ViewVC Help
View File | Revision Log | Show Annotations | Download File
/cvs/CBOR-XS/Changes
Revision: 1.99
Committed: Fri Sep 8 20:16:09 2023 UTC (7 months, 3 weeks ago) by root
Branch: MAIN
CVS Tags: rel-1_87, HEAD
Changes since 1.98: +3 -4 lines
Log Message:
*** empty log message ***

File Contents

# Content
1 Revision history for Perl extension CBOR::XS
2
3 TODO: pack_keys?
4 TODO: document encode_cbor_sharing?
5 TODO: large negative integers
6 TODO: type cast tests.
7 TODO: round-tripping of types, such as float16 - maybe types::Serialiser support?
8 TODO: possibly implement https://peteroupc.github.io/CBOR/extended.html, but NaNs are nonportable. rely on libecb?
9 TODO: https://github.com/svaarala/cbor-specs/blob/master/cbor-nonutf8-string-tags.rst, but maybe that is overkill?
10
11 1.87 Fri 08 Sep 2023 22:14:18 CEST
12 - shared references were not decoded correctly: instead of getting
13 multiple references to the same object, you got the same
14 reference to the same object, causing a number of issues. For
15 example, modifying the reference would modify all places the
16 reference was used, and encoding the decoded structure would
17 unshare the previously shared hashes, as trheir reference count
18 would be 1. Fixing this was rather involved, as perl lacks the
19 ability to easily swap or copy arrays and hashes.
20 - \0, \1, \undef do not work, and were not intended to ever work, as
21 special values, despite being mentioned in the documentation (reported
22 by nuclightq).
23 - new feature: allow_weak_cycles.
24
25 1.86 Thu 04 Nov 2021 17:48:16 CET
26 - fixed a wrong printf format specifier (reported by Petr Písař).
27
28 1.85 Sat 23 Oct 2021 04:59:56 CEST
29 - left debugging printf in code, need a test for that :(
30
31 1.84 Thu 21 Oct 2021 03:11:52 CEST
32 - fix a bug in validate_utf8 where we call perl's is_utf8_string with
33 a lenght of zero for empty strings, but perl interprets that as
34 "calculate length", causing spurious validation errors for
35 empty strings.
36 - include validate_utf8 in new_safe.
37 - avoid some warnings.
38
39 1.83 Tue Dec 8 09:27:06 CET 2020
40 - add CBOR::XS::as_map cast function.
41
42 1.82 Tue Dec 1 02:47:40 CET 2020
43 - add CBOR::XS::as_bool cast function.
44
45 1.81 Mon Nov 30 19:29:33 CET 2020
46 - cast functions were broken due to last-minute renaming. thats
47 what you get for not having a tessuite.
48 - Math::BigInt and Math::BigFloat are pretty broken (again),
49 so disable some tests. (try printing the bigfloat
50 799999999999999999998E99999999999999999998).
51
52 1.8 Sun Nov 29 22:35:13 CET 2020
53 - experimental support for some type casts, as well as embedding
54 raw cbor data.
55
56 1.71 Thu Nov 15 20:52:13 CET 2018
57 - work around what smells like a perl bug w.r.t. exceptions
58 thrown in callbacks.
59 - update libecb.
60
61 1.7 Tue Jun 27 04:02:23 CEST 2017
62 - SECURITY FIX: fix two bugs found by american fuzzy lop,
63 upgrade is advised if you accept data from untrusted
64 sources.
65 - an out-of bound sharedref or stringref index could cause an
66 out of bounds access - might be exploitable.
67 - a decoding error during indefinite array or hash decoding
68 could cause an endless loop.
69
70 1.6 Wed Dec 7 15:13:23 CET 2016
71 - greatly expand the SECURITY IMPLICATIONS and similar sections.
72 - new constructor new_safe, to create a secure CBOR::XS object.
73 - new option forbid_objects, to disallow serialisation.
74 - new CBOR::XS::safe_filter functionality.
75 - fix a crash when decoding a cyclic data structure using
76 stringref/pack_strings when allow_cycles is disabled.
77 - fix a crash when decoding hash keys with length >= 2**31.
78 - avoid unreasonably long decoding times for certain
79 types of (corrupt) cbor texts.
80 - support arrays and hashes with >= 2**31 members.
81 - avoid overflow on pointer arithmetic when checking whether enough
82 data is available.
83 - fix a memory leak that occured when decoding failed while decoding
84 a tagged value.
85 - do not leak the partially constructed result when stringifying
86 a hash key throws an exception.
87 - various code size and efficiency optimizations (reduced code
88 from 42 to 40kB on my system, despite the new features).
89
90 1.5 Wed Apr 27 11:38:39 CEST 2016
91 - Math::BigFloat madness workaround, see
92 http://blog.schmorp.de/2016-04-23-mathbigfloat-maintainer-fail.html
93 (bugreport by zdm@softvisio.net).
94 - add text_keys and text_strings options to force CBOR text encoding
95 for perl hash keys or all strings, as a result of discussions
96 with Fredrik Ljunggren.
97 - implement support for arbitrary-exponent numbers (see
98 http://peteroupc.github.io/CBOR/bigfrac.html, tags 264 and 265)
99 for both en- and decoding.
100 - implement support for rational numbers (see
101 http://peteroupc.github.io/CBOR/rational.html, tag 30) for both
102 en- and decoding.
103 - the above effectively implements all registered CBOR extensions
104 in a sensible manner.
