|
Line 0
Link Here
|
| 0 |
- |
1 |
package Koha::ExternalContent::OCLC; |
|
|
2 |
|
| 3 |
# This file is part of Koha. |
| 4 |
# |
| 5 |
# Copyright (C) 2025 Dataly Tech |
| 6 |
# |
| 7 |
# Koha is free software; you can redistribute it and/or modify it |
| 8 |
# under the terms of the GNU General Public License as published by |
| 9 |
# the Free Software Foundation; either version 3 of the License, or |
| 10 |
# (at your option) any later version. |
| 11 |
# |
| 12 |
# Koha is distributed in the hope that it will be useful, but |
| 13 |
# WITHOUT ANY WARRANTY; without even the implied warranty of |
| 14 |
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
| 15 |
# GNU General Public License for more details. |
| 16 |
# |
| 17 |
# You should have received a copy of the GNU General Public License |
| 18 |
# along with Koha; if not, see <http://www.gnu.org/licenses>. |
| 19 |
|
| 20 |
use Modern::Perl; |
| 21 |
|
| 22 |
use C4::Context; |
| 23 |
|
| 24 |
use Koha::Caches; |
| 25 |
|
| 26 |
use Encode qw( decode ); |
| 27 |
use JSON qw( from_json ); |
| 28 |
use LWP::UserAgent; |
| 29 |
use MIME::Base64 qw( encode_base64 ); |
| 30 |
|
| 31 |
=head1 NAME |
| 32 |
|
| 33 |
Koha::ExternalContent::OCLC |
| 34 |
|
| 35 |
=head1 SYNOPSIS |
| 36 |
|
| 37 |
A class that provides helper functions related to OCLC's APIs. |
| 38 |
|
| 39 |
For more information on these APIs, please visit the following pages: |
| 40 |
|
| 41 |
L<https://www.oclc.org/developer/api/oclc-apis.en.html> |
| 42 |
|
| 43 |
L<https://developer.api.oclc.org/> |
| 44 |
|
| 45 |
=head1 DESCRIPTION |
| 46 |
|
| 47 |
This class contains common functions used when accessing OCLC's APIs. |
| 48 |
|
| 49 |
Examples of such APIs include, but are not limited to: |
| 50 |
|
| 51 |
=over |
| 52 |
|
| 53 |
=item * Citations API |
| 54 |
|
| 55 |
=item * Dewey Linked Data API |
| 56 |
|
| 57 |
=item * Dewey Lookup API |
| 58 |
|
| 59 |
=item * Discovery Library Profiles API |
| 60 |
|
| 61 |
=item * Entity Search API |
| 62 |
|
| 63 |
=item * FAST API |
| 64 |
|
| 65 |
=item * New Titles API |
| 66 |
|
| 67 |
=item * Suggestions API |
| 68 |
|
| 69 |
=item * VIAF API |
| 70 |
|
| 71 |
=item * WorldCat Discovery API |
| 72 |
|
| 73 |
=item * WorldCat Entity Data |
| 74 |
|
| 75 |
=item * WorldCat Metadata API |
| 76 |
|
| 77 |
=back |
| 78 |
|
| 79 |
Currently, this class has functions that allow use of the Dewey Linked Data and Dewey Lookup APIs. |
| 80 |
|
| 81 |
=head1 FUNCTIONS |
| 82 |
|
| 83 |
=head3 get_access_token |
| 84 |
|
| 85 |
my $access_token = get_access_token( |
| 86 |
{ |
| 87 |
scope => $scope, |
| 88 |
client_id => $client_id, |
| 89 |
client_secret => $client_secret, |
| 90 |
} |
| 91 |
); |
| 92 |
|
| 93 |
Requests an API access token from OCLC's OAuth 2.0 server, for a particular scope (such as C<deweyLinkedData>, C<WorldCatMetadataAPI>, etc.). |
| 94 |
|
| 95 |
Tokens are valid for a specific time period, and to ensure re-usability within this timeframe they will be cached (in C<memcached>) using a unique key per scope and with an appropriate expiry set. |
