Merge pull request #4323 from oleibman/comppcre

Use Composer/Pcre Part 1 of Many
PHPOffice · Jan 21, 2025 · 3f15579 · 3f15579
2 parents 8871095 + e7ae646
commit 3f15579
Show file tree

Hide file tree

Showing 20 changed files with 384 additions and 188 deletions.
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -29,6 +29,7 @@ and this project adheres to [Semantic Versioning](https://semver.org).
 - Ods Reader Sheet Names with Period. [Issue #4311](https://github.com/PHPOffice/PhpSpreadsheet/issues/4311) [PR #4313](https://github.com/PHPOffice/PhpSpreadsheet/pull/4313)
 - Mpdf and Tcpdf Hidden Columns and Merged Cells. [Issue #4319](https://github.com/PHPOffice/PhpSpreadsheet/issues/4319) [PR #4320](https://github.com/PHPOffice/PhpSpreadsheet/pull/4320)
 - Html Writer Allow mailto. [Issue #4316](https://github.com/PHPOffice/PhpSpreadsheet/issues/4316) [PR #4322](https://github.com/PHPOffice/PhpSpreadsheet/pull/4322)
+- Use composer/pcre rather than preg_* in Writer. [PR #4323](https://github.com/PHPOffice/PhpSpreadsheet/pull/4323)
 
 ## 2025-01-11 - 3.8.0
 

diff --git a/composer.json b/composer.json
@@ -79,6 +79,7 @@
         "ext-xmlwriter": "*",
         "ext-zip": "*",
         "ext-zlib": "*",
+        "composer/pcre": "^3.3",
         "maennchen/zipstream-php": "^2.1 || ^3.0",
         "markbaker/complex": "^3.0",
         "markbaker/matrix": "^3.0",

diff --git a/composer.lock b/composer.lock
diff --git a/phpstan.neon.dist b/phpstan.neon.dist
@@ -2,6 +2,7 @@ includes:
     - phpstan-baseline.neon
     - vendor/phpstan/phpstan-phpunit/extension.neon
     - vendor/phpstan/phpstan-phpunit/rules.neon
+    - vendor/composer/pcre/extension.neon
 
 parameters:
     level: 8

diff --git a/samples/Pdf/21b_Pdf.php b/samples/Pdf/21b_Pdf.php
@@ -1,5 +1,6 @@
 <?php
 
+use PhpOffice\PhpSpreadsheet\Exception as SpreadsheetException;
 use PhpOffice\PhpSpreadsheet\IOFactory;
 use PhpOffice\PhpSpreadsheet\Worksheet\PageSetup;
 use PhpOffice\PhpSpreadsheet\Writer\Pdf\Dompdf;
@@ -21,7 +22,7 @@ function replaceBody(string $html): string
         </body>
         EOF;
 
-    return preg_replace($bodystring, $bodyrepl, $html) ?? '';
+    return preg_replace($bodystring, $bodyrepl, $html) ?? throw new SpreadsheetException('preg failed');
 }
 
 require __DIR__ . '/../Header.php';

diff --git a/samples/Pdf/21c_Pdf.php b/samples/Pdf/21c_Pdf.php
@@ -1,5 +1,6 @@
 <?php
 
+use PhpOffice\PhpSpreadsheet\Exception as SpreadsheetException;
 use PhpOffice\PhpSpreadsheet\IOFactory;
 use PhpOffice\PhpSpreadsheet\Spreadsheet;
 use PhpOffice\PhpSpreadsheet\Writer\Pdf\Mpdf;
@@ -16,7 +17,7 @@ function addHeadersFootersMpdf2000(string $html): string
         odd-footer-name: html_myFooter2;
 
         EOF;
-    $html = preg_replace('/@page page0 {/', $pagerepl, $html) ?? '';
+    $html = preg_replace('/@page page0 {/', $pagerepl, $html) ?? throw new SpreadsheetException('preg 1 failed');
     $bodystring = '/<body>/';
     $simulatedBodyStart = Mpdf::SIMULATED_BODY_START;
     $bodyrepl = <<<EOF
@@ -40,7 +41,7 @@ function addHeadersFootersMpdf2000(string $html): string
 
         EOF;
 
-    return preg_replace($bodystring, $bodyrepl, $html) ?? '';
+    return preg_replace($bodystring, $bodyrepl, $html) ?? throw new SpreadsheetException('preg 2 failed');
 }
 
