We could not synchronize checks via GitHub's checks API since Scrutinizer's GitHub App is not installed for this repository.
1 | <?php |
||
2 | |||
3 | /** |
||
4 | * (c) Kitodo. Key to digital objects e.V. <[email protected]> |
||
5 | * |
||
6 | * This file is part of the Kitodo and TYPO3 projects. |
||
7 | * |
||
8 | * @license GNU General Public License version 3 or later. |
||
9 | * For the full copyright and license information, please read the |
||
10 | * LICENSE.txt file that was distributed with this source code. |
||
11 | */ |
||
12 | |||
13 | namespace Kitodo\Dlf\Format; |
||
14 | |||
15 | /** |
||
16 | * Fulltext ALTO format class for the 'dlf' extension |
||
17 | * |
||
18 | * ** This currently supports ALTO 2.x / 3.x / 4.x ** |
||
19 | * |
||
20 | * @package TYPO3 |
||
21 | * @subpackage dlf |
||
22 | * |
||
23 | * @access public |
||
24 | */ |
||
25 | class Alto implements \Kitodo\Dlf\Common\FulltextInterface |
||
26 | { |
||
27 | /** |
||
28 | * This extracts the fulltext data from ALTO XML |
||
29 | * |
||
30 | * @access public |
||
31 | * |
||
32 | * @param \SimpleXMLElement $xml The XML to extract the raw text from |
||
33 | * |
||
34 | * @return string The raw unformatted fulltext |
||
35 | */ |
||
36 | public function getRawText(\SimpleXMLElement $xml): string |
||
37 | { |
||
38 | $rawText = ''; |
||
39 | |||
40 | // register ALTO namespace depending on document |
||
41 | $this->registerAltoNamespace($xml); |
||
42 | |||
43 | // Get all (presumed) words of the text. |
||
44 | $strings = $xml->xpath('./alto:Layout/alto:Page/alto:PrintSpace//alto:TextBlock/alto:TextLine/alto:String'); |
||
45 | $words = []; |
||
46 | if (!empty($strings)) { |
||
47 | for ($i = 0; $i < count($strings); $i++) { |
||
0 ignored issues
–
show
|
|||
48 | $attributes = $strings[$i]->attributes(); |
||
49 | if (isset($attributes['SUBS_TYPE'])) { |
||
50 | if ($attributes['SUBS_TYPE'] == 'HypPart1') { |
||
51 | $i++; |
||
52 | $words[] = $attributes['SUBS_CONTENT']; |
||
53 | } |
||
54 | } else { |
||
55 | $words[] = $attributes['CONTENT']; |
||
56 | } |
||
57 | } |
||
58 | $rawText = implode(' ', $words); |
||
59 | } |
||
60 | return $rawText; |
||
61 | } |
||
62 | |||
63 | /** |
||
64 | * This extracts the fulltext data from ALTO XML and returns it in MiniOCR format |
||
65 | * |
||
66 | * @access public |
||
67 | * |
||
68 | * @param \SimpleXMLElement $xml The XML to extract the raw text from |
||
69 | * |
||
70 | * @return string The unformatted fulltext in MiniOCR format |
||
71 | */ |
||
72 | public function getTextAsMiniOcr(\SimpleXMLElement $xml): string |
||
73 | { |
||
74 | // register ALTO namespace depending on document |
||
75 | $this->registerAltoNamespace($xml); |
||
76 | |||
77 | // get all text blocks |
||
78 | $blocks = $xml->xpath('./alto:Layout/alto:Page/alto:PrintSpace//alto:TextBlock'); |
||
79 | |||
80 | if (empty($blocks)) { |
||
81 | return ''; |
||
82 | } |
||
83 | |||
84 | $miniOcr = new \SimpleXMLElement("<ocr></ocr>"); |
||
85 | |||
86 | foreach ($blocks as $block) { |
||
87 | $newBlock = $miniOcr->addChild('b'); |
||
88 | foreach ($block->children() as $key => $value) { |
