… | |
… | |
26 | substr $many_cbor_strings, 0, $length, ""; # remove decoded cbor string |
26 | substr $many_cbor_strings, 0, $length, ""; # remove decoded cbor string |
27 | } |
27 | } |
28 | |
28 | |
29 | =head1 DESCRIPTION |
29 | =head1 DESCRIPTION |
30 | |
30 | |
31 | WARNING! This module is very new, and not very well tested (that's up |
|
|
32 | to you to do). Furthermore, details of the implementation might change |
|
|
33 | freely before version 1.0. And lastly, most extensions depend on an IANA |
|
|
34 | assignment, and until that assignment is official, this implementation is |
|
|
35 | not interoperable with other implementations (even future versions of this |
|
|
36 | module) until the assignment is done. |
|
|
37 | |
|
|
38 | You are still invited to try out CBOR, and this module. |
|
|
39 | |
|
|
40 | This module converts Perl data structures to the Concise Binary Object |
31 | This module converts Perl data structures to the Concise Binary Object |
41 | Representation (CBOR) and vice versa. CBOR is a fast binary serialisation |
32 | Representation (CBOR) and vice versa. CBOR is a fast binary serialisation |
42 | format that aims to use a superset of the JSON data model, i.e. when you |
33 | format that aims to use an (almost) superset of the JSON data model, i.e. |
43 | can represent something in JSON, you should be able to represent it in |
34 | when you can represent something useful in JSON, you should be able to |
44 | CBOR. |
35 | represent it in CBOR. |
45 | |
36 | |
46 | In short, CBOR is a faster and very compact binary alternative to JSON, |
37 | In short, CBOR is a faster and quite compact binary alternative to JSON, |
47 | with the added ability of supporting serialisation of Perl objects. (JSON |
38 | with the added ability of supporting serialisation of Perl objects. (JSON |
48 | often compresses better than CBOR though, so if you plan to compress the |
39 | often compresses better than CBOR though, so if you plan to compress the |
49 | data later you might want to compare both formats first). |
40 | data later and speed is less important you might want to compare both |
|
|
41 | formats first). |
|
|
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. |
50 | |
45 | |
51 | 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, |
52 | 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 |
53 | 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 |
54 | data, the worse L<Storable> performs in comparison. |
49 | data, the worse L<Storable> performs in comparison. |
55 | |
50 | |
56 | As for compactness, C<CBOR::XS> encoded data structures are usually about |
51 | Regarding compactness, C<CBOR::XS>-encoded data structures are usually |
57 | 20% smaller than the same data encoded as (compact) JSON or L<Storable>. |
52 | about 20% smaller than the same data encoded as (compact) JSON or |
|
|
53 | L<Storable>. |
58 | |
54 | |
59 | In addition to the core CBOR data format, this module implements a number |
55 | In addition to the core CBOR data format, this module implements a |
60 | of extensions, to support cyclic and self-referencing data structures |
56 | number of extensions, to support cyclic and shared data structures |
61 | (see C<allow_sharing>), string deduplication (see C<allow_stringref>) and |
57 | (see C<allow_sharing> and C<allow_cycles>), string deduplication (see |
62 | scalar references (always enabled). |
58 | C<pack_strings>) and scalar references (always enabled). |
63 | |
|
|
64 | 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 | |
59 | |
67 | See MAPPING, below, on how CBOR::XS maps perl values to CBOR values and |
60 | See MAPPING, below, on how CBOR::XS maps perl values to CBOR values and |
68 | vice versa. |
61 | vice versa. |
69 | |
62 | |
70 | =cut |
63 | =cut |
71 | |
64 | |
72 | package CBOR::XS; |
65 | package CBOR::XS; |
73 | |
66 | |
74 | use common::sense; |
67 | use common::sense; |
75 | |
68 | |
76 | our $VERSION = 0.09; |
69 | our $VERSION = 1.71; |
77 | our @ISA = qw(Exporter); |
70 | our @ISA = qw(Exporter); |
78 | |
71 | |
79 | our @EXPORT = qw(encode_cbor decode_cbor); |
72 | our @EXPORT = qw(encode_cbor decode_cbor); |
80 | |
73 | |
81 | use Exporter; |
74 | use Exporter; |
… | |
… | |
119 | |
112 | |
120 | The mutators for flags all return the CBOR object again and thus calls can |
113 | The mutators for flags all return the CBOR object again and thus calls can |
121 | be chained: |
114 | be chained: |
122 | |
115 | |
123 | my $cbor = CBOR::XS->new->encode ({a => [1,2]}); |
116 | my $cbor = CBOR::XS->new->encode ({a => [1,2]}); |
|
|
117 | |
|
|
118 | =item $cbor = new_safe CBOR::XS |
|
|
119 | |
|
|
120 | Create a new, safe/secure CBOR::XS object. This is similar to C<new>, |
|
|
121 | but configures the coder object to be safe to use with untrusted |
|
|
122 | data. Currently, this is equivalent to: |
|
|
123 | |
|
|
124 | my $cbor = CBOR::XS |
|
|
125 | ->new |
|
|
126 | ->forbid_objects |
|
|
127 | ->filter (\&CBOR::XS::safe_filter) |
|
|
128 | ->max_size (1e8); |
|
|
129 | |
|
|
130 | But is more future proof (it is better to crash because of a change than |
|
|
131 | to be exploited in other ways). |
|
|
132 | |
|
|
133 | =cut |
|
|
134 | |
|
|
135 | sub new_safe { |
|
|
136 | CBOR::XS |
|
|
137 | ->new |
|
|
138 | ->forbid_objects |
|
|
139 | ->filter (\&CBOR::XS::safe_filter) |
|
|
140 | ->max_size (1e8) |
|
|
141 | } |
124 | |
142 | |
125 | =item $cbor = $cbor->max_depth ([$maximum_nesting_depth]) |
143 | =item $cbor = $cbor->max_depth ([$maximum_nesting_depth]) |
126 | |
144 | |
127 | =item $max_depth = $cbor->get_max_depth |
145 | =item $max_depth = $cbor->get_max_depth |
128 | |
146 | |
… | |
… | |
144 | |
162 | |
145 | Note that nesting is implemented by recursion in C. The default value has |
163 | Note that nesting is implemented by recursion in C. The default value has |
146 | been chosen to be as large as typical operating systems allow without |
164 | been chosen to be as large as typical operating systems allow without |
147 | crashing. |
165 | crashing. |
148 | |
166 | |
149 | See SECURITY CONSIDERATIONS, below, for more info on why this is useful. |
167 | See L<SECURITY CONSIDERATIONS>, below, for more info on why this is useful. |
150 | |
168 | |
151 | =item $cbor = $cbor->max_size ([$maximum_string_size]) |
169 | =item $cbor = $cbor->max_size ([$maximum_string_size]) |
152 | |
170 | |
153 | =item $max_size = $cbor->get_max_size |
171 | =item $max_size = $cbor->get_max_size |
154 | |
172 | |
… | |
… | |
159 | effect on C<encode> (yet). |
177 | effect on C<encode> (yet). |
160 | |
178 | |
161 | If no argument is given, the limit check will be deactivated (same as when |
179 | If no argument is given, the limit check will be deactivated (same as when |
162 | C<0> is specified). |
180 | C<0> is specified). |
163 | |
181 | |
164 | See SECURITY CONSIDERATIONS, below, for more info on why this is useful. |
182 | See L<SECURITY CONSIDERATIONS>, below, for more info on why this is useful. |
165 | |
183 | |
166 | =item $cbor = $cbor->allow_unknown ([$enable]) |
184 | =item $cbor = $cbor->allow_unknown ([$enable]) |
167 | |
185 | |
168 | =item $enabled = $cbor->get_allow_unknown |
186 | =item $enabled = $cbor->get_allow_unknown |
169 | |
187 | |
… | |
… | |
186 | as an array, is referenced multiple times), but instead will emit a |
204 | as an array, is referenced multiple times), but instead will emit a |
187 | reference to the earlier value. |
205 | reference to the earlier value. |
188 | |
206 | |
189 | This means that such values will only be encoded once, and will not result |
207 | This means that such values will only be encoded once, and will not result |
190 | in a deep cloning of the value on decode, in decoders supporting the value |
208 | in a deep cloning of the value on decode, in decoders supporting the value |
191 | sharing extension. |
209 | sharing extension. This also makes it possible to encode cyclic data |
|
|
210 | structures (which need C<allow_cycles> to be enabled to be decoded by this |
|
|
211 | module). |
192 | |
212 | |
193 | It is recommended to leave it off unless you know your |
213 | It is recommended to leave it off unless you know your |
194 | communication partner supports the value sharing extensions to CBOR |
214 | communication partner supports the value sharing extensions to CBOR |
195 | (http://cbor.schmorp.de/value-sharing). |
215 | (L<http://cbor.schmorp.de/value-sharing>), as without decoder support, the |
|
|
216 | resulting data structure might be unusable. |
