… | |
… | |
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<allow_stringref>) 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 = 0.08; |
69 | our $VERSION = 1.41; |
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; |
… | |
… | |
186 | as an array, is referenced multiple times), but instead will emit a |
179 | as an array, is referenced multiple times), but instead will emit a |
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. |
184 | sharing extension. This also makes it possible to encode cyclic data |
|
|
185 | structures (which need C<allow_cycles> to ne enabled to be decoded by this |
|
|
186 | module). |
192 | |
187 | |
193 | It is recommended to leave it off unless you know your |
188 | It is recommended to leave it off unless you know your |
194 | communication partner supports the value sharing extensions to CBOR |
189 | communication partner supports the value sharing extensions to CBOR |
195 | (http://cbor.schmorp.de/value-sharing). |
190 | (L<http://cbor.schmorp.de/value-sharing>), as without decoder support, the |
|
|
191 | resulting data structure might be unusable. |
196 | |
192 | |
197 | Detecting shared values incurs a runtime overhead when values are encoded |
193 | Detecting shared values incurs a runtime overhead when values are encoded |
198 | that have a reference counter large than one, and might unnecessarily |
194 | that have a reference counter large than one, and might unnecessarily |
199 | increase the encoded size, as potentially shared values are encode as |
195 | increase the encoded size, as potentially shared values are encode as |
200 | sharable whether or not they are actually shared. |
196 | shareable whether or not they are actually shared. |
201 | |
197 | |
202 | 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, |
203 | 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 |
204 | 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 |
205 | 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 |
206 | for L<Storable>). |
202 | with L<Storable>). |
207 | |
203 | |
208 | If C<$enable> is false (the default), then C<encode> will encode |
204 | If C<$enable> is false (the default), then C<encode> will encode shared |
209 | exception when it encounters anything it cannot encode as CBOR. |
205 | data structures repeatedly, unsharing them in the process. Cyclic data |
|
|
206 | structures cannot be encoded in this mode. |
210 | |
207 | |
211 | 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 |
212 | references will always be decoded properly if present. |
209 | references will always be decoded properly if present. |
213 | |
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 | FUTURE DIRECTION: the motivation behind this option is to avoid I<real> |
|
|
224 | cycles - future versions of this module might chose to decode cyclic data |
|
|
225 | structures using weak references when this option is off, instead of |
|
|
226 | throwing an error. |
|
|
227 | |
|
|
228 | This option does not affect C<encode> in any way - shared values and |
|
|
229 | references will always be encoded properly if present. |
|
|
230 | |
214 | =item $cbor = $cbor->allow_stringref ([$enable]) |
231 | =item $cbor = $cbor->pack_strings ([$enable]) |
215 | |
232 | |
216 | =item $enabled = $cbor->get_allow_stringref |
233 | =item $enabled = $cbor->get_pack_strings |
217 | |
234 | |
218 | If C<$enable> is true (or missing), then C<encode> will try not to encode |
235 | If C<$enable> is true (or missing), then C<encode> will try not to encode |
219 | the same string twice, but will instead encode a reference to the string |
236 | the same string twice, but will instead encode a reference to the string |
220 | instead. Depending on your data format. this can save a lot of space, but |
237 | instead. Depending on your data format, this can save a lot of space, but |
221 | also results in a very large runtime overhead (expect encoding times to be |
238 | also results in a very large runtime overhead (expect encoding times to be |
222 | 2-4 times as high as without). |
239 | 2-4 times as high as without). |
223 | |
240 | |
224 | It is recommended to leave it off unless you know your |
241 | It is recommended to leave it off unless you know your |
225 | communications partner supports the stringref extension to CBOR |
242 | communications partner supports the stringref extension to CBOR |
226 | (http://cbor.schmorp.de/stringref). |
243 | (L<http://cbor.schmorp.de/stringref>), as without decoder support, the |
