|
1
|
|
|
<?php |
|
2
|
|
|
/* |
|
3
|
|
|
* This file is part of dispositif/wikibot application (@github) |
|
4
|
|
|
* 2019-2023 © Philippe M./Irønie <[email protected]> |
|
5
|
|
|
* For the full copyright and MIT license information, view the license file. |
|
6
|
|
|
*/ |
|
7
|
|
|
|
|
8
|
|
|
declare(strict_types=1); |
|
9
|
|
|
|
|
10
|
|
|
namespace App\Domain\Publisher; |
|
11
|
|
|
|
|
12
|
|
|
use App\Domain\Utils\ArrayProcessTrait; |
|
13
|
|
|
use DomainException; |
|
14
|
|
|
use Exception; |
|
15
|
|
|
|
|
16
|
|
|
/** |
|
17
|
|
|
* Static methods for Google Books URL parsing. |
|
18
|
|
|
* identique à https://www.google.fr/books/edition/_/43cIAQAAMAAJ?gbpv=1&dq=orgues+basilique+saint+quentin |
|
19
|
|
|
*/ |
|
20
|
|
|
abstract class GoogleBooksUtil |
|
21
|
|
|
{ |
|
22
|
|
|
use ArrayProcessTrait; |
|
23
|
|
|
|
|
24
|
|
|
final public const DEFAULT_GOOGLEBOOKS_URL = 'https://books.google.com/books'; |
|
25
|
|
|
/** |
|
26
|
|
|
* todo refac regex with end of URL |
|
27
|
|
|
*/ |
|
28
|
|
|
final public const GOOGLEBOOKS_START_URL_PATTERN = '(?:https?://(?:books|play)\.google\.[a-z\.]{2,6}/(?:books)?(?:books/[^\?]+\.html)?(?:/reader)?\?(?:[a-zA-Z=&]+&)?(?:[&=A-Z0-9-_%\+]+&)?(?:id|isbn)=|https://www\.google\.[a-z\.]{2,6}/books/edition/[^/]+/)'; |
|
29
|
|
|
|
|
30
|
|
|
final public const GOOGLEBOOKS_NEW_START_URL_PATTERN = 'https://www\.google\.[a-z.]{2,6}/books/edition/[^/]+/'; |
|
31
|
|
|
|
|
32
|
|
|
final public const GOOGLEBOOKS_ID_REGEX = '[0-9A-Za-z_\-]{12}'; |
|
33
|
|
|
|
|
34
|
|
|
/** |
|
35
|
|
|
* todo : add frontcover ? |
|
36
|
|
|
* q : keywords search (may be empty) / dq : quoted phrase search |
|
37
|
|
|
*/ |
|
38
|
|
|
final public const GOOGLEBOOKS_KEEP_PARAMETERS = ['id', 'isbn', 'pg', 'printsec', 'q', 'dq', 'gbpv']; |
|
39
|
|
|
|
|
40
|
|
|
final public const TRACKING_PARAMETERS = [ |
|
41
|
|
|
'xtor', |
|
42
|
|
|
'ved', |
|
43
|
|
|
'ots', |
|
44
|
|
|
'sig', |
|
45
|
|
|
'source', |
|
46
|
2 |
|
'utm_source', |
|
47
|
|
|
'utm_medium', |
|
48
|
2 |
|
'utm_campaign', |
|
49
|
2 |
|
'utm_term', |
|
50
|
2 |
|
'utm_content', |
|
51
|
2 |
|
]; |
|
52
|
|
|
|
|
53
|
|
|
/** |
|
54
|
|
|
* Check if URL contains tracking parameters. |
|
55
|
1 |
|
*/ |
|
56
|
|
|
public static function isTrackingUrl(string $url): bool |
|
57
|
|
|
{ |
|
58
|
|
|
$urlData = self::parseGoogleBookQuery($url); |
|
59
|
|
|
|
|
60
|
|
|
return !empty(array_intersect_key(array_flip(self::TRACKING_PARAMETERS), $urlData)); |
|
61
|
|
|
} |
|
62
|
|
|
|
|
63
|
|
|
/** |
|
64
|
|
|
* Parse URL argument from ?query and #fragment. |
|
65
|
22 |
|