196 | |
217 | |
197 | Detecting shared values incurs a runtime overhead when values are encoded |
218 | Detecting shared values incurs a runtime overhead when values are encoded |
198 | that have a reference counter large than one, and might unnecessarily |
219 | that have a reference counter large than one, and might unnecessarily |
199 | increase the encoded size, as potentially shared values are encode as |
220 | increase the encoded size, as potentially shared values are encoded as |
200 | sharable whether or not they are actually shared. |
221 | shareable whether or not they are actually shared. |
201 | |
222 | |
202 | 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, |
203 | 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 |
204 | 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 |
205 | not impossible to create in Perl, are not supported (this is the same as |
226 | not impossible to create in Perl, are not supported (this is the same as |
206 | for L<Storable>). |
227 | with L<Storable>). |
207 | |
228 | |
208 | If C<$enable> is false (the default), then C<encode> will encode |
229 | If C<$enable> is false (the default), then C<encode> will encode shared |
209 | exception when it encounters anything it cannot encode as CBOR. |
230 | data structures repeatedly, unsharing them in the process. Cyclic data |
|
|
231 | structures cannot be encoded in this mode. |
210 | |
232 | |
211 | This option does not affect C<decode> in any way - shared values and |
233 | This option does not affect C<decode> in any way - shared values and |
212 | references will always be decoded properly if present. |
234 | references will always be decoded properly if present. |
213 | |
235 | |
|
|
236 | =item $cbor = $cbor->allow_cycles ([$enable]) |
|
|
237 | |
|
|
238 | =item $enabled = $cbor->get_allow_cycles |
|
|
239 | |
|
|
240 | If C<$enable> is true (or missing), then C<decode> will happily decode |
|
|
241 | self-referential (cyclic) data structures. By default these will not be |
|
|
242 | decoded, as they need manual cleanup to avoid memory leaks, so code that |
|
|
243 | isn't prepared for this will not leak memory. |
|
|
244 | |
|
|
245 | If C<$enable> is false (the default), then C<decode> will throw an error |
|
|
246 | when it encounters a self-referential/cyclic data structure. |
|
|
247 | |
|
|
248 | FUTURE DIRECTION: the motivation behind this option is to avoid I<real> |
|
|
249 | cycles - future versions of this module might chose to decode cyclic data |
|
|
250 | structures using weak references when this option is off, instead of |
|
|
251 | throwing an error. |
|
|
252 | |
|
|
253 | This option does not affect C<encode> in any way - shared values and |
|
|
254 | references will always be encoded properly if present. |
|
|
255 | |
|
|
256 | =item $cbor = $cbor->forbid_objects ([$enable]) |
|
|
257 | |
|
|
258 | =item $enabled = $cbor->get_forbid_objects |
|
|
259 | |
|
|
260 | Disables the use of the object serialiser protocol. |
|
|
261 | |
|
|
262 | If C<$enable> is true (or missing), then C<encode> will will throw an |
|
|
263 | exception when it encounters perl objects that would be encoded using the |
|
|
264 | perl-object tag (26). When C<decode> encounters such tags, it will fall |
|
|
265 | back to the general filter/tagged logic as if this were an unknown tag (by |
|
|
266 | default resulting in a C<CBOR::XC::Tagged> object). |
|
|
267 | |
|
|
268 | If C<$enable> is false (the default), then C<encode> will use the |
|
|
269 | L<Types::Serialiser> object serialisation protocol to serialise objects |
|
|
270 | into perl-object tags, and C<decode> will do the same to decode such tags. |
|
|
271 | |
|
|
272 | See L<SECURITY CONSIDERATIONS>, below, for more info on why forbidding this |
|
|
273 | protocol can be useful. |
|
|
274 | |
214 | =item $cbor = $cbor->allow_stringref ([$enable]) |
275 | =item $cbor = $cbor->pack_strings ([$enable]) |
215 | |
276 | |
216 | =item $enabled = $cbor->get_allow_stringref |
277 | =item $enabled = $cbor->get_pack_strings |
217 | |
278 | |
218 | If C<$enable> is true (or missing), then C<encode> will try not to encode |
279 | If C<$enable> is true (or missing), then C<encode> will try not to encode |
219 | the same string twice, but will instead encode a reference to the string |
280 | the same string twice, but will instead encode a reference to the string |
220 | instead. Depending on your data format. this can save a lot of space, but |
281 | instead. Depending on your data format, this can save a lot of space, but |
221 | also results in a very large runtime overhead (expect encoding times to be |
282 | also results in a very large runtime overhead (expect encoding times to be |
222 | 2-4 times as high as without). |
283 | 2-4 times as high as without). |
223 | |
284 | |
224 | It is recommended to leave it off unless you know your |
285 | It is recommended to leave it off unless you know your |
225 | communications partner supports the stringref extension to CBOR |
286 | communications partner supports the stringref extension to CBOR |
226 | (http://cbor.schmorp.de/stringref). |
287 | (L<http://cbor.schmorp.de/stringref>), as without decoder support, the |
|
|
288 | resulting data structure might not be usable. |
227 | |
289 | |
228 | If C<$enable> is false (the default), then C<encode> will encode |
290 | If C<$enable> is false (the default), then C<encode> will encode strings |
229 | exception when it encounters anything it cannot encode as CBOR. |
291 | the standard CBOR way. |
230 | |
292 | |
231 | This option does not affect C<decode> in any way - string references will |
293 | This option does not affect C<decode> in any way - string references will |
232 | always be decoded properly if present. |
294 | always be decoded properly if present. |
|
|
295 | |
|
|
296 | =item $cbor = $cbor->text_keys ([$enable]) |
|
|
297 | |
|
|
298 | =item $enabled = $cbor->get_text_keys |
|
|
299 | |
|
|
300 | If C<$enabled> is true (or missing), then C<encode> will encode all |
|
|
301 | perl hash keys as CBOR text strings/UTF-8 string, upgrading them as needed. |
|
|
302 | |
|
|
303 | If C<$enable> is false (the default), then C<encode> will encode hash keys |
|
|
304 | normally - upgraded perl strings (strings internally encoded as UTF-8) as |
|
|
305 | CBOR text strings, and downgraded perl strings as CBOR byte strings. |
|
|
306 | |
|
|
307 | This option does not affect C<decode> in any way. |
|
|
308 | |
|
|
309 | This option is useful for interoperability with CBOR decoders that don't |
|
|
310 | treat byte strings as a form of text. It is especially useful as Perl |
|
|
311 | gives very little control over hash keys. |
|
|
312 | |
|
|
313 | Enabling this option can be slow, as all downgraded hash keys that are |
|
|
314 | encoded need to be scanned and converted to UTF-8. |
|
|
315 | |
|
|
316 | =item $cbor = $cbor->text_strings ([$enable]) |
|
|
317 | |
|
|
318 | =item $enabled = $cbor->get_text_strings |
|
|
319 | |
|
|
320 | This option works similar to C<text_keys>, above, but works on all strings |
|
|
321 | (including hash keys), so C<text_keys> has no further effect after |
|
|
322 | enabling C<text_strings>. |
|
|
323 | |
|
|
324 | If C<$enabled> is true (or missing), then C<encode> will encode all perl |
|
|
325 | strings as CBOR text strings/UTF-8 strings, upgrading them as needed. |
|
|
326 | |
|
|
327 | If C<$enable> is false (the default), then C<encode> will encode strings |
|
|
328 | normally (but see C<text_keys>) - upgraded perl strings (strings |
|
|
329 | internally encoded as UTF-8) as CBOR text strings, and downgraded perl |
|
|
330 | strings as CBOR byte strings. |
|
|
331 | |
|
|
332 | This option does not affect C<decode> in any way. |
|
|
333 | |
|
|
334 | This option has similar advantages and disadvantages as C<text_keys>. In |
|
|
335 | addition, this option effectively removes the ability to encode byte |
|
|
336 | strings, which might break some C<FREEZE> and C<TO_CBOR> methods that rely |
|
|
337 | on this, such as bignum encoding, so this option is mainly useful for very |
|
|
338 | simple data. |
|
|
339 | |
|
|
340 | =item $cbor = $cbor->validate_utf8 ([$enable]) |
|
|
341 | |
|
|
342 | =item $enabled = $cbor->get_validate_utf8 |
|
|
343 | |
|
|
344 | If C<$enable> is true (or missing), then C<decode> will validate that |
|
|
345 | elements (text strings) containing UTF-8 data in fact contain valid UTF-8 |
|
|
346 | data (instead of blindly accepting it). This validation obviously takes |
