Merge pull request #3341 from HuongNV13/php82-str-split-function-return

Fix PHP8.2 str_split function returns empty arrays for empty strings
PHPOffice · Feb 28, 2025 · 027a5c6 · 027a5c6
2 parents a936254 + 93169c6
commit 027a5c6
Show file tree

Hide file tree

Showing 12 changed files with 30 additions and 22 deletions.
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -36,6 +36,7 @@ and this project adheres to [Semantic Versioning](https://semver.org).
 - Ignore ignoredErrors when not applicable. [Issue #4375](https://github.com/PHPOffice/PhpSpreadsheet/issues/4375) [PR #4377](https://github.com/PHPOffice/PhpSpreadsheet/pull/4377)
 - Better handling of defined names on sheets whose titles include apostrophes. [Issue #4356](https://github.com/PHPOffice/PhpSpreadsheet/issues/4356) [Issue #4362](https://github.com/PHPOffice/PhpSpreadsheet/issues/4362) [Issue #4376](https://github.com/PHPOffice/PhpSpreadsheet/issues/4376) [PR #4360](https://github.com/PHPOffice/PhpSpreadsheet/pull/4360)
 - Partial solution for removing rows or columns that include edge ranges. [Issue #1449](https://github.com/PHPOffice/PhpSpreadsheet/issues/1449) [PR #3528](https://github.com/PHPOffice/PhpSpreadsheet/pull/3528)
+- Prefer mb_str_split to str_split. [PR #3341](https://github.com/PHPOffice/PhpSpreadsheet/pull/3341)
 
 ## 2025-02-08 - 4.0.0
 

diff --git a/src/PhpSpreadsheet/Calculation/Engineering/ConvertHex.php b/src/PhpSpreadsheet/Calculation/Engineering/ConvertHex.php
@@ -96,7 +96,7 @@ public static function toDecimal($value)
         }
 
         $binX = '';
-        foreach (str_split($value) as $char) {
+        foreach (mb_str_split($value, 1, 'UTF-8') as $char) {
             $binX .= str_pad(base_convert($char, 16, 2), 4, '0', STR_PAD_LEFT);
         }
         if (strlen($binX) == 40 && $binX[0] == '1') {

diff --git a/src/PhpSpreadsheet/Calculation/Engineering/ConvertOctal.php b/src/PhpSpreadsheet/Calculation/Engineering/ConvertOctal.php
@@ -96,7 +96,7 @@ public static function toDecimal($value)
         }
 
         $binX = '';
-        foreach (str_split($value) as $char) {
+        foreach (mb_str_split($value, 1, 'UTF-8') as $char) {
             $binX .= str_pad(decbin((int) $char), 3, '0', STR_PAD_LEFT);
         }
         if (strlen($binX) == 30 && $binX[0] == '1') {

diff --git a/src/PhpSpreadsheet/Calculation/MathTrig/Arabic.php b/src/PhpSpreadsheet/Calculation/MathTrig/Arabic.php
@@ -74,11 +74,14 @@ public static function evaluate(mixed $roman): array|int|string
         // Convert the roman numeral to an arabic number
         $negativeNumber = $roman[0] === '-';
         if ($negativeNumber) {
-            $roman = substr($roman, 1);
+            $roman = trim(substr($roman, 1));
+            if ($roman === '') {
+                return ExcelError::NAN();
+            }
         }
 
         try {
-            $arabic = self::calculateArabic(str_split($roman));
+            $arabic = self::calculateArabic(mb_str_split($roman, 1, 'UTF-8'));
         } catch (Exception) {
             return ExcelError::VALUE(); // Invalid character detected
         }

diff --git a/src/PhpSpreadsheet/Reader/Csv/Delimiter.php b/src/PhpSpreadsheet/Reader/Csv/Delimiter.php
@@ -55,7 +55,7 @@ protected function countPotentialDelimiters(): void
 
     protected function countDelimiterValues(string $line, array $delimiterKeys): void
     {
-        $splitString = str_split($line, 1);
+        $splitString = mb_str_split($line, 1, 'UTF-8');
         $distribution = array_count_values($splitString);
         $countLine = array_intersect_key($distribution, $delimiterKeys);
 

diff --git a/src/PhpSpreadsheet/Reader/Security/XmlScanner.php b/src/PhpSpreadsheet/Reader/Security/XmlScanner.php
@@ -87,7 +87,7 @@ private function findCharSet(string $xml): string
     public function scan($xml): string
     {
         // Don't rely purely on libxml_disable_entity_loader()
-        $pattern = '/\0*' . implode('\0*', str_split($this->pattern)) . '\0*/';
+        $pattern = '/\0*' . implode('\0*', mb_str_split($this->pattern, 1, 'UTF-8')) . '\0*/';
 