|
|
244 | resulting data structure might not be usable. |
227 | |
245 | |
228 | If C<$enable> is false (the default), then C<encode> will encode |
246 | If C<$enable> is false (the default), then C<encode> will encode strings |
229 | exception when it encounters anything it cannot encode as CBOR. |
247 | the standard CBOR way. |
230 | |
248 | |
231 | This option does not affect C<decode> in any way - string references will |
249 | This option does not affect C<decode> in any way - string references will |
232 | always be decoded properly if present. |
250 | always be decoded properly if present. |
233 | |
251 | |
|
|
252 | =item $cbor = $cbor->validate_utf8 ([$enable]) |
|
|
253 | |
|
|
254 | =item $enabled = $cbor->get_validate_utf8 |
|
|
255 | |
|
|
256 | If C<$enable> is true (or missing), then C<decode> will validate that |
|
|
257 | elements (text strings) containing UTF-8 data in fact contain valid UTF-8 |
|
|
258 | data (instead of blindly accepting it). This validation obviously takes |
|
|
259 | extra time during decoding. |
|
|
260 | |
|
|
261 | The concept of "valid UTF-8" used is perl's concept, which is a superset |
|
|
262 | of the official UTF-8. |
|
|
263 | |
|
|
264 | If C<$enable> is false (the default), then C<decode> will blindly accept |
|
|
265 | UTF-8 data, marking them as valid UTF-8 in the resulting data structure |
|
|
266 | regardless of whether thats true or not. |
|
|
267 | |
|
|
268 | Perl isn't too happy about corrupted UTF-8 in strings, but should |
|
|
269 | generally not crash or do similarly evil things. Extensions might be not |
|
|
270 | so forgiving, so it's recommended to turn on this setting if you receive |
|
|
271 | untrusted CBOR. |
|
|
272 | |
|
|
273 | This option does not affect C<encode> in any way - strings that are |
|
|
274 | supposedly valid UTF-8 will simply be dumped into the resulting CBOR |
|
|
275 | string without checking whether that is, in fact, true or not. |
|
|
276 | |
234 | =item $cbor = $cbor->filter ([$cb->($tag, $value)]) |
277 | =item $cbor = $cbor->filter ([$cb->($tag, $value)]) |
235 | |
278 | |
236 | =item $cb_or_undef = $cbor->get_filter |
279 | =item $cb_or_undef = $cbor->get_filter |
237 | |
280 | |
238 | TODO |
281 | Sets or replaces the tagged value decoding filter (when C<$cb> is |
|
|
282 | specified) or clears the filter (if no argument or C<undef> is provided). |
|
|
283 | |
|
|
284 | The filter callback is called only during decoding, when a non-enforced |
|
|
285 | tagged value has been decoded (see L<TAG HANDLING AND EXTENSIONS> for a |
|
|
286 | list of enforced tags). For specific tags, it's often better to provide a |
|
|
287 | default converter using the C<%CBOR::XS::FILTER> hash (see below). |
|
|
288 | |
|
|
289 | The first argument is the numerical tag, the second is the (decoded) value |
|
|
290 | that has been tagged. |
|
|
291 | |
|
|
292 | The filter function should return either exactly one value, which will |
|
|
293 | replace the tagged value in the decoded data structure, or no values, |
|
|
294 | which will result in default handling, which currently means the decoder |
|
|
295 | creates a C<CBOR::XS::Tagged> object to hold the tag and the value. |
|
|
296 | |
|
|
297 | When the filter is cleared (the default state), the default filter |
|
|
298 | function, C<CBOR::XS::default_filter>, is used. This function simply looks |
|
|
299 | up the tag in the C<%CBOR::XS::FILTER> hash. If an entry exists it must be |
|
|
300 | a code reference that is called with tag and value, and is responsible for |
|
|
301 | decoding the value. If no entry exists, it returns no values. |
|
|
302 | |
|
|
303 | Example: decode all tags not handled internally into C<CBOR::XS::Tagged> |
|
|
304 | objects, with no other special handling (useful when working with |
|
|
305 | potentially "unsafe" CBOR data). |
|
|
306 | |
|
|
307 | CBOR::XS->new->filter (sub { })->decode ($cbor_data); |
|
|
308 | |
|
|
309 | Example: provide a global filter for tag 1347375694, converting the value |
|
|
310 | into some string form. |
|
|
311 | |
|
|
312 | $CBOR::XS::FILTER{1347375694} = sub { |
|
|
313 | my ($tag, $value); |
|
|
314 | |
|
|
315 | "tag 1347375694 value $value" |
|
|
316 | }; |
239 | |
317 | |
240 | =item $cbor_data = $cbor->encode ($perl_scalar) |
318 | =item $cbor_data = $cbor->encode ($perl_scalar) |
241 | |
319 | |
242 | Converts the given Perl data structure (a scalar value) to its CBOR |