 $spreadsheet = new Spreadsheet();

diff --git a/src/PhpSpreadsheet/Shared/StringHelper.php b/src/PhpSpreadsheet/Shared/StringHelper.php
@@ -638,4 +638,9 @@ public static function testStringAsNumeric(string $textValue): float|string
 
         return (is_numeric(substr($textValue, 0, strlen((string) $v)))) ? $v : $textValue;
     }
+
+    public static function strlenAllowNull(?string $string): int
+    {
+        return strlen("$string");
+    }
 }
diff --git a/src/PhpSpreadsheet/Writer/Html.php b/src/PhpSpreadsheet/Writer/Html.php
@@ -2,6 +2,7 @@
 
 namespace PhpOffice\PhpSpreadsheet\Writer;
 
+use Composer\Pcre\Preg;
 use PhpOffice\PhpSpreadsheet\Calculation\Calculation;
 use PhpOffice\PhpSpreadsheet\Cell\Cell;
 use PhpOffice\PhpSpreadsheet\Cell\Coordinate;
@@ -646,13 +647,13 @@ private function writeImageInCell(string $coordinates): string
                 $filename = $drawing->getPath();
 
                 // Strip off eventual '.'
-                $filename = (string) preg_replace('/^[.]/', '', $filename);
+                $filename = Preg::replace('/^[.]/', '', $filename);
 
                 // Prepend images root
                 $filename = $this->getImagesRoot() . $filename;
 
                 // Strip off eventual '.' if followed by non-/
-                $filename = (string) preg_replace('@^[.]([^/])@', '$1', $filename);
+                $filename = Preg::replace('@^[.]([^/])@', '$1', $filename);
 
                 // Convert UTF8 data to PCDATA
                 $filename = htmlspecialchars($filename, Settings::htmlEntityFlags());
@@ -1411,7 +1412,7 @@ private function generateRowCellData(Worksheet $worksheet, null|Cell|string $cel
 
             // Converts the cell content so that spaces occuring at beginning of each new line are replaced by &nbsp;
             // Example: "  Hello\n to the world" is converted to "&nbsp;&nbsp;Hello\n&nbsp;to the world"
-            $cellData = (string) preg_replace('/(?m)(?:^|\\G) /', '&nbsp;', $cellData);
+            $cellData = Preg::replace('/(?m)(?:^|\\G) /', '&nbsp;', $cellData);
 
             // convert newline "\n" to '<br>'
             $cellData = nl2br($cellData);
@@ -1599,9 +1600,9 @@ private function generateRow(Worksheet $worksheet, array $values, int $row, stri
             if ($worksheet->hyperlinkExists($coordinate) && !$worksheet->getHyperlink($coordinate)->isInternal()) {
                 $url = $worksheet->getHyperlink($coordinate)->getUrl();
                 $urlDecode1 = html_entity_decode($url, ENT_QUOTES | ENT_SUBSTITUTE, 'UTF-8');
-                $urlTrim = preg_replace('/^\\s+/u', '', $urlDecode1) ?? $urlDecode1;
-                $parseScheme = preg_match('/^([\\w\\s]+):/u', strtolower($urlTrim), $matches);
-                if ($parseScheme === 1 && !in_array($matches[1], ['http', 'https', 'file', 'ftp', 'mailto', 's3'], true)) {
+                $urlTrim = Preg::replace('/^\\s+/u', '', $urlDecode1);
+                $parseScheme = Preg::isMatch('/^([\\w\\s]+):/u', strtolower($urlTrim), $matches);
+                if ($parseScheme && !in_array($matches[1], ['http', 'https', 'file', 'ftp', 'mailto', 's3'], true)) {
                     $cellData = htmlspecialchars($url, Settings::htmlEntityFlags());
                 } else {
                     $tooltip = $worksheet->getHyperlink($coordinate)->getTooltip();
@@ -1755,7 +1756,7 @@ public static function formatColorStatic(string $value, string $format): string
         $matches = [];
 
         $color_regex = '/^\\[[a-zA-Z]+\\]/';
-        if (preg_match($color_regex, $format, $matches)) {
+        if (Preg::isMatch($color_regex, $format, $matches)) {
             $color = str_replace(['[', ']'], '', $matches[0]);
             $color = strtolower($color);
         }

diff --git a/src/PhpSpreadsheet/Writer/Ods/Content.php b/src/PhpSpreadsheet/Writer/Ods/Content.php
@@ -2,6 +2,7 @@
 
 namespace PhpOffice\PhpSpreadsheet\Writer\Ods;
 
+use Composer\Pcre\Preg;
 use PhpOffice\PhpSpreadsheet\Calculation\Calculation;
 use PhpOffice\PhpSpreadsheet\Calculation\Exception as CalculationException;
 use PhpOffice\PhpSpreadsheet\Cell\Cell;
@@ -227,16 +228,15 @@ private function writeCells(XMLWriter $objWriter, RowCellIterator $cells): void
                         }
                     }
                     if (isset($attributes['ref'])) {
-                        if (preg_match('/^([A-Z]{1,3})([0-9]{1,7})(:([A-Z]{1,3})([0-9]{1,7}))?$/', (string) $attributes['ref'], $matches) == 1) {
+                        if (Preg::isMatch('/^([A-Z]{1,3})([0-9]{1,7})(:([A-Z]{1,3})([0-9]{1,7}))?$/', (string) $attributes['ref'], $matches)) {
                             $matrixRowSpan = 1;
                             $matrixColSpan = 1;
                             if (isset($matches[3])) {
                                 $minRow = (int) $matches[2];
-                                // https://github.com/phpstan/phpstan/issues/11602
-                                $maxRow = (int) $matches[5]; // @phpstan-ignore-line
+                                $maxRow = (int) $matches[5];
                                 $matrixRowSpan = $maxRow - $minRow + 1;
                                 $minCol = Coordinate::columnIndexFromString($matches[1]);
-                                $maxCol = Coordinate::columnIndexFromString($matches[4]); // @phpstan-ignore-line
+                                $maxCol = Coordinate::columnIndexFromString($matches[4]);
                                 $matrixColSpan = $maxCol - $minCol + 1;
                             }
                             $objWriter->writeAttribute('table:number-matrix-columns-spanned', "$matrixColSpan");

diff --git a/src/PhpSpreadsheet/Writer/Ods/Formula.php b/src/PhpSpreadsheet/Writer/Ods/Formula.php
@@ -2,8 +2,10 @@
 
 namespace PhpOffice\PhpSpreadsheet\Writer\Ods;
 
+use Composer\Pcre\Preg;
 use PhpOffice\PhpSpreadsheet\Calculation\Calculation;
 use PhpOffice\PhpSpreadsheet\DefinedName;
+use PhpOffice\PhpSpreadsheet\Shared\StringHelper;
 
 class Formula
 {
@@ -33,14 +35,13 @@ public function convertFormula(string $formula, string $worksheetName = ''): str
 
     private function convertDefinedNames(string $formula): string
     {
-        $splitCount = preg_match_all(
+        $splitCount = Preg::matchAllWithOffsets(
             '/' . Calculation::CALCULATION_REGEXP_DEFINEDNAME . '/mui',
             $formula,
-            $splitRanges,
-            PREG_OFFSET_CAPTURE
+            $splitRanges
         );
 
-        $lengths = array_map('strlen', array_column($splitRanges[0], 0));
+        $lengths = array_map([StringHelper::class, 'strlenAllowNull'], array_column($splitRanges[0], 0));
         $offsets = array_column($splitRanges[0], 1);
         $values = array_column($splitRanges[0], 0);
 
@@ -60,14 +61,13 @@ private function convertDefinedNames(string $formula): string
 
     private function convertCellReferences(string $formula, string $worksheetName): string
     {
-        $splitCount = preg_match_all(
+        $splitCount = Preg::matchAllWithOffsets(
             '/' . Calculation::CALCULATION_REGEXP_CELLREF_RELATIVE . '/mui',
             $formula,
-            $splitRanges,
-            PREG_OFFSET_CAPTURE
+            $splitRanges
         );
 
-        $lengths = array_map('strlen', array_column($splitRanges[0], 0));
+        $lengths = array_map([StringHelper::class, 'strlenAllowNull'], array_column($splitRanges[0], 0));
         $offsets = array_column($splitRanges[0], 1);
 
         $worksheets = $splitRanges[2];
@@ -76,7 +76,7 @@ private function convertCellReferences(string $formula, string $worksheetName):
 
         // Replace any commas in the formula with semi-colons for Ods
         // If by chance there are commas in worksheet names, then they will be "fixed" again in the loop
-        //    because we've already extracted worksheet names with our preg_match_all()
+        //    because we've already extracted worksheet names with our Preg::matchAllWithOffsets()
         $formula = str_replace(',', ';', $formula);
         while ($splitCount > 0) {
             --$splitCount;