         $xml = "$xml";
         if (preg_match($pattern, $xml)) {

diff --git a/src/PhpSpreadsheet/Style/ConditionalFormatting/ConditionalFormattingRuleExtension.php b/src/PhpSpreadsheet/Style/ConditionalFormatting/ConditionalFormattingRuleExtension.php
@@ -34,7 +34,7 @@ public function __construct(?string $id = null, string $cfRule = self::CONDITION
 
     private function generateUuid(): string
     {
-        $chars = str_split('xxxxxxxx-xxxx-4xxx-yxxx-xxxxxxxxxxxx');
+        $chars = mb_str_split('xxxxxxxx-xxxx-4xxx-yxxx-xxxxxxxxxxxx', 1, 'UTF-8');
 
         foreach ($chars as $i => $char) {
             if ($char === 'x') {

diff --git a/src/PhpSpreadsheet/Style/NumberFormat/DateFormatter.php b/src/PhpSpreadsheet/Style/NumberFormat/DateFormatter.php
@@ -207,6 +207,6 @@ private static function setLowercaseCallback(array $matches): string
 
     private static function escapeQuotesCallback(array $matches): string
     {
-        return '\\' . implode('\\', str_split($matches[1]));
+        return '\\' . implode('\\', mb_str_split($matches[1], 1, 'UTF-8'));
     }
 }
diff --git a/tests/data/Calculation/Engineering/HEX2DEC.php b/tests/data/Calculation/Engineering/HEX2DEC.php
@@ -12,6 +12,7 @@
     ['4886718345', '123456789'],
     [ExcelError::NAN(), '123.45'],
     ['0', '0'],
+    ['0', ''],
     [ExcelError::NAN(), 'G3579A'],
     [ExcelError::VALUE(), true],
     [ExcelError::VALUE(), false],

diff --git a/tests/data/Calculation/Engineering/OCT2DEC.php b/tests/data/Calculation/Engineering/OCT2DEC.php
@@ -15,6 +15,7 @@
     [ExcelError::NAN(), '3579'],
     ['44', '54'],
     ['-165', '7777777533'], // 2's Complement
+    ['0', ''],
     [ExcelError::NAN(), '37777777770'], // too many digits
     ['536870911', '3777777777'], // highest positive
     ['-536870912', '4000000000'], // lowest negative

diff --git a/tests/data/Calculation/MathTrig/ARABIC.php b/tests/data/Calculation/MathTrig/ARABIC.php
@@ -59,4 +59,12 @@
         '#VALUE!',
         'WRONG',
     ],
+    [
+        0,
+        '',
+    ],
+    [
+        '#NUM!',
+        '-',
+    ],
 ];
diff --git a/tests/data/Style/NumberFormatDates.php b/tests/data/Style/NumberFormatDates.php
@@ -8,38 +8,32 @@
         22269.0625,
         'dd-mm-yyyy hh:mm:ss',
     ],
-    // Oasis uses upper-case
-    [
+    'Oasis uses upper-case' => [
         '12/19/1960 01:30:00',
         22269.0625,
         'MM/DD/YYYY HH:MM:SS',
     ],
-    // Date with plaintext escaped with a \
-    [
+    'plaintext escaped with backslash' => [
         '1960-12-19T01:30:00',
         22269.0625,
         'yyyy-mm-dd\Thh:mm:ss',
     ],
-    // Date with plaintext in quotes
-    [
+    'plaintext in quotes' => [
         '1960-12-19T01:30:00 Z',
         22269.0625,
         'yyyy-mm-dd"T"hh:mm:ss \Z',
     ],
-    // Date with quoted formatting characters
-    [
+    'quoted formatting characters' => [
         'y-m-d 1960-12-19 h:m:s 01:30:00',
         22269.0625,
         '"y-m-d" yyyy-mm-dd "h:m:s" hh:mm:ss',
     ],
-    // Date with quoted formatting characters
-    [
-        'y-m-d 1960-12-19 h:m:s 01:30:00',
+    'quoted formatting non-ascii characters' => [
+        '§1960-12-19',
         22269.0625,
-        '"y-m-d "yyyy-mm-dd" h:m:s "hh:mm:ss',
+        '"§"yyyy-mm-dd',
     ],
-    // Date with fractional/decimal time
-    [
+    'fractional/decimal time' => [
         '2023/02/28 0:00:00.000',
         44985,
         'yyyy/mm/dd\ h:mm:ss.000',