CBOR value can be tagged with |
576 |
|
|
a numeric 64 bit number, which are centrally administered. |
577 |
|
|
|
578 |
|
|
C<CBOR::XS> handles a few tags internally when en- or decoding. You can |
579 |
|
|
also create tags yourself by encoding C<CBOR::XS::Tagged> objects, and the |
580 |
|
|
decoder will create C<CBOR::XS::Tagged> objects itself when it hits an |
581 |
|
|
unknown tag. |
582 |
|
|
|
583 |
|
|
These objects are simply blessed array references - the first member of |
584 |
|
|
the array being the numerical tag, the second being the value. |
585 |
|
|
|
586 |
|
|
You can interact with C<CBOR::XS::Tagged> objects in the following ways: |
587 |
|
|
|
588 |
|
|
=over 4 |
589 |
|
|
|
590 |
|
|
=item $tagged = CBOR::XS::tag $tag, $value |
591 |
|
|
|
592 |
|
|
This function(!) creates a new C<CBOR::XS::Tagged> object using the given |
593 |
|
|
C<$tag> (0..2**64-1) to tag the given C<$value> (which can be any Perl |
594 |
|
|
value that can be encoded in CBOR, including serialisable Perl objects and |
595 |
|
|
C<CBOR::XS::Tagged> objects). |
596 |
|
|
|
597 |
|
|
=item $tagged->[0] |
598 |
|
|
|
599 |
|
|
=item $tagged->[0] = $new_tag |
600 |
|
|
|
601 |
|
|
=item $tag = $tagged->tag |
602 |
|
|
|
603 |
|
|
=item $new_tag = $tagged->tag ($new_tag) |
604 |
|
|
|
605 |
|
|
Access/mutate the tag. |
606 |
|
|
|
607 |
|
|
=item $tagged->[1] |
608 |
|
|
|
609 |
|
|
=item $tagged->[1] = $new_value |
610 |
|
|
|
611 |
|
|
=item $value = $tagged->value |
612 |
|
|
|
613 |
|
|
=item $new_value = $tagged->value ($new_value) |
614 |
|
|
|
615 |
|
|
Access/mutate the tagged value. |
616 |
|
|
|
617 |
|
|
=back |
618 |
|
|
|
619 |
|
|
=cut |
620 |
|
|
|
621 |
|
|
sub tag($$) { |
622 |
|
|
bless [@_], CBOR::XS::Tagged::; |
623 |
|
|
} |
624 |
|
|
|
625 |
|
|
sub CBOR::XS::Tagged::tag { |
626 |
|
|
$_[0][0] = $_[1] if $#_; |
627 |
|
|
$_[0][0] |
628 |
|
|
} |
629 |
|
|
|
630 |
|
|
sub CBOR::XS::Tagged::value { |
631 |
|
|
$_[0][1] = $_[1] if $#_; |
632 |
|
|
$_[0][1] |
633 |
|
|
} |
634 |
|
|
|
635 |
root |
1.13 |
=head2 EXAMPLES |
636 |
|
|
|
637 |
|
|
Here are some examples of C<CBOR::XS::Tagged> uses to tag objects. |
638 |
|
|
|
639 |
|
|
You can look up CBOR tag value and emanings in the IANA registry at |
640 |
|
|
L<http://www.iana.org/assignments/cbor-tags/cbor-tags.xhtml>. |
641 |
|
|
|
642 |
|
|
Prepend a magic header (C<$CBOR::XS::MAGIC>): |
643 |
|
|
|
644 |
|
|
my $cbor = encode_cbor CBOR::XS::tag 55799, $value; |
645 |
|
|
# same as: |
646 |
|
|
my $cbor = $CBOR::XS::MAGIC . encode_cbor $value; |
647 |
|
|
|
648 |
|
|
Serialise some URIs and a regex in an array: |
649 |
|
|
|
650 |
|
|
my $cbor = encode_cbor [ |
651 |
|
|
(CBOR::XS::tag 32, "http://www.nethype.de/"), |
652 |
|
|
(CBOR::XS::tag 32, "http://software.schmorp.de/"), |
653 |
|
|
(CBOR::XS::tag 35, "^[Pp][Ee][Rr][lL]\$"), |
654 |
|
|
]; |
655 |
|
|
|
656 |
|
|
Wrap CBOR data in CBOR: |
657 |
|
|
|
658 |
|
|
my $cbor_cbor = encode_cbor |
659 |
|
|
CBOR::XS::tag 24, |
660 |
|
|
encode_cbor [1, 2, 3]; |
661 |
|
|
|
662 |
root |
1.19 |
=head1 TAG HANDLING AND EXTENSIONS |
663 |
|
|
|
664 |
root |
1.22 |
This section describes how this module handles specific tagged values |
665 |
|
|
and extensions. If a tag is not mentioned here and no additional filters |
666 |
|
|
are provided for it, then the default handling applies (creating a |
667 |
|
|
CBOR::XS::Tagged object on decoding, and only encoding the tag when |
668 |
|
|
explicitly requested). |
669 |
root |
1.19 |
|
670 |
root |
1.23 |
Tags not handled specifically are currently converted into a |
671 |
|
|
L<CBOR::XS::Tagged> object, which is simply a blessed array reference |
672 |
|
|
consisting of the numeric tag value followed by the (decoded) CBOR value. |
673 |
|
|
|
674 |
root |
1.19 |
Future versions of this module reserve the right to special case |
675 |
root |
1.22 |
additional tags (such as base64url). |
676 |
|
|
|
677 |
|
|
=head2 ENFORCED TAGS |
678 |
|
|
|
679 |
|
|
These tags are always handled when decoding, and their handling cannot be |
680 |
|
|
overriden by the user. |
681 |
root |
1.19 |
|
682 |
|
|
=over 4 |
683 |
|
|
|
684 |
root |
1.26 |
=item 26 (perl-object, L<http://cbor.schmorp.de/perl-object>) |
685 |
root |
1.19 |
|
686 |
root |
1.23 |
These tags are automatically created (and decoded) for serialisable |
687 |
|
|
objects using the C<FREEZE/THAW> methods (the L<Types::Serialier> object |
688 |
|
|
serialisation protocol). See L<OBJECT SERIALISATION> for details. |
689 |
root |
1.19 |
|
690 |
root |
1.26 |
=item 28, 29 (sharable, sharedref, L <http://cbor.schmorp.de/value-sharing>) |
691 |
root |
1.19 |
|
692 |
|
|
These tags are automatically decoded when encountered, resulting in |
693 |
|
|
shared values in the decoded object. They are only encoded, however, when |
694 |
|
|
C<allow_sharable> is enabled. |
695 |
|
|
|
696 |
root |
1.26 |
=item 256, 25 (stringref-namespace, stringref, L <http://cbor.schmorp.de/stringref>) |
697 |
root |
1.21 |
|
698 |
|
|
These tags are automatically decoded when encountered. They are only |
699 |
root |
1.25 |
encoded, however, when C<pack_strings> is enabled. |
700 |
root |
1.21 |
|
701 |
root |
1.19 |
=item 22098 (indirection, L<http://cbor.schmorp.de/indirection>) |
702 |
|
|
|
703 |
|
|
This tag is automatically generated when a reference are encountered (with |
704 |
|
|
the exception of hash and array refernces). It is converted to a reference |
705 |
|
|
when decoding. |
706 |
|
|
|
707 |
|
|
=item 55799 (self-describe CBOR, RFC 7049) |
708 |
|
|
|
709 |
|
|
This value is not generated on encoding (unless explicitly requested by |
710 |
|
|
the user), and is simply ignored when decoding. |
711 |
|
|
|
712 |
|
|
=back |
713 |
|
|
|
714 |
root |
1.24 |
=head2 NON-ENFORCED TAGS |
715 |
root |
1.22 |
|
716 |
|
|
These tags have default filters provided when decoding. Their handling can |
717 |
|
|
be overriden by changing the C<%CBOR::XS::FILTER> entry for the tag, or by |
718 |
root |
1.24 |
providing a custom C<filter> callback when decoding. |
719 |
root |
1.22 |
|
720 |
|
|
When they result in decoding into a specific Perl class, the module |
721 |
|
|
usually provides a corresponding C<TO_CBOR> method as well. |
722 |
|
|
|
723 |
|
|
When any of these need to load additional modules that are not part of the |
724 |
|
|
perl core distribution (e.g. L<URI>), it is (currently) up to the user to |
725 |
|
|
provide these modules. The decoding usually fails with an exception if the |
726 |
|
|
required module cannot be loaded. |
727 |
|
|
|
728 |
|
|
=over 4 |
729 |
|
|
|
730 |
|
|
=item 2, 3 (positive/negative bignum) |
731 |
|
|
|
732 |
|
|
These tags are decoded into L<Math::BigInt> objects. The corresponding |
733 |
|
|
C<Math::BigInt::TO_CBOR> method encodes "small" bigints into normal CBOR |
734 |
|
|
integers, and others into positive/negative CBOR bignums. |
735 |
|
|
|
736 |
|
|
=item 4, 5 (decimal fraction/bigfloat) |
737 |
|
|
|
738 |
|
|
Both decimal fractions and bigfloats are decoded into L<Math::BigFloat> |
739 |
|
|
objects. The corresponding C<Math::BigFloat::TO_CBOR> method I<always> |
740 |
|
|
encodes into a decimal fraction. |
741 |
|
|
|
742 |
|
|
CBOR cannot represent bigfloats with I<very> large exponents - conversion |
743 |
|
|
of such big float objects is undefined. |
744 |
|
|
|
745 |
|
|
Also, NaN and infinities are not encoded properly. |
746 |
|
|
|
747 |
|
|
=item 21, 22, 23 (expected later JSON conversion) |
748 |
|
|
|
749 |
|
|
CBOR::XS is not a CBOR-to-JSON converter, and will simply ignore these |
750 |
|
|
tags. |
751 |
|
|
|
752 |
|
|
=item 32 (URI) |
753 |
|
|
|
754 |
|
|
These objects decode into L<URI> objects. The corresponding |
755 |
|
|
C<URI::TO_CBOR> method again results in a CBOR URI value. |
756 |
|
|
|
757 |
|
|
=back |
758 |
|
|
|
759 |
|
|
=cut |
760 |
|
|
|
761 |
|
|
our %FILTER = ( |
762 |
|
|
# 0 # rfc4287 datetime, utf-8 |
763 |
|
|
# 1 # unix timestamp, any |
764 |
|
|
|
765 |
|
|
2 => sub { # pos bigint |
766 |
|
|
require Math::BigInt; |
767 |
|
|
Math::BigInt->new ("0x" . unpack "H*", pop) |
768 |
|
|
}, |
769 |
|
|
|
770 |
|
|
3 => sub { # neg bigint |
771 |
|
|
require Math::BigInt; |
772 |
|
|
-Math::BigInt->new ("0x" . unpack "H*", pop) |
773 |
|
|
}, |
774 |
|
|
|
775 |
|
|
4 => sub { # decimal fraction, array |
776 |
|
|
require Math::BigFloat; |
777 |
|
|
Math::BigFloat->new ($_[1][1] . "E" . $_[1][0]) |
778 |
|
|
}, |
779 |
|
|
|
780 |
|
|
5 => sub { # bigfloat, array |
781 |
|
|
require Math::BigFloat; |
782 |
|
|
scalar Math::BigFloat->new ($_[1][1])->blsft ($_[1][0], 2) |
783 |
|
|
}, |
784 |
|
|
|
785 |
|
|
21 => sub { pop }, # expected conversion to base64url encoding |
786 |
|
|
22 => sub { pop }, # expected conversion to base64 encoding |
787 |
|
|
23 => sub { pop }, # expected conversion to base16 encoding |
788 |
|
|
|
789 |
|
|
# 24 # embedded cbor, byte string |
790 |
|
|
|
791 |
|
|
32 => sub { |
792 |
|
|
require URI; |
793 |
|
|
URI->new (pop) |
794 |
|
|
}, |
795 |
|
|
|
796 |
|
|
# 33 # base64url rfc4648, utf-8 |
797 |
|
|
# 34 # base64 rfc46484, utf-8 |
798 |
|
|
# 35 # regex pcre/ecma262, utf-8 |
799 |
|
|
# 36 # mime message rfc2045, utf-8 |
800 |
|
|
); |
801 |
|
|
|
802 |
root |
1.19 |
|
803 |
root |
1.7 |
=head1 CBOR and JSON |
804 |
root |
1.1 |
|
805 |
root |
1.4 |
CBOR is supposed to implement a superset of the JSON data model, and is, |
806 |
|
|
with some coercion, able to represent all JSON texts (something that other |
807 |
|
|
"binary JSON" formats such as BSON generally do not support). |
808 |
|
|
|
809 |
|
|
CBOR implements some extra hints and support for JSON interoperability, |
810 |
|
|
and the spec offers further guidance for conversion between CBOR and |
811 |
|
|
JSON. None of this is currently implemented in CBOR, and the guidelines |
812 |
|
|
in the spec do not result in correct round-tripping of data. If JSON |
813 |
|
|
interoperability is improved in the future, then the goal will be to |
814 |
|
|
ensure that decoded JSON data will round-trip encoding and decoding to |
815 |
|
|
CBOR intact. |
816 |
root |
1.1 |
|
817 |
|
|
|
818 |
|
|
=head1 SECURITY CONSIDERATIONS |