320 | Converts the given Perl data structure (a scalar value) to its CBOR |
243 | representation. |
321 | representation. |
… | |
… | |
257 | and you need to know where the first CBOR string ends amd the next one |
335 | and you need to know where the first CBOR string ends amd the next one |
258 | starts. |
336 | starts. |
259 | |
337 | |
260 | CBOR::XS->new->decode_prefix ("......") |
338 | CBOR::XS->new->decode_prefix ("......") |
261 | => ("...", 3) |
339 | => ("...", 3) |
|
|
340 | |
|
|
341 | =back |
|
|
342 | |
|
|
343 | =head2 INCREMENTAL PARSING |
|
|
344 | |
|
|
345 | In some cases, there is the need for incremental parsing of JSON |
|
|
346 | texts. While this module always has to keep both CBOR text and resulting |
|
|
347 | Perl data structure in memory at one time, it does allow you to parse a |
|
|
348 | CBOR stream incrementally, using a similar to using "decode_prefix" to see |
|
|
349 | if a full CBOR object is available, but is much more efficient. |
|
|
350 | |
|
|
351 | It basically works by parsing as much of a CBOR string as possible - if |
|
|
352 | the CBOR data is not complete yet, the pasrer will remember where it was, |
|
|
353 | to be able to restart when more data has been accumulated. Once enough |
|
|
354 | data is available to either decode a complete CBOR value or raise an |
|
|
355 | error, a real decode will be attempted. |
|
|
356 | |
|
|
357 | A typical use case would be a network protocol that consists of sending |
|
|
358 | and receiving CBOR-encoded messages. The solution that works with CBOR and |
|
|
359 | about anything else is by prepending a length to every CBOR value, so the |
|
|
360 | receiver knows how many octets to read. More compact (and slightly slower) |
|
|
361 | would be to just send CBOR values back-to-back, as C<CBOR::XS> knows where |
|
|
362 | a CBOR value ends, and doesn't need an explicit length. |
|
|
363 | |
|
|
364 | The following methods help with this: |
|
|
365 | |
|
|
366 | =over 4 |
|
|
367 | |
|
|
368 | =item @decoded = $cbor->incr_parse ($buffer) |
|
|
369 | |
|
|
370 | This method attempts to decode exactly one CBOR value from the beginning |
|
|
371 | of the given C<$buffer>. The value is removed from the C<$buffer> on |
|
|
372 | success. When C<$buffer> doesn't contain a complete value yet, it returns |
|
|
373 | nothing. Finally, when the C<$buffer> doesn't start with something |
|
|
374 | that could ever be a valid CBOR value, it raises an exception, just as |
|
|
375 | C<decode> would. In the latter case the decoder state is undefined and |
|
|
376 | must be reset before being able to parse further. |
|
|
377 | |
|
|
378 | This method modifies the C<$buffer> in place. When no CBOR value can be |
|
|
379 | decoded, the decoder stores the current string offset. On the next call, |
|
|
380 | continues decoding at the place where it stopped before. For this to make |
|
|
381 | sense, the C<$buffer> must begin with the same octets as on previous |
|
|
382 | unsuccessful calls. |
|
|
383 | |
|
|
384 | You can call this method in scalar context, in which case it either |
|
|
385 | returns a decoded value or C<undef>. This makes it impossible to |
|
|
386 | distinguish between CBOR null values (which decode to C<undef>) and an |
|
|
387 | unsuccessful decode, which is often acceptable. |
|
|
388 | |
|
|
389 | =item @decoded = $cbor->incr_parse_multiple ($buffer) |
|
|
390 | |
|
|
391 | Same as C<incr_parse>, but attempts to decode as many CBOR values as |
|
|
392 | possible in one go, instead of at most one. Calls to C<incr_parse> and |
|
|
393 | C<incr_parse_multiple> can be interleaved. |
|
|
394 | |
|
|
395 | =item $cbor->incr_reset |
|
|
396 | |
|
|
397 | Resets the incremental decoder. This throws away any saved state, so that |
|
|
398 | subsequent calls to C<incr_parse> or C<incr_parse_multiple> start to parse |
|
|
399 | a new CBOR value from the beginning of the C<$buffer> again. |
|
|
400 | |
|
|
401 | This method can be caled at any time, but it I<must> be called if you want |
|
|
402 | to change your C<$buffer> or there was a decoding error and you want to |
|
|
403 | reuse the C<$cbor> object for future incremental parsings. |
262 | |
404 | |
263 | =back |
405 | =back |
264 | |
406 | |
265 | |
407 | |
266 | =head1 MAPPING |
408 | =head1 MAPPING |
… | |
… | |
284 | CBOR integers become (numeric) perl scalars. On perls without 64 bit |
426 | CBOR integers become (numeric) perl scalars. On perls without 64 bit |
285 | support, 64 bit integers will be truncated or otherwise corrupted. |
