1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152<?php
/**
* Make a sample PDF for testing, with no libraries.
*
* WHY IS THIS HERE?
*
* The test site needs PDFs to find, and committing binary sample files to a repo is
* a habit worth avoiding. A PDF is a text format with a byte-offset index at the end,
* so a usable one can be built in a page of PHP.
*
* WHAT IT IS GOOD FOR AND WHAT IT IS NOT
*
* These PDFs have real page objects and real text, which is exactly what the plugin's
* page counter reads, so they are perfect for testing discovery, page counting, the
* 25-page rule, and the whole queue.
*
* They are NOT good conversions. Equalify Iris will make something dull out of them,
* because there is nothing in them but a heading and some lines. To test the quality
* of a conversion, use a real document.
*
* USAGE
*
* php make-sample-pdf.php <output-path> <pages> "<title>"
*
* EXAMPLES
*
* php make-sample-pdf.php short.pdf 3 "Accessibility Report"
* php make-sample-pdf.php long.pdf 30 "Too Long To Convert"
*/
// phpcs:disable WordPress.Security.EscapeOutput, WordPress.PHP.DevelopmentFunctions
if ( 'cli' !== PHP_SAPI ) {
exit( 1 );
}
$path = $argv[1] ?? 'sample.pdf';
$pages = max( 1, (int) ( $argv[2] ?? 3 ) );
$title = $argv[3] ?? 'Sample Document';
/**
* Escape a string for use inside a PDF text object.
*
* Parentheses delimit strings in PDF, so an unescaped one in the title would end the
* string early and corrupt the file.
*/
function pdf_string( string $text ): string {
return str_replace( array( '\\', '(', ')' ), array( '\\\\', '\\(', '\\)' ), $text );
}
/**
* The drawing instructions for one page.
*
* BT begins a text object, Tf picks a font and size, Td moves to a position measured
* in points from the bottom left, Tj draws the string, ET ends the object.
*/
function page_content( int $page_number, int $total, string $title ): string {
$heading = 1 === $page_number ? $title : sprintf( '%s — page %d', $title, $page_number );
$lines = array(
sprintf( 'This is page %d of %d in a sample PDF.', $page_number, $total ),
'It was generated for testing the Equalify Iris WordPress plugin.',
'',
'Section ' . $page_number,
'',
'The text on this page exists so that a conversion has something to',
'return. It is deliberately dull. If you are checking how good a real',
'conversion looks, test with a real document instead.',
);
$stream = "BT\n/F1 20 Tf\n72 720 Td\n(" . pdf_string( $heading ) . ") Tj\nET\n";
$y = 680;
foreach ( $lines as $line ) {
if ( '' !== $line ) {
$stream .= "BT\n/F1 11 Tf\n72 {$y} Td\n(" . pdf_string( $line ) . ") Tj\nET\n";
}
$y -= 18;
}
return $stream;
}
/*
* Build the objects.
*
* Object 1 is the catalog (the root), object 2 is the page tree, object 3 is the font.
* After that come two objects per page: the page itself and its content stream. Object
* numbers are worked out first because the page tree has to list them all.
*/
$objects = array();
$page_refs = array();
for ( $i = 0; $i < $pages; $i++ ) {
$page_refs[] = ( 4 + ( $i * 2 ) ) . ' 0 R';
}
$objects[1] = "<< /Type /Catalog /Pages 2 0 R >>";
$objects[2] = "<< /Type /Pages /Kids [" . implode( ' ', $page_refs ) . "] /Count {$pages} >>";
$objects[3] = "<< /Type /Font /Subtype /Type1 /BaseFont /Helvetica /Encoding /WinAnsiEncoding >>";
for ( $i = 0; $i < $pages; $i++ ) {
$page_object = 4 + ( $i * 2 );
$content_object = $page_object + 1;
$content = page_content( $i + 1, $pages, $title );
$objects[ $page_object ] = "<< /Type /Page /Parent 2 0 R /MediaBox [0 0 612 792] "
. "/Resources << /Font << /F1 3 0 R >> >> /Contents {$content_object} 0 R >>";
$objects[ $content_object ] = "<< /Length " . strlen( $content ) . " >>\nstream\n{$content}endstream";
}
/*
* Write the file, recording where each object starts.
*
* Those byte offsets are the whole reason this cannot be a simple string of text: the
* cross-reference table at the end lists the exact position of every object, and a
* reader that finds the wrong offset rejects the file.
*/
$pdf = "%PDF-1.4\n";
$offsets = array();
ksort( $objects );
foreach ( $objects as $number => $body ) {
$offsets[ $number ] = strlen( $pdf );
$pdf .= "{$number} 0 obj\n{$body}\nendobj\n";
}
$xref_position = strlen( $pdf );
$object_count = count( $objects ) + 1; // +1 for the required free entry 0.
$pdf .= "xref\n0 {$object_count}\n";
$pdf .= "0000000000 65535 f \n";
foreach ( $offsets as $offset ) {
$pdf .= sprintf( "%010d 00000 n \n", $offset );
}
$pdf .= "trailer\n<< /Size {$object_count} /Root 1 0 R >>\n";
$pdf .= "startxref\n{$xref_position}\n%%EOF\n";
if ( false === file_put_contents( $path, $pdf ) ) {
fwrite( STDERR, "Could not write {$path}\n" );
exit( 1 );
}
printf( "Wrote %s — %d page(s), %s\n", $path, $pages, number_format( strlen( $pdf ) ) . ' bytes' );