Merge branch 'PHP-8.5'

* PHP-8.5:
  [intl] Fix empty-needle grapheme_strpos offsets
  Fix GH-23444: ODBC_ATTR_ASSUME_UTF8 corrupts Unicode data outside Windows (#23445)

Author: iliaal

NEWS

@@ -13,6 +13,8 @@ PHP                                                                        NEWS
     className/classList writes and attribute removals. (Ilia Alshanetsky)
 
 - Intl:
+  . Fixed grapheme_strpos() and grapheme_strrpos() with an empty needle
+    returning UTF-16 offsets instead of grapheme offsets. (Ilia Alshanetsky)
   . Fixed a memory leak when dumping IntlCalendar instances. (Ilia Alshanetsky)
   . Fixed Collator::sortWithSortKeys() allocating fixed 2MiB buffers
     regardless of array size. (Ilia Alshanetsky)

ext/intl/grapheme/grapheme_util.cpp

@@ -139,7 +139,11 @@ U_CFUNC int32_t grapheme_strpos_utf16(char *haystack, size_t haystack_len, char
 			ret_pos = -1;
 			goto finish;
 		}
-		ret_pos = last && offset >= 0 ? uhaystack_len : offset_pos;
+		if (last && offset >= 0) {
+			ret_pos = grapheme_count_graphemes(bi, uhaystack, uhaystack_len);
+		} else {
+			ret_pos = grapheme_count_graphemes(bi, uhaystack, offset_pos);
+		}
 		goto finish;
 	}
 

ext/intl/tests/grapheme_empty_offset_multibyte.phpt

@@ -0,0 +1,34 @@
+--TEST--
+grapheme_strpos() family with empty needle and offset on multi-code-unit graphemes
+--EXTENSIONS--
+intl
+--FILE--
+<?php
+
+var_dump(grapheme_strpos("😀x", ""));
+var_dump(grapheme_strpos("😀x", "", 0));
+var_dump(grapheme_strpos("😀x", "", 1));
+var_dump(grapheme_stripos("😀x", "", 1));
+var_dump(grapheme_strpos("😀x", "", -1));
+var_dump(grapheme_strrpos("😀x", ""));
+var_dump(grapheme_strrpos("😀x", "", 1));
+var_dump(grapheme_strripos("😀x", "", 1));
+var_dump(grapheme_strrpos("😀x", "", -1));
+try {
+    var_dump(grapheme_strpos("😀x", "", 5));
+} catch (Throwable $e) {
+    echo $e::class, ': ', $e->getMessage(), "\n";
+}
+
+?>
+--EXPECT--
+int(0)
+int(0)
+int(1)
+int(1)
+int(1)
+int(2)
+int(2)
+int(2)
+int(1)
+ValueError: grapheme_strpos(): Argument #3 ($offset) must be contained in argument #1 ($haystack)

ext/pdo_odbc/odbc_stmt.c

@@ -35,6 +35,7 @@ enum pdo_odbc_conv_result {
 
 static int pdo_odbc_sqltype_is_unicode(pdo_odbc_stmt *S, SQLSMALLINT sqltype)
 {
+#ifdef PHP_WIN32
 	if (!S->assume_utf8) return 0;
 	switch (sqltype) {
 #ifdef SQL_WCHAR
@@ -52,6 +53,9 @@ static int pdo_odbc_sqltype_is_unicode(pdo_odbc_stmt *S, SQLSMALLINT sqltype)
 		default:
 			return 0;
 	}
+#else
+	return 0;
+#endif
 }
 
 static int pdo_odbc_utf82ucs2(pdo_stmt_t *stmt, int is_unicode, const char *buf,
@@ -544,7 +548,15 @@ static int odbc_stmt_param_hook(pdo_stmt_t *stmt, struct pdo_bound_param_data *p
 								break;
 						}
 					} else {
-						P->len = SQL_LEN_DATA_AT_EXEC(Z_STRLEN_P(parameter));
+						zend_ulong ulen;
+						if (pdo_odbc_utf82ucs2(stmt, P->is_unicode,
+									Z_STRVAL_P(parameter),
+									Z_STRLEN_P(parameter),
+									&ulen) == PDO_ODBC_CONV_OK) {
+							P->len = SQL_LEN_DATA_AT_EXEC(ulen);
+						} else {
+							P->len = SQL_LEN_DATA_AT_EXEC(Z_STRLEN_P(parameter));
+						}
 					}
 				}
 				return 1;

ext/pdo_odbc/tests/gh23444.phpt

@@ -0,0 +1,34 @@
+--TEST--
+GH-23444 (Unicode data is corrupted with ODBC_ATTR_ASSUME_UTF8)
+--EXTENSIONS--
+pdo_odbc
+--SKIPIF--
+<?php
+require 'ext/pdo/tests/pdo_test.inc';
+PDOTest::skip();
+?>
+--FILE--
+<?php
+require 'ext/pdo/tests/pdo_test.inc';
+$db = PDOTest::test_factory(__DIR__ . '/common.phpt');
+$db->exec("CREATE TABLE gh23444 (v NVARCHAR(100))");
+
+// 13 bytes as UTF-8, so an unconverted parameter is an odd number of bytes
+$string = "\u{6e2c}\u{8a66}\u{4e2d}\u{1f418}";
+
+$db->setAttribute(PDO::ODBC_ATTR_ASSUME_UTF8, true);
+$stmt = $db->prepare("INSERT INTO gh23444 VALUES(?)");
+$stmt->execute([$string]);
+
+$stmt = $db->prepare("SELECT v FROM gh23444 WHERE v = ?");
+$stmt->execute([$string]);
+var_dump($stmt->fetchColumn() === $string);
+?>
+--CLEAN--
+<?php
+require 'ext/pdo/tests/pdo_test.inc';
+$db = PDOTest::test_factory(__DIR__ . '/common.phpt');
+$db->exec("DROP TABLE IF EXISTS gh23444");
+?>
+--EXPECT--
+bool(true)