427 | support, 64 bit integers will be truncated or otherwise corrupted. |
286 | |
428 | |
287 | =item byte strings |
429 | =item byte strings |
288 | |
430 | |
289 | Byte strings will become octet strings in Perl (the byte values 0..255 |
431 | Byte strings will become octet strings in Perl (the Byte values 0..255 |
290 | will simply become characters of the same value in Perl). |
432 | will simply become characters of the same value in Perl). |
291 | |
433 | |
292 | =item UTF-8 strings |
434 | =item UTF-8 strings |
293 | |
435 | |
294 | UTF-8 strings in CBOR will be decoded, i.e. the UTF-8 octets will be |
436 | UTF-8 strings in CBOR will be decoded, i.e. the UTF-8 octets will be |
… | |
… | |
317 | =item tagged values |
459 | =item tagged values |
318 | |
460 | |
319 | Tagged items consists of a numeric tag and another CBOR value. |
461 | Tagged items consists of a numeric tag and another CBOR value. |
320 | |
462 | |
321 | See L<TAG HANDLING AND EXTENSIONS> and the description of C<< ->filter >> |
463 | See L<TAG HANDLING AND EXTENSIONS> and the description of C<< ->filter >> |
322 | for details. |
464 | for details on which tags are handled how. |
323 | |
465 | |
324 | =item anything else |
466 | =item anything else |
325 | |
467 | |
326 | Anything else (e.g. unsupported simple values) will raise a decoding |
468 | Anything else (e.g. unsupported simple values) will raise a decoding |
327 | error. |
469 | error. |
… | |
… | |
330 | |
472 | |
331 | |
473 | |
332 | =head2 PERL -> CBOR |
474 | =head2 PERL -> CBOR |
333 | |
475 | |
334 | The mapping from Perl to CBOR is slightly more difficult, as Perl is a |
476 | The mapping from Perl to CBOR is slightly more difficult, as Perl is a |
335 | truly typeless language, so we can only guess which CBOR type is meant by |
477 | typeless language. That means this module can only guess which CBOR type |
336 | a Perl value. |
478 | is meant by a perl value. |
337 | |
479 | |
338 | =over 4 |
480 | =over 4 |
339 | |
481 | |
340 | =item hash references |
482 | =item hash references |
341 | |
483 | |
342 | Perl hash references become CBOR maps. As there is no inherent ordering in |
484 | Perl hash references become CBOR maps. As there is no inherent ordering in |
343 | hash keys (or CBOR maps), they will usually be encoded in a pseudo-random |
485 | hash keys (or CBOR maps), they will usually be encoded in a pseudo-random |
344 | order. |
486 | order. This order can be different each time a hahs is encoded. |
345 | |
487 | |
346 | Currently, tied hashes will use the indefinite-length format, while normal |
488 | Currently, tied hashes will use the indefinite-length format, while normal |
347 | hashes will use the fixed-length format. |
489 | hashes will use the fixed-length format. |
348 | |
490 | |
349 | =item array references |
491 | =item array references |
350 | |
492 | |
351 | Perl array references become fixed-length CBOR arrays. |
493 | Perl array references become fixed-length CBOR arrays. |
352 | |
494 | |
353 | =item other references |
495 | =item other references |
354 | |
496 | |
355 | Other unblessed references are generally not allowed and will cause an |
497 | Other unblessed references will be represented using |
356 | exception to be thrown, except for references to the integers C<0> and |
498 | the indirection tag extension (tag value C<22098>, |
357 | C<1>, which get turned into false and true in CBOR. |
499 | L<http://cbor.schmorp.de/indirection>). CBOR decoders are guaranteed |
|
|
500 | to be able to decode these values somehow, by either "doing the right |
|
|
501 | thing", decoding into a generic tagged object, simply ignoring the tag, or |
|
|
502 | something else. |
358 | |
503 | |
359 | =item CBOR::XS::Tagged objects |
504 | =item CBOR::XS::Tagged objects |
360 | |
505 | |
361 | Objects of this type must be arrays consisting of a single C<[tag, value]> |
506 | Objects of this type must be arrays consisting of a single C<[tag, value]> |
362 | pair. The (numerical) tag will be encoded as a CBOR tag, the value will |
507 | pair. The (numerical) tag will be encoded as a CBOR tag, the value will |
363 | be encoded as appropriate for the value. You cna use C<CBOR::XS::tag> to |
508 | be encoded as appropriate for the value. You must use C<CBOR::XS::tag> to |
364 | create such objects. |
509 | create such objects. |
365 | |
510 | |
366 | =item Types::Serialiser::true, Types::Serialiser::false, Types::Serialiser::error |