||
89 | if ($key === "TextLine") { |
||
90 | $newLine = $newBlock->addChild('l'); |
||
91 | foreach ($value->children() as $wordKey => $word) { |
||
92 | if ($wordKey == "String") { |
||
93 | $attributes = $word->attributes(); |
||
94 | $newWord = $newLine->addChild('w', $this->getWord($attributes)); |
||
95 | $newWord->addAttribute('x', $this->getCoordinates($attributes)); |
||
96 | } |
||
97 | } |
||
98 | } |
||
99 | } |
||
100 | } |
||
101 | |||
102 | $miniOcrXml = $miniOcr->asXml(); |
||
103 | if (\is_string($miniOcrXml)) { |
||
104 | return $miniOcrXml; |
||
105 | } |
||
106 | return ''; |
||
107 | } |
||
108 | |||
109 | /** |
||
110 | * This extracts and parses the word from attribute |
||
111 | * |
||
112 | * @access private |
||
113 | * |
||
114 | * @param \SimpleXMLElement $attributes The XML to extract the word |
||
115 | * |
||
116 | * @return string The parsed word extracted from attribute |
||
117 | */ |
||
118 | private function getWord(\SimpleXMLElement $attributes): string |
||
119 | { |
||
120 | if (!empty($attributes['SUBS_CONTENT'])) { |
||
121 | if ($attributes['SUBS_TYPE'] == 'HypPart1') { |
||
122 | return htmlspecialchars((string) $attributes['SUBS_CONTENT']); |
||
123 | } |
||
124 | return ' '; |
||
125 | } |
||
126 | return htmlspecialchars((string) $attributes['CONTENT']) . ' '; |
||
127 | } |
||
128 | |||
129 | /** |
||
130 | * This extracts and parses the word coordinates from attributes |
||
131 | * |
||
132 | * @access private |
||
133 | * |
||
134 | * @param \SimpleXMLElement $attributes The XML to extract the word coordinates |
||
135 | * |
||
136 | * @return string The parsed word coordinates extracted from attribute |
||
137 | */ |
||
138 | private function getCoordinates(\SimpleXMLElement $attributes): string |
||
139 | { |
||
140 | return (string) $attributes['HPOS'] . ' ' . (string) $attributes['VPOS'] . ' ' . (string) $attributes['WIDTH'] . ' ' . (string) $attributes['HEIGHT']; |
||
141 | } |
||
142 | |||
143 | /** |
||
144 | * This registers the necessary ALTO namespace for the current ALTO-XML |
||
145 | * |
||
146 | * @access private |
||
147 | * |
||
148 | * @param \SimpleXMLElement &$xml: The XML to register the namespace for |
||
149 | */ |
||
150 | private function registerAltoNamespace(\SimpleXMLElement &$xml) |
||
151 | { |
||
152 | $namespace = $xml->getDocNamespaces(); |
||
153 | |||
154 | if (in_array('http://www.loc.gov/standards/alto/ns-v2#', $namespace, true)) { |
||
155 | $xml->registerXPathNamespace('alto', 'http://www.loc.gov/standards/alto/ns-v2#'); |
||
156 | } elseif (in_array('http://www.loc.gov/standards/alto/ns-v3#', $namespace, true)) { |
||
157 | $xml->registerXPathNamespace('alto', 'http://www.loc.gov/standards/alto/ns-v3#'); |
||
158 | } elseif (in_array('http://www.loc.gov/standards/alto/ns-v4#', $namespace, true)) { |
||
159 | $xml->registerXPathNamespace('alto', 'http://www.loc.gov/standards/alto/ns-v4#'); |
||
160 | } |
||
161 | } |
||
162 | } |
||
163 |
If the size of the collection does not change during the iteration, it is generally a good practice to compute it beforehand, and not on each iteration: