… | |
… | |
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). |
50 | |
42 | |
51 | To give you a general idea about speed, with texts in the megabyte range, |
43 | 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 |
44 | 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 |
45 | L<JSON::XS> and decodes about 15%-30% faster than those. The shorter the |
54 | data, the worse L<Storable> performs in comparison. |
46 | data, the worse L<Storable> performs in comparison. |
55 | |
47 | |
56 | As for compactness, C<CBOR::XS> encoded data structures are usually about |
48 | Regarding compactness, C<CBOR::XS>-encoded data structures are usually |
57 | 20% smaller than the same data encoded as (compact) JSON or L<Storable>. |
49 | about 20% smaller than the same data encoded as (compact) JSON or |
|
|
50 | L<Storable>. |
58 | |
51 | |
59 | In addition to the core CBOR data format, this module implements a number |
52 | In addition to the core CBOR data format, this module implements a |
60 | of extensions, to support cyclic and self-referencing data structures |
53 | number of extensions, to support cyclic and shared data structures |
61 | (see C<allow_sharing>), string deduplication (see C<pack_strings>) and |
54 | (see C<allow_sharing> and C<allow_cycles>), string deduplication (see |
62 | scalar references (always enabled). |
55 | C<pack_strings>) and scalar references (always enabled). |
63 | |
56 | |
64 | The primary goal of this module is to be I<correct> and the secondary goal |
57 | 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. |
58 | 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 |
… | |
… | |
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 = 1.12; |
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; |
… | |
… | |
187 | reference to the earlier value. |
180 | reference to the earlier value. |
188 | |
181 | |
189 | This means that such values will only be encoded once, and will not result |
182 | 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 |
183 | in a deep cloning of the value on decode, in decoders supporting the value |
191 | sharing extension. This also makes it possible to encode cyclic data |
184 | sharing extension. This also makes it possible to encode cyclic data |
192 | structures. |
185 | structures (which need C<allow_cycles> to ne enabled to be decoded by this |
|
|
186 | module). |
193 | |
187 | |
194 | It is recommended to leave it off unless you know your |
188 | It is recommended to leave it off unless you know your |
195 | communication partner supports the value sharing extensions to CBOR |
189 | communication partner supports the value sharing extensions to CBOR |
196 | (L<http://cbor.schmorp.de/value-sharing>), as without decoder support, the |
190 | (L<http://cbor.schmorp.de/value-sharing>), as without decoder support, the |
197 | resulting data structure might be unusable. |
191 | resulting data structure might be unusable. |
198 | |
192 | |
199 | Detecting shared values incurs a runtime overhead when values are encoded |
193 | Detecting shared values incurs a runtime overhead when values are encoded |
200 | that have a reference counter large than one, and might unnecessarily |
194 | that have a reference counter large than one, and might unnecessarily |
201 | increase the encoded size, as potentially shared values are encode as |
195 | increase the encoded size, as potentially shared values are encode as |
202 | sharable whether or not they are actually shared. |
196 | shareable whether or not they are actually shared. |
203 | |
197 | |
204 | At the moment, only targets of references can be shared (e.g. scalars, |
198 | 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 |
199 | 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 |
200 | 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 |
201 | not impossible to create in Perl, are not supported (this is the same as |
… | |
… | |
212 | structures cannot be encoded in this mode. |
206 | structures cannot be encoded in this mode. |
213 | |
207 | |
214 | This option does not affect C<decode> in any way - shared values and |
208 | This option does not affect C<decode> in any way - shared values and |
215 | references will always be decoded properly if present. |
209 | references will always be decoded properly if present. |
216 | |
210 | |
|
|
211 | =item $cbor = $cbor->allow_cycles ([$enable]) |
|
|
212 | |
|
|
213 | =item $enabled = $cbor->get_allow_cycles |
|
|
214 | |
|
|
215 | If C<$enable> is true (or missing), then C<decode> will happily decode |
|
|
216 | self-referential (cyclic) data structures. By default these will not be |
|
|
217 | decoded, as they need manual cleanup to avoid memory leaks, so code that |
|
|
218 | isn't prepared for this will not leak memory. |
|
|
219 | |
|
|
220 | If C<$enable> is false (the default), then C<decode> will throw an error |
|
|
221 | when it encounters a self-referential/cyclic data structure. |
|
|
222 | |
|
|
223 | This option does not affect C<encode> in any way - shared values and |
|
|
224 | references will always be decoded properly if present. |
|
|
225 | |
217 | =item $cbor = $cbor->pack_strings ([$enable]) |
226 | =item $cbor = $cbor->pack_strings ([$enable]) |
218 | |
227 | |
219 | =item $enabled = $cbor->get_pack_strings |
228 | =item $enabled = $cbor->get_pack_strings |
220 | |
229 | |
221 | If C<$enable> is true (or missing), then C<encode> will try not to encode |
230 | If C<$enable> is true (or missing), then C<encode> will try not to encode |
… | |
… | |
233 | the standard CBOR way. |
242 | the standard CBOR way. |
234 | |
243 | |
235 | This option does not affect C<decode> in any way - string references will |
244 | This option does not affect C<decode> in any way - string references will |
236 | always be decoded properly if present. |
245 | always be decoded properly if present. |
237 | |
246 | |
|
|
247 | =item $cbor = $cbor->validate_utf8 ([$enable]) |
|
|
248 | |
|
|
249 | =item $enabled = $cbor->get_validate_utf8 |
|
|
250 | |
|
|
251 | If C<$enable> is true (or missing), then C<decode> will validate that |
|
|
252 | elements (text strings) containing UTF-8 data in fact contain valid UTF-8 |
|
|
253 | data (instead of blindly accepting it). This validation obviously takes |
|
|
254 | extra time during decoding. |
|
|
255 | |
|
|
256 | The concept of "valid UTF-8" used is perl's concept, which is a superset |
|
|
257 | of the official UTF-8. |
|
|
258 | |
|
|
259 | If C<$enable> is false (the default), then C<decode> will blindly accept |
|
|
260 | UTF-8 data, marking them as valid UTF-8 in the resulting data structure |
|
|
261 | regardless of whether thats true or not. |
|
|
262 | |
|
|
263 | Perl isn't too happy about corrupted UTF-8 in strings, but should |
|
|
264 | generally not crash or do similarly evil things. Extensions might be not |
|
|
265 | so forgiving, so it's recommended to turn on this setting if you receive |
|
|
266 | untrusted CBOR. |
|
|
267 | |
|
|
268 | This option does not affect C<encode> in any way - strings that are |
|
|
269 | supposedly valid UTF-8 will simply be dumped into the resulting CBOR |
|
|
270 | string without checking whether that is, in fact, true or not. |
|
|
271 | |
238 | =item $cbor = $cbor->filter ([$cb->($tag, $value)]) |
272 | =item $cbor = $cbor->filter ([$cb->($tag, $value)]) |
239 | |
273 | |
240 | =item $cb_or_undef = $cbor->get_filter |
274 | =item $cb_or_undef = $cbor->get_filter |
241 | |