511 | =item Types::Serialiser::true, Types::Serialiser::false, Types::Serialiser::error |
367 | |
512 | |
368 | These special values become CBOR true, CBOR false and CBOR undefined |
513 | These special values become CBOR true, CBOR false and CBOR undefined |
… | |
… | |
385 | # dump as number |
530 | # dump as number |
386 | encode_cbor [2] # yields [2] |
531 | encode_cbor [2] # yields [2] |
387 | encode_cbor [-3.0e17] # yields [-3e+17] |
532 | encode_cbor [-3.0e17] # yields [-3e+17] |
388 | my $value = 5; encode_cbor [$value] # yields [5] |
533 | my $value = 5; encode_cbor [$value] # yields [5] |
389 | |
534 | |
390 | # used as string, so dump as string |
535 | # used as string, so dump as string (either byte or text) |
391 | print $value; |
536 | print $value; |
392 | encode_cbor [$value] # yields ["5"] |
537 | encode_cbor [$value] # yields ["5"] |
393 | |
538 | |
394 | # undef becomes null |
539 | # undef becomes null |
395 | encode_cbor [undef] # yields [null] |
540 | encode_cbor [undef] # yields [null] |
… | |
… | |
398 | |
543 | |
399 | my $x = 3.1; # some variable containing a number |
544 | my $x = 3.1; # some variable containing a number |
400 | "$x"; # stringified |
545 | "$x"; # stringified |
401 | $x .= ""; # another, more awkward way to stringify |
546 | $x .= ""; # another, more awkward way to stringify |
402 | print $x; # perl does it for you, too, quite often |
547 | print $x; # perl does it for you, too, quite often |
|
|
548 | |
|
|
549 | You can force whether a string ie encoded as byte or text string by using |
|
|
550 | C<utf8::upgrade> and C<utf8::downgrade>): |
|
|
551 | |
|
|
552 | utf8::upgrade $x; # encode $x as text string |
|
|
553 | utf8::downgrade $x; # encode $x as byte string |
|
|
554 | |
|
|
555 | Perl doesn't define what operations up- and downgrade strings, so if the |
|
|
556 | difference between byte and text is important, you should up- or downgrade |
|
|
557 | your string as late as possible before encoding. |
403 | |
558 | |
404 | You can force the type to be a CBOR number by numifying it: |
559 | You can force the type to be a CBOR number by numifying it: |
405 | |
560 | |
406 | my $x = "3"; # some variable containing a string |
561 | my $x = "3"; # some variable containing a string |
407 | $x += 0; # numify it, ensuring it will be dumped as a number |
562 | $x += 0; # numify it, ensuring it will be dumped as a number |
… | |
… | |
420 | |
575 | |
421 | =back |
576 | =back |
422 | |
577 | |
423 | =head2 OBJECT SERIALISATION |
578 | =head2 OBJECT SERIALISATION |
424 | |
579 | |
|
|
580 | This module implements both a CBOR-specific and the generic |
|
|
581 | L<Types::Serialier> object serialisation protocol. The following |
|
|
582 | subsections explain both methods. |
|
|
583 | |
|
|
584 | =head3 ENCODING |
|
|
585 | |
425 | This module knows two way to serialise a Perl object: The CBOR-specific |
586 | This module knows two way to serialise a Perl object: The CBOR-specific |
426 | way, and the generic way. |
587 | way, and the generic way. |
427 | |
588 | |
428 | Whenever the encoder encounters a Perl object that it cnanot serialise |
589 | Whenever the encoder encounters a Perl object that it cannot serialise |
429 | directly (most of them), it will first look up the C<TO_CBOR> method on |
590 | directly (most of them), it will first look up the C<TO_CBOR> method on |
430 | it. |
591 | it. |
431 | |
592 | |
432 | If it has a C<TO_CBOR> method, it will call it with the object as only |
593 | If it has a C<TO_CBOR> method, it will call it with the object as only |
433 | argument, and expects exactly one return value, which it will then |
594 | argument, and expects exactly one return value, which it will then |
… | |
… | |
439 | |
600 | |
440 | The C<FREEZE> method can return any number of values (i.e. zero or |
601 | The C<FREEZE> method can return any number of values (i.e. zero or |
441 | more). These will be encoded as CBOR perl object, together with the |
602 | more). These will be encoded as CBOR perl object, together with the |
442 | classname. |
603 | classname. |
443 | |
604 | |
|
|
605 | These methods I<MUST NOT> change the data structure that is being |
|
|
606 | serialised. Failure to comply to this can result in memory corruption - |
|
|
607 | and worse. |
|
|
608 | |
444 | If an object supports neither C<TO_CBOR> nor C<FREEZE>, encoding will fail |
609 | If an object supports neither C<TO_CBOR> nor C<FREEZE>, encoding will fail |
445 | with an error. |
