1
|
|
|
<?php |
2
|
|
|
|
3
|
|
|
/** |
4
|
|
|
* @file This file is part of the PdfParser library. |
5
|
|
|
* |
6
|
|
|
* @author Konrad Abicht <[email protected]> |
7
|
|
|
* @date 2020-06-01 |
8
|
|
|
* |
9
|
|
|
* @author Sébastien MALOT <[email protected]> |
10
|
|
|
* @date 2017-01-03 |
11
|
|
|
* |
12
|
|
|
* @license LGPLv3 |
13
|
|
|
* @url <https://github.com/smalot/pdfparser> |
14
|
|
|
* |
15
|
|
|
* PdfParser is a pdf library written in PHP, extraction oriented. |
16
|
|
|
* Copyright (C) 2017 - Sébastien MALOT <[email protected]> |
17
|
|
|
* |
18
|
|
|
* This program is free software: you can redistribute it and/or modify |
19
|
|
|
* it under the terms of the GNU Lesser General Public License as published by |
20
|
|
|
* the Free Software Foundation, either version 3 of the License, or |
21
|
|
|
* (at your option) any later version. |
22
|
|
|
* |
23
|
|
|
* This program is distributed in the hope that it will be useful, |
24
|
|
|
* but WITHOUT ANY WARRANTY; without even the implied warranty of |
25
|
|
|
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
26
|
|
|
* GNU Lesser General Public License for more details. |
27
|
|
|
* |
28
|
|
|
* You should have received a copy of the GNU Lesser General Public License |
29
|
|
|
* along with this program. |
30
|
|
|
* If not, see <http://www.pdfparser.org/sites/default/LICENSE.txt>. |
31
|
|
|
*/ |
32
|
|
|
|
33
|
|
|
namespace Tests\Smalot\PdfParser\Integration; |
34
|
|
|
|
35
|
|
|
use Smalot\PdfParser\Element; |
36
|
|
|
use Smalot\PdfParser\Element\ElementMissing; |
37
|
|
|
use Smalot\PdfParser\Element\ElementName; |
38
|
|
|
use Smalot\PdfParser\Header; |
39
|
|
|
use Smalot\PdfParser\Page; |
40
|
|
|
use Smalot\PdfParser\PDFObject; |
41
|
|
|
use Tests\Smalot\PdfParser\TestCase; |
42
|
|
|
|
43
|
|
|
/** |
44
|
|
|
* Class Header |
45
|
|
|
*/ |
46
|
|
|
class HeaderTest extends TestCase |
47
|
|
|
{ |
48
|
|
|
/** |
49
|
|
|
* Checks that init function is called for each element. |
50
|
|
|
*/ |
51
|
|
|
public function testInitHappyPath() |
52
|
|
|
{ |
53
|
|
|
$element = $this->createMock(Element::class); |
54
|
|
|
$element->expects($this->exactly(1))->method('init'); |
55
|
|
|
|
56
|
|
|
$fixture = new Header([$element]); |
57
|
|
|
$fixture->init(); |
58
|
|
|
} |
59
|
|
|
|
60
|
|
|
/** |
61
|
|
|
* Checks buggy behavior if an element was given which is not of type Element. |
62
|
|
|
* |
63
|
|
|
* Problem was, it always called $element::init(), even if its not an object at all. |
64
|
|
|
* |
65
|
|
|
* @see https://github.com/smalot/pdfparser/issues/367 |
66
|
|
|
* |
67
|
|
|
* @doesNotPerformAssertions |
68
|
|
|
*/ |
69
|
|
|
public function testInitInvalidElement() |
70
|
|
|
{ |
71
|
|
|
$element = false; |
72
|
|
|
|
73
|
|
|
$fixture = new Header([$element]); |
|
|
|
|
74
|
|
|
$fixture->init(); |
75
|
|
|
} |
76
|
|
|
|
77
|
|
|
public function testParse() |
78
|
|
|
{ |
79
|
|
|
$document = $this->getDocumentInstance(); |
80
|
|
|
|
81
|
|
|
$content = '<</Type/Page/SubType/Text>>foo'; |
82
|
|
|
$position = 0; |
83
|
|
|
$header = Header::parse($content, $document, $position); |
84
|
|
|
|
85
|
|
|
$this->assertTrue($header instanceof Header); |
86
|
|
|
$this->assertEquals(27, $position); |
87
|
|
|
$this->assertEquals(2, \count($header->getElements())); |
88
|
|
|
|
89
|
|
|
// No header to parse |
90
|
|
|
$this->assertEquals('Page', (string) $header->get('Type')); |
91
|
|
|
$content = 'foo'; |
92
|
|
|
$position = 0; |
93
|
|
|
$header = Header::parse($content, $document, $position); |
94
|
|
|
|
95
|
|
|