275 | |
242 | Sets or replaces the tagged value decoding filter (when C<$cb> is |
276 | Sets or replaces the tagged value decoding filter (when C<$cb> is |
… | |
… | |
259 | function, C<CBOR::XS::default_filter>, is used. This function simply looks |
293 | function, C<CBOR::XS::default_filter>, is used. This function simply looks |
260 | up the tag in the C<%CBOR::XS::FILTER> hash. If an entry exists it must be |
294 | up the tag in the C<%CBOR::XS::FILTER> hash. If an entry exists it must be |
261 | a code reference that is called with tag and value, and is responsible for |
295 | a code reference that is called with tag and value, and is responsible for |
262 | decoding the value. If no entry exists, it returns no values. |
296 | decoding the value. If no entry exists, it returns no values. |
263 | |
297 | |
264 | Example: decode all tags not handled internally into CBOR::XS::Tagged |
298 | Example: decode all tags not handled internally into C<CBOR::XS::Tagged> |
265 | objects, with no other special handling (useful when working with |
299 | objects, with no other special handling (useful when working with |
266 | potentially "unsafe" CBOR data). |
300 | potentially "unsafe" CBOR data). |
267 | |
301 | |
268 | CBOR::XS->new->filter (sub { })->decode ($cbor_data); |
302 | CBOR::XS->new->filter (sub { })->decode ($cbor_data); |
269 | |
303 | |
… | |
… | |
296 | and you need to know where the first CBOR string ends amd the next one |
330 | and you need to know where the first CBOR string ends amd the next one |
297 | starts. |
331 | starts. |
298 | |
332 | |
299 | CBOR::XS->new->decode_prefix ("......") |
333 | CBOR::XS->new->decode_prefix ("......") |
300 | => ("...", 3) |
334 | => ("...", 3) |
|
|
335 | |
|
|
336 | =back |
|
|
337 | |
|
|
338 | =head2 INCREMENTAL PARSING |
|
|
339 | |
|
|
340 | In some cases, there is the need for incremental parsing of JSON |
|
|
341 | texts. While this module always has to keep both CBOR text and resulting |
|
|
342 | Perl data structure in memory at one time, it does allow you to parse a |
|
|
343 | CBOR stream incrementally, using a similar to using "decode_prefix" to see |
|
|
344 | if a full CBOR object is available, but is much more efficient. |
|
|
345 | |
|
|
346 | It basically works by parsing as much of a CBOR string as possible - if |
|
|
347 | the CBOR data is not complete yet, the pasrer will remember where it was, |
|
|
348 | to be able to restart when more data has been accumulated. Once enough |
|
|
349 | data is available to either decode a complete CBOR value or raise an |
|
|
350 | error, a real decode will be attempted. |
|
|
351 | |
|
|
352 | A typical use case would be a network protocol that consists of sending |
|
|
353 | and receiving CBOR-encoded messages. The solution that works with CBOR and |
|
|
354 | about anything else is by prepending a length to every CBOR value, so the |
|
|
355 | receiver knows how many octets to read. More compact (and slightly slower) |
|
|
356 | would be to just send CBOR values back-to-back, as C<CBOR::XS> knows where |
|
|
357 | a CBOR value ends, and doesn't need an explicit length. |
|
|
358 | |
|
|
359 | The following methods help with this: |
|
|
360 | |
|
|
361 | =over 4 |
|
|
362 | |
|
|
363 | =item @decoded = $cbor->incr_parse ($buffer) |
|
|
364 | |
|
|
365 | This method attempts to decode exactly one CBOR value from the beginning |
|
|
366 | of the given C<$buffer>. The value is removed from the C<$buffer> on |
|
|
367 | success. When C<$buffer> doesn't contain a complete value yet, it returns |
|
|
368 | nothing. Finally, when the C<$buffer> doesn't start with something |
|
|
369 | that could ever be a valid CBOR value, it raises an exception, just as |
|
|
370 | C<decode> would. In the latter case the decoder state is undefined and |
|
|
371 | must be reset before being able to parse further. |
|
|
372 | |
|
|
373 | This method modifies the C<$buffer> in place. When no CBOR value can be |
|
|
374 | decoded, the decoder stores the current string offset. On the next call, |
|
|
375 | continues decoding at the place where it stopped before. For this to make |
|
|
376 | sense, the C<$buffer> must begin with the same octets as on previous |
|
|
377 | unsuccessful calls. |
|
|
378 | |
|
|
379 | You can call this method in scalar context, in which case it either |
|
|
380 | returns a decoded value or C<undef>. This makes it impossible to |
|
|
381 | distinguish between CBOR null values (which decode to C<undef>) and an |
|
|
382 | unsuccessful decode, which is often acceptable. |
|
|
383 | |
|
|
384 | =item @decoded = $cbor->incr_parse_multiple ($buffer) |
|
|
385 | |
|
|
386 | Same as C<incr_parse>, but attempts to decode as many CBOR values as |
|
|
387 | possible in one go, instead of at most one. Calls to C<incr_parse> and |
|
|
388 | C<incr_parse_multiple> can be interleaved. |
|
|
389 | |
|
|
390 | =item $cbor->incr_reset |
|
|
391 | |
|
|
392 | Resets the incremental decoder. This throws away any saved state, so that |
|
|
393 | subsequent calls to C<incr_parse> or C<incr_parse_multiple> start to parse |
|
|
394 | a new CBOR value from the beginning of the C<$buffer> again. |
|
|
395 | |
|
|
396 | This method can be caled at any time, but it I<must> be called if you want |
|
|
397 | to change your C<$buffer> or there was a decoding error and you want to |
|
|
398 | reuse the C<$cbor> object for future incremental parsings. |
301 | |
399 | |
302 | =back |
400 | =back |
303 | |
401 | |
304 | |
402 | |
305 | =head1 MAPPING |
403 | =head1 MAPPING |
… | |
… | |
356 | =item tagged values |
454 | =item tagged values |
357 | |
455 | |
358 | Tagged items consists of a numeric tag and another CBOR value. |
456 | Tagged items consists of a numeric tag and another CBOR value. |
359 | |
457 | |
360 | See L<TAG HANDLING AND EXTENSIONS> and the description of C<< ->filter >> |
458 | See L<TAG HANDLING AND EXTENSIONS> and the description of C<< ->filter >> |
361 | for details. |
459 | for details on which tags are handled how. |
362 | |
460 | |
363 | =item anything else |
461 | =item anything else |
364 | |
462 | |
365 | Anything else (e.g. unsupported simple values) will raise a decoding |
463 | Anything else (e.g. unsupported simple values) will raise a decoding |
366 | error. |
464 | error. |
… | |
… | |
369 | |
467 | |
370 | |
468 | |
371 | =head2 PERL -> CBOR |
469 | =head2 PERL -> CBOR |
372 | |
470 | |
373 | The mapping from Perl to CBOR is slightly more difficult, as Perl is a |
471 | 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 |
472 | typeless language. That means this module can only guess which CBOR type |
375 | a Perl value. |
473 | is meant by a perl value. |
376 | |
474 | |
377 | =over 4 |
475 | =over 4 |
378 | |
476 | |
379 | =item hash references |
477 | =item hash references |
380 | |
478 | |
381 | Perl hash references become CBOR maps. As there is no inherent ordering in |
479 | 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 |
480 | hash keys (or CBOR maps), they will usually be encoded in a pseudo-random |
383 | order. |
481 | order. This order can be different each time a hahs is encoded. |
384 | |
482 | |
385 | Currently, tied hashes will use the indefinite-length format, while normal |
483 | Currently, tied hashes will use the indefinite-length format, while normal |
386 | hashes will use the fixed-length format. |
484 | hashes will use the fixed-length format. |
387 | |
485 | |
388 | =item array references |
486 | =item array references |
389 | |
487 | |
390 | Perl array references become fixed-length CBOR arrays. |
488 | Perl array references become fixed-length CBOR arrays. |
391 | |
489 | |
392 | =item other references |
490 | =item other references |
393 | |
491 | |
394 | Other unblessed references are generally not allowed and will cause an |
492 | Other unblessed references will be represented using |
395 | exception to be thrown, except for references to the integers C<0> and |
493 | the indirection tag extension (tag value C<22098>, |
396 | C<1>, which get turned into false and true in CBOR. |
494 | L<http://cbor.schmorp.de/indirection>). CBOR decoders are guaranteed |
|
|
495 | to be able to decode these values somehow, by either "doing the right |
|
|
496 | thing", decoding into a generic tagged object, simply ignoring the tag, or |
|
|
497 | something else. |
397 | |
498 | |
398 | =item CBOR::XS::Tagged objects |
499 | =item CBOR::XS::Tagged objects |
399 | |
500 | |
400 | Objects of this type must be arrays consisting of a single C<[tag, value]> |
501 | Objects of this type must be arrays consisting of a single C<[tag, value]> |
401 | pair. The (numerical) tag will be encoded as a CBOR tag, the value will |
502 | 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 |
503 | be encoded as appropriate for the value. You must use C<CBOR::XS::tag> to |
403 | create such objects. |
504 | create such objects. |
404 | |
505 | |
405 | =item Types::Serialiser::true, Types::Serialiser::false, Types::Serialiser::error |
506 | =item Types::Serialiser::true, Types::Serialiser::false, Types::Serialiser::error |
406 | |
507 | |
407 | These special values become CBOR true, CBOR false and CBOR undefined |
508 | These special values become CBOR true, CBOR false and CBOR undefined |
… | |
… | |
469 | |
570 | |
470 | =back |
571 | =back |
471 | |
572 | |
472 | =head2 OBJECT SERIALISATION |
573 | =head2 OBJECT SERIALISATION |
473 | |
574 | |
|
|
575 | This module implements both a CBOR-specific and the generic |
|
|
576 | L<Types::Serialier> object serialisation protocol. The following |
|
|
577 | subsections explain both methods. |
|
|
578 | |
|
|
579 | =head3 ENCODING |
|
|
580 | |
474 | This module knows two way to serialise a Perl object: The CBOR-specific |
581 | This module knows two way to serialise a Perl object: The CBOR-specific |
475 | way, and the generic way. |
582 | way, and the generic way. |
476 | |
583 | |
477 | Whenever the encoder encounters a Perl object that it cnanot serialise |
584 | Whenever the encoder encounters a Perl object that it cannot serialise |
478 | directly (most of them), it will first look up the C<TO_CBOR> method on |
585 | directly (most of them), it will first look up the C<TO_CBOR> method on |
479 | it. |
586 | it. |
480 | |
587 | |
481 | If it has a C<TO_CBOR> method, it will call it with the object as only |
588 | If it has a C<TO_CBOR> method, it will call it with the object as only |
482 | argument, and expects exactly one return value, which it will then |
589 | argument, and expects exactly one return value, which it will then |
… | |
… | |
488 | |
595 | |
489 | The C<FREEZE> method can return any number of values (i.e. zero or |
596 | The C<FREEZE> method can return any number of values (i.e. zero or |
490 | more). These will be encoded as CBOR perl object, together with the |
597 | more). These will be encoded as CBOR perl object, together with the |
491 | classname. |
598 | classname. |
492 | |
599 | |
|
|
600 | These methods I<MUST NOT> change the data structure that is being |
|
|
601 | serialised. Failure to comply to this can result in memory corruption - |
|
|
602 | and worse. |
|
|
603 | |