819 |
|
|
|
820 |
|
|
When you are using CBOR in a protocol, talking to untrusted potentially |
821 |
|
|
hostile creatures requires relatively few measures. |
822 |
|
|
|
823 |
|
|
First of all, your CBOR decoder should be secure, that is, should not have |
824 |
|
|
any buffer overflows. Obviously, this module should ensure that and I am |
825 |
|
|
trying hard on making that true, but you never know. |
826 |
|
|
|
827 |
|
|
Second, you need to avoid resource-starving attacks. That means you should |
828 |
|
|
limit the size of CBOR data you accept, or make sure then when your |
829 |
|
|
resources run out, that's just fine (e.g. by using a separate process that |
830 |
|
|
can crash safely). The size of a CBOR string in octets is usually a good |
831 |
|
|
indication of the size of the resources required to decode it into a Perl |
832 |
|
|
structure. While CBOR::XS can check the size of the CBOR text, it might be |
833 |
|
|
too late when you already have it in memory, so you might want to check |
834 |
|
|
the size before you accept the string. |
835 |
|
|
|
836 |
|
|
Third, CBOR::XS recurses using the C stack when decoding objects and |
837 |
|
|
arrays. The C stack is a limited resource: for instance, on my amd64 |
838 |
|
|
machine with 8MB of stack size I can decode around 180k nested arrays but |
839 |
|
|
only 14k nested CBOR objects (due to perl itself recursing deeply on croak |
840 |
|
|
to free the temporary). If that is exceeded, the program crashes. To be |
841 |
|
|
conservative, the default nesting limit is set to 512. If your process |
842 |
|
|
has a smaller stack, you should adjust this setting accordingly with the |
843 |
|
|
C<max_depth> method. |
844 |
|
|
|
845 |
|
|
Something else could bomb you, too, that I forgot to think of. In that |
846 |
|
|
case, you get to keep the pieces. I am always open for hints, though... |
847 |
|
|
|
848 |
|
|
Also keep in mind that CBOR::XS might leak contents of your Perl data |
849 |
|
|
structures in its error messages, so when you serialise sensitive |
850 |
|
|
information you might want to make sure that exceptions thrown by CBOR::XS |
851 |
|
|
will not end up in front of untrusted eyes. |
852 |
|
|
|
853 |
|
|
=head1 CBOR IMPLEMENTATION NOTES |
854 |
|
|
|
855 |
|
|
This section contains some random implementation notes. They do not |
856 |
|
|
describe guaranteed behaviour, but merely behaviour as-is implemented |
857 |
|
|
right now. |
858 |
|
|
|
859 |
|
|
64 bit integers are only properly decoded when Perl was built with 64 bit |
860 |
|
|
support. |
861 |
|
|
|
862 |
|
|
Strings and arrays are encoded with a definite length. Hashes as well, |
863 |
|
|
unless they are tied (or otherwise magical). |
864 |
|
|
|
865 |
|
|
Only the double data type is supported for NV data types - when Perl uses |
866 |
|
|
long double to represent floating point values, they might not be encoded |
867 |
|
|
properly. Half precision types are accepted, but not encoded. |
868 |
|
|
|
869 |
|
|
Strict mode and canonical mode are not implemented. |
870 |
|
|
|
871 |
|
|
|
872 |
|
|
=head1 THREADS |
873 |
|
|
|
874 |
|
|
This module is I<not> guaranteed to be thread safe and there are no |
875 |
|
|
plans to change this until Perl gets thread support (as opposed to the |
876 |
|
|
horribly slow so-called "threads" which are simply slow and bloated |
877 |
|
|
