1 |
NAME |
2 |
CBOR::XS - Concise Binary Object Representation (CBOR, RFC7049) |
3 |
|
4 |
SYNOPSIS |
5 |
use CBOR::XS; |
6 |
|
7 |
$binary_cbor_data = encode_cbor $perl_value; |
8 |
$perl_value = decode_cbor $binary_cbor_data; |
9 |
|
10 |
# OO-interface |
11 |
|
12 |
$coder = CBOR::XS->new; |
13 |
#TODO |
14 |
|
15 |
DESCRIPTION |
16 |
WARNING! THIS IS A PRE-ALPHA RELEASE! IT WILL CRASH, CORRUPT YOUR DATA |
17 |
AND EAT YOUR CHILDREN! |
18 |
|
19 |
This module converts Perl data structures to CBOR and vice versa. Its |
20 |
primary goal is to be *correct* and its secondary goal is to be *fast*. |
21 |
To reach the latter goal it was written in C. |
22 |
|
23 |
See MAPPING, below, on how CBOR::XS maps perl values to CBOR values and |
24 |
vice versa. |
25 |
|
26 |
FUNCTIONAL INTERFACE |
27 |
The following convenience methods are provided by this module. They are |
28 |
exported by default: |
29 |
|
30 |
$cbor_data = encode_cbor $perl_scalar |
31 |
Converts the given Perl data structure to CBOR representation. |
32 |
Croaks on error. |
33 |
|
34 |
$perl_scalar = decode_cbor $cbor_data |
35 |
The opposite of "encode_cbor": expects a valid CBOR string to parse, |
36 |
returning the resulting perl scalar. Croaks on error. |
37 |
|
38 |
OBJECT-ORIENTED INTERFACE |
39 |
The object oriented interface lets you configure your own encoding or |
40 |
decoding style, within the limits of supported formats. |
41 |
|
42 |
$cbor = new CBOR::XS |
43 |
Creates a new CBOR::XS object that can be used to de/encode CBOR |
44 |
strings. All boolean flags described below are by default |
45 |
*disabled*. |
46 |
|
47 |
The mutators for flags all return the CBOR object again and thus |
48 |
calls can be chained: |
49 |
|
50 |
#TODO my $cbor = CBOR::XS->new->encode ({a => [1,2]}); |
51 |
|
52 |
$cbor = $cbor->max_depth ([$maximum_nesting_depth]) |
53 |
$max_depth = $cbor->get_max_depth |
54 |
Sets the maximum nesting level (default 512) accepted while encoding |
55 |
or decoding. If a higher nesting level is detected in CBOR data or a |
56 |
Perl data structure, then the encoder and decoder will stop and |
57 |
croak at that point. |
58 |
|
59 |
Nesting level is defined by number of hash- or arrayrefs that the |
60 |
encoder needs to traverse to reach a given point or the number of |
61 |
"{" or "[" characters without their matching closing parenthesis |
62 |
crossed to reach a given character in a string. |
63 |
|
64 |
Setting the maximum depth to one disallows any nesting, so that |
65 |
ensures that the object is only a single hash/object or array. |
66 |
|
67 |
If no argument is given, the highest possible setting will be used, |
68 |
which is rarely useful. |
69 |
|
70 |
Note that nesting is implemented by recursion in C. The default |
71 |
value has been chosen to be as large as typical operating systems |
72 |
allow without crashing. |
73 |
|
74 |
See SECURITY CONSIDERATIONS, below, for more info on why this is |
75 |
useful. |
76 |
|
77 |
$cbor = $cbor->max_size ([$maximum_string_size]) |
78 |
$max_size = $cbor->get_max_size |
79 |
Set the maximum length a CBOR string may have (in bytes) where |
80 |
decoding is being attempted. The default is 0, meaning no limit. |
81 |
When "decode" is called on a string that is longer then this many |
82 |
bytes, it will not attempt to decode the string but throw an |
83 |
exception. This setting has no effect on "encode" (yet). |
84 |
|
85 |
If no argument is given, the limit check will be deactivated (same |
86 |
as when 0 is specified). |
87 |
|
88 |
See SECURITY CONSIDERATIONS, below, for more info on why this is |
89 |
useful. |
90 |
|
91 |
$cbor_data = $cbor->encode ($perl_scalar) |
92 |
Converts the given Perl data structure (a scalar value) to its CBOR |
93 |
representation. |
94 |
|
95 |
$perl_scalar = $cbor->decode ($cbor_data) |
96 |
The opposite of "encode": expects CBOR data and tries to parse it, |
97 |
returning the resulting simple scalar or reference. Croaks on error. |
98 |
|
99 |
($perl_scalar, $octets) = $cbor->decode_prefix ($cbor_data) |
100 |
This works like the "decode" method, but instead of raising an |
101 |
exception when there is trailing garbage after the CBOR string, it |
102 |
will silently stop parsing there and return the number of characters |
103 |
consumed so far. |
104 |
|
105 |
This is useful if your CBOR texts are not delimited by an outer |
106 |
protocol and you need to know where the first CBOR string ends amd |