$this->assertTrue($header instanceof Header); |
96
|
|
|
$this->assertEquals(0, $position); |
97
|
|
|
$this->assertEquals(0, \count($header->getElements())); |
98
|
|
|
|
99
|
|
|
$position = 0; |
100
|
|
|
$content = "<</CreationDate(D:20100309184803+01'00')/Author(Utilisateur)/Creator(PScript5.dll Version 5.2.2)/Producer(Acrobat Distiller 7.0.5 \(Windows\))/ModDate(D:20100310104810+01'00')/Title(Microsoft Word - CLEMI.docx)>>"; |
101
|
|
|
Header::parse($content, $document, $position); |
102
|
|
|
$this->assertEquals(212, $position); |
103
|
|
|
|
104
|
|
|
$position = 0; |
105
|
|
|
$content = '[5 0 R ] foo'; |
106
|
|
|
$header = Header::parse($content, $document, $position); |
107
|
|
|
$this->assertEquals(8, $position); |
108
|
|
|
$this->assertEquals(1, \count($header->getElements())); |
109
|
|
|
} |
110
|
|
|
|
111
|
|
|
public function testGetElements() |
112
|
|
|
{ |
113
|
|
|
$document = $this->getDocumentInstance(); |
114
|
|
|
|
115
|
|
|
$content = '<</Type/Page/Subtype/Text>>foo'; |
116
|
|
|
$position = 0; |
117
|
|
|
$header = Header::parse($content, $document, $position); |
118
|
|
|
|
119
|
|
|
$elements = $header->getElements(); |
120
|
|
|
$this->assertEquals(2, \count($elements)); |
121
|
|
|
$this->assertTrue(current($elements) instanceof ElementName); |
122
|
|
|
|
123
|
|
|
$types = $header->getElementTypes(); |
124
|
|
|
$this->assertTrue(\is_array($types)); |
125
|
|
|
$this->assertEquals(ElementName::class, $types['Type']); |
126
|
|
|
$this->assertEquals(ElementName::class, $types['Subtype']); |
127
|
|
|
} |
128
|
|
|
|
129
|
|
|
public function testHas() |
130
|
|
|
{ |
131
|
|
|
$document = $this->getDocumentInstance(); |
132
|
|
|
|
133
|
|
|
$content = '<</Type/Page/SubType/Text/Font 5 0 R>>foo'; |
134
|
|
|
$position = 0; |
135
|
|
|
$header = Header::parse($content, $document, $position); |
136
|
|
|
|
137
|
|
|
$this->assertTrue($header->has('Type')); |
138
|
|
|
$this->assertTrue($header->has('SubType')); |
139
|
|
|
$this->assertTrue($header->has('Font')); |
140
|
|
|
$this->assertFalse($header->has('Text')); |
141
|
|
|
} |
142
|
|
|
|
143
|
|
|
public function testGet() |
144
|
|
|
{ |
145
|
|
|
$document = $this->getDocumentInstance(); |
146
|
|
|
|
147
|
|
|
$content = '<</Type/Page/SubType/Text/Font 5 0 R/Resources 8 0 R>>foo'; |
148
|
|
|
$position = 0; |
149
|
|
|
$header = Header::parse($content, $document, $position); |
150
|
|
|
$object = new Page($document, $header); |
151
|
|
|
$document->setObjects(['5_0' => $object]); |
152
|
|
|
|
153
|
|
|
$this->assertTrue($header->get('Type') instanceof ElementName); |
154
|
|
|
$this->assertTrue($header->get('SubType') instanceof ElementName); |
155
|
|
|
$this->assertTrue($header->get('Font') instanceof Page); |
156
|
|
|
$this->assertTrue($header->get('Image') instanceof ElementMissing); |
157
|
|
|
$this->assertTrue($header->get('Resources') instanceof ElementMissing); |
158
|
|
|
} |
159
|
|
|
|
160
|
|
|
public function testResolveXRef() |
161
|
|
|
{ |
162
|
|
|
$document = $this->getDocumentInstance(); |
163
|
|
|
$content = '<</Type/Page/SubType/Text/Font 5 0 R/Resources 8 0 R>>foo'; |
164
|
|
|
$position = 0; |
165
|
|
|
$header = Header::parse($content, $document, $position); |
166
|
|
|
$object = new Page($document, $header); |
167
|
|
|
$document->setObjects(['5_0' => $object]); |
168
|
|
|
|
169
|
|
|
$this->assertTrue($header->get('Font') instanceof PDFObject); |
170
|
|
|
|
171
|
|
|
$header = $header->get('Resources'); |
172
|
|
|
$this->assertTrue($header instanceof ElementMissing); |
173
|
|
|
} |
174
|
|
|
} |
175
|
|
|
|