|
|
347 | extra time during decoding. |
|
|
348 | |
|
|
349 | The concept of "valid UTF-8" used is perl's concept, which is a superset |
|
|
350 | of the official UTF-8. |
|
|
351 | |
|
|
352 | If C<$enable> is false (the default), then C<decode> will blindly accept |
|
|
353 | UTF-8 data, marking them as valid UTF-8 in the resulting data structure |
|
|
354 | regardless of whether that's true or not. |
|
|
355 | |
|
|
356 | Perl isn't too happy about corrupted UTF-8 in strings, but should |
|
|
357 | generally not crash or do similarly evil things. Extensions might be not |
|
|
358 | so forgiving, so it's recommended to turn on this setting if you receive |
|
|
359 | untrusted CBOR. |
|
|
360 | |
|
|
361 | This option does not affect C<encode> in any way - strings that are |
|
|
362 | supposedly valid UTF-8 will simply be dumped into the resulting CBOR |
|
|
363 | string without checking whether that is, in fact, true or not. |
233 | |
364 | |
234 | =item $cbor = $cbor->filter ([$cb->($tag, $value)]) |
365 | =item $cbor = $cbor->filter ([$cb->($tag, $value)]) |
235 | |
366 | |
236 | =item $cb_or_undef = $cbor->get_filter |
367 | =item $cb_or_undef = $cbor->get_filter |
237 | |
368 | |
… | |
… | |
250 | replace the tagged value in the decoded data structure, or no values, |
381 | replace the tagged value in the decoded data structure, or no values, |
251 | which will result in default handling, which currently means the decoder |
382 | which will result in default handling, which currently means the decoder |
252 | creates a C<CBOR::XS::Tagged> object to hold the tag and the value. |
383 | creates a C<CBOR::XS::Tagged> object to hold the tag and the value. |
253 | |
384 | |
254 | When the filter is cleared (the default state), the default filter |
385 | When the filter is cleared (the default state), the default filter |
255 | function, C<CBOR::XS::default_filter>, is used. This function simply looks |
386 | function, C<CBOR::XS::default_filter>, is used. This function simply |
256 | up the tag in the C<%CBOR::XS::FILTER> hash. If an entry exists it must be |
387 | looks up the tag in the C<%CBOR::XS::FILTER> hash. If an entry exists |
257 | a code reference that is called with tag and value, and is responsible for |
388 | it must be a code reference that is called with tag and value, and is |
258 | decoding the value. If no entry exists, it returns no values. |
389 | responsible for decoding the value. If no entry exists, it returns no |
|
|
390 | values. C<CBOR::XS> provides a number of default filter functions already, |
|
|
391 | the the C<%CBOR::XS::FILTER> hash can be freely extended with more. |
259 | |
392 | |
|
|
393 | C<CBOR::XS> additionally provides an alternative filter function that is |
|
|
394 | supposed to be safe to use with untrusted data (which the default filter |
|
|
395 | might not), called C<CBOR::XS::safe_filter>, which works the same as |
|
|
396 | the C<default_filter> but uses the C<%CBOR::XS::SAFE_FILTER> variable |
|
|
397 | instead. It is prepopulated with the tag decoding functions that are |
|
|
398 | deemed safe (basically the same as C<%CBOR::XS::FILTER> without all |
|
|
399 | the bignum tags), and can be extended by user code as wlel, although, |
|
|
400 | obviously, one should be very careful about adding decoding functions |
|
|
401 | here, since the expectation is that they are safe to use on untrusted |
|
|
402 | data, after all. |
|
|
403 | |
260 | Example: decode all tags not handled internally into CBOR::XS::Tagged |
404 | Example: decode all tags not handled internally into C<CBOR::XS::Tagged> |
261 | objects, with no other special handling (useful when working with |
405 | objects, with no other special handling (useful when working with |
262 | potentially "unsafe" CBOR data). |
406 | potentially "unsafe" CBOR data). |
263 | |
407 | |
264 | CBOR::XS->new->filter (sub { })->decode ($cbor_data); |
408 | CBOR::XS->new->filter (sub { })->decode ($cbor_data); |
265 | |
409 | |
… | |
… | |
269 | $CBOR::XS::FILTER{1347375694} = sub { |
413 | $CBOR::XS::FILTER{1347375694} = sub { |
270 | my ($tag, $value); |
414 | my ($tag, $value); |
271 | |
415 | |
272 | "tag 1347375694 value $value" |
416 | "tag 1347375694 value $value" |
273 | }; |
417 | }; |
|
|
418 | |
|
|
419 | Example: provide your own filter function that looks up tags in your own |
|
|
420 | hash: |
|
|
421 | |
|
|
422 | my %my_filter = ( |
|
|
423 | 998347484 => sub { |
|
|
424 | my ($tag, $value); |
|
|
425 | |
|
|
426 | "tag 998347484 value $value" |
|
|
427 | }; |
|
|
428 | ); |
|
|
429 | |
|
|
430 | my $coder = CBOR::XS->new->filter (sub { |
|
|
431 | &{ $my_filter{$_[0]} or return } |
|
|
432 | }); |
|
|
433 | |
|
|
434 | |
|
|
435 | Example: use the safe filter function (see L<SECURITY CONSIDERATIONS> for |
|
|
436 | more considerations on security). |
|
|
437 | |
|
|
438 | CBOR::XS->new->filter (\&CBOR::XS::safe_filter)->decode ($cbor_data); |
274 | |
439 | |
275 | =item $cbor_data = $cbor->encode ($perl_scalar) |
440 | =item $cbor_data = $cbor->encode ($perl_scalar) |
276 | |
441 | |
277 | Converts the given Perl data structure (a scalar value) to its CBOR |
442 | Converts the given Perl data structure (a scalar value) to its CBOR |
278 | representation. |
443 | representation. |
… | |
… | |
288 | when there is trailing garbage after the CBOR string, it will silently |
453 | when there is trailing garbage after the CBOR string, it will silently |
289 | stop parsing there and return the number of characters consumed so far. |
454 | stop parsing there and return the number of characters consumed so far. |
290 | |
455 | |
291 | This is useful if your CBOR texts are not delimited by an outer protocol |
456 | This is useful if your CBOR texts are not delimited by an outer protocol |
292 | and you need to know where the first CBOR string ends amd the next one |
457 | and you need to know where the first CBOR string ends amd the next one |
293 | starts. |
458 | starts - CBOR strings are self-delimited, so it is possible to concatenate |
|
|
459 | CBOR strings without any delimiters or size fields and recover their data. |
294 | |
460 | |
295 | CBOR::XS->new->decode_prefix ("......") |
461 | CBOR::XS->new->decode_prefix ("......") |
296 | => ("...", 3) |
462 | => ("...", 3) |
|
|
463 | |
|
|
464 | =back |
|
|
465 | |
|
|
466 | =head2 INCREMENTAL PARSING |
|
|
467 | |
|
|
468 | In some cases, there is the need for incremental parsing of JSON |
|
|
469 | texts. While this module always has to keep both CBOR text and resulting |
|
|
470 | Perl data structure in memory at one time, it does allow you to parse a |
|
|
471 | CBOR stream incrementally, using a similar to using "decode_prefix" to see |
|
|
472 | if a full CBOR object is available, but is much more efficient. |
|
|
473 | |
|
|
474 | It basically works by parsing as much of a CBOR string as possible - if |
|
|
475 | the CBOR data is not complete yet, the pasrer will remember where it was, |
|
|
476 | to be able to restart when more data has been accumulated. Once enough |
|
|
477 | data is available to either decode a complete CBOR value or raise an |
|
|
478 | error, a real decode will be attempted. |
|
|
479 | |
|
|
480 | A typical use case would be a network protocol that consists of sending |
|
|
481 | and receiving CBOR-encoded messages. The solution that works with CBOR and |
|
|
482 | about anything else is by prepending a length to every CBOR value, so the |
|
|
483 | receiver knows how many octets to read. More compact (and slightly slower) |
|
|
484 | would be to just send CBOR values back-to-back, as C<CBOR::XS> knows where |
|
|
485 | a CBOR value ends, and doesn't need an explicit length. |
|
|
486 | |
|
|
487 | The following methods help with this: |
|
|
488 | |
|
|
489 | =over 4 |
|
|
490 | |
|
|
491 | =item @decoded = $cbor->incr_parse ($buffer) |
|
|
492 | |
|
|
493 | This method attempts to decode exactly one CBOR value from the beginning |
|
|
494 | of the given C<$buffer>. The value is removed from the C<$buffer> on |
|
|
495 | success. When C<$buffer> doesn't contain a complete value yet, it returns |
|
|
496 | nothing. Finally, when the C<$buffer> doesn't start with something |
|
|
497 | that could ever be a valid CBOR value, it raises an exception, just as |
|
|
498 | C<decode> would. In the latter case the decoder state is undefined and |
|
|
499 | must be reset before being able to parse further. |
|
|
500 | |
|
|
501 | This method modifies the C<$buffer> in place. When no CBOR value can be |
|
