|
1
|
|
|
<?php declare(strict_types=1); |
|
2
|
|
|
namespace html_go\indexing; |
|
3
|
|
|
|
|
4
|
|
|
use InvalidArgumentException; |
|
5
|
|
|
use html_go\exceptions\InternalException; |
|
6
|
|
|
|
|
7
|
|
|
abstract class AbstractIndexer |
|
8
|
|
|
{ |
|
9
|
|
|
protected string $parentDir; |
|
10
|
|
|
protected string $commonDir; |
|
11
|
|
|
protected string $userDataDir; |
|
12
|
|
|
|
|
13
|
|
|
protected string $pageInxFile; |
|
14
|
|
|
protected string $catInxFile; |
|
15
|
|
|
protected string $postInxFile; |
|
16
|
|
|
protected string $tagInxFile; |
|
17
|
|
|
protected string $tag2postInxFile; |
|
18
|
|
|
protected string $cat2postInxFile; |
|
19
|
|
|
protected string $menuInxFile; |
|
20
|
|
|
|
|
21
|
|
|
public function __construct(string $parentDir) { |
|
22
|
|
|
if (($path = \realpath($parentDir)) === false) { |
|
23
|
|
|
throw new InternalException("realpath() function failed on [$parentDir]"); // @codeCoverageIgnore |
|
24
|
|
|
} |
|
25
|
|
|
$this->parentDir = $path; |
|
26
|
|
|
|
|
27
|
|
|
$this->commonDir = $path.DS.'content'.DS.'common'; |
|
28
|
|
|
if (\is_dir($this->commonDir) === false) { |
|
29
|
|
|
throw new InvalidArgumentException("The content/common directory cannot be found [$this->commonDir]"); |
|
30
|
|
|
} |
|
31
|
|
|
|
|
32
|
|
|
$this->userDataDir = $path.DS.'content'.DS.'user-data'; |
|
33
|
|
|
if (\is_dir($this->userDataDir) === false) { |
|
34
|
|
|
throw new InvalidArgumentException("The content/user-data directory cannot be found [$this->userDataDir]"); |
|
35
|
|
|
} |
|
36
|
|
|
$indexDir = $path.DS.'cache'.DS.'indexes'; |
|
37
|
|
|
$this->pageInxFile = $indexDir.DS.'page.inx'; |
|
38
|
|
|
$this->catInxFile = $indexDir.DS.'category.inx'; |
|
39
|
|
|
$this->postInxFile = $indexDir.DS.'post.inx'; |
|
40
|
|
|
$this->tagInxFile = $indexDir.DS.'tag.inx'; |
|
41
|
|
|
$this->tag2postInxFile = $indexDir.DS.'tag2post.inx'; |
|
42
|
|
|
$this->cat2postInxFile = $indexDir.DS.'cat2post.inx'; |
|
43
|
|
|
$this->menuInxFile = $indexDir.DS.'menu.inx'; |
|
44
|
|
|
} |
|
45
|
|
|
|
|
46
|
|
|
/** |
|
47
|
|
|
* Load the given index file. |
|
48
|
|
|
* @param string $filename |
|
49
|
|
|
* @throws InternalException |
|
50
|
|
|
* @throws InvalidArgumentException |
|
51
|
|
|
* @return array<mixed> |
|
52
|
|
|
*/ |
|
53
|
|
|
protected function loadIndex(string $filename): array { |
|
54
|
|
|
if (\file_exists($filename) === false) { |
|
55
|
|
|
throw new InvalidArgumentException("Index file does not exist [$filename]. Call 'redindex()'"); // @codeCoverageIgnore |
|
56
|
|
|
} |
|
57
|
|
|
if (($data = \file_get_contents($filename)) === false) { |
|
58
|
|
|
throw new InternalException("file_get_contents() failed [$filename]"); // @codeCoverageIgnore |
|
59
|
|
|
} |
|
60
|
|
|
if (($data = \unserialize($data)) === false) { |
|
61
|
|
|
throw new InternalException("unserialize() failed [$filename]"); // @codeCoverageIgnore |
|
62
|
|
|
} |
|
63
|
|
|
return $data; |
|
64
|
|
|
} |
|
65
|
|
|
|
|
66
|
|
|
/** |
|
67
|
|
|
* Recursively scans a folder heirarchy returning the all the files and folders |
|
68
|
|
|
* in an array. |
|
69
|
|
|
* @return array<int, string> |
|
70
|
|
|
* @throws InternalException |
|
71
|
|
|
*/ |
|
72
|
|
|
protected function scanDirectory(string $rootDir): array { |
|
73
|
|
|
static $files = []; |
|
74
|
|
|
if (($handle = \opendir($rootDir)) === false) { |
|
75
|
|
|
throw new InternalException("opendir() failed [$rootDir]"); // @codeCoverageIgnore |
|
76
|
|
|
} |
|
77
|
|
|
while (($entry = \readdir($handle)) !== false) { |
