Code Coverage |
||||||||||
Lines |
Functions and Methods |
Classes and Traits |
||||||||
| Total | |
100.00% |
61 / 61 |
|
100.00% |
3 / 3 |
CRAP | |
100.00% |
1 / 1 |
| Merger | |
100.00% |
61 / 61 |
|
100.00% |
3 / 3 |
13 | |
100.00% |
1 / 1 |
| mergeFiles | |
100.00% |
8 / 8 |
|
100.00% |
1 / 1 |
4 | |||
| mergeData | |
100.00% |
6 / 6 |
|
100.00% |
1 / 1 |
3 | |||
| mergeSources | |
100.00% |
47 / 47 |
|
100.00% |
1 / 1 |
6 | |||
| 1 | <?php |
| 2 | declare(strict_types=1); |
| 3 | /** |
| 4 | * Pop PHP Framework (https://www.popphp.org/) |
| 5 | * |
| 6 | * @link https://github.com/popphp/popphp-framework |
| 7 | * @author Nick Sagona, III <nick@popphp.org> |
| 8 | * @copyright Copyright (c) 2009-2026 Nick Sagona, III |
| 9 | * @license https://www.popphp.org/license New BSD License |
| 10 | */ |
| 11 | |
| 12 | /** |
| 13 | * @namespace |
| 14 | */ |
| 15 | namespace Pop\Pdf\Build; |
| 16 | |
| 17 | use Pop\Pdf\Document\AbstractDocument; |
| 18 | use Pop\Pdf\Extract\Document as ExtractDocument; |
| 19 | use Pop\Pdf\Extract\Value; |
| 20 | |
| 21 | /** |
| 22 | * Pdf merger class |
| 23 | * |
| 24 | * Combines whole PDF documents into one, natively - no external |
| 25 | * dependencies. Each source is read via the same ObjectGraphReader used by |
| 26 | * Build\Parser, at an increasing per-source object-number offset, then each |
| 27 | * source's entire original /Pages subtree is spliced under one new master |
| 28 | * /Pages node. |
| 29 | * |
| 30 | * @category Pop |
| 31 | * @package Pop\Pdf |
| 32 | * @author Nick Sagona, III <nick@popphp.org> |
| 33 | * @copyright Copyright (c) 2009-2026 Nick Sagona, III |
| 34 | * @license https://www.popphp.org/license New BSD License |
| 35 | * @version 6.0.0 |
| 36 | */ |
| 37 | class Merger |
| 38 | { |
| 39 | |
| 40 | /** |
| 41 | * Merge PDF files into one document |
| 42 | * |
| 43 | * @param array $files |
| 44 | * @throws Exception |
| 45 | * @return AbstractDocument |
| 46 | */ |
| 47 | public function mergeFiles(array $files): AbstractDocument |
| 48 | { |
| 49 | $sources = []; |
| 50 | |
| 51 | foreach ($files as $file) { |
| 52 | if (!file_exists($file)) { |
| 53 | throw new Exception("Error: The PDF file '{$file}' does not exist."); |
| 54 | } |
| 55 | try { |
| 56 | $sources[] = ExtractDocument::fromFile($file); |
| 57 | } catch (\Pop\Pdf\Extract\Exception $e) { |
| 58 | throw new Exception($e->getMessage(), $e->getCode(), $e); |
| 59 | } |
| 60 | } |
| 61 | |
| 62 | return $this->mergeSources($sources); |
| 63 | } |
| 64 | |
| 65 | /** |
| 66 | * Merge raw PDF data streams into one document |
| 67 | * |
| 68 | * @param array $dataList |
| 69 | * @throws Exception |
| 70 | * @return AbstractDocument |
| 71 | */ |
| 72 | public function mergeData(array $dataList): AbstractDocument |
| 73 | { |
| 74 | $sources = []; |
| 75 | |
| 76 | foreach ($dataList as $data) { |
| 77 | try { |
| 78 | $sources[] = new ExtractDocument($data); |
| 79 | } catch (\Pop\Pdf\Extract\Exception $e) { |
| 80 | throw new Exception($e->getMessage(), $e->getCode(), $e); |
| 81 | } |
| 82 | } |
| 83 | |
| 84 | return $this->mergeSources($sources); |
| 85 | } |
| 86 | |
| 87 | /** |
| 88 | * Merge a set of already-parsed Extract\Document sources |
| 89 | * |
| 90 | * @param array $sources |
| 91 | * @throws Exception |
| 92 | * @return AbstractDocument |
| 93 | */ |
| 94 | protected function mergeSources(array $sources): AbstractDocument |
| 95 | { |
| 96 | if (count($sources) < 2) { |
| 97 | throw new Exception('Error: Merging requires at least 2 source PDF documents.'); |
| 98 | } |
| 99 | |