process simulations - use fork, it's I<much> faster, cheaper, better). |
878 |
|
|
|
879 |
|
|
(It might actually work, but you have been warned). |
880 |
|
|
|
881 |
|
|
|
882 |
|
|
=head1 BUGS |
883 |
|
|
|
884 |
|
|
While the goal of this module is to be correct, that unfortunately does |
885 |
|
|
not mean it's bug-free, only that I think its design is bug-free. If you |
886 |
|
|
keep reporting bugs they will be fixed swiftly, though. |
887 |
|
|
|
888 |
|
|
Please refrain from using rt.cpan.org or any other bug reporting |
889 |
|
|
service. I put the contact address into my modules for a reason. |
890 |
|
|
|
891 |
|
|
=cut |
892 |
|
|
|
893 |
root |
1.22 |
our %FILTER = ( |
894 |
|
|
# 0 # rfc4287 datetime, utf-8 |
895 |
|
|
# 1 # unix timestamp, any |
896 |
|
|
|
897 |
|
|
2 => sub { # pos bigint |
898 |
|
|
require Math::BigInt; |
899 |
|
|
Math::BigInt->new ("0x" . unpack "H*", pop) |
900 |
|
|
}, |
901 |
|
|
|
902 |
|
|
3 => sub { # neg bigint |
903 |
|
|
require Math::BigInt; |
904 |
|
|
-Math::BigInt->new ("0x" . unpack "H*", pop) |
905 |
|
|
}, |
906 |
|
|
|
907 |
|
|
4 => sub { # decimal fraction, array |
908 |
|
|
require Math::BigFloat; |
909 |
|
|
Math::BigFloat->new ($_[1][1] . "E" . $_[1][0]) |
910 |
|
|
}, |
911 |
|
|
|
912 |
|
|
5 => sub { # bigfloat, array |
913 |
|
|
require Math::BigFloat; |
914 |
|
|
scalar Math::BigFloat->new ($_[1][1])->blsft ($_[1][0], 2) |
915 |
|
|
}, |
916 |
|
|
|
917 |
|
|
21 => sub { pop }, # expected conversion to base64url encoding |
918 |
|
|
22 => sub { pop }, # expected conversion to base64 encoding |
919 |
|
|
23 => sub { pop }, # expected conversion to base16 encoding |
920 |
|
|
|
921 |
|
|
# 24 # embedded cbor, byte string |
922 |
|
|
|
923 |
|
|
32 => sub { |
924 |
|
|
require URI; |
925 |
|
|
URI->new (pop) |
926 |
|
|
}, |
927 |
|
|
|
928 |
|
|
# 33 # base64url rfc4648, utf-8 |
929 |
|
|
# 34 # base64 rfc46484, utf-8 |
930 |
|
|
# 35 # regex pcre/ecma262, utf-8 |
931 |
|
|
# 36 # mime message rfc2045, utf-8 |
932 |
|
|
); |
933 |
|
|
|
934 |
|
|
sub CBOR::XS::default_filter { |
935 |
|
|
&{ $FILTER{$_[0]} or return } |
936 |
|
|
} |
937 |
|
|
|
938 |
|
|
sub URI::TO_CBOR { |
939 |
|
|
my $uri = $_[0]->as_string; |
940 |
|
|
utf8::upgrade $uri; |
941 |
|
|
CBOR::XS::tag 32, $uri |
942 |
|
|
} |
943 |
|
|
|
944 |
|
|
sub Math::BigInt::TO_CBOR { |
945 |
|
|
if ($_[0] >= -2147483648 && $_[0] <= 2147483647) { |
946 |
|
|
$_[0]->numify |
947 |
|
|
} else { |
948 |
|
|
my $hex = substr $_[0]->as_hex, 2; |
949 |
|
|
$hex = "0$hex" if 1 & length $hex; # sigh |
950 |
|
|
CBOR::XS::tag $_[0] >= 0 ? 2 : 3, pack "H*", $hex |
951 |
|
|
} |
952 |
|
|
} |
953 |
|
|
|
954 |
|
|
sub Math::BigFloat::TO_CBOR { |
955 |
|
|
my ($m, $e) = $_[0]->parts; |
956 |
|
|
CBOR::XS::tag 4, [$e->numify, $m] |
957 |
|
|
} |
958 |
|
|
|
959 |
root |
1.1 |
XSLoader::load "CBOR::XS", $VERSION; |
960 |
|
|
|
961 |
|
|
=head1 SEE ALSO |
962 |
|
|
|
963 |
|
|
The L<JSON> and L<JSON::XS> modules that do similar, but human-readable, |
964 |
|
|
serialisation. |
965 |
|
|
|
966 |
root |
1.6 |
The L<Types::Serialiser> module provides the data model for true, false |
967 |
|
|
and error values. |
968 |
|
|
|
969 |
root |
1.1 |
=head1 AUTHOR |
970 |
|
|
|
971 |
|
|
Marc Lehmann <schmorp@schmorp.de> |
972 |
|
|
http://home.schmorp.de/ |
973 |
|
|
|
974 |
|
|
=cut |
975 |
|
|
|
976 |
root |
1.6 |
1 |
977 |
|
|
|