|
78
|
|
|
$path = $rootDir.DS.$entry; |
|
79
|
|
|
if (\is_dir($path)) { |
|
80
|
|
|
if ($entry === '.' || $entry === '..') { |
|
81
|
|
|
continue; |
|
82
|
|
|
} |
|
83
|
|
|
$this->scanDirectory($path); |
|
84
|
|
|
continue; |
|
85
|
|
|
} |
|
86
|
|
|
$files[] = $path; |
|
87
|
|
|
} |
|
88
|
|
|
\closedir($handle); |
|
89
|
|
|
return $files; |
|
90
|
|
|
} |
|
91
|
|
|
|
|
92
|
|
|
/** |
|
93
|
|
|
* @return array<int, string> |
|
94
|
|
|
* @throws InternalException |
|
95
|
|
|
*/ |
|
96
|
|
|
protected function parseDirectory(string $pattern): array { |
|
97
|
|
|
if (($files = \glob($pattern, GLOB_NOSORT)) === false) { |
|
98
|
|
|
throw new InternalException("glob() failed [$pattern]"); // @codeCoverageIgnore |
|
99
|
|
|
} |
|
100
|
|
|
return $files; |
|
101
|
|
|
} |
|
102
|
|
|
|
|
103
|
|
|
/** |
|
104
|
|
|
* Writes data to an index file, creating the file if necessary. |
|
105
|
|
|
* @param string $filepath |
|
106
|
|
|
* @param array<mixed> $index |
|
107
|
|
|
* @throws InternalException |
|
108
|
|
|
*/ |
|
109
|
|
|
protected function writeIndex(string $filepath, array $index): void { |
|
110
|
|
|
$index = \serialize($index); |
|
111
|
|
|
if (\file_put_contents($filepath, print_r($index, true)) === false) { |
|
112
|
|
|
throw new InternalException("file_put_contents() failed [$filepath]"); // @codeCoverageIgnore |
|
113
|
|
|
} |
|
114
|
|
|
} |
|
115
|
|
|
|
|
116
|
|
|
/** |
|
117
|
|
|
* Creates and populates an index Element class. |
|
118
|
|
|
* @param string $key The index key |
|
119
|
|
|
* @param string $path The filepath |
|
120
|
|
|
* @param string $section 'pages', 'posts', 'categories' or 'tags' |
|
121
|
|
|
* @param string $optional When populating with variable arguments, use the |
|
122
|
|
|
* following <b>named parameters<b>: |
|
123
|
|
|
* <ul> |
|
124
|
|
|
* <li>type:</li> |
|
125
|
|
|
* <li>category:</li> |
|
126
|
|
|
* <li>username:</li> |
|
127
|
|
|
* <li>date:</li> |
|
128
|
|
|
* <li>tags:</li> |
|
129
|
|
|
* </ul> |
|
130
|
|
|
* @return \stdClass |
|
131
|
|
|
*/ |
|
132
|
|
|
protected function createElementClass(string $key, string $path, string $section, string ...$optional): \stdClass { |
|
133
|
|
|
if (\in_array($section, [CATEGORY_SECTION, TAG_SECTION, PAGE_SECTION, POST_SECTION]) === false) { |
|
134
|
|
|
throw new \InvalidArgumentException("Unknown section [$section]"); |
|
135
|
|
|
} |
|
136
|
|
|
$obj = new \stdClass(); |
|
137
|
|
|
$obj->key = $key; |
|
138
|
|
|
$obj->path = $path; |
|
139
|
|
|
$obj->section = $section; |
|
140
|
|
|
$obj->type = $this->checkSetOrDefault($optional, 'type', EMPTY_VALUE); |
|
141
|
|
|
$obj->category = $this->checkSetOrDefault($optional, 'category', EMPTY_VALUE); |
|
142
|
|
|
$obj->username = $this->checkSetOrDefault($optional, 'username', EMPTY_VALUE); |
|
143
|
|
|
$obj->timestamp = $this->checkSetOrDefault($optional, 'timestamp', EMPTY_VALUE); |
|
144
|
|
|
|
|
145
|
|
|
$tags = []; |
|
146
|
|
|
if (!empty($optional['tags'])) { |
|
147
|
|
|
$tags = \explode(',', $optional['tags']); |
|
148
|
|
|
} |
|
149
|
|
|
|
|
150
|
|
|
$obj->tags = $tags; |
|
151
|
|
|
return $obj; |
|
152
|
|
|
} |
|
153
|
|
|
|
|
154
|
|
|
/** |
|
155
|
|
|
* Create an Element object. The type of element and what properties are poplutated and |
|
156
|
|
|
* persisted to the index is determined by the <code>section</code>. |
|
157
|
|
|
* @param string $key |
|
158
|
|
|
* @param string $filepath |
|
159
|
|
|
* @param string $section |
|
160
|
|
|
* @throws InternalException |
|