107 |
the next one starts. |
108 |
|
109 |
CBOR::XS->new->decode_prefix ("......") |
110 |
=> ("...", 3) |
111 |
|
112 |
MAPPING |
113 |
This section describes how CBOR::XS maps Perl values to CBOR values and |
114 |
vice versa. These mappings are designed to "do the right thing" in most |
115 |
circumstances automatically, preserving round-tripping characteristics |
116 |
(what you put in comes out as something equivalent). |
117 |
|
118 |
For the more enlightened: note that in the following descriptions, |
119 |
lowercase *perl* refers to the Perl interpreter, while uppercase *Perl* |
120 |
refers to the abstract Perl language itself. |
121 |
|
122 |
CBOR -> PERL |
123 |
True, False |
124 |
These CBOR values become "CBOR::XS::true" and "CBOR::XS::false", |
125 |
respectively. They are overloaded to act almost exactly like the |
126 |
numbers 1 and 0. You can check whether a scalar is a CBOR boolean by |
127 |
using the "CBOR::XS::is_bool" function. |
128 |
|
129 |
Null, Undefined |
130 |
CBOR Null and Undefined values becomes "undef" in Perl (in the |
131 |
future, Undefined may raise an exception). |
132 |
|
133 |
PERL -> CBOR |
134 |
The mapping from Perl to CBOR is slightly more difficult, as Perl is a |
135 |
truly typeless language, so we can only guess which CBOR type is meant |
136 |
by a Perl value. |
137 |
|
138 |
hash references |
139 |
Perl hash references become CBOR maps. As there is no inherent |
140 |
ordering in hash keys (or CBOR maps), they will usually be encoded |
141 |
in a pseudo-random order. |
142 |
|
143 |
array references |
144 |
Perl array references become CBOR arrays. |
145 |
|
146 |
other references |
147 |
Other unblessed references are generally not allowed and will cause |
148 |
an exception to be thrown, except for references to the integers 0 |
149 |
and 1, which get turned into "False" and "True" in CBOR. |
150 |
|
151 |
CBOR::XS::true, CBOR::XS::false |
152 |
These special values become CBOR True and CBOR False values, |
153 |
respectively. You can also use "\1" and "\0" directly if you want. |
154 |
|
155 |
blessed objects |
156 |
Blessed objects are not directly representable in CBOR. TODO See the |
157 |
"allow_blessed" and "convert_blessed" methods on various options on |
158 |
how to deal with this: basically, you can choose between throwing an |
159 |
exception, encoding the reference as if it weren't blessed, or |
160 |
provide your own serialiser method. |
161 |
|
162 |
simple scalars |
163 |
TODO Simple Perl scalars (any scalar that is not a reference) are |
164 |
the most difficult objects to encode: CBOR::XS will encode undefined |
165 |
scalars as CBOR "Null" values, scalars that have last been used in a |
166 |
string context before encoding as CBOR strings, and anything else as |
167 |
number value: |
168 |
|
169 |
# dump as number |
170 |
encode_cbor [2] # yields [2] |
171 |
encode_cbor [-3.0e17] # yields [-3e+17] |
172 |
my $value = 5; encode_cbor [$value] # yields [5] |
173 |
|
174 |
# used as string, so dump as string |
175 |
print $value; |
176 |
encode_cbor [$value] # yields ["5"] |
177 |
|
178 |
# undef becomes null |
179 |
encode_cbor [undef] # yields [null] |
180 |
|
181 |
You can force the type to be a CBOR string by stringifying it: |
182 |
|
183 |
my $x = 3.1; # some variable containing a number |
184 |
"$x"; # stringified |
185 |
$x .= ""; # another, more awkward way to stringify |
186 |
print $x; # perl does it for you, too, quite often |
187 |
|
188 |
You can force the type to be a CBOR number by numifying it: |
189 |
|
190 |
my $x = "3"; # some variable containing a string |
191 |
$x += 0; # numify it, ensuring it will be dumped as a number |
192 |
$x *= 1; # same thing, the choice is yours. |
193 |
|
194 |
You can not currently force the type in other, less obscure, ways. |
195 |
Tell me if you need this capability (but don't forget to explain why |
196 |
it's needed :). |
197 |
|
198 |
Note that numerical precision has the same meaning as under Perl (so |
199 |
binary to decimal conversion follows the same rules as in Perl, |
200 |
which can differ to other languages). Also, your perl interpreter |
201 |
might expose extensions to the floating point numbers of your |
202 |
platform, such as infinities or NaN's - these cannot be represented |
203 |
in CBOR, and it is an error to pass those in. |
204 |
|
205 |
MAGIC HEADER |
206 |
There is no way to distinguish CBOR from other formats programmatically. |
