Fix TEXT and TIMEVALUE Functions 1.29 Branch

Fix #4249. Technically speaking, only the 1.29 branch needs fixing, and only for TEXT. It was fixed for the other branches by PR #3898. However, in adding test cases for the fix, it became apparent that PhpSpreadsheet's parsing in TIMEVALUE (which is called from TEXT in the original issue) did not really match Excel's. There are probably still edge cases where it doesn't, but, in the absence of a spec for how it operates, this will do for now.

We do not usually backport fixes from the master branch. Because this is more of a forward port from the earlier branch, there is an equivalent PR for each active branch.

phpstan-baseline.neon

@@ -60,11 +60,6 @@ parameters:
 			count: 1
 			path: src/PhpSpreadsheet/Calculation/DateTimeExcel/Helpers.php
 
-		-
-			message: "#^Binary operation \"%%\" between string and 24 results in an error\\.$#"
-			count: 1
-			path: src/PhpSpreadsheet/Calculation/DateTimeExcel/TimeValue.php
-
 		-
 			message: "#^Binary operation \"\\-\" between 1 and array\\|float\\|string results in an error\\.$#"
 			count: 1

src/PhpSpreadsheet/Calculation/DateTimeExcel/TimeValue.php

@@ -2,6 +2,7 @@
 
 namespace PhpOffice\PhpSpreadsheet\Calculation\DateTimeExcel;
 
+use Composer\Pcre\Preg;
 use Datetime;
 use PhpOffice\PhpSpreadsheet\Calculation\ArrayEnabled;
 use PhpOffice\PhpSpreadsheet\Calculation\Functions;
@@ -12,6 +13,19 @@ class TimeValue
 {
     use ArrayEnabled;
 
+    private const EXTRACT_TIME = '/\b'
+        . '(\d+)' // match[1] - hour
+        . '(:' // start of match[2] (rest of string) - colon
+        . '(\d+' // start of match[3] - minute
+        . '(:\d+' // start of match[4] - colon and seconds
+        . '([.]\d+)?' // match[5] - optional decimal point followed by fractional seconds
+        . ')?' // end of match[4], which is optional
+        . ')' // end of match 3
+        // Excel does not require 'm' to trail 'a' or 'p'; Php does
+        . '(\s*(a|p))?' // match[6] optional whitespace followed by optional match[7] a or p
+        . ')' // end of match[2]
+        . '/i';
+
     /**
      * TIMEVALUE.
      *
@@ -48,12 +62,15 @@ public static function fromString($timeValue)
         }
 
         $timeValue = trim($timeValue ?? '', '"');
-        $timeValue = str_replace(['/', '.'], '-', $timeValue);
-
-        $arraySplit = preg_split('/[\/:\-\s]/', $timeValue) ?: [];
-        if ((count($arraySplit) == 2 || count($arraySplit) == 3) && $arraySplit[0] > 24) {
-            $arraySplit[0] = ($arraySplit[0] % 24);
-            $timeValue = implode(':', $arraySplit);
+        if (Preg::isMatch(self::EXTRACT_TIME, $timeValue, $matches)) {
+            if (empty($matches[6])) { // am/pm
+                $hour = (int) $matches[1];
+                $timeValue = ($hour % 24) . $matches[2];
+            } elseif ($matches[6] === $matches[7]) { // Excel wants space before am/pm
+                return ExcelError::VALUE();
+            } else {
+                $timeValue = $matches[0] . 'm';
+            }
         }
 
         $PHPDateArray = Helpers::dateParse($timeValue);

src/PhpSpreadsheet/Calculation/TextData/Format.php

@@ -2,6 +2,7 @@
 
 namespace PhpOffice\PhpSpreadsheet\Calculation\TextData;
 
+use Composer\Pcre\Preg;
 use DateTimeInterface;
 use PhpOffice\PhpSpreadsheet\Calculation\ArrayEnabled;
 use PhpOffice\PhpSpreadsheet\Calculation\Calculation;
@@ -128,8 +129,11 @@ public static function TEXTFORMAT($value, $format)
         $value = Helpers::extractString($value);
         $format = Helpers::extractString($format);
 
-        if (!is_numeric($value) && Date::isDateTimeFormatCode($format)) {
-            $value = DateTimeExcel\DateValue::fromString($value) + DateTimeExcel\TimeValue::fromString($value);
+        if (!is_numeric($value) && Date::isDateTimeFormatCode($format) && !Preg::isMatch('/^\s*\d+(\s+\d+)+\s*$/', $value)) {
+            $value1 = DateTimeExcel\DateValue::fromString($value);
+            $value2 = DateTimeExcel\TimeValue::fromString($value);
+            /** @var float|int|string */
+            $value = (is_numeric($value1) && is_numeric($value2)) ? ($value1 + $value2) : (is_numeric($value1) ? $value1 : (is_numeric($value2) ? $value2 : $value));
         }
 
         return (string) NumberFormat::toFormattedString($value, $format);

tests/data/Calculation/TextData/TEXT.php

@@ -71,6 +71,56 @@
         '2014-02-15 16:17',
         'dd-mmm-yyyy HH:MM:SS AM/PM',
     ],
+    'datetime integer' => [
+        '1900-01-06 00:00',
+        6,
+        'yyyy-mm-dd hh:mm',
+    ],
+    'datetime integer as string' => [
+        '1900-01-06 00:00',
+        '6',
+        'yyyy-mm-dd hh:mm',
+    ],
+    'datetime 2 integers without date delimiters' => [
+        '5 6',
+        '5 6',
+        'yyyy-mm-dd hh:mm',
+    ],
+    'datetime 2 integers separated by hyphen' => [
+        (new DateTimeImmutable())->format('Y') . '-05-13 00:00',
+        '5-13',
+        'yyyy-mm-dd hh:mm',
+    ],
+    'datetime string date only' => [
+        '1951-01-23 00:00',
+        'January 23, 1951',
+        'yyyy-mm-dd hh:mm',
+    ],
+    'datetime string time followed by date' => [
+        '1952-05-02 03:54',
+        '3:54 May 2, 1952',
+        'yyyy-mm-dd hh:mm',
+    ],
+    'datetime string date followed by time pm' => [
+        '1952-05-02 15:54',
+        'May 2, 1952 3:54 pm',
+        'yyyy-mm-dd hh:mm',
+    ],
+    'datetime string date followed by time p' => [
+        '1952-05-02 15:54',
+        'May 2, 1952 3:54 p',
+        'yyyy-mm-dd hh:mm',
+    ],
+    'datetime decimal string interpreted as time' => [
+        '1900-01-02 12:00',
+        '2.5',
+        'yyyy-mm-dd hh:mm',
+    ],
+    'datetime unparseable string' => [
+        'xyz',
+        'xyz',
+        'yyyy-mm-dd hh:mm',
+    ],
     [
         '1 3/4',
         1.75,