|
502 | decoded, the decoder stores the current string offset. On the next call, |
|
|
503 | continues decoding at the place where it stopped before. For this to make |
|
|
504 | sense, the C<$buffer> must begin with the same octets as on previous |
|
|
505 | unsuccessful calls. |
|
|
506 | |
|
|
507 | You can call this method in scalar context, in which case it either |
|
|
508 | returns a decoded value or C<undef>. This makes it impossible to |
|
|
509 | distinguish between CBOR null values (which decode to C<undef>) and an |
|
|
510 | unsuccessful decode, which is often acceptable. |
|
|
511 | |
|
|
512 | =item @decoded = $cbor->incr_parse_multiple ($buffer) |
|
|
513 | |
|
|
514 | Same as C<incr_parse>, but attempts to decode as many CBOR values as |
|
|
515 | possible in one go, instead of at most one. Calls to C<incr_parse> and |
|
|
516 | C<incr_parse_multiple> can be interleaved. |
|
|
517 | |
|
|
518 | =item $cbor->incr_reset |
|
|
519 | |
|
|
520 | Resets the incremental decoder. This throws away any saved state, so that |
|
|
521 | subsequent calls to C<incr_parse> or C<incr_parse_multiple> start to parse |
|
|
522 | a new CBOR value from the beginning of the C<$buffer> again. |
|
|
523 | |
|
|
524 | This method can be called at any time, but it I<must> be called if you want |
|
|
525 | to change your C<$buffer> or there was a decoding error and you want to |
|
|
526 | reuse the C<$cbor> object for future incremental parsings. |
297 | |
527 | |
298 | =back |
528 | =back |
299 | |
529 | |
300 | |
530 | |
301 | =head1 MAPPING |
531 | =head1 MAPPING |
… | |
… | |
319 | CBOR integers become (numeric) perl scalars. On perls without 64 bit |
549 | CBOR integers become (numeric) perl scalars. On perls without 64 bit |
320 | support, 64 bit integers will be truncated or otherwise corrupted. |
550 | support, 64 bit integers will be truncated or otherwise corrupted. |
321 | |
551 | |
322 | =item byte strings |
552 | =item byte strings |
323 | |
553 | |
324 | Byte strings will become octet strings in Perl (the byte values 0..255 |
554 | Byte strings will become octet strings in Perl (the Byte values 0..255 |
325 | will simply become characters of the same value in Perl). |
555 | will simply become characters of the same value in Perl). |
326 | |
556 | |
327 | =item UTF-8 strings |
557 | =item UTF-8 strings |
328 | |
558 | |
329 | UTF-8 strings in CBOR will be decoded, i.e. the UTF-8 octets will be |
559 | UTF-8 strings in CBOR will be decoded, i.e. the UTF-8 octets will be |
… | |
… | |
352 | =item tagged values |
582 | =item tagged values |
353 | |
583 | |
354 | Tagged items consists of a numeric tag and another CBOR value. |
584 | Tagged items consists of a numeric tag and another CBOR value. |
355 | |
585 | |
356 | See L<TAG HANDLING AND EXTENSIONS> and the description of C<< ->filter >> |
586 | See L<TAG HANDLING AND EXTENSIONS> and the description of C<< ->filter >> |
357 | for details. |
587 | for details on which tags are handled how. |
358 | |
588 | |
359 | =item anything else |
589 | =item anything else |
360 | |
590 | |
361 | Anything else (e.g. unsupported simple values) will raise a decoding |
591 | Anything else (e.g. unsupported simple values) will raise a decoding |
362 | error. |
592 | error. |
… | |
… | |
365 | |
595 | |
366 | |
596 | |
367 | =head2 PERL -> CBOR |
597 | =head2 PERL -> CBOR |
368 | |
598 | |
369 | The mapping from Perl to CBOR is slightly more difficult, as Perl is a |
599 | The mapping from Perl to CBOR is slightly more difficult, as Perl is a |
370 | truly typeless language, so we can only guess which CBOR type is meant by |
600 | typeless language. That means this module can only guess which CBOR type |
371 | a Perl value. |
601 | is meant by a perl value. |
372 | |
602 | |
373 | =over 4 |
603 | =over 4 |
374 | |
604 | |
375 | =item hash references |
605 | =item hash references |
376 | |
606 | |
377 | Perl hash references become CBOR maps. As there is no inherent ordering in |
607 | Perl hash references become CBOR maps. As there is no inherent ordering in |
378 | hash keys (or CBOR maps), they will usually be encoded in a pseudo-random |
608 | hash keys (or CBOR maps), they will usually be encoded in a pseudo-random |
379 | order. |
609 | order. This order can be different each time a hash is encoded. |
380 | |
610 | |
381 | Currently, tied hashes will use the indefinite-length format, while normal |
611 | Currently, tied hashes will use the indefinite-length format, while normal |
382 | hashes will use the fixed-length format. |
612 | hashes will use the fixed-length format. |
383 | |
613 | |
384 | =item array references |
614 | =item array references |
385 | |
615 | |
386 | Perl array references become fixed-length CBOR arrays. |
616 | Perl array references become fixed-length CBOR arrays. |
387 | |
617 | |
388 | =item other references |
618 | =item other references |
389 | |
619 | |
390 | Other unblessed references are generally not allowed and will cause an |
620 | Other unblessed references will be represented using |
391 | exception to be thrown, except for references to the integers C<0> and |
621 | the indirection tag extension (tag value C<22098>, |
392 | C<1>, which get turned into false and true in CBOR. |
622 | L<http://cbor.schmorp.de/indirection>). CBOR decoders are guaranteed |
|
|
623 | to be able to decode these values somehow, by either "doing the right |
|
|
624 | thing", decoding into a generic tagged object, simply ignoring the tag, or |
|
|
625 | something else. |
393 | |
626 | |
394 | =item CBOR::XS::Tagged objects |
627 | =item CBOR::XS::Tagged objects |
395 | |
628 | |
396 | Objects of this type must be arrays consisting of a single C<[tag, value]> |
629 | Objects of this type must be arrays consisting of a single C<[tag, value]> |
397 | pair. The (numerical) tag will be encoded as a CBOR tag, the value will |
630 | pair. The (numerical) tag will be encoded as a CBOR tag, the value will |
398 | be encoded as appropriate for the value. You cna use C<CBOR::XS::tag> to |
631 | be encoded as appropriate for the value. You must use C<CBOR::XS::tag> to |
399 | create such objects. |
632 | create such objects. |
400 | |
633 | |
401 | =item Types::Serialiser::true, Types::Serialiser::false, Types::Serialiser::error |
634 | =item Types::Serialiser::true, Types::Serialiser::false, Types::Serialiser::error |
402 | |
635 | |
403 | These special values become CBOR true, CBOR false and CBOR undefined |
636 | These special values become CBOR true, CBOR false and CBOR undefined |
… | |
… | |
420 | # dump as number |
653 | # dump as number |
421 | encode_cbor [2] # yields [2] |
654 | encode_cbor [2] # yields [2] |
422 | encode_cbor [-3.0e17] # yields [-3e+17] |
655 | encode_cbor [-3.0e17] # yields [-3e+17] |
423 | my $value = 5; encode_cbor [$value] # yields [5] |
656 | my $value = 5; encode_cbor [$value] # yields [5] |
424 | |
657 | |
425 | # used as string, so dump as string |
658 | # used as string, so dump as string (either byte or text) |
426 | print $value; |
659 | print $value; |
427 | encode_cbor [$value] # yields ["5"] |
660 | encode_cbor [$value] # yields ["5"] |
428 | |
661 | |
429 | # undef becomes null |
662 | # undef becomes null |
430 | encode_cbor [undef] # yields [null] |
663 | encode_cbor [undef] # yields [null] |
… | |
… | |
433 | |
666 | |
434 | my $x = 3.1; # some variable containing a number |
667 | my $x = 3.1; # some variable containing a number |
435 | "$x"; # stringified |
668 | "$x"; # stringified |
436 | $x .= ""; # another, more awkward way to stringify |
669 | $x .= ""; # another, more awkward way to stringify |
437 | print $x; # perl does it for you, too, quite often |
670 | print $x; # perl does it for you, too, quite often |
|
|
671 | |
|
|
672 | You can force whether a string is encoded as byte or text string by using |
|
|
673 | C<utf8::upgrade> and C<utf8::downgrade> (if C<text_strings> is disabled): |
|
|
674 | |
|
|
675 | utf8::upgrade $x; # encode $x as text string |
|
|
676 | utf8::downgrade $x; # encode $x as byte string |
|
|
677 | |
|
|
678 | Perl doesn't define what operations up- and downgrade strings, so if the |
|
|
679 | difference between byte and text is important, you should up- or downgrade |
|
|
680 | your string as late as possible before encoding. You can also force the |
|
|
681 | use of CBOR text strings by using C<text_keys> or C<text_strings>. |
438 | |
682 | |
439 | You can force the type to be a CBOR number by numifying it: |
683 | You can force the type to be a CBOR number by numifying it: |
440 | |
684 | |
441 | my $x = "3"; # some variable containing a string |
685 | my $x = "3"; # some variable containing a string |
442 | $x += 0; # numify it, ensuring it will be dumped as a number |
686 | $x += 0; # numify it, ensuring it will be dumped as a number |
… | |
… | |
455 | |
699 | |
456 | =back |
700 | =back |
457 | |
701 | |
458 | =head2 OBJECT SERIALISATION |
702 | =head2 OBJECT SERIALISATION |
459 | |
703 | |
|
|
704 | This module implements both a CBOR-specific and the generic |
|
|
705 | L<Types::Serialier> object serialisation protocol. The following |
|
|
706 | subsections explain both methods. |
|
|
707 | |
|
|
708 | =head3 ENCODING |
|
|
709 | |
460 | This module knows two way to serialise a Perl object: The CBOR-specific |
710 | This module knows two way to serialise a Perl object: The CBOR-specific |
461 | way, and the generic way. |
711 | way, and the generic way. |
462 | |
712 | |
463 | Whenever the encoder encounters a Perl object that it cnanot serialise |
713 | Whenever the encoder encounters a Perl object that it cannot serialise |
464 | directly (most of them), it will first look up the C<TO_CBOR> method on |
714 | directly (most of them), it will first look up the C<TO_CBOR> method on |
465 | it. |
715 | it. |
466 | |
716 | |
467 | If it has a C<TO_CBOR> method, it will call it with the object as only |
717 | If it has a C<TO_CBOR> method, it will call it with the object as only |
468 | argument, and expects exactly one return value, which it will then |
718 | argument, and expects exactly one return value, which it will then |
… | |
… | |
474 | |
724 | |
475 | The C<FREEZE> method can return any number of values (i.e. zero or |
725 | The C<FREEZE> method can return any number of values (i.e. zero or |
476 | more). These will be encoded as CBOR perl object, together with the |
726 | more). These will be encoded as CBOR perl object, together with the |
477 | classname. |
727 | classname. |
478 | |
728 | |
|
|
729 | These methods I<MUST NOT> change the data structure that is being |
|
|
730 | serialised. Failure to comply to this can result in memory corruption - |
|
|
731 | and worse. |
|
|
732 | |
479 | If an object supports neither C<TO_CBOR> nor C<FREEZE>, encoding will fail |
733 | If an object supports neither C<TO_CBOR> nor C<FREEZE>, encoding will fail |
480 | with an error. |
734 | with an error. |
481 | |
735 | |
|
|
736 | =head3 DECODING |
|
|
737 | |
482 | Objects encoded via C<TO_CBOR> cannot be automatically decoded, but |
738 | Objects encoded via C<TO_CBOR> cannot (normally) be automatically decoded, |
483 | objects encoded via C<FREEZE> can be decoded using the following protocol: |
739 | but objects encoded via C<FREEZE> can be decoded using the following |
|
|
740 | protocol: |
484 | |
741 | |
485 | When an encoded CBOR perl object is encountered by the decoder, it will |
742 | When an encoded CBOR perl object is encountered by the decoder, it will |
486 | look up the C<THAW> method, by using the stored classname, and will fail |
743 | look up the C<THAW> method, by using the stored classname, and will fail |
487 | if the method cannot be found. |
744 | if the method cannot be found. |
488 | |
745 | |
489 | After the lookup it will call the C<THAW> method with the stored classname |
746 | After the lookup it will call the C<THAW> method with the stored classname |
490 | as first argument, the constant string C<CBOR> as second argument, and all |
747 | as first argument, the constant string C<CBOR> as second argument, and all |
491 | values returned by C<FREEZE> as remaining arguments. |
748 | values returned by C<FREEZE> as remaining arguments. |
492 | |
749 | |
493 | =head4 EXAMPLES |
750 | =head3 EXAMPLES |
494 | |
751 | |
495 | Here is an example C<TO_CBOR> method: |
752 | Here is an example C<TO_CBOR> method: |
496 | |
753 | |
497 | sub My::Object::TO_CBOR { |
754 | sub My::Object::TO_CBOR { |
498 | my ($obj) = @_; |
755 | my ($obj) = @_; |
… | |
… | |
509 | |
766 | |
510 | sub URI::TO_CBOR { |
767 | sub URI::TO_CBOR { |
511 | my ($self) = @_; |
768 | my ($self) = @_; |
512 | my $uri = "$self"; # stringify uri |
769 | my $uri = "$self"; # stringify uri |
513 | utf8::upgrade $uri; # make sure it will be encoded as UTF-8 string |
770 | utf8::upgrade $uri; # make sure it will be encoded as UTF-8 string |
514 | CBOR::XS::tagged 32, "$_[0]" |
771 | CBOR::XS::tag 32, "$_[0]" |
515 | } |
772 | } |
516 | |
773 | |
517 | This will encode URIs as a UTF-8 string with tag 32, which indicates an |
774 | This will encode URIs as a UTF-8 string with tag 32, which indicates an |
518 | URI. |
775 | URI. |
519 | |
776 | |
… | |
… | |
530 | "$self" # encode url string |
787 | "$self" # encode url string |
531 | } |
788 | } |
532 | |
789 | |
533 | sub URI::THAW { |
790 | sub URI::THAW { |
534 | my ($class, $serialiser, $uri) = @_; |
791 | my ($class, $serialiser, $uri) = @_; |
535 | |
|
|
536 | $class->new ($uri) |
792 | $class->new ($uri) |
537 | } |
793 | } |
538 | |
794 | |
539 | Unlike C<TO_CBOR>, multiple values can be returned by C<FREEZE>. For |
795 | Unlike C<TO_CBOR>, multiple values can be returned by C<FREEZE>. For |
540 | example, a C<FREEZE> method that returns "type", "id" and "variant" values |
796 | example, a C<FREEZE> method that returns "type", "id" and "variant" values |
… | |
… | |
671 | additional tags (such as base64url). |
927 | additional tags (such as base64url). |
672 | |
928 | |
673 | =head2 ENFORCED TAGS |
929 | =head2 ENFORCED TAGS |
674 | |
930 | |
675 | These tags are always handled when decoding, and their handling cannot be |
931 | These tags are always handled when decoding, and their handling cannot be |
676 | overriden by the user. |
932 | overridden by the user. |
677 | |
933 | |
678 | =over 4 |
934 | =over 4 |
679 | |
935 | |
680 | =item <unassigned> (perl-object, L<http://cbor.schmorp.de/perl-object>) |
936 | =item 26 (perl-object, L<http://cbor.schmorp.de/perl-object>) |
681 | |
937 | |
682 | These tags are automatically created (and decoded) for serialisable |
938 | These tags are automatically created (and decoded) for serialisable |
683 | objects using the C<FREEZE/THAW> methods (the L<Types::Serialier> object |
939 | objects using the C<FREEZE/THAW> methods (the L<Types::Serialier> object |
684 | serialisation protocol). See L<OBJECT SERIALISATION> for details. |
940 | serialisation protocol). See L<OBJECT SERIALISATION> for details. |
685 | |
941 | |
686 | =item <unassigned>, <unassigned> (sharable, sharedref, L <http://cbor.schmorp.de/value-sharing>) |
942 | =item 28, 29 (shareable, sharedref, L<http://cbor.schmorp.de/value-sharing>) |
687 | |
943 | |
688 | These tags are automatically decoded when encountered, resulting in |
944 | These tags are automatically decoded when encountered (and they do not |
|
|
945 | result in a cyclic data structure, see C<allow_cycles>), resulting in |
689 | shared values in the decoded object. They are only encoded, however, when |
946 | shared values in the decoded object. They are only encoded, however, when |
690 | C<allow_sharable> is enabled. |
947 | C<allow_sharing> is enabled. |
691 | |
948 | |
|
|
949 | Not all shared values can be successfully decoded: values that reference |
|
|
950 | themselves will I<currently> decode as C<undef> (this is not the same |
|
|
951 | as a reference pointing to itself, which will be represented as a value |
|
|
952 | that contains an indirect reference to itself - these will be decoded |
|
|
953 | properly). |
|
|
954 | |
|
|
955 | Note that considerably more shared value data structures can be decoded |
|
|
956 | than will be encoded - currently, only values pointed to by references |
|
|
957 | will be shared, others will not. While non-reference shared values can be |
|
|
958 | generated in Perl with some effort, they were considered too unimportant |
|
|
959 | to be supported in the encoder. The decoder, however, will decode these |
|
|
960 | values as shared values. |
|
|
961 | |
692 | =item <unassigned>, <unassigned> (stringref-namespace, stringref, L <http://cbor.schmorp.de/stringref>) |
962 | =item 256, 25 (stringref-namespace, stringref, L<http://cbor.schmorp.de/stringref>) |
693 | |
963 | |
694 | These tags are automatically decoded when encountered. They are only |
964 | These tags are automatically decoded when encountered. They are only |
695 | encoded, however, when C<allow_stringref> is enabled. |
965 | encoded, however, when C<pack_strings> is enabled. |
696 | |
966 | |
697 | =item 22098 (indirection, L<http://cbor.schmorp.de/indirection>) |
967 | =item 22098 (indirection, L<http://cbor.schmorp.de/indirection>) |
698 | |
968 | |
699 | This tag is automatically generated when a reference are encountered (with |
969 | This tag is automatically generated when a reference are encountered (with |
700 | the exception of hash and array refernces). It is converted to a reference |
970 | the exception of hash and array references). It is converted to a reference |
701 | when decoding. |
971 | when decoding. |
702 | |
972 | |
703 | =item 55799 (self-describe CBOR, RFC 7049) |
973 | =item 55799 (self-describe CBOR, RFC 7049) |
704 | |
974 | |
705 | This value is not generated on encoding (unless explicitly requested by |
975 | This value is not generated on encoding (unless explicitly requested by |
… | |
… | |
708 | =back |
978 | =back |
709 | |
979 | |
710 | =head2 NON-ENFORCED TAGS |
980 | =head2 NON-ENFORCED TAGS |
711 | |
981 | |
712 | These tags have default filters provided when decoding. Their handling can |
982 | These tags have default filters provided when decoding. Their handling can |
713 | be overriden by changing the C<%CBOR::XS::FILTER> entry for the tag, or by |
983 | be overridden by changing the C<%CBOR::XS::FILTER> entry for the tag, or by |
714 | providing a custom C<filter> callback when decoding. |
984 | providing a custom C<filter> callback when decoding. |
715 | |
985 | |
716 | When they result in decoding into a specific Perl class, the module |
986 | When they result in decoding into a specific Perl class, the module |
717 | usually provides a corresponding C<TO_CBOR> method as well. |
987 | usually provides a corresponding C<TO_CBOR> method as well. |
718 | |
988 | |
… | |
… | |
721 | provide these modules. The decoding usually fails with an exception if the |
991 | provide these modules. The decoding usually fails with an exception if the |
722 | required module cannot be loaded. |
992 | required module cannot be loaded. |
723 | |
993 | |
724 | =over 4 |
994 | =over 4 |
725 | |
995 | |
|
|
996 | =item 0, 1 (date/time string, seconds since the epoch) |
|
|
997 | |
|
|
998 | These tags are decoded into L<Time::Piece> objects. The corresponding |
|
|
999 | C<Time::Piece::TO_CBOR> method always encodes into tag 1 values currently. |
|
|
1000 | |
|
|
1001 | The L<Time::Piece> API is generally surprisingly bad, and fractional |
|
|
1002 | seconds are only accidentally kept intact, so watch out. On the plus side, |
|
|
1003 | the module comes with perl since 5.10, which has to count for something. |
|
|
1004 | |
726 | =item 2, 3 (positive/negative bignum) |
1005 | =item 2, 3 (positive/negative bignum) |
727 | |
1006 | |
728 | These tags are decoded into L<Math::BigInt> objects. The corresponding |
1007 | These tags are decoded into L<Math::BigInt> objects. The corresponding |
729 | C<Math::BigInt::TO_CBOR> method encodes "small" bigints into normal CBOR |
1008 | C<Math::BigInt::TO_CBOR> method encodes "small" bigints into normal CBOR |
730 | integers, and others into positive/negative CBOR bignums. |
1009 | integers, and others into positive/negative CBOR bignums. |
731 | |
1010 | |
732 | =item 4, 5 (decimal fraction/bigfloat) |
1011 | =item 4, 5, 264, 265 (decimal fraction/bigfloat) |
733 | |
1012 | |
734 | Both decimal fractions and bigfloats are decoded into L<Math::BigFloat> |
1013 | Both decimal fractions and bigfloats are decoded into L<Math::BigFloat> |
735 | objects. The corresponding C<Math::BigFloat::TO_CBOR> method I<always> |
1014 | objects. The corresponding C<Math::BigFloat::TO_CBOR> method I<always> |
736 | encodes into a decimal fraction. |
1015 | encodes into a decimal fraction (either tag 4 or 264). |
737 | |
1016 | |
738 | CBOR cannot represent bigfloats with I<very> large exponents - conversion |
1017 | NaN and infinities are not encoded properly, as they cannot be represented |
739 | of such big float objects is undefined. |
1018 | in CBOR. |
740 | |
1019 | |
741 | Also, NaN and infinities are not encoded properly. |
1020 | See L<BIGNUM SECURITY CONSIDERATIONS> for more info. |
|
|
1021 | |
|
|
1022 | =item 30 (rational numbers) |
|
|
1023 | |
|
|
1024 | These tags are decoded into L<Math::BigRat> objects. The corresponding |
|
|
1025 | C<Math::BigRat::TO_CBOR> method encodes rational numbers with denominator |
|
|
1026 | C<1> via their numerator only, i.e., they become normal integers or |
|
|
1027 | C<bignums>. |
|
|
1028 | |
|
|
1029 | See L<BIGNUM SECURITY CONSIDERATIONS> for more info. |
742 | |
1030 | |
743 | =item 21, 22, 23 (expected later JSON conversion) |
1031 | =item 21, 22, 23 (expected later JSON conversion) |
744 | |
1032 | |
745 | CBOR::XS is not a CBOR-to-JSON converter, and will simply ignore these |
1033 | CBOR::XS is not a CBOR-to-JSON converter, and will simply ignore these |
746 | tags. |
1034 | tags. |
… | |
… | |
751 | C<URI::TO_CBOR> method again results in a CBOR URI value. |
1039 | C<URI::TO_CBOR> method again results in a CBOR URI value. |
752 | |
1040 | |
753 | =back |
1041 | =back |
754 | |
1042 | |
755 | =cut |
1043 | =cut |
756 | |
|
|
757 | our %FILTER = ( |
|
|
758 | # 0 # rfc4287 datetime, utf-8 |
|
|
759 | # 1 # unix timestamp, any |
|
|
760 | |
|
|
761 | 2 => sub { # pos bigint |
|
|
762 | require Math::BigInt; |
|
|
763 | Math::BigInt->new ("0x" . unpack "H*", pop) |
|
|
764 | }, |
|
|
765 | |
|
|
766 | 3 => sub { # neg bigint |
|
|
767 | require Math::BigInt; |
|
|
768 | -Math::BigInt->new ("0x" . unpack "H*", pop) |
|
|
769 | }, |
|
|
770 | |
|
|
771 | 4 => sub { # decimal fraction, array |
|
|
772 | require Math::BigFloat; |
|
|
773 | Math::BigFloat->new ($_[1][1] . "E" . $_[1][0]) |
|
|
774 | }, |
|
|
775 | |
|
|
776 | 5 => sub { # bigfloat, array |
|
|
777 | require Math::BigFloat; |
|
|
778 | scalar Math::BigFloat->new ($_[1][1])->blsft ($_[1][0], 2) |
|
|
779 | }, |
|
|
780 | |
|
|
781 | 21 => sub { pop }, # expected conversion to base64url encoding |
|
|
782 | 22 => sub { pop }, # expected conversion to base64 encoding |
|
|
783 | 23 => sub { pop }, # expected conversion to base16 encoding |
|
|
784 | |
|
|
785 | # 24 # embedded cbor, byte string |
|
|
786 | |
|
|
787 | 32 => sub { |
|
|
788 | require URI; |
|
|
789 | URI->new (pop) |
|
|
790 | }, |
|
|
791 | |
|
|
792 | # 33 # base64url rfc4648, utf-8 |
|
|
793 | # 34 # base64 rfc46484, utf-8 |
|
|
794 | # 35 # regex pcre/ecma262, utf-8 |
|
|
795 | # 36 # mime message rfc2045, utf-8 |
|
|
796 | ); |
|
|
797 | |
|
|
798 | |
1044 | |
799 | =head1 CBOR and JSON |
1045 | =head1 CBOR and JSON |
800 | |
1046 | |
801 | CBOR is supposed to implement a superset of the JSON data model, and is, |
1047 | CBOR is supposed to implement a superset of the JSON data model, and is, |
802 | with some coercion, able to represent all JSON texts (something that other |
1048 | with some coercion, able to represent all JSON texts (something that other |
… | |
… | |
811 | CBOR intact. |
1057 | CBOR intact. |
812 | |
1058 | |
813 | |
1059 | |
814 | =head1 SECURITY CONSIDERATIONS |
1060 | =head1 SECURITY CONSIDERATIONS |
815 | |
1061 | |
816 | When you are using CBOR in a protocol, talking to untrusted potentially |
1062 | Tl;dr... if you want to decode or encode CBOR from untrusted sources, you |
817 | hostile creatures requires relatively few measures. |
1063 | should start with a coder object created via C<new_safe> (which implements |
|
|
1064 | the mitigations explained below): |
818 | |
1065 | |
|
|
1066 | my $coder = CBOR::XS->new_safe; |
|
|
1067 | |
|
|
1068 | my $data = $coder->decode ($cbor_text); |
|
|
1069 | my $cbor = $coder->encode ($data); |
|
|
1070 | |
|
|
1071 | Longer version: When you are using CBOR in a protocol, talking to |
|
|
1072 | untrusted potentially hostile creatures requires some thought: |
|
|
1073 | |
|
|
1074 | =over 4 |
|
|
1075 | |
|
|
1076 | =item Security of the CBOR decoder itself |
|
|
1077 | |
819 | First of all, your CBOR decoder should be secure, that is, should not have |
1078 | First and foremost, your CBOR decoder should be secure, that is, should |
|
|
1079 | not have any buffer overflows or similar bugs that could potentially be |
820 | any buffer overflows. Obviously, this module should ensure that and I am |
1080 | exploited. Obviously, this module should ensure that and I am trying hard |
821 | trying hard on making that true, but you never know. |
1081 | on making that true, but you never know. |
822 | |
1082 | |
|
|
1083 | =item CBOR::XS can invoke almost arbitrary callbacks during decoding |
|
|
1084 | |
|
|
1085 | CBOR::XS supports object serialisation - decoding CBOR can cause calls |
|
|
1086 | to I<any> C<THAW> method in I<any> package that exists in your process |
|
|
1087 | (that is, CBOR::XS will not try to load modules, but any existing C<THAW> |
|
|
1088 | method or function can be called, so they all have to be secure). |
|
|
1089 | |
|
|
1090 | Less obviously, it will also invoke C<TO_CBOR> and C<FREEZE> methods - |
|
|
1091 | even if all your C<THAW> methods are secure, encoding data structures from |
|
|
1092 | untrusted sources can invoke those and trigger bugs in those. |
|
|
1093 | |
|
|
1094 | So, if you are not sure about the security of all the modules you |
|
|
1095 | have loaded (you shouldn't), you should disable this part using |
|
|
1096 | C<forbid_objects> or using C<new_safe>. |
|
|
1097 | |
|
|
1098 | =item CBOR can be extended with tags that call library code |
|
|
1099 | |
|
|
1100 | CBOR can be extended with tags, and C<CBOR::XS> has a registry of |
|
|
1101 | conversion functions for many existing tags that can be extended via |
|
|
1102 | third-party modules (see the C<filter> method). |
|
|
1103 | |
|
|
1104 | If you don't trust these, you should configure the "safe" filter function, |
|
|
1105 | C<CBOR::XS::safe_filter> (C<new_safe> does this), which by default only |
|
|
1106 | includes conversion functions that are considered "safe" by the author |
|
|
1107 | (but again, they can be extended by third party modules). |
|
|
1108 | |
|
|
1109 | Depending on your level of paranoia, you can use the "safe" filter: |
|
|
1110 | |
|
|
1111 | $cbor->filter (\&CBOR::XS::safe_filter); |
|
|
1112 | |
|
|
1113 | ... your own filter... |
|
|
1114 | |
|
|
1115 | $cbor->filter (sub { ... do your stuffs here ... }); |
|
|
1116 | |
|
|
1117 | ... or even no filter at all, disabling all tag decoding: |
|
|
1118 | |
|
|
1119 | $cbor->filter (sub { }); |
|
|
1120 | |
|
|
1121 | This is never a problem for encoding, as the tag mechanism only exists in |
|
|
1122 | CBOR texts. |
|
|
1123 | |
|
|
1124 | =item Resource-starving attacks: object memory usage |
|
|
1125 | |
823 | Second, you need to avoid resource-starving attacks. That means you should |
1126 | You need to avoid resource-starving attacks. That means you should limit |
824 | limit the size of CBOR data you accept, or make sure then when your |
1127 | the size of CBOR data you accept, or make sure then when your resources |
825 | resources run out, that's just fine (e.g. by using a separate process that |
1128 | run out, that's just fine (e.g. by using a separate process that can |
826 | can crash safely). The size of a CBOR string in octets is usually a good |
1129 | crash safely). The size of a CBOR string in octets is usually a good |
827 | indication of the size of the resources required to decode it into a Perl |
1130 | indication of the size of the resources required to decode it into a Perl |
828 | structure. While CBOR::XS can check the size of the CBOR text, it might be |
1131 | structure. While CBOR::XS can check the size of the CBOR text (using |
829 | too late when you already have it in memory, so you might want to check |
1132 | C<max_size> - done by C<new_safe>), it might be too late when you already |
830 | the size before you accept the string. |
1133 | have it in memory, so you might want to check the size before you accept |
|
|
1134 | the string. |
831 | |
1135 | |
|
|
1136 | As for encoding, it is possible to construct data structures that are |
|
|
1137 | relatively small but result in large CBOR texts (for example by having an |
|
|
1138 | array full of references to the same big data structure, which will all be |
|
|
1139 | deep-cloned during encoding by default). This is rarely an actual issue |
|
|
1140 | (and the worst case is still just running out of memory), but you can |
|
|
1141 | reduce this risk by using C<allow_sharing>. |
|
|
1142 | |
|
|
1143 | =item Resource-starving attacks: stack overflows |
|
|
1144 | |
832 | Third, CBOR::XS recurses using the C stack when decoding objects and |
1145 | CBOR::XS recurses using the C stack when decoding objects and arrays. The |
833 | arrays. The C stack is a limited resource: for instance, on my amd64 |
1146 | C stack is a limited resource: for instance, on my amd64 machine with 8MB |
834 | machine with 8MB of stack size I can decode around 180k nested arrays but |
1147 | of stack size I can decode around 180k nested arrays but only 14k nested |
835 | only 14k nested CBOR objects (due to perl itself recursing deeply on croak |
1148 | CBOR objects (due to perl itself recursing deeply on croak to free the |
836 | to free the temporary). If that is exceeded, the program crashes. To be |
1149 | temporary). If that is exceeded, the program crashes. To be conservative, |
837 | conservative, the default nesting limit is set to 512. If your process |
1150 | the default nesting limit is set to 512. If your process has a smaller |
838 | has a smaller stack, you should adjust this setting accordingly with the |
1151 | stack, you should adjust this setting accordingly with the C<max_depth> |
839 | C<max_depth> method. |
1152 | method. |
|
|
1153 | |
|
|
1154 | =item Resource-starving attacks: CPU en-/decoding complexity |
|
|
1155 | |
|
|
1156 | CBOR::XS will use the L<Math::BigInt>, L<Math::BigFloat> and |
|
|
1157 | L<Math::BigRat> libraries to represent encode/decode bignums. These can be |
|
|
1158 | very slow (as in, centuries of CPU time) and can even crash your program |
|
|
1159 | (and are generally not very trustworthy). See the next section on bignum |
|
|
1160 | security for details. |
|
|
1161 | |
|
|
1162 | =item Data breaches: leaking information in error messages |
|
|
1163 | |
|
|
1164 | CBOR::XS might leak contents of your Perl data structures in its error |
|
|
1165 | messages, so when you serialise sensitive information you might want to |
|
|
1166 | make sure that exceptions thrown by CBOR::XS will not end up in front of |
|
|
1167 | untrusted eyes. |
|
|
1168 | |
|
|
1169 | =item Something else... |
840 | |
1170 | |
841 | Something else could bomb you, too, that I forgot to think of. In that |
1171 | Something else could bomb you, too, that I forgot to think of. In that |
842 | case, you get to keep the pieces. I am always open for hints, though... |
1172 | case, you get to keep the pieces. I am always open for hints, though... |
843 | |
1173 | |
844 | Also keep in mind that CBOR::XS might leak contents of your Perl data |
1174 | =back |
845 | structures in its error messages, so when you serialise sensitive |
1175 | |
846 | information you might want to make sure that exceptions thrown by CBOR::XS |
1176 | |
847 | will not end up in front of untrusted eyes. |
1177 | =head1 BIGNUM SECURITY CONSIDERATIONS |
|
|
1178 | |
|
|
1179 | CBOR::XS provides a C<TO_CBOR> method for both L<Math::BigInt> and |
|
|
1180 | L<Math::BigFloat> that tries to encode the number in the simplest possible |
|
|
1181 | way, that is, either a CBOR integer, a CBOR bigint/decimal fraction (tag |
|
|
1182 | 4) or an arbitrary-exponent decimal fraction (tag 264). Rational numbers |
|
|
1183 | (L<Math::BigRat>, tag 30) can also contain bignums as members. |
|
|
1184 | |
|
|
1185 | CBOR::XS will also understand base-2 bigfloat or arbitrary-exponent |
|
|
1186 | bigfloats (tags 5 and 265), but it will never generate these on its own. |
|
|
1187 | |
|
|
1188 | Using the built-in L<Math::BigInt::Calc> support, encoding and decoding |
|
|
1189 | decimal fractions is generally fast. Decoding bigints can be slow for very |
|
|
1190 | big numbers (tens of thousands of digits, something that could potentially |
|
|
1191 | be caught by limiting the size of CBOR texts), and decoding bigfloats or |
|
|
1192 | arbitrary-exponent bigfloats can be I<extremely> slow (minutes, decades) |
|
|
1193 | for large exponents (roughly 40 bit and longer). |
|
|
1194 | |
|
|
1195 | Additionally, L<Math::BigInt> can take advantage of other bignum |
|
|
1196 | libraries, such as L<Math::GMP>, which cannot handle big floats with large |
|
|
1197 | exponents, and might simply abort or crash your program, due to their code |
|
|
1198 | quality. |
|
|
1199 | |
|
|
1200 | This can be a concern if you want to parse untrusted CBOR. If it is, you |
|
|
1201 | might want to disable decoding of tag 2 (bigint) and 3 (negative bigint) |
|
|
1202 | types. You should also disable types 5 and 265, as these can be slow even |
|
|
1203 | without bigints. |
|
|
1204 | |
|
|
1205 | Disabling bigints will also partially or fully disable types that rely on |
|
|
1206 | them, e.g. rational numbers that use bignums. |
|
|
1207 | |
848 | |
1208 | |
849 | =head1 CBOR IMPLEMENTATION NOTES |
1209 | =head1 CBOR IMPLEMENTATION NOTES |
850 | |
1210 | |
851 | This section contains some random implementation notes. They do not |
1211 | This section contains some random implementation notes. They do not |
852 | describe guaranteed behaviour, but merely behaviour as-is implemented |
1212 | describe guaranteed behaviour, but merely behaviour as-is implemented |
… | |
… | |
861 | Only the double data type is supported for NV data types - when Perl uses |
1221 | Only the double data type is supported for NV data types - when Perl uses |
862 | long double to represent floating point values, they might not be encoded |
1222 | long double to represent floating point values, they might not be encoded |
863 | properly. Half precision types are accepted, but not encoded. |
1223 | properly. Half precision types are accepted, but not encoded. |
864 | |
1224 | |
865 | Strict mode and canonical mode are not implemented. |
1225 | Strict mode and canonical mode are not implemented. |
|
|
1226 | |
|
|
1227 | |
|
|
1228 | =head1 LIMITATIONS ON PERLS WITHOUT 64-BIT INTEGER SUPPORT |
|
|
1229 | |
|
|
1230 | On perls that were built without 64 bit integer support (these are rare |
|
|
1231 | nowadays, even on 32 bit architectures, as all major Perl distributions |
|
|
1232 | are built with 64 bit integer support), support for any kind of 64 bit |
|
|
1233 | integer in CBOR is very limited - most likely, these 64 bit values will |
|
|
1234 | be truncated, corrupted, or otherwise not decoded correctly. This also |
|
|
1235 | includes string, array and map sizes that are stored as 64 bit integers. |
866 | |
1236 | |
867 | |
1237 | |
868 | =head1 THREADS |
1238 | =head1 THREADS |
869 | |
1239 | |
870 | This module is I<not> guaranteed to be thread safe and there are no |
1240 | This module is I<not> guaranteed to be thread safe and there are no |
… | |
… | |
884 | Please refrain from using rt.cpan.org or any other bug reporting |
1254 | Please refrain from using rt.cpan.org or any other bug reporting |
885 | service. I put the contact address into my modules for a reason. |
1255 | service. I put the contact address into my modules for a reason. |
886 | |
1256 | |
887 | =cut |
1257 | =cut |
888 | |
1258 | |
|
|
1259 | # clumsy and slow hv_store-in-hash helper function |
|
|
1260 | sub _hv_store { |
|
|
1261 | $_[0]{$_[1]} = $_[2]; |
|
|
1262 | } |
|
|
1263 | |
889 | our %FILTER = ( |
1264 | our %FILTER = ( |
890 | # 0 # rfc4287 datetime, utf-8 |
1265 | 0 => sub { # rfc4287 datetime, utf-8 |
891 | # 1 # unix timestamp, any |
1266 | require Time::Piece; |
|
|
1267 | # Time::Piece::Strptime uses the "incredibly flexible date parsing routine" |
|
|
1268 | # from FreeBSD, which can't parse ISO 8601, RFC3339, RFC4287 or much of anything |
|
|
1269 | # else either. Whats incredibe over standard strptime totally escapes me. |
|
|
1270 | # doesn't do fractional times, either. sigh. |
|
|
1271 | # In fact, it's all a lie, it uses whatever strptime it wants, and of course, |
|
|
1272 | # they are all incompatible. The openbsd one simply ignores %z (but according to the |
|
|
1273 | # docs, it would be much more incredibly flexible indeed. If it worked, that is.). |
|
|
1274 | scalar eval { |
|
|
1275 | my $s = $_[1]; |
|
|
1276 | |
|
|
1277 | $s =~ s/Z$/+00:00/; |
|
|
1278 | $s =~ s/(\.[0-9]+)?([+-][0-9][0-9]):([0-9][0-9])$// |
|
|
1279 | or die; |
|
|
1280 | |
|
|
1281 | my $b = $1 - ($2 * 60 + $3) * 60; # fractional part + offset. hopefully |
|
|
1282 | my $d = Time::Piece->strptime ($s, "%Y-%m-%dT%H:%M:%S"); |
|
|
1283 | |
|
|
1284 | Time::Piece::gmtime ($d->epoch + $b) |
|
|
1285 | } || die "corrupted CBOR date/time string ($_[0])"; |
|
|
1286 | }, |
|
|
1287 | |
|
|
1288 | 1 => sub { # seconds since the epoch, possibly fractional |
|
|
1289 | require Time::Piece; |
|
|
1290 | scalar Time::Piece::gmtime (pop) |
|
|
1291 | }, |
892 | |
1292 | |
893 | 2 => sub { # pos bigint |
1293 | 2 => sub { # pos bigint |
894 | require Math::BigInt; |
1294 | require Math::BigInt; |
895 | Math::BigInt->new ("0x" . unpack "H*", pop) |
1295 | Math::BigInt->new ("0x" . unpack "H*", pop) |
896 | }, |
1296 | }, |
… | |
… | |
903 | 4 => sub { # decimal fraction, array |
1303 | 4 => sub { # decimal fraction, array |
904 | require Math::BigFloat; |
1304 | require Math::BigFloat; |
905 | Math::BigFloat->new ($_[1][1] . "E" . $_[1][0]) |
1305 | Math::BigFloat->new ($_[1][1] . "E" . $_[1][0]) |
906 | }, |
1306 | }, |
907 | |
1307 | |
|
|
1308 | 264 => sub { # decimal fraction with arbitrary exponent |
|
|
1309 | require Math::BigFloat; |
|
|
1310 | Math::BigFloat->new ($_[1][1] . "E" . $_[1][0]) |
|
|
1311 | }, |
|
|
1312 | |
908 | 5 => sub { # bigfloat, array |
1313 | 5 => sub { # bigfloat, array |
909 | require Math::BigFloat; |
1314 | require Math::BigFloat; |
910 | scalar Math::BigFloat->new ($_[1][1])->blsft ($_[1][0], 2) |
1315 | scalar Math::BigFloat->new ($_[1][1]) * Math::BigFloat->new (2)->bpow ($_[1][0]) |
|
|
1316 | }, |
|
|
1317 | |
|
|
1318 | 265 => sub { # bigfloat with arbitrary exponent |
|
|
1319 | require Math::BigFloat; |
|
|
1320 | scalar Math::BigFloat->new ($_[1][1]) * Math::BigFloat->new (2)->bpow ($_[1][0]) |
|
|
1321 | }, |
|
|
1322 | |
|
|
1323 | 30 => sub { # rational number |
|
|
1324 | require Math::BigRat; |
|
|
1325 | Math::BigRat->new ("$_[1][0]/$_[1][1]") # separate parameters only work in recent versons |
911 | }, |
1326 | }, |
912 | |
1327 | |
913 | 21 => sub { pop }, # expected conversion to base64url encoding |
1328 | 21 => sub { pop }, # expected conversion to base64url encoding |
914 | 22 => sub { pop }, # expected conversion to base64 encoding |
1329 | 22 => sub { pop }, # expected conversion to base64 encoding |
915 | 23 => sub { pop }, # expected conversion to base16 encoding |
1330 | 23 => sub { pop }, # expected conversion to base16 encoding |
… | |
… | |
925 | # 34 # base64 rfc46484, utf-8 |
1340 | # 34 # base64 rfc46484, utf-8 |
926 | # 35 # regex pcre/ecma262, utf-8 |
1341 | # 35 # regex pcre/ecma262, utf-8 |
927 | # 36 # mime message rfc2045, utf-8 |
1342 | # 36 # mime message rfc2045, utf-8 |
928 | ); |
1343 | ); |
929 | |
1344 | |
930 | sub CBOR::XS::default_filter { |
1345 | sub default_filter { |
931 | &{ $FILTER{$_[0]} or return } |
1346 | &{ $FILTER{$_[0]} or return } |
|
|
1347 | } |
|
|
1348 | |
|
|
1349 | our %SAFE_FILTER = map { $_ => $FILTER{$_} } 0, 1, 21, 22, 23, 32; |
|
|
1350 | |
|
|
1351 | sub safe_filter { |
|
|
1352 | &{ $SAFE_FILTER{$_[0]} or return } |
932 | } |
1353 | } |
933 | |
1354 | |
934 | sub URI::TO_CBOR { |
1355 | sub URI::TO_CBOR { |
935 | my $uri = $_[0]->as_string; |
1356 | my $uri = $_[0]->as_string; |
936 | utf8::upgrade $uri; |
1357 | utf8::upgrade $uri; |
937 | CBOR::XS::tag 32, $uri |
1358 | tag 32, $uri |
938 | } |
1359 | } |
939 | |
1360 | |
940 | sub Math::BigInt::TO_CBOR { |
1361 | sub Math::BigInt::TO_CBOR { |
941 | if ($_[0] >= -2147483648 && $_[0] <= 2147483647) { |
1362 | if (-2147483648 <= $_[0] && $_[0] <= 2147483647) { |
942 | $_[0]->numify |
1363 | $_[0]->numify |
943 | } else { |
1364 | } else { |
944 | my $hex = substr $_[0]->as_hex, 2; |
1365 | my $hex = substr $_[0]->as_hex, 2; |
945 | $hex = "0$hex" if 1 & length $hex; # sigh |
1366 | $hex = "0$hex" if 1 & length $hex; # sigh |
946 | CBOR::XS::tag $_[0] >= 0 ? 2 : 3, pack "H*", $hex |
1367 | tag $_[0] >= 0 ? 2 : 3, pack "H*", $hex |
947 | } |
1368 | } |
948 | } |
1369 | } |
949 | |
1370 | |
950 | sub Math::BigFloat::TO_CBOR { |
1371 | sub Math::BigFloat::TO_CBOR { |
951 | my ($m, $e) = $_[0]->parts; |
1372 | my ($m, $e) = $_[0]->parts; |
|
|
1373 | |
|
|
1374 | -9223372036854775808 <= $e && $e <= 18446744073709551615 |
952 | CBOR::XS::tag 4, [$e->numify, $m] |
1375 | ? tag 4, [$e->numify, $m] |
|
|
1376 | : tag 264, [$e, $m] |
|
|
1377 | } |
|
|
1378 | |
|
|
1379 | sub Math::BigRat::TO_CBOR { |
|
|
1380 | my ($n, $d) = $_[0]->parts; |
|
|
1381 | |
|
|
1382 | # older versions of BigRat need *1, as they not always return numbers |
|
|
1383 | |
|
|
1384 | $d*1 == 1 |
|
|
1385 | ? $n*1 |
|
|
1386 | : tag 30, [$n*1, $d*1] |
|
|
1387 | } |
|
|
1388 | |
|
|
1389 | sub Time::Piece::TO_CBOR { |
|
|
1390 | tag 1, 0 + $_[0]->epoch |
953 | } |
1391 | } |
954 | |
1392 | |
955 | XSLoader::load "CBOR::XS", $VERSION; |
1393 | XSLoader::load "CBOR::XS", $VERSION; |
956 | |
1394 | |
957 | =head1 SEE ALSO |
1395 | =head1 SEE ALSO |