493 | If an object supports neither C<TO_CBOR> nor C<FREEZE>, encoding will fail |
604 | If an object supports neither C<TO_CBOR> nor C<FREEZE>, encoding will fail |
494 | with an error. |
605 | with an error. |
495 | |
606 | |
|
|
607 | =head3 DECODING |
|
|
608 | |
496 | Objects encoded via C<TO_CBOR> cannot be automatically decoded, but |
609 | Objects encoded via C<TO_CBOR> cannot (normally) be automatically decoded, |
497 | objects encoded via C<FREEZE> can be decoded using the following protocol: |
610 | but objects encoded via C<FREEZE> can be decoded using the following |
|
|
611 | protocol: |
498 | |
612 | |
499 | When an encoded CBOR perl object is encountered by the decoder, it will |
613 | When an encoded CBOR perl object is encountered by the decoder, it will |
500 | look up the C<THAW> method, by using the stored classname, and will fail |
614 | look up the C<THAW> method, by using the stored classname, and will fail |
501 | if the method cannot be found. |
615 | if the method cannot be found. |
502 | |
616 | |
503 | After the lookup it will call the C<THAW> method with the stored classname |
617 | After the lookup it will call the C<THAW> method with the stored classname |
504 | as first argument, the constant string C<CBOR> as second argument, and all |
618 | as first argument, the constant string C<CBOR> as second argument, and all |
505 | values returned by C<FREEZE> as remaining arguments. |
619 | values returned by C<FREEZE> as remaining arguments. |
506 | |
620 | |
507 | =head4 EXAMPLES |
621 | =head3 EXAMPLES |
508 | |
622 | |
509 | Here is an example C<TO_CBOR> method: |
623 | Here is an example C<TO_CBOR> method: |
510 | |
624 | |
511 | sub My::Object::TO_CBOR { |
625 | sub My::Object::TO_CBOR { |
512 | my ($obj) = @_; |
626 | my ($obj) = @_; |
… | |
… | |
523 | |
637 | |
524 | sub URI::TO_CBOR { |
638 | sub URI::TO_CBOR { |
525 | my ($self) = @_; |
639 | my ($self) = @_; |
526 | my $uri = "$self"; # stringify uri |
640 | my $uri = "$self"; # stringify uri |
527 | utf8::upgrade $uri; # make sure it will be encoded as UTF-8 string |
641 | utf8::upgrade $uri; # make sure it will be encoded as UTF-8 string |
528 | CBOR::XS::tagged 32, "$_[0]" |
642 | CBOR::XS::tag 32, "$_[0]" |
529 | } |
643 | } |
530 | |
644 | |
531 | This will encode URIs as a UTF-8 string with tag 32, which indicates an |
645 | This will encode URIs as a UTF-8 string with tag 32, which indicates an |
532 | URI. |
646 | URI. |
533 | |
647 | |
… | |
… | |
695 | |
809 | |
696 | These tags are automatically created (and decoded) for serialisable |
810 | These tags are automatically created (and decoded) for serialisable |
697 | objects using the C<FREEZE/THAW> methods (the L<Types::Serialier> object |
811 | objects using the C<FREEZE/THAW> methods (the L<Types::Serialier> object |
698 | serialisation protocol). See L<OBJECT SERIALISATION> for details. |
812 | serialisation protocol). See L<OBJECT SERIALISATION> for details. |
699 | |
813 | |
700 | =item 28, 29 (sharable, sharedref, L <http://cbor.schmorp.de/value-sharing>) |
814 | =item 28, 29 (shareable, sharedref, L <http://cbor.schmorp.de/value-sharing>) |
701 | |
815 | |
702 | These tags are automatically decoded when encountered, resulting in |
816 | These tags are automatically decoded when encountered (and they do not |
|
|
817 | result in a cyclic data structure, see C<allow_cycles>), resulting in |
703 | shared values in the decoded object. They are only encoded, however, when |
818 | shared values in the decoded object. They are only encoded, however, when |
704 | C<allow_sharable> is enabled. |
819 | C<allow_sharing> is enabled. |
|
|
820 | |
|
|
821 | Not all shared values can be successfully decoded: values that reference |
|
|