610 | with an error. |
446 | |
611 | |
|
|
612 | =head3 DECODING |
|
|
613 | |
447 | Objects encoded via C<TO_CBOR> cannot be automatically decoded, but |
614 | Objects encoded via C<TO_CBOR> cannot (normally) be automatically decoded, |
448 | objects encoded via C<FREEZE> can be decoded using the following protocol: |
615 | but objects encoded via C<FREEZE> can be decoded using the following |
|
|
616 | protocol: |
449 | |
617 | |
450 | When an encoded CBOR perl object is encountered by the decoder, it will |
618 | When an encoded CBOR perl object is encountered by the decoder, it will |
451 | look up the C<THAW> method, by using the stored classname, and will fail |
619 | look up the C<THAW> method, by using the stored classname, and will fail |
452 | if the method cannot be found. |
620 | if the method cannot be found. |
453 | |
621 | |
454 | After the lookup it will call the C<THAW> method with the stored classname |
622 | After the lookup it will call the C<THAW> method with the stored classname |
455 | as first argument, the constant string C<CBOR> as second argument, and all |
623 | as first argument, the constant string C<CBOR> as second argument, and all |
456 | values returned by C<FREEZE> as remaining arguments. |
624 | values returned by C<FREEZE> as remaining arguments. |
457 | |
625 | |
458 | =head4 EXAMPLES |
626 | =head3 EXAMPLES |
459 | |
627 | |
460 | Here is an example C<TO_CBOR> method: |
628 | Here is an example C<TO_CBOR> method: |
461 | |
629 | |
462 | sub My::Object::TO_CBOR { |
630 | sub My::Object::TO_CBOR { |
463 | my ($obj) = @_; |
631 | my ($obj) = @_; |
… | |
… | |
474 | |
642 | |
475 | sub URI::TO_CBOR { |
643 | sub URI::TO_CBOR { |
476 | my ($self) = @_; |
644 | my ($self) = @_; |
477 | my $uri = "$self"; # stringify uri |
645 | my $uri = "$self"; # stringify uri |
478 | utf8::upgrade $uri; # make sure it will be encoded as UTF-8 string |
646 | utf8::upgrade $uri; # make sure it will be encoded as UTF-8 string |
479 | CBOR::XS::tagged 32, "$_[0]" |
647 | CBOR::XS::tag 32, "$_[0]" |
480 | } |
648 | } |
481 | |
649 | |
482 | This will encode URIs as a UTF-8 string with tag 32, which indicates an |
650 | This will encode URIs as a UTF-8 string with tag 32, which indicates an |
483 | URI. |
651 | URI. |
484 | |
652 | |
… | |
… | |
640 | These tags are always handled when decoding, and their handling cannot be |
808 | These tags are always handled when decoding, and their handling cannot be |
641 | overriden by the user. |
809 | overriden by the user. |
642 | |
810 | |
643 | =over 4 |
811 | =over 4 |
644 | |
812 | |
645 | =item <unassigned> (perl-object, L<http://cbor.schmorp.de/perl-object>) |
813 | =item 26 (perl-object, L<http://cbor.schmorp.de/perl-object>) |
646 | |
814 | |
647 | These tags are automatically created (and decoded) for serialisable |
815 | These tags are automatically created (and decoded) for serialisable |
648 | objects using the C<FREEZE/THAW> methods (the L<Types::Serialier> object |
816 | objects using the C<FREEZE/THAW> methods (the L<Types::Serialier> object |
649 | serialisation protocol). See L<OBJECT SERIALISATION> for details. |
817 | serialisation protocol). See L<OBJECT SERIALISATION> for details. |
650 | |
818 | |
651 | =item <unassigned>, <unassigned> (sharable, sharedref, L <http://cbor.schmorp.de/value-sharing>) |
819 | =item 28, 29 (shareable, sharedref, L<http://cbor.schmorp.de/value-sharing>) |
652 | |
820 | |
653 | These tags are automatically decoded when encountered, resulting in |
821 | These tags are automatically decoded when encountered (and they do not |
|
|
822 | result in a cyclic data structure, see C<allow_cycles>), resulting in |
654 | shared values in the decoded object. They are only encoded, however, when |
823 | shared values in the decoded object. They are only encoded, however, when |
655 | C<allow_sharable> is enabled. |
824 | C<allow_sharing> is enabled. |
656 | |
825 | |
|
|
826 | Not all shared values can be successfully decoded: values that reference |
|
|
827 | themselves will I<currently> decode as C<undef> (this is not the same |
|
|
828 | as a reference pointing to itself, which will be represented as a value |
|
|
829 | that contains an indirect reference to itself - these will be decoded |
|
|
830 | properly). |
|
|
831 | |
|
|
832 | Note that considerably more shared value data structures can be decoded |
|
|
833 | than will be encoded - currently, only values pointed to by references |
|
|