207 |
To make it easier to distinguish CBOR from other formats, the CBOR |
208 |
specification has a special "magic string" that can be prepended to any |
209 |
CBOR string without changing it's meaning. |
210 |
|
211 |
This string is available as $CBOR::XS::MAGIC. This module does not |
212 |
prepend this string tot he CBOR data it generates, but it will ignroe it |
213 |
if present, so users can prepend this string as a "file type" indicator |
214 |
as required. |
215 |
|
216 |
CBOR and JSON |
217 |
TODO |
218 |
|
219 |
SECURITY CONSIDERATIONS |
220 |
When you are using CBOR in a protocol, talking to untrusted potentially |
221 |
hostile creatures requires relatively few measures. |
222 |
|
223 |
First of all, your CBOR decoder should be secure, that is, should not |
224 |
have any buffer overflows. Obviously, this module should ensure that and |
225 |
I am trying hard on making that true, but you never know. |
226 |
|
227 |
Second, you need to avoid resource-starving attacks. That means you |
228 |
should limit the size of CBOR data you accept, or make sure then when |
229 |
your resources run out, that's just fine (e.g. by using a separate |
230 |
process that can crash safely). The size of a CBOR string in octets is |
231 |
usually a good indication of the size of the resources required to |
232 |
decode it into a Perl structure. While CBOR::XS can check the size of |
233 |
the CBOR text, it might be too late when you already have it in memory, |
234 |
so you might want to check the size before you accept the string. |
235 |
|
236 |
Third, CBOR::XS recurses using the C stack when decoding objects and |
237 |
arrays. The C stack is a limited resource: for instance, on my amd64 |
238 |
machine with 8MB of stack size I can decode around 180k nested arrays |
239 |
but only 14k nested CBOR objects (due to perl itself recursing deeply on |
240 |
croak to free the temporary). If that is exceeded, the program crashes. |
241 |
To be conservative, the default nesting limit is set to 512. If your |
242 |
process has a smaller stack, you should adjust this setting accordingly |
243 |
with the "max_depth" method. |
244 |
|
245 |
Something else could bomb you, too, that I forgot to think of. In that |
246 |
case, you get to keep the pieces. I am always open for hints, though... |
247 |
|
248 |
Also keep in mind that CBOR::XS might leak contents of your Perl data |
249 |
structures in its error messages, so when you serialise sensitive |
250 |
information you might want to make sure that exceptions thrown by |
251 |
CBOR::XS will not end up in front of untrusted eyes. |
252 |
|
253 |
CBOR IMPLEMENTATION NOTES |
254 |
This section contains some random implementation notes. They do not |
255 |
describe guaranteed behaviour, but merely behaviour as-is implemented |
256 |
right now. |
257 |
|
258 |
64 bit integers are only properly decoded when Perl was built with 64 |
259 |
bit support. |
260 |
|
261 |
Strings and arrays are encoded with a definite length. Hashes as well, |
262 |
unless they are tied (or otherwise magical). |
263 |
|
264 |
Only the double data type is supported for NV data types - when Perl |
265 |
uses long double to represent floating point values, they might not be |
266 |
encoded properly. Half precision types are accepted, but not encoded. |
267 |
|
268 |
Strict mode and canonical mode are not implemented. |
269 |
|
270 |
THREADS |
271 |
This module is *not* guaranteed to be thread safe and there are no plans |
272 |
to change this until Perl gets thread support (as opposed to the |
273 |
horribly slow so-called "threads" which are simply slow and bloated |
274 |
process simulations - use fork, it's *much* faster, cheaper, better). |
275 |
|
276 |
(It might actually work, but you have been warned). |
277 |
|
278 |
BUGS |
279 |
While the goal of this module is to be correct, that unfortunately does |
280 |
not mean it's bug-free, only that I think its design is bug-free. If you |
281 |
keep reporting bugs they will be fixed swiftly, though. |
282 |
|
283 |
Please refrain from using rt.cpan.org or any other bug reporting |
284 |
service. I put the contact address into my modules for a reason. |
285 |
|
286 |
SEE ALSO |
287 |
The JSON and JSON::XS modules that do similar, but human-readable, |
288 |
serialisation. |
289 |
|
290 |
AUTHOR |
291 |
Marc Lehmann <schmorp@schmorp.de> |
292 |
http://home.schmorp.de/ |
293 |
|