105 - remove some weird dead code that was duplicated (%FILTER).
106 - add t/58_hv.t, which tests hashes and the new text_* flags.
107 hashes apparently were not encoded at all in any of the existing
108 tests.
109 - document Math::BigFloat base-2 performance/crash issues.
110 - use stability canary.
111
112 1.41 Thu 25 Feb 15:22:03 CET 2016
113 - avoid perl panics on nested FREEZE/THAW calls (testcase by
114 Victor Efimov).
115
116 1.4 Mon Feb 8 05:10:15 CET 2016
117 - buffer overflow fix: a fast path during decoding did not check
118 remaining length when decoding hash keys, found by fuzzing.
119 This can potentially leak information in the error message
120 or crash the process.
121 - use C style { 0 } struct initializer.
122 - upgrade libecb.
123
124 1.3 Mon Apr 27 22:21:04 CEST 2015
125 - the incremental parser didn't properly parse tagged values
126 (testcase by Mons Anderson).
127 - slightly speed up encoding of plain (nonmagical) arrays.
128 - try to clarify further that effectively all 32 bit architectures
129 have 64 bit integer support.
130 - upgrade libecb.
131
132 1.26 Sat Oct 25 08:35:44 CEST 2014
133 - update the t/57_incr.t subtest that would rely on 64 bit ints.
134 - disable t/50_rfc.t test that fails because of broken data::dumper.
135
136 1.25 Sun Jan 5 15:19:14 CET 2014
137 - map key decoding was pretty much botched due to the recent cleanups.
138 - work around Time::Piece->epoch returning a string value, avoid encoding
139 this as a tag 1 string.
140 - enable more testcases in t/50_rfc.t, now that they work :)
141
142 1.2 Tue Dec 10 22:06:42 CET 2013
143 - implement an incremental decoder.
144
145 1.12 Tue Dec 3 11:23:22 CET 2013
146 - work around broken Time::Piece (in old versions of the module, %z doesn't
147 work as documented, gives different results on different platforms(!)).
148
149 1.11 Sun Dec 1 18:00:00 CET 2013
150 - new setting: validate_utf8, for when you can't trust your cbor data.
151 - do not leak memory on decoding errors, when allow_cycles is enabled.
152 - add default filters for tags 0 and 1, using Time::Piece.
153 - more tests added.
154
155 1.1 Sat Nov 30 19:14:27 CET 2013
156 - INCOMPATIBLE CHANGE: new decoder setting: allow_cyclic, needed to decode
157 cyclic data structures (to avoid memleaks in unsuspecting code).
158 - no longer "share" references that aren't, i.e. true/false/null/error/tagged.
159 - fix stringref w.r.t. indefinite-length strings.
160 - verify indefinite-length string chunk types.
161 - do not allow extremely large arrays - assume an array element
162 requires at least one CBOR byte, to avoid memory exhaustion attacks.
163 - major code overhaul.
164
165 1.0 Thu Nov 28 16:43:31 CET 2013
166 - use the now official tag values for extensions. remove the
167 experimental notice. it's the real thing now, with real bugs.
168 - renamed allow_stringref to pack_strings.
169 - port to perl <= 5.16.
170 - slightly improve the documentation.
171
172 0.09 Fri Nov 22 16:54:18 CET 2013
173 - bignum/bigfloat/decimal support.
174 - uri support.
175 - tag filter functions support for decoding.
176 - do not support reference-to-1/0/undef anymore, you need to use
177 the Types::Serialiser objects now.
178 - experimental sharable extension support (http://cbor.schmorp.de/value-sharing).
179 - experimental stringref extension support (http://cbor.schmorp.de/stringref).
180 - implement indirection tag (http://cbor.schmorp.de/indirection).
181
182 0.08 Wed Oct 30 11:10:43 CET 2013
183 - defused another too fragile test.
184
185 0.07 Tue Oct 29 23:04:07 CET 2013
186 - don't crash in decode when silly values are passed in.
187 - considerably speed up map decoding when map keys
188 are utf-8 or byte strings.
189 - raising an exception in THAW should now work without
190 leaking.
191
192 0.06 Tue Oct 29 16:56:07 CET 2013
193 - do not leak when deserialiasing via THAW.
194 - implement and document CBOR::XS creation/access/mutate
195 methods.
196
197 0.05 Mon Oct 28 22:27:47 CET 2013
198 - do not leak hash keys on decoding.
199
200 0.04 Sun Oct 27 23:47:47 CET 2013
201 - implement TO_CBOR/FREEZE/THAW serialisation protocols.
202 - requested perl-object and generic-object tags from iana.
203 - switched to Types::Serialiser for true, false and error.
204 - disabled some fragile tests (thanks, andk).
205
206 0.03 Sun Oct 27 00:28:41 CEST 2013
207 - improve 32 bit platform compatibility.
208 - take more advantage of ecb.h.
209 - preliminary and bare-bones tagged support.
210 - improved docs.
211
212 0.02 Sat Oct 26 13:08:05 CEST 2013
213 - no aborts left.
214 - add $CBOR::XS::MAGIC.
215 - preliminary tagged decoding to arrayref.
216 - indefinite encoding fixed.
217 - half float decoding implemented.
218 - t/50_rfc.t adds test vectors from the rfc, which
219 are checked as applicable.
220
221 0.01 Fri Oct 25 21:39:56 CEST 2013
222 - original version; cloned from JSON-XS
223