| 96 |
|
| 97 |
This request follows the Client Credentials Grant OAuth pattern, documented here: L<https://www.oclc.org/developer/api/keys/oauth/client-credentials-grant.en.html> |
| 98 |
|
| 99 |
Essentially, an HTTP POST request to C<https://oauth.oclc.org/token> is performed, with the Base64-encoded Client ID and Client Secret components included in an C<Authorization: Basic> header. |
| 100 |
|
| 101 |
WSKeys (or Web Service Keys) are documented here: L<https://www.oclc.org/developer/develop/authentication/what-is-a-wskey.en.html> |
| 102 |
|
| 103 |
Instructions on how to request such a key can be found here: L<https://help.oclc.org/Librarian_Toolbox/OCLC_APIs/Get_started/Request_an_API_WSKey> |
| 104 |
|
| 105 |
=head4 Parameters |
| 106 |
|
| 107 |
=over 4 |
| 108 |
|
| 109 |
=item C<options> (required) - a hashref containing the following keys: |
| 110 |
|
| 111 |
=over 8 |
| 112 |
|
| 113 |
=item C<scope> (string, required) - the service (i.e. scope) for which the client is requesting access (e.g. C<deweyLinkedData>, C<WorldCatMetadataAPI>, etc.) |
| 114 |
|
| 115 |
=item C<client_id> (string, optional) - the Client ID component of your WSKey (falls back to the contents of the C<OCLCAPIWSKeyClientID> System Preference) |
| 116 |
|
| 117 |
=item C<client_secret> (string, optional) - the Client Secret component of your WSKey (falls back to the contents of the C<OCLCAPIWSKeyClientSecret> System Preference) |
| 118 |
|
| 119 |
=back |
| 120 |
|
| 121 |
=back |
| 122 |
|
| 123 |
=head4 Returns |
| 124 |
|
| 125 |
=over 4 |
| 126 |
|
| 127 |
=item C<access_token> (string) - the API access token, or C<undef> if no token could be obtained |
| 128 |
|
| 129 |
=item C<msg> (string) - the source of the token (API, cache), or the error message returned if no token could be obtained |
| 130 |
|
| 131 |
=back |
| 132 |
|
| 133 |
=head4 Related System Preferences |
| 134 |
|
| 135 |
=over 4 |
| 136 |
|
| 137 |
=item C<OCLCAPIWSKeyClientID> |
| 138 |
|
| 139 |
=item C<OCLCAPIWSKeyClientSecret> |
| 140 |
|
| 141 |
=back |
| 142 |
|
| 143 |
=cut |
| 144 |
|
| 145 |
sub get_access_token { |
| 146 |
|
| 147 |
my ( $self, $args ) = @_; |
| 148 |
|
| 149 |
my $scope = $args->{scope}; |
| 150 |
my $client_id = $args->{client_id} ||= C4::Context->preference('OCLCAPIWSKeyClientID'); |
| 151 |
my $client_secret = $args->{client_secret} ||= C4::Context->preference('OCLCAPIWSKeyClientSecret'); |
| 152 |
my $cache = Koha::Caches->get_instance; |
| 153 |
|
| 154 |
my $cached_access_token = $cache->get_from_cache( 'OCLC_access_token-' . $scope ); |
| 155 |
if ($cached_access_token) { |
| 156 |
return ( $cached_access_token->{token}, 'cache' ); |
| 157 |
} else { |
| 158 |
my $basic_auth = encode_base64( "$client_id:$client_secret", '' ); |
| 159 |
my $ua = LWP::UserAgent->new; |
| 160 |
my $method = 'POST'; |
| 161 |
my $uri = "https://oauth.oclc.org/token?grant_type=client_credentials&scope=$scope"; |