834 | will be shared, others will not. While non-reference shared values can be |
|
|
835 | generated in Perl with some effort, they were considered too unimportant |
|
|
836 | to be supported in the encoder. The decoder, however, will decode these |
|
|
837 | values as shared values. |
|
|
838 | |
657 | =item <unassigned>, <unassigned> (stringref-namespace, stringref, L <http://cbor.schmorp.de/stringref>) |
839 | =item 256, 25 (stringref-namespace, stringref, L<http://cbor.schmorp.de/stringref>) |
658 | |
840 | |
659 | These tags are automatically decoded when encountered. They are only |
841 | These tags are automatically decoded when encountered. They are only |
660 | encoded, however, when C<allow_stringref> is enabled. |
842 | encoded, however, when C<pack_strings> is enabled. |
661 | |
843 | |
662 | =item 22098 (indirection, L<http://cbor.schmorp.de/indirection>) |
844 | =item 22098 (indirection, L<http://cbor.schmorp.de/indirection>) |
663 | |
845 | |
664 | This tag is automatically generated when a reference are encountered (with |
846 | This tag is automatically generated when a reference are encountered (with |
665 | the exception of hash and array refernces). It is converted to a reference |
847 | the exception of hash and array refernces). It is converted to a reference |
… | |
… | |
670 | This value is not generated on encoding (unless explicitly requested by |
852 | This value is not generated on encoding (unless explicitly requested by |
671 | the user), and is simply ignored when decoding. |
853 | the user), and is simply ignored when decoding. |
672 | |
854 | |
673 | =back |
855 | =back |
674 | |
856 | |
675 | =head2 OPTIONAL TAGS |
857 | =head2 NON-ENFORCED TAGS |
676 | |
858 | |
677 | These tags have default filters provided when decoding. Their handling can |
859 | These tags have default filters provided when decoding. Their handling can |
678 | be overriden by changing the C<%CBOR::XS::FILTER> entry for the tag, or by |
860 | be overriden by changing the C<%CBOR::XS::FILTER> entry for the tag, or by |
679 | providing a custom C<filter> function when decoding. |
861 | providing a custom C<filter> callback when decoding. |
680 | |
862 | |
681 | When they result in decoding into a specific Perl class, the module |
863 | When they result in decoding into a specific Perl class, the module |
682 | usually provides a corresponding C<TO_CBOR> method as well. |
864 | usually provides a corresponding C<TO_CBOR> method as well. |
683 | |
865 | |
684 | When any of these need to load additional modules that are not part of the |
866 | When any of these need to load additional modules that are not part of the |
685 | perl core distribution (e.g. L<URI>), it is (currently) up to the user to |
867 | perl core distribution (e.g. L<URI>), it is (currently) up to the user to |
686 | provide these modules. The decoding usually fails with an exception if the |
868 | provide these modules. The decoding usually fails with an exception if the |
687 | required module cannot be loaded. |
869 | required module cannot be loaded. |
688 | |
870 | |
689 | =over 4 |
871 | =over 4 |
|
|
872 | |
|
|
873 | =item 0, 1 (date/time string, seconds since the epoch) |
|
|
874 | |
|
|
875 | These tags are decoded into L<Time::Piece> objects. The corresponding |
|
|
876 | C<Time::Piece::TO_CBOR> method always encodes into tag 1 values currently. |
|
|
877 | |
|
|
878 | The L<Time::Piece> API is generally surprisingly bad, and fractional |
|
|
879 | seconds are only accidentally kept intact, so watch out. On the plus side, |
|
|
880 | the module comes with perl since 5.10, which has to count for something. |
690 | |
881 | |
691 | =item 2, 3 (positive/negative bignum) |
882 | =item 2, 3 (positive/negative bignum) |
692 | |
883 | |
693 | These tags are decoded into L<Math::BigInt> objects. The corresponding |
884 | These tags are decoded into L<Math::BigInt> objects. The corresponding |
694 | C<Math::BigInt::TO_CBOR> method encodes "small" bigints into normal CBOR |
885 | C<Math::BigInt::TO_CBOR> method encodes "small" bigints into normal CBOR |
… | |
… | |
828 | properly. Half precision types are accepted, but not encoded. |
1019 | properly. Half precision types are accepted, but not encoded. |
829 | |
1020 | |
830 | Strict mode and canonical mode are not implemented. |
1021 | Strict mode and canonical mode are not implemented. |
831 | |
1022 | |
832 | |
1023 | |
|
|
1024 | =head1 LIMITATIONS ON PERLS WITHOUT 64-BIT INTEGER SUPPORT |
|
|
1025 | |
|
|
1026 | On perls that were built without 64 bit integer support (these are rare |
|
|
1027 | nowadays, even on 32 bit architectures, as all major Perl distributions |
|
|
1028 | are built with 64 bit integer support), support for any kind of 64 bit |
|
|
1029 | integer in CBOR is very limited - most likely, these 64 bit values will |
|
|
1030 | be truncated, corrupted, or otherwise not decoded correctly. This also |
|
|
1031 | includes string, array and map sizes that are stored as 64 bit integers. |
|
|
1032 | |
|
|
1033 | |
833 | =head1 THREADS |
1034 | =head1 THREADS |
834 | |
1035 | |
835 | This module is I<not> guaranteed to be thread safe and there are no |
1036 | This module is I<not> guaranteed to be thread safe and there are no |
836 | plans to change this until Perl gets thread support (as opposed to the |
1037 | plans to change this until Perl gets thread support (as opposed to the |
837 | horribly slow so-called "threads" which are simply slow and bloated |
1038 | horribly slow so-called "threads" which are simply slow and bloated |
… | |
… | |
850 | service. I put the contact address into my modules for a reason. |
1051 | service. I put the contact address into my modules for a reason. |
851 | |
1052 | |
852 | =cut |
1053 | =cut |
853 | |
1054 | |
854 | our %FILTER = ( |
1055 | our %FILTER = ( |
855 | # 0 # rfc4287 datetime, utf-8 |
1056 | 0 => sub { # rfc4287 datetime, utf-8 |
856 | # 1 # unix timestamp, any |
1057 | require Time::Piece; |
|
|
1058 | # Time::Piece::Strptime uses the "incredibly flexible date parsing routine" |
|
|
1059 | # from FreeBSD, which can't parse ISO 8601, RFC3339, RFC4287 or much of anything |
|
|
1060 | # else either. Whats incredibe over standard strptime totally escapes me. |
|
|
1061 | # doesn't do fractional times, either. sigh. |
|
|
1062 | # In fact, it's all a lie, it uses whatever strptime it wants, and of course, |
|
|
1063 | # they are all incompatible. The openbsd one simply ignores %z (but according to the |
|
|
1064 | # docs, it would be much more incredibly flexible indeed. If it worked, that is.). |
|
|
1065 | scalar eval { |
|
|
1066 | my $s = $_[1]; |
|
|
1067 | |
|
|
1068 | $s =~ s/Z$/+00:00/; |
|
|
1069 | $s =~ s/(\.[0-9]+)?([+-][0-9][0-9]):([0-9][0-9])$// |
|
|
1070 | or die; |
|
|
1071 | |
|
|
1072 | my $b = $1 - ($2 * 60 + $3) * 60; # fractional part + offset. hopefully |
|
|
1073 | my $d = Time::Piece->strptime ($s, "%Y-%m-%dT%H:%M:%S"); |
|
|
1074 | |
|
|
1075 | Time::Piece::gmtime ($d->epoch + $b) |
|
|
1076 | } || die "corrupted CBOR date/time string ($_[0])"; |
|
|
1077 | }, |
|
|
1078 | |
|
|
1079 | 1 => sub { # seconds since the epoch, possibly fractional |
|
|
1080 | require Time::Piece; |
|
|
1081 | scalar Time::Piece::gmtime (pop) |
|
|
1082 | }, |
857 | |
1083 | |
858 | 2 => sub { # pos bigint |
1084 | 2 => sub { # pos bigint |
859 | require Math::BigInt; |
1085 | require Math::BigInt; |
860 | Math::BigInt->new ("0x" . unpack "H*", pop) |
1086 | Math::BigInt->new ("0x" . unpack "H*", pop) |
861 | }, |
1087 | }, |
… | |
… | |
897 | } |
1123 | } |
898 | |
1124 | |
899 | sub URI::TO_CBOR { |
1125 | sub URI::TO_CBOR { |
900 | my $uri = $_[0]->as_string; |
1126 | my $uri = $_[0]->as_string; |
901 | utf8::upgrade $uri; |
1127 | utf8::upgrade $uri; |
902 | CBOR::XS::tag 32, $uri |
1128 | tag 32, $uri |
903 | } |
1129 | } |
904 | |
1130 | |
905 | sub Math::BigInt::TO_CBOR { |
1131 | sub Math::BigInt::TO_CBOR { |
906 | if ($_[0] >= -2147483648 && $_[0] <= 2147483647) { |
1132 | if ($_[0] >= -2147483648 && $_[0] <= 2147483647) { |
907 | $_[0]->numify |
1133 | $_[0]->numify |
908 | } else { |
1134 | } else { |
909 | my $hex = substr $_[0]->as_hex, 2; |
1135 | my $hex = substr $_[0]->as_hex, 2; |
910 | $hex = "0$hex" if 1 & length $hex; # sigh |
1136 | $hex = "0$hex" if 1 & length $hex; # sigh |
911 | CBOR::XS::tag $_[0] >= 0 ? 2 : 3, pack "H*", $hex |
1137 | tag $_[0] >= 0 ? 2 : 3, pack "H*", $hex |
912 | } |
1138 | } |
913 | } |
1139 | } |
914 | |
1140 | |
915 | sub Math::BigFloat::TO_CBOR { |
1141 | sub Math::BigFloat::TO_CBOR { |
916 | my ($m, $e) = $_[0]->parts; |
1142 | my ($m, $e) = $_[0]->parts; |
917 | CBOR::XS::tag 4, [$e->numify, $m] |
1143 | tag 4, [$e->numify, $m] |
|
|
1144 | } |
|
|
1145 | |
|
|
1146 | sub Time::Piece::TO_CBOR { |
|
|
1147 | tag 1, 0 + $_[0]->epoch |
918 | } |
1148 | } |
919 | |
1149 | |
920 | XSLoader::load "CBOR::XS", $VERSION; |
1150 | XSLoader::load "CBOR::XS", $VERSION; |
921 | |
1151 | |
922 | =head1 SEE ALSO |
1152 | =head1 SEE ALSO |