* Do not remove empty values. |
|
66
|
|
|
*/ |
|
67
|
|
|
public static function parseGoogleBookQuery(string $url): array |
|
68
|
22 |
|
{ |
|
69
|
22 |
|
$queryData = parse_url($url, PHP_URL_QUERY); // after ? |
|
70
|
|
|
$fragmentData = parse_url($url, PHP_URL_FRAGMENT); // after # |
|
71
|
22 |
|
// queryData precedence over fragmentData |
|
72
|
|
|
parse_str(implode('&', [$fragmentData, $queryData]), $urlData); |
|
73
|
22 |
|
|
|
74
|
|
|
return self::arrayKeysToLower($urlData); |
|
75
|
|
|
} |
|
76
|
|
|
|
|
77
|
|
|
/** |
|
78
|
|
|
* TODO refac (responsability). |
|
79
|
|
|
* |
|
80
|
|
|
* Clean the google book old URL : delete tracking and user optional params, |
|
81
|
|
|
* also redondat search query params. |
|
82
|
|
|
* Skip the process for new URL 2019 format. |
|
83
|
|
|
* |
|
84
|
13 |
|
* @throws Exception |
|
85
|
|
|
*/ |
|
86
|
13 |
|
public static function simplifyGoogleUrl(string $url): string |
|
87
|
|
|
{ |
|
88
|
|
|
if (!self::isGoogleBookURL($url)) { |
|
89
|
|
|
// not DomainException for live testing with OuvrageOptimize |
|
90
|
|
|
throw new Exception('not a Google Book URL'); |
|
91
|
13 |
|
} |
|
92
|
13 |
|
|
|
93
|
|
|
if (self::isNewGoogleBookUrl($url)) { |
|
94
|
|
|
if (!self::getIDFromNewGBurl($url)) { |
|
95
|
13 |
|
throw new DomainException('no Google Book ID in URL'); |
|
96
|
|
|
} |
|
97
|
|
|
|
|
98
|
|
|
return $url; |
|
99
|
13 |
|
} |
|
100
|
|
|
|
|
101
|
|
|
$gooDat = self::parseGoogleBookQuery($url); |
|
102
|
|
|
|
|
103
|
13 |
|
if (empty($gooDat['id']) && empty($gooDat['isbn'])) { |
|
104
|
13 |
|
throw new DomainException("no GoogleBook 'id' or 'isbn' in URL"); |
|
105
|
13 |
|
} |
|
106
|
13 |
|
if (isset($gooDat['id']) && !self::validateGoogleBooksId($gooDat['id'])) { |
|
107
|
|
|
throw new DomainException("GoogleBook 'id' malformed"); |
|
108
|
|
|
} |
|
109
|
|
|
|
|
110
|
|
|
$dat = self::parseAndCleanParams($gooDat); |
|
111
|
|
|
$googleURL = self::modifyGoogleDomainURL($url); |
|
112
|
|
|
|
|
113
|
|
|
// todo verify http_build_query() enc_type parameter |
|
114
|
|
|
// todo http_build_query() process an urlencode, but a not encoded q= value ("fu+bar") is beautiful |
|
115
|
|
|
return $googleURL . '?' . http_build_query($dat); |
|
116
|
|
|
} |
|
117
|
|
|
|
|
118
|
|
|
/** |
|
119
|
|
|
* Check google URL pattern. |
|
120
|
|
|
*/ |
|
121
|
|
|
public static function isGoogleBookURL(string $text): bool |
|
122
|
|
|
{ |
|
123
|
|
|
return preg_match('#^' . self::GOOGLEBOOKS_START_URL_PATTERN . '[^>\]} \n]+$#i', $text) > 0; |
|
124
|
|
|
} |
|
125
|
13 |
|
|
|
126
|
|
|
/** |
|
127
|
6 |
|
* Extract domain from google URL. |
|
128
|
4 |
|
* return '.fr', '.com,'.co.uk', '.co.ma' or null |
|
129
|
|
|
*/ |
|
130
|
|
|
private static function extractGoogleDomain(string $url): ?string |
|
131
|
2 |
|
{ |
|
132
|
|
|
$host = parse_url($url, PHP_URL_HOST); // "books.google.fr" |
|
133
|
|
|
if (!empty($host) && preg_match('#google((?:\.[a-z]{2,3})?\.[a-z]{2,3})$#', $host, $matches) > 0) { |
|
134
|
13 |
|
|
|
135
|
12 |
|
return $matches[1] ?? null; // .fr |
|
136
|
|
|
} |
|
137
|
13 |
|
|
|
138
|
8 |
|
return null; |
|
139
|
|
|
} |
|
140
|
|
|
|
|
141
|
13 |
|
/** |
|
142
|
|
|
* Google style url_encode(). No UTF-8 encoding. |
|
143
|
|
|
*/ |
|
144
|
13 |
|
public static function googleUrlEncode(string $str): string |
|
145
|
13 |
|
{ |
|
146
|
13 |
|
return str_replace(' ', '+', trim(urldecode($str))); |
|
147
|
|
|
} |
|
148
|
|
|
|
|
149
|
|
|
/** |
|
150
|
13 |
|
* New Google Books format (nov 2019). |
|
151
|
|
|
* Example : https://www.google.fr/books/edition/_/U4NmPwAACAAJ?hl=en |
|
152
|
|
|
*/ |
|
153
|
|
|
private static function isNewGoogleBookUrl(string $url): bool |
|
154
|
|
|
{ |
|
155
|
|
|
return (bool)preg_match( |
|
156
|
|
|
'#^' . self::GOOGLEBOOKS_NEW_START_URL_PATTERN . self::GOOGLEBOOKS_ID_REGEX . '(?:&.+)?#', |
|
157
|
|
|
$url |
|
158
|
|
|
); |
|
159
|
|
|
} |
|
160
|
24 |
|
|
|
161
|
|
|
/** |
|
162
|
24 |
|
* @param string[] $gooDat |
|
163
|
23 |
|
* |
|
164
|
|
|
* @return string[] |
|
165
|
|
|
*/ |
|
166
|
1 |
|
protected static function parseAndCleanParams(array $gooDat): array |
|
167
|
|
|
{ |
|
168
|
|
|
$dat = []; |
|
169
|
|
|
// keep only a few parameters (+'q' ?) |
|
170
|
|
|
// q : keywords search / dq : quoted phrase search |
|
171
|
|
|
// q can be empty !!!! |
|
172
|
|
|
foreach (self::GOOGLEBOOKS_KEEP_PARAMETERS as $keep) { |
|
173
|
|
|
if (isset($gooDat[$keep])) { |
|
174
|
|
|
$dat[$keep] = $gooDat[$keep]; |
|
175
|
|
|
} |
|
176
|
13 |
|
} |
|
177
|
|
|
|
|
178
|
13 |
|
// 1 exemple : https://fr.wikipedia.org/w/index.php?title=Foudre_de_Catatumbo&diff=next&oldid=168721836&diffmode=source |
|
179
|
13 |
|
// 1. mettre URL &dq= pour final |
|
180
|
|
|
// |
|
181
|
13 |
|
// 2. si q!=dq (changement ultérieur formulaire recherche) alors q= prévaut pour résultat final |
|
182
|
|
|
// 2. mettre URL &q pour final |
|
183
|
|
|
// |
|
184
|
|
|
// 3. Recherche global sur http://books.google.fr => pg= dq= (#q= avec q==dq) |
|
185
|
|
|
// 3. dans ce cas (q==dq), url final avec seulement dq= donne résultat OK |
|
186
|
|
|
// |
|
187
|
|
|
// 4 . if you use a url without &redir_esc=y#v=onepage for a book with "Preview" available, |
|
188
|
|
|
// usually &dq shows the highlighted text in full page view whereas &q shows the snippet view (so you have to |
|
189
|
|
|
// click on the snippet to see the full page). |
|
190
|
|
|
// &dq allows highlighting in books where there is "Preview" available and &pg=PTx is in the URL |
|
191
|
|
|
// |
|
192
|
|
|
// #v=onepage ou #v=snippet |
|
193
|
|
|
if (isset($dat['q']) && isset($dat['dq'])) { |
|
194
|
1 |
|
// si q==dq alors dq prévaut pour affichage (sinon affichage différent avec url seulement q=) |
|
195
|
|
|
if ($dat['q'] === $dat['dq']) { |
|
196
|
1 |
|
unset($dat['q']); |
|
197
|
|
|
} // si q!=dq (exemple : nouveaux mots clés dans formulaire recherche) alors q= prévaut pour résultat final |
|
198
|
|
|
else { |
|
199
|
|
|
unset($dat['dq']); |
|
200
|
|
|
} |
|
201
|
|
|
} |
|
202
|
|
|
if (empty($dat['q'])) { |
|
203
|
|
|
unset($dat['q']); |
|
204
|
|
|
} |
|
205
|
|
|
if (empty($dat['dq'])) { |
|
206
|
|
|
unset($dat['dq']); |
|
207
|
|
|
} |
|
208
|
|
|
|
|
209
|
|
|
return $dat; |
|
210
|
|
|
} |
|
211
|
|
|
|
|
212
|
|
|
/** |
|
213
|
|
|
* Changed : do not replace '.com' Googledomain name. This method is useless. |
|
214
|
|
|
* Naive replacement of Google domain name. |
|
215
|
|
|
*/ |
|
216
|
|
|
protected static function modifyGoogleDomainURL(string $url): string |
|
217
|
|
|
{ |
|
218
|
|
|
$defaultGoogleDomainURL = self::DEFAULT_GOOGLEBOOKS_URL; |
|
219
|
|
|
$gooDomain = self::extractGoogleDomain($url); // '.fr', '.co.uk'… |
|
220
|
|
|
|
|
221
|
|
|
if ($gooDomain) { |
|
222
|
|
|
$defaultGoogleDomainURL = str_replace('.com', $gooDomain, $defaultGoogleDomainURL); |
|
223
|
|
|
} |
|
224
|
|
|
|
|
225
|
|
|
return $defaultGoogleDomainURL; |
|
226
|
|
|
} |
|
227
|
|
|
|
|
228
|
|
|
/** |
|
229
|
|
|
* Extract ID from new Google Books URL. |
|
230
|
|
|
* https://www.google.fr/books/edition/_/U4NmPwAACAAJ?hl=en => U4NmPwAACAAJ |
|
231
|
|
|
*/ |
|
232
|
|
|
private static function getIDFromNewGBurl(string $url): ?string |
|
233
|
|
|
{ |
|
234
|
|
|
if (preg_match( |
|
235
|
|
|
'#^' . self::GOOGLEBOOKS_NEW_START_URL_PATTERN . '(' . self::GOOGLEBOOKS_ID_REGEX . ')(?:&.+)?#', |
|
236
|
|
|
$url, |
|
237
|
|
|
$matches |
|
238
|
|
|
) |
|
239
|
|
|
) { |
|
240
|
|
|
return $matches[1]; |
|
241
|
|
|
} |
|
242
|
|
|
|
|
243
|
|
|
return null; |
|
244
|
|
|
} |
|
245
|
|
|
|
|
246
|
|
|
protected static function validateGoogleBooksId(string $id): bool |
|
247
|
|
|
{ |
|
248
|
|
|
return preg_match('#' . self::GOOGLEBOOKS_ID_REGEX . '#', $id) > 0; |
|
249
|
|
|
} |
|
250
|
|
|
} |
|
251
|
|
|
|