822 | themselves will I<currently> decode as C<undef> (this is not the same |
|
|
823 | as a reference pointing to itself, which will be represented as a value |
|
|
824 | that contains an indirect reference to itself - these will be decoded |
|
|
825 | properly). |
|
|
826 | |
|
|
827 | Note that considerably more shared value data structures can be decoded |
|
|
828 | than will be encoded - currently, only values pointed to by references |
|
|
829 | will be shared, others will not. While non-reference shared values can be |
|
|
830 | generated in Perl with some effort, they were considered too unimportant |
|
|
831 | to be supported in the encoder. The decoder, however, will decode these |
|
|
832 | values as shared values. |
705 | |
833 | |
706 | =item 256, 25 (stringref-namespace, stringref, L <http://cbor.schmorp.de/stringref>) |
834 | =item 256, 25 (stringref-namespace, stringref, L <http://cbor.schmorp.de/stringref>) |
707 | |
835 | |
708 | These tags are automatically decoded when encountered. They are only |
836 | These tags are automatically decoded when encountered. They are only |
709 | encoded, however, when C<pack_strings> is enabled. |
837 | encoded, however, when C<pack_strings> is enabled. |
… | |
… | |
734 | perl core distribution (e.g. L<URI>), it is (currently) up to the user to |
862 | perl core distribution (e.g. L<URI>), it is (currently) up to the user to |
735 | provide these modules. The decoding usually fails with an exception if the |
863 | provide these modules. The decoding usually fails with an exception if the |
736 | required module cannot be loaded. |
864 | required module cannot be loaded. |
737 | |
865 | |
738 | =over 4 |
866 | =over 4 |
|
|
867 | |
|
|
868 | =item 0, 1 (date/time string, seconds since the epoch) |
|
|
869 | |
|
|
870 | These tags are decoded into L<Time::Piece> objects. The corresponding |
|
|
871 | C<Time::Piece::TO_CBOR> method always encodes into tag 1 values currently. |
|
|
872 | |
|
|
873 | The L<Time::Piece> API is generally surprisingly bad, and fractional |
|
|
874 | seconds are only accidentally kept intact, so watch out. On the plus side, |
|
|
875 | the module comes with perl since 5.10, which has to count for something. |
739 | |
876 | |
740 | =item 2, 3 (positive/negative bignum) |
877 | =item 2, 3 (positive/negative bignum) |
741 | |
878 | |
742 | These tags are decoded into L<Math::BigInt> objects. The corresponding |
879 | These tags are decoded into L<Math::BigInt> objects. The corresponding |
743 | C<Math::BigInt::TO_CBOR> method encodes "small" bigints into normal CBOR |
880 | C<Math::BigInt::TO_CBOR> method encodes "small" bigints into normal CBOR |
… | |
… | |
877 | properly. Half precision types are accepted, but not encoded. |
1014 | properly. Half precision types are accepted, but not encoded. |
878 | |
1015 | |
879 | Strict mode and canonical mode are not implemented. |
1016 | Strict mode and canonical mode are not implemented. |
880 | |
1017 | |
881 | |
1018 | |
|
|
1019 | =head1 LIMITATIONS ON PERLS WITHOUT 64-BIT INTEGER SUPPORT |
|
|
1020 | |
|
|
1021 | On perls that were built without 64 bit integer support (these are rare |
|
|
1022 | nowadays, even on 32 bit architectures), support for any kind of 64 bit |
|
|
1023 | integer in CBOR is very limited - most likely, these 64 bit values will |
|
|
1024 | be truncated, corrupted, or otherwise not decoded correctly. This also |
|
|
1025 | includes string, array and map sizes that are stored as 64 bit integers. |
|
|
1026 | |
|
|
1027 | |
882 | =head1 THREADS |
1028 | =head1 THREADS |
883 | |
1029 | |
884 | This module is I<not> guaranteed to be thread safe and there are no |
1030 | This module is I<not> guaranteed to be thread safe and there are no |