| 162 |
my $header = [ 'Authorization' => "Basic $basic_auth" ]; |
| 163 |
my $request = HTTP::Request->new( $method, $uri, $header ); |
| 164 |
my $response = $ua->request($request); |
| 165 |
my $content = Encode::decode( "utf8", $response->content ); |
| 166 |
my $json = from_json($content); |
| 167 |
|
| 168 |
if ( $response->is_success ) { |
| 169 |
$cache->set_in_cache( |
| 170 |
"OCLC_access_token-$scope", |
| 171 |
{ |
| 172 |
token => $json->{access_token}, |
| 173 |
expires_at => $json->{expires_at}, |
| 174 |
}, |
| 175 |
{ expiry => $json->{expires_in} } |
| 176 |
); |
| 177 |
return ( $json->{access_token}, 'API' ); |
| 178 |
} else { |
| 179 |
return ( undef, $json->{code} . ' ' . $json->{message} ); |
| 180 |
} |
| 181 |
} |
| 182 |
|
| 183 |
} |
| 184 |
|
| 185 |
=head3 get_ddc_uri |
| 186 |
|
| 187 |
my $ddc_uri = get_ddc_uri( |
| 188 |
{ |
| 189 |
access_token => $access_token, |
| 190 |
ddc_number => $ddc_number, |
| 191 |
} |
| 192 |
); |
| 193 |
|
| 194 |
Requests a Linked Data ID (URI) for a particular DDC number from OCLC's Dewey Linked Data API. |
| 195 |
|
| 196 |
The API is documented here: L<https://developer.api.oclc.org/dewey-ld> |
| 197 |
|
| 198 |
=head4 Parameters |
| 199 |
|
| 200 |
=over 4 |
| 201 |
|
| 202 |
=item C<options> (required) - a hashref containing the following keys: |
| 203 |
|
| 204 |
=over 8 |
| 205 |
|
| 206 |
=item C<access_token> (string, required) - the access_token to use when making the API request |
| 207 |
|
| 208 |
=item C<ddc_number> (string, required) - the DDC number to look up |
| 209 |
|
| 210 |
=back |
| 211 |
|
| 212 |
=back |
| 213 |
|
| 214 |
=head4 Returns |
| 215 |
|
| 216 |
=over 4 |
| 217 |
|
| 218 |
=item C<uri> (string) - the Linked Data ID (URI) associated with the DDC number requested, or C<undef> if no URI could be obtained |
| 219 |
|
| 220 |
=item C<msg> (string) - the source of the URI (API, cache), or the error message returned if no token could be obtained |
| 221 |
|
| 222 |
=back |
| 223 |
|
| 224 |
=cut |
| 225 |
|
| 226 |
sub get_ddc_uri { |
| 227 |
|
| 228 |
my ( $self, $args ) = @_; |
| 229 |
|
| 230 |
my $access_token = $args->{access_token}; |
| 231 |
my $ddc_number = $args->{ddc_number}; |
| 232 |
my $cache = Koha::Caches->get_instance; |
| 233 |
|
| 234 |
my $cached_uri = $cache->get_from_cache("$ddc_number-uri"); |
| 235 |
if ($cached_uri) { |
| 236 |
return ( $cached_uri, 'cache' ); |
| 237 |
} else { |
| 238 |
my $ua = LWP::UserAgent->new; |
| 239 |
my $method = 'GET'; |
| 240 |
my $url = "https://id.oclc.org/worldcat/ddc/api/id?ddc=$ddc_number"; |
| 241 |
my $header = [ 'Authorization' => "Bearer $access_token" ]; |
| 242 |
my $request = HTTP::Request->new( $method, $url, $header ); |
| 243 |
my $response = $ua->request($request); |
| 244 |
my $content = Encode::decode( "utf8", $response->content ); |
| 245 |
my $json = from_json($content); |
| 246 |
|
| 247 |
if ( $response->is_success ) { |
| 248 |
if ( $json->{$ddc_number} ) { |
| 249 |
$cache->set_in_cache( "$ddc_number-uri", $json->{$ddc_number} ); |
| 250 |
return ( $json->{$ddc_number}, 'API' ); |
| 251 |
} else { |
| 252 |
return ( undef, 'not found' ); |
| 253 |
} |
| 254 |
} else { |
| 255 |
return ( undef, $json->{message} ); |
| 256 |
} |
| 257 |
} |
| 258 |
|
| 259 |
} |
| 260 |
|
| 261 |
=head3 get_ddc_url |
| 262 |
|
| 263 |
my $ddc_url = get_ddc_url( |
| 264 |
{ |
| 265 |
access_token => $access_token, |
| 266 |
ddc_number => $ddc_number, |
| 267 |
} |
| 268 |
); |
| 269 |
|
| 270 |
Requests a Linked Data URL for a particular DDC number from OCLC's Dewey Linked Data API. |
| 271 |
|
| 272 |
The API is documented here: L<https://developer.api.oclc.org/dewey-ld> |
| 273 |
|
| 274 |
=head4 Parameters |
| 275 |
|
| 276 |
=over 4 |
| 277 |
|
| 278 |
=item C<options> (required) - a hashref containing the following keys: |
| 279 |
|
| 280 |
=over 8 |
| 281 |
|
| 282 |
=item C<access_token> (string, required) - the access_token to use when making the API request |
| 283 |
|
| 284 |
=item C<ddc_number> (string, required) - the DDC number to look up |
| 285 |
|
| 286 |
=back |
| 287 |
|
| 288 |
=back |
| 289 |
|
| 290 |
=head4 Returns |
| 291 |
|
| 292 |
=over 4 |
| 293 |
|
| 294 |
=item C<url> (string) - the Linked Data URL associated with the DDC number requested, or C<undef> if no URL could be obtained |
| 295 |
|
| 296 |
=item C<msg> (string) - the source of the URL (API, cache), or the error message returned if no token could be obtained |
| 297 |
|
| 298 |
=back |
| 299 |
|
| 300 |
=cut |
| 301 |
|
| 302 |
sub get_ddc_url { |
| 303 |
|
| 304 |
my ( $self, $args ) = @_; |
| 305 |
|
| 306 |
my $access_token = $args->{access_token}; |
| 307 |
my $ddc_number = $args->{ddc_number}; |
| 308 |
my $cache = Koha::Caches->get_instance; |
| 309 |
|
| 310 |
my $cached_url = $cache->get_from_cache("$ddc_number-url"); |
| 311 |
if ($cached_url) { |
| 312 |
return ( $cached_url, 'cache' ); |
| 313 |
} else { |
| 314 |
my $ua = LWP::UserAgent->new; |
| 315 |
my $method = 'GET'; |
| 316 |
my $url = "https://id.oclc.org/worldcat/ddc/api/url?ddc=$ddc_number"; |
| 317 |
my $header = [ 'Authorization' => "Bearer $access_token" ]; |
| 318 |
my $request = HTTP::Request->new( $method, $url, $header ); |
| 319 |
my $response = $ua->request($request); |
| 320 |
my $content = Encode::decode( "utf8", $response->content ); |
| 321 |
my $json = from_json($content); |
| 322 |
|
| 323 |
if ( $response->is_success ) { |
| 324 |
$cache->set_in_cache( "$ddc_number-url", $json->{$ddc_number} ); |
| 325 |
return ( $json->{$ddc_number}, 'API' ); |
| 326 |
} else { |
| 327 |
return ( undef, $json->{message} ); |
| 328 |
} |
| 329 |
} |
| 330 |
|
| 331 |
} |
| 332 |
|
| 333 |
=head3 get_ddc_description |
| 334 |
|
| 335 |
my $ddc_description = get_ddc_description( |
| 336 |
{ |
| 337 |
access_token => $access_token, |
| 338 |
ddc_uri => $ddc_uri, |
| 339 |
[ language => $language, ] |
| 340 |
} |
| 341 |
); |
| 342 |
|
| 343 |
Requests a Linked Data Description for a particular DDC number from OCLC's Dewey Linked Data API. |
| 344 |
|
| 345 |
The API is documented here: L<https://developer.api.oclc.org/dewey-ld> |
| 346 |
|
| 347 |
=head4 Parameters |
| 348 |
|
| 349 |
=over 4 |
| 350 |
|
| 351 |
=item C<options> (required) - a hashref containing the following keys: |
| 352 |
|
| 353 |
=over 8 |
| 354 |
|
| 355 |
=item C<access_token> (string, required) - the access_token to use when making the API request |
| 356 |
|
| 357 |
=item C<ddc_uri> (string, required) - the DDC URI to look up |
| 358 |
|
| 359 |
=item C<language> (string, optional) - the language of the description to obtain (falls back to the setting of the C<OCLCDeweyLinkedDataLanguage> System Preference) |
| 360 |
|
| 361 |
=back |
| 362 |
|
| 363 |
=back |
| 364 |
|
| 365 |
=head4 Returns |
| 366 |
|
| 367 |
=over 4 |
| 368 |
|
| 369 |
=item C<description> (string) - the Linked Data Description associated with the DDC URI requested, or C<undef> if no Description could be obtained |
| 370 |
|
| 371 |
=item C<msg> (string) - the source of the LD Description (API, cache), or the error message returned if no token could be obtained |
| 372 |
|
| 373 |
=back |
| 374 |
|
| 375 |
=head4 Related System Preferences |
| 376 |
|
| 377 |
=over 4 |
| 378 |
|
| 379 |
=item C<OCLCDeweyLinkedDataLanguage> |
| 380 |
|
| 381 |
=back |
| 382 |
|
| 383 |
=cut |
| 384 |
|
| 385 |
sub get_ddc_description { |
| 386 |
|
| 387 |
my ( $self, $args ) = @_; |
| 388 |
|
| 389 |
my $access_token = $args->{access_token}; |
| 390 |
my $ddc_uri = $args->{ddc_uri}; |
| 391 |
my $language = $args->{language} ||= C4::Context->preference('OCLCDeweyLinkedDataLanguage'); |
| 392 |
my $cache = Koha::Caches->get_instance; |
| 393 |
|
| 394 |
my $cached_description = $cache->get_from_cache("$ddc_uri-description-$language"); |
| 395 |
if ($cached_description) { |
| 396 |
return ( $cached_description, 'cache' ); |
| 397 |
} else { |
| 398 |
my $ua = LWP::UserAgent->new; |
| 399 |
my $method = 'GET'; |
| 400 |
my $url = "https://id.oclc.org/worldcat/ddc/$ddc_uri"; |
| 401 |
my $header = [ 'Authorization' => "Bearer $access_token" ]; |
| 402 |
my $request = HTTP::Request->new( $method, $url, $header ); |
| 403 |
my $response = $ua->request($request); |
| 404 |
my $content = Encode::decode( "utf8", $response->content ); |
| 405 |
my $json = from_json($content); |
| 406 |
|
| 407 |
if ( $response->is_success ) { |
| 408 |
|
| 409 |
# Some DDC numbers (e.g. 348.481) do not have descriptions in all |
| 410 |
# possible languages. We take that into account here. |
| 411 |
my $description = $json->{'prefLabel'}->{$language} || "No description found for language code '$language'"; |
| 412 |
$cache->set_in_cache( "$ddc_uri-description-$language", $description ); |
| 413 |
return ( $description, 'API' ); |
| 414 |
|
| 415 |
} else { |
| 416 |
return ( undef, $json->{message} ); |
| 417 |
} |
| 418 |
} |
| 419 |
|
| 420 |
} |
| 421 |
|
| 422 |
1; |