1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377
|
package JSON::Validator::Util;
use Mojo::Base -strict;
use B;
use Carp ();
use Exporter 'import';
use JSON::Validator::Error;
use List::Util;
use Mojo::Collection;
use Mojo::JSON;
use Mojo::Loader;
use Mojo::Util;
use Scalar::Util 'blessed';
use constant SEREAL_SUPPORT => !$ENV{JSON_VALIDATOR_NO_SEREAL} && eval 'use Sereal::Encoder 4.00;1';
our @EXPORT_OK = (
qw(E data_checksum data_section data_type is_type negotiate_content_type),
qw(schema_extract json_pointer prefix_errors schema_type),
);
sub E { JSON::Validator::Error->new(@_) }
my $serializer = SEREAL_SUPPORT ? \&_sereal_encode : \&_yaml_dump;
sub data_checksum {
return Mojo::Util::md5_sum(ref $_[0] ? $serializer->($_[0]) : defined $_[0] ? qq('$_[0]') : 'undef');
}
sub data_section {
my ($class, $file, $params) = @_;
state $skip_re = qr{(^JSON::Validator|^Mojo::Base$|^Mojolicious$|\w+::_Dynamic)};
my @classes = $class ? ([$class]) : ();
unless (@classes) {
my $i = 0;
while ($class = caller($i++)) {
push @classes, [$class] unless $class =~ $skip_re;
}
}
for my $group (@classes) {
push @$group, grep { !/$skip_re/ } do { no strict 'refs'; @{"$group->[0]\::ISA"} };
for my $class (@$group) {
next unless my $text = Mojo::Loader::data_section($class, $file);
return Mojo::Util::encode($params->{encoding}, $text) if $params->{encoding};
return $text;
}
}
return undef unless $params->{confess};
my $err = Mojo::JSON::encode_json([map { @$_ == 1 ? $_->[0] : $_ } @classes]);
Carp::confess(qq(Could not find "$file" in __DATA__ section of $err.));
}
sub data_type {
my $ref = ref $_[0];
my $blessed = blessed $_[0];
return 'object' if $ref eq 'HASH';
return lc $ref if $ref and !$blessed;
return 'null' if !defined $_[0];
return 'boolean' if $blessed and ("$_[0]" eq "1" or !"$_[0]");
if (is_type($_[0], 'NUM')) {
return 'integer' if grep { ($_->{type} // '') eq 'integer' } @{$_[1] || []};
return 'number';
}
return $blessed || 'string';
}
sub is_type {
my $type = $_[1];
if ($type eq 'BOOL') {
return blessed $_[0] && ($_[0]->isa('JSON::PP::Boolean') || "$_[0]" eq "1" || !$_[0]);
}
# NUM
if ($type eq 'NUM') {
return B::svref_2object(\$_[0])->FLAGS & (B::SVp_IOK | B::SVp_NOK) && 0 + $_[0] eq $_[0] && $_[0] * 0 == 0;
}
# Class or data type
return blessed $_[0] ? $_[0]->isa($type) : ref $_[0] eq $type;
}
sub negotiate_content_type {
my ($accepts, $header) = @_;
return '' unless $header;
my %header_map;
/^\s*([^,; ]+)(?:\s*\;\s*q\s*=\s*(\d+(?:\.\d+)?))?\s*$/i and $header_map{lc $1} = $2 // 1 for split /,/, $header;
my @headers = sort { $header_map{$b} <=> $header_map{$a} } sort keys %header_map;
# Check for exact match
for my $ct (@$accepts) {
return $ct if $header_map{$ct};
}
# Check for closest match
for my $re (map { my $re = "$_"; $re =~ s!\*!.*!g; $re = qr{$re}; [$_, $re] } grep {/\*/} @$accepts) {
for my $ct (@headers) {
return $re->[0] if $ct =~ $re->[1];
}
}
for my $re (map { local $_ = "$_"; s!\*!.*!g; qr{$_} } grep {/\*/} @headers) {
for my $ct (@$accepts) {
return $ct if $ct =~ $re;
}
}
# Could not find any valid content type
return '';
}
sub schema_extract {
my ($data, $p, $cb) = @_;
$p = [ref $p ? @$p : length $p ? split('/', $p, -1) : $p];
shift @$p if @$p and defined $p->[0] and !length $p->[0];
_schema_extract($data, $p, '', $cb);
}
sub json_pointer {
local $_ = $_[1];
s!~!~0!g;
s!/!~1!g;
"$_[0]/$_";
}
sub prefix_errors {
