Strom \0 from document content

jplitza · jplitza · commit 8d1f7dc1ab30 · 2025-07-28T09:16:13.000+02:00
Apparently the text from PDFs can contain null bytes, which PostgreSQL rejects in TEXT columns, so we have to strip  them
diff --git a/lib/Platform/SQLPlatform.php b/lib/Platform/SQLPlatform.php
@@ -224,7 +224,7 @@ public function indexDocument(IIndexDocument $document): IIndex {
 					$pdf = $parser->parseContent($content);
 					$content = $pdf->getText();
 				}
-				$indexDocument->setContent($content);
+				$indexDocument->setContent(str_replace("\0", "", $content));
 
 				if ($indexDocument->getId()) {
 					$this->indexDocumentMapper->update($indexDocument);

Original file line number	Diff line number	Diff line change
`@@ -224,7 +224,7 @@ public function indexDocument(IIndexDocument $document): IIndex {`
`224`	`224`	`$pdf = $parser->parseContent($content);`
`225`	`225`	`$content = $pdf->getText();`
`226`	`226`	`}`
`227`		`- $indexDocument->setContent($content);`
	`227`	`+ $indexDocument->setContent(str_replace("\0", "", $content));`
`228`	`228`
`229`	`229`	`if ($indexDocument->getId()) {`
`230`	`230`	`$this->indexDocumentMapper->update($indexDocument);`