161
|
|
|
* @throws InvalidArgumentException |
|
162
|
|
|
* @return \stdClass |
|
163
|
|
|
*/ |
|
164
|
|
|
protected function createElement(string $key, string $filepath, string $section): \stdClass { |
|
165
|
|
|
if (empty($key) || empty($section)) { |
|
166
|
|
|
throw new \InvalidArgumentException("A parameter is empty for [$key][$filepath][$section]"); // @codeCoverageIgnore |
|
167
|
|
|
} |
|
168
|
|
|
|
|
169
|
|
|
if( $section === POST_SECTION) { |
|
170
|
|
|
$uriDateStringTagList = $this->getPostUriDateStringAndTagListFromIndexKey($key); |
|
171
|
|
|
$typeCatUsername = $this->getTypeCategoryUsernameFromFilepath($filepath); |
|
172
|
|
|
return $this->createElementClass($uriDateStringTagList[0], $filepath, POST_SECTION, type: $typeCatUsername[0], category: $typeCatUsername[1], username: $typeCatUsername[2], timestamp: $uriDateStringTagList[1], tags: $uriDateStringTagList[2]); |
|
173
|
|
|
} |
|
174
|
|
|
return $this->createElementClass($key, $filepath, $section); |
|
175
|
|
|
} |
|
176
|
|
|
|
|
177
|
|
|
/** |
|
178
|
|
|
* Checks if the given key is set in the given array. If so, returns the value, |
|
179
|
|
|
* otherwise returns the default value. |
|
180
|
|
|
* @param array<mixed> $ar |
|
181
|
|
|
* @param string $key |
|
182
|
|
|
* @param mixed $default |
|
183
|
|
|
* @return mixed |
|
184
|
|
|
*/ |
|
185
|
|
|
private function checkSetOrDefault(array $ar, string $key, mixed $default): mixed { |
|
186
|
|
|
if (isset($ar[$key])) { |
|
187
|
|
|
return $ar[$key]; |
|
188
|
|
|
} |
|
189
|
|
|
return $default; |
|
190
|
|
|
} |
|
191
|
|
|
|
|
192
|
|
|
/** |
|
193
|
|
|
* |
|
194
|
|
|
* @param string $key |
|
195
|
|
|
* @throws InvalidArgumentException |
|
196
|
|
|
* @return array<string> |
|
197
|
|
|
*/ |
|
198
|
|
|
private function getPostUriDateStringAndTagListFromIndexKey(string $key): array { |
|
199
|
|
|
if (\strlen($key) < 17) { |
|
200
|
|
|
throw new InvalidArgumentException("Post content filename is too short [$key]"); // @codeCoverageIgnore |
|
201
|
|
|
} |
|
202
|
|
|
$dateString = \substr($key, 0, TIMESTAMP_LEN); |
|
203
|
|
|
$start = 15; |
|
204
|
|
|
if (($end = \strpos($key, '_', $start)) === false) { |
|
205
|
|
|
throw new InvalidArgumentException("Post content filename syntax error [$key]"); // @codeCoverageIgnore |
|
206
|
|
|
} |
|
207
|
|
|
$tagList = \substr($key, $start, $end - $start); |
|
208
|
|
|
$title = \substr($key, $end + 1); |
|
209
|
|
|
$year = \substr($dateString, 0, 4); |
|
210
|
|
|
$month = \substr($dateString, 4, 2); |
|
211
|
|
|
$uri = $year.FWD_SLASH.$month.FWD_SLASH.$title; |
|
212
|
|
|
return [$uri, $dateString, $tagList]; |
|
213
|
|
|
} |
|
214
|
|
|
|
|
215
|
|
|
/** |
|
216
|
|
|
* Extract the post's type, category and username from its filepath. |
|
217
|
|
|
* @param string $filepath |
|
218
|
|
|
* @return array<string> value index: [0] = type, [1] = category, [2] = username |
|
219
|
|
|
*/ |
|
220
|
|
|
private function getTypeCategoryUsernameFromFilepath(string $filepath): array { |
|
221
|
|
|
$pathinfo = \pathinfo($filepath); |
|
222
|
|
|
$parts = \explode(DS, $pathinfo['dirname']); |
|
223
|
|
|
$cnt = \count($parts); |
|
224
|
|
|
// type, category, username |
|
225
|
|
|
return [$parts[$cnt - 1], CATEGORY_SECTION.FWD_SLASH.$parts[$cnt - 2], $parts[$cnt - 4]]; |
|
226
|
|
|
} |
|
227
|
|
|
|
|
228
|
|
|
/** |
|
229
|
|
|
* Reindex the whole system. Generally called when new content has been added. |
|
230
|
|
|
*/ |
|
231
|
|
|
public abstract function reindex(): void; |
|
232
|
|
|
} |
|
233
|
|
|
|