my ($type, @errors_with_index) = @_;
my @errors;
for my $e (@errors_with_index) {
my $index = shift @$e;
push @errors, map {
my $msg = sprintf '/%s/%s %s', $type, $index, $_->message;
$msg =~ s!(\d+)\s/!$1/!g;
E +{%$_, message => $msg}; # preserve 'details', for later introspection
} @$e;
}
return @errors;
}
sub schema_type {
return '' if ref $_[0] ne 'HASH';
return $_[0]->{type} if $_[0]->{type};
return _guessed_right(object => $_[1]) if $_[0]->{additionalProperties};
return _guessed_right(object => $_[1]) if $_[0]->{patternProperties};
return _guessed_right(object => $_[1]) if $_[0]->{properties};
return _guessed_right(object => $_[1]) if exists $_[0]->{propertyNames};
return _guessed_right(object => $_[1]) if $_[0]->{required};
return _guessed_right(object => $_[1])
if $_[0]->{dependencies}
or $_[0]->{dependentSchemas}
or $_[0]->{dependentRequired};
return _guessed_right(object => $_[1]) if defined $_[0]->{maxProperties} or defined $_[0]->{minProperties};
# additionalItems is intentionally omitted - it requires 'items' to take effect
return _guessed_right(array => $_[1]) if exists $_[0]->{items};
return _guessed_right(array => $_[1]) if $_[0]->{uniqueItems};
return _guessed_right(array => $_[1]) if exists $_[0]->{contains};
return _guessed_right(array => $_[1]) if exists $_[0]->{maxItems} or exists $_[0]->{minItems};
return _guessed_right(string => $_[1]) if $_[0]->{pattern};
return _guessed_right(string => $_[1]) if exists $_[0]->{maxLength} or defined $_[0]->{minLength};
return _guessed_right(number => $_[1]) if $_[0]->{multipleOf};
return _guessed_right(number => $_[1])
if defined $_[0]->{maximum}
or defined $_[0]->{minimum}
or defined $_[0]->{exclusiveMaximum}
or defined $_[0]->{exclusiveMinimum};
return 'const' if exists $_[0]->{const};
return '';
}
# _guessed_right($type, $data);
sub _guessed_right {
return $_[0] if !defined $_[1];
return $_[0] if $_[0] eq data_type $_[1], [{type => $_[0]}];
return '';
}
sub _schema_extract {
my ($data, $path, $pos, $cb) = @_, my $tied;
while (@$path) {
my $p = shift @$path;
unless (defined $p) {
my $i = 0;
return Mojo::Collection->new(map { _schema_extract($_->[0], [@$path], json_pointer($pos, $_->[1]), $cb) }
ref $data eq 'ARRAY' ? map { [$_, $i++] }
@$data : ref $data eq 'HASH' ? map { [$data->{$_}, $_] } sort keys %$data : [$data, '']);
}
$p =~ s!~1!/!g;
$p =~ s/~0/~/g;
$pos = json_pointer $pos, $p if $cb;
if (ref $data eq 'HASH' and exists $data->{$p}) {
$data = $data->{$p};
}
elsif (ref $data eq 'ARRAY' and $p =~ /^\d+$/ and @$data > $p) {
$data = $data->[$p];
}
else {
return undef;
}
$data = $tied->schema while ref $data eq 'HASH' and $tied = tied %$data;
}
return $cb->($data, $pos) if $cb;
return $data;
}
sub _sereal_encode {
state $s = Sereal::Encoder->new({canonical => 1});
return $s->encode($_[0]);
}
BEGIN {
if (eval 'use YAML::XS 0.67;1') {
*_yaml_dump = sub { local $YAML::XS::Boolean = 'JSON::PP'; YAML::XS::Dump(@_) };
*_yaml_load = sub { local $YAML::XS::Boolean = 'JSON::PP'; YAML::XS::Load(@_) };
}
else {
require YAML::PP;
my $pp = YAML::PP->new(boolean => 'JSON::PP');
*_yaml_dump = sub { $pp->dump_string(@_) };
*_yaml_load = sub { $pp->load_string(@_) };
}
}
1;
=encoding utf8
=head1 NAME
JSON::Validator::Util - Utility functions for JSON::Validator
=head1 DESCRIPTION
L<JSON::Validator::Util> is a package containing utility functions for
L<JSON::Validator>. Each of the L</FUNCTIONS> can be imported.
=head1 FUNCTIONS
=head2 data_checksum
$str = data_checksum $any;
Will create a checksum for any data structure stored in C<$any>.
=head2 data_section
$str = data_section "Some::Module", "file.json";
$str = data_section "Some::Module", "file.json", {encode => 'UTF-8'};
Same as L<Mojo::Loader/data_section>, but will also look up the file in any
inherited class.
=head2 data_type
$str = data_type $any;
$str = data_type $any, [@schemas];
$str = data_type $any, [{type => "integer", ...}];
Returns the JSON type for C<$any>. C<$str> can be array, boolean, integer,
null, number object or string. Note that a list of schemas need to be provided
to differentiate between "integer" and "number".
=head2 is_type
$bool = is_type $any, $class;
$bool = is_type $any, $type; # $type = "ARRAY", "BOOL", "HASH", "NUM" ...
Checks if C<$any> is a, or inherits from, C<$class> or C<$type>. Two special
types can be checked:
=over 2
=item * BOOL
Checks if C<$any> is a boolean value. C<$any> is considered boolean if it is an
object inheriting from L<JSON::PP::Boolean> or is another object that
stringifies to "1" or "0".
=item * NUM
Checks if C<$any> is indeed a number.
=back
=head2 json_pointer
$str = json_pointer $path, $append;
Will concat C<$append> on to C<$path>, but will also escape the two special
characters "~" and "/" in C<$append>.
=head2 negotiate_content_type
$content_type = negotiate_content_type($header, \@content_types);
This method can take a "Content-Type" or "Accept" header and find the closest
matching content type in C<@content_types>. C<@content_types> can contain
wildcards, meaning "*/*" will match anything.
=head2 prefix_errors
@errors = prefix_errors $prefix, @errors;
Consider this internal for now.
=head2 schema_extract
$data = schema_extract $any, $json_pointer;
$data = schema_extract $any, "/x/cool_beans/y";
$collection = schema_extract $any, ["x", undef, "y"];
schema_extract $any, $json_pointer, sub { my ($data, $json_pointer) = @_ };
The basic usage is to extract data from C<$any>, using a C<$json_pointer> -
L<RFC 6901|http://tools.ietf.org/html/rfc6901>. It can however be used in a
more complex way by passing in an array-ref, instead of a plain string. The
array-ref can contain C<undef()> values, will result in extracting any element
on that point, regardsless of value. In that case a L<Mojo::Collection> will
be returned.
A callback can also be given. This callback will be called each time the
C<$json_pointer> matches some data, and will pass in the C<$json_pointer> at
that place.
In addition, if the C<$json_pointer> points to a L<JSON::Validator::Ref> at any
point, the "$ref" will be followed, while if you used L<Mojo::JSON::Pointer>,
it would return either the L<JSON::Validator::Ref> or C<undef()>.
Even though L</schema_extract> has special capabilities for handling a
JSON-Schema, it can be used for any data-structure, just like
L<Mojo::JSON::Pointer>.
=head2 schema_type
$str = schema_type $hash_ref;
$str = schema_type $hash_ref, $any;
Looks at C<$hash_ref> and tries to figure out what kind of type the schema
represents. C<$str> can be "array", "const", "number", "object", "string", or
fallback to empty string if the correct type could not be figured out.
C<$any> can be provided to double check the type, so if C<$hash_ref> describes
an "object", but C<$any> is an array-ref, then C<$str> will become an empty
string. Example:
# $str = "";
$str = schema {additionalProperties => false}, [];
# $str = "object"
$str = schema {additionalProperties => false};
$str = schema {additionalProperties => false}, {};
Note that this process is relatively slow, so it will make your validation
faster if you specify "type". Both of the two below is valid, but the one with
"type" will be faster.
{"type": "object", "properties": {}} # Faster
{"properties": {}} # Slower
=head1 SEE ALSO
L<JSON::Validator>.
=cut
|