| 100 | try { |
| 101 | $graphs = []; |
| 102 | $objectLists = []; |
| 103 | $pageObjectLists = []; |
| 104 | $offset = 0; |
| 105 | |
| 106 | foreach ($sources as $source) { |
| 107 | $graph = Import\ObjectGraphReader::read($source, $offset); |
| 108 | $graphs[] = $graph; |
| 109 | $objectLists[] = $graph['objects']; |
| 110 | $pageObjectLists[] = $graph['pageObjects']; |
| 111 | $offset = $graph['nextOffset']; |
| 112 | } |
| 113 | |
| 114 | // Object arrays are keyed by object number, and array_merge() |
| 115 | // would renumber integer keys, so a union preserving those keys |
| 116 | // is required. array_replace() lets a later argument overwrite |
| 117 | // an earlier one on key collision, which is the opposite of the |
| 118 | // '+=' union semantics being replicated here (first source |
| 119 | // wins), so the collected lists are combined in reverse order. |
| 120 | $allObjects = array_replace(...array_reverse($objectLists)); |
| 121 | $allPages = array_merge(...$pageObjectLists); |
| 122 | } catch (\Pop\Pdf\Extract\Exception $e) { |
| 123 | throw new Exception($e->getMessage(), $e->getCode(), $e); |
| 124 | } |
| 125 | |
| 126 | $masterObjNum = $offset + 1; |
| 127 | $rootObjNum = $masterObjNum + 1; |
| 128 | $infoObjNum = $rootObjNum + 1; |
| 129 | $masterKids = []; |
| 130 | |
| 131 | foreach ($graphs as $graph) { |
| 132 | $dict = $graph['topPagesDict']; |
| 133 | $dict['Parent'] = new Value\Reference($masterObjNum, 0); |
| 134 | |
| 135 | $streamObject = new PdfObject\StreamObject($graph['topPagesObjNum']); |
| 136 | $streamObject->setDefinition(Import\ObjectSerializer::serializeDict($dict)); |
| 137 | $streamObject->setImported(true); |
| 138 | |
| 139 | $allObjects[$graph['topPagesObjNum']] = $streamObject; |
| 140 | $masterKids[] = $graph['topPagesObjNum']; |
| 141 | } |
| 142 | |
| 143 | $masterParent = new PdfObject\ParentObject($masterObjNum); |
| 144 | $masterParent->setKids($masterKids); |
| 145 | $masterParent->setCount(count($allPages)); |
| 146 | $masterParent->setImported(true); |
| 147 | $allObjects[$masterObjNum] = $masterParent; |
| 148 | |
| 149 | // Compiler::setDocument() synthesizes its own default RootObject |
| 150 | // (hardcoded index 1, pointing at hardcoded Pages index 2) and |
| 151 | // InfoObject (hardcoded index 3) whenever the document's imported |
| 152 | // objects don't already include one - unconditionally overwriting |
| 153 | // whatever real merged object landed at that number. Both must be |
| 154 | // supplied explicitly, at guaranteed-collision-free numbers beyond |
| 155 | // everything already allocated (this exact bug was found and fixed |
| 156 | // in Build\Parser during Task 6's review - Merger repeats the same |
| 157 | // assembly shape and needs the same fix built in from the start). |
| 158 | $root = new PdfObject\RootObject($rootObjNum); |
| 159 | $root->setParentIndex($masterObjNum); |
| 160 | $root->setImported(true); |
| 161 | $allObjects[$rootObjNum] = $root; |
| 162 | |
| 163 | $info = new PdfObject\InfoObject($infoObjNum); |
| 164 | $info->setImported(true); |
| 165 | $allObjects[$infoObjNum] = $info; |
| 166 | |
| 167 | $document = new \Pop\Pdf\Document(); |
| 168 | $document->importObjects($allObjects); |
| 169 | |
| 170 | foreach ($allPages as $pageObject) { |
| 171 | $page = new \Pop\Pdf\Document\Page($pageObject->getWidth(), $pageObject->getHeight(), $pageObject->getIndex()); |
| 172 | $page->importPageObject($pageObject); |
| 173 | $document->addPage($page); |
| 174 | } |
| 175 | |
| 176 | return $document; |
| 177 | } |
| 178 | |
| 179 | } |