885 | plans to change this until Perl gets thread support (as opposed to the |
1031 | plans to change this until Perl gets thread support (as opposed to the |
886 | horribly slow so-called "threads" which are simply slow and bloated |
1032 | horribly slow so-called "threads" which are simply slow and bloated |
… | |
… | |
899 | service. I put the contact address into my modules for a reason. |
1045 | service. I put the contact address into my modules for a reason. |
900 | |
1046 | |
901 | =cut |
1047 | =cut |
902 | |
1048 | |
903 | our %FILTER = ( |
1049 | our %FILTER = ( |
904 | # 0 # rfc4287 datetime, utf-8 |
1050 | 0 => sub { # rfc4287 datetime, utf-8 |
905 | # 1 # unix timestamp, any |
1051 | require Time::Piece; |
|
|
1052 | # Time::Piece::Strptime uses the "incredibly flexible date parsing routine" |
|
|
1053 | # from FreeBSD, which can't parse ISO 8601, RFC3339, RFC4287 or much of anything |
|
|
1054 | # else either. Whats incredibe over standard strptime totally escapes me. |
|
|
1055 | # doesn't do fractional times, either. sigh. |
|
|
1056 | # In fact, it's all a lie, it uses whatever strptime it wants, and of course, |
|
|
1057 | # they are all incomptible. The openbsd one simply ignores %z (but according to the |
|
|
1058 | # docs, it would be much more incredibly flexible indeed. If it worked, that is.). |
|
|
1059 | scalar eval { |
|
|
1060 | my $s = $_[1]; |
|
|
1061 | |
|
|
1062 | $s =~ s/Z$/+00:00/; |
|
|
1063 | $s =~ s/(\.[0-9]+)?([+-][0-9][0-9]):([0-9][0-9])$// |
|
|
1064 | or die; |
|
|
1065 | |
|
|
1066 | my $b = $1 - ($2 * 60 + $3) * 60; # fractional part + offset. hopefully |
|
|
1067 | my $d = Time::Piece->strptime ($s, "%Y-%m-%dT%H:%M:%S"); |
|
|
1068 | |
|
|
1069 | Time::Piece::gmtime ($d->epoch + $b) |
|
|
1070 | } || die "corrupted CBOR date/time string ($_[0])"; |
|
|
1071 | }, |
|
|
1072 | |
|
|
1073 | 1 => sub { # seconds since the epoch, possibly fractional |
|
|
1074 | require Time::Piece; |
|
|
1075 | scalar Time::Piece::gmtime (pop) |
|
|
1076 | }, |
906 | |
1077 | |
907 | 2 => sub { # pos bigint |
1078 | 2 => sub { # pos bigint |
908 | require Math::BigInt; |
1079 | require Math::BigInt; |
909 | Math::BigInt->new ("0x" . unpack "H*", pop) |
1080 | Math::BigInt->new ("0x" . unpack "H*", pop) |
910 | }, |
1081 | }, |
… | |
… | |
946 | } |
1117 | } |
947 | |
1118 | |
948 | sub URI::TO_CBOR { |
1119 | sub URI::TO_CBOR { |
949 | my $uri = $_[0]->as_string; |
1120 | my $uri = $_[0]->as_string; |
950 | utf8::upgrade $uri; |
1121 | utf8::upgrade $uri; |
951 | CBOR::XS::tag 32, $uri |
1122 | tag 32, $uri |
952 | } |
1123 | } |
953 | |
1124 | |
954 | sub Math::BigInt::TO_CBOR { |
1125 | sub Math::BigInt::TO_CBOR { |
955 | if ($_[0] >= -2147483648 && $_[0] <= 2147483647) { |
1126 | if ($_[0] >= -2147483648 && $_[0] <= 2147483647) { |
956 | $_[0]->numify |
1127 | $_[0]->numify |
957 | } else { |
1128 | } else { |
958 | my $hex = substr $_[0]->as_hex, 2; |
1129 | my $hex = substr $_[0]->as_hex, 2; |
959 | $hex = "0$hex" if 1 & length $hex; # sigh |
1130 | $hex = "0$hex" if 1 & length $hex; # sigh |
960 | CBOR::XS::tag $_[0] >= 0 ? 2 : 3, pack "H*", $hex |
1131 | tag $_[0] >= 0 ? 2 : 3, pack "H*", $hex |
961 | } |
1132 | } |
962 | } |
1133 | } |
963 | |
1134 | |
964 | sub Math::BigFloat::TO_CBOR { |
1135 | sub Math::BigFloat::TO_CBOR { |
965 | my ($m, $e) = $_[0]->parts; |
1136 | my ($m, $e) = $_[0]->parts; |
966 | CBOR::XS::tag 4, [$e->numify, $m] |
1137 | tag 4, [$e->numify, $m] |
|
|
1138 | } |
|
|
1139 | |
|
|
1140 | sub Time::Piece::TO_CBOR { |
|
|
1141 | tag 1, $_[0]->epoch |
967 | } |
1142 | } |
968 | |
1143 | |
969 | XSLoader::load "CBOR::XS", $VERSION; |
1144 | XSLoader::load "CBOR::XS", $VERSION; |
970 | |
1145 | |
971 | =head1 SEE ALSO |
1146 | =head1 SEE ALSO |