mbstring: deprecate mb_internal_encoding() and mb_http_output()

ext/mbstring/mbstring.c

@@ -24,6 +24,7 @@
 #include "php.h"
 #include "php_ini.h"
 #include "php_variables.h"
+#include "zend_attributes.h"
 #include "mbstring.h"
 #include "ext/standard/php_string.h"
 #include "ext/standard/php_mail.h"

ext/mbstring/mbstring.stub.php

@@ -55,6 +55,7 @@
 function mb_language(?string $language = null): string|bool {}
 
 /** @refcount 1 */
+ #[\Deprecated(since: '8.5', message: 'use internal_encoding INI settings instead')]
 function mb_internal_encoding(?string $encoding = null): string|bool {}
 
 /**
@@ -64,6 +65,7 @@ function mb_internal_encoding(?string $encoding = null): string|bool {}
 function mb_http_input(?string $type = null): array|string|false {}
 
 /** @refcount 1 */
+#[\Deprecated(since: '8.5', message: 'use output_encoding INI settings instead')]
 function mb_http_output(?string $encoding = null): string|bool {}
 
 /**

ext/mbstring/tests/bug45239.phpt

@@ -12,5 +12,6 @@ mb_internal_encoding("utf-8");
 mb_parse_str("a=%fc", $dummy);
 var_dump(mb_http_input());
 ?>
---EXPECT--
+--EXPECTF--
+Deprecated: Function mb_internal_encoding() is deprecated since 8.5, use internal_encoding INI settings instead in %s on line %d
 string(5) "UTF-8"

ext/mbstring/tests/bug48697.phpt

@@ -18,8 +18,22 @@ var_dump(mb_internal_encoding());
 ?>
 --EXPECTF--
 Deprecated: ini_set(): Use of mbstring.internal_encoding is deprecated in %s on line %d
+
+Deprecated: Function mb_internal_encoding() is deprecated since 8.5, use internal_encoding INI settings instead in %s on line %d
 string(11) "ISO-8859-15"
+
+Deprecated: Function mb_internal_encoding() is deprecated since 8.5, use internal_encoding INI settings instead in %s on line %d
+
+Deprecated: Function mb_internal_encoding() is deprecated since 8.5, use internal_encoding INI settings instead in %s on line %d
 string(5) "UTF-8"
+
+Deprecated: Function mb_internal_encoding() is deprecated since 8.5, use internal_encoding INI settings instead in %s on line %d
 string(5) "UTF-8"
+
+Deprecated: Function mb_internal_encoding() is deprecated since 8.5, use internal_encoding INI settings instead in %s on line %d
+
+Deprecated: Function mb_internal_encoding() is deprecated since 8.5, use internal_encoding INI settings instead in %s on line %d
 string(5) "UTF-8"
+
+Deprecated: Function mb_internal_encoding() is deprecated since 8.5, use internal_encoding INI settings instead in %s on line %d
 string(5) "UTF-8"

ext/mbstring/tests/bug69079.phpt

@@ -2,6 +2,8 @@
 Bug #69079 (enhancement for mb_substitute_character)
 --EXTENSIONS--
 mbstring
+--INI--
+error_reporting=E_ALL & ~E_DEPRECATED
 --FILE--
 <?php
 

ext/mbstring/tests/bug69086.phpt

@@ -2,6 +2,8 @@
 Request #69086 (enhancement for mb_convert_encoding)
 --EXTENSIONS--
 mbstring
+--INI--
+error_reporting=E_ALL & ~E_DEPRECATED
 --FILE--
 <?php
 mb_substitute_character(0xfffd);

ext/mbstring/tests/casefold.phpt

@@ -12,7 +12,8 @@ output_handler=
     print mb_strtoupper( "äëï\n" );
     print mb_convert_case( "äëï\n", MB_CASE_TITLE );
 ?>
---EXPECT--
+--EXPECTF--
+Deprecated: Function mb_internal_encoding() is deprecated since 8.5, use internal_encoding INI settings instead in %s on line %d
 abcdefghijklmnopqrstuvwxyz
 ABCDEFGHIJKLMNOPQRSTUVWXYZ
 ÄËÏ

ext/mbstring/tests/casemapping.phpt

@@ -65,7 +65,7 @@ echo mb_convert_case("ΚΑΛΗΣΠΕΡΑ ΣΑΣ", MB_CASE_LOWER, "UTF-8"), "\n";
 echo mb_convert_case("ΚΑΛΗΣΠΕΡΑ ΣΑΣ", MB_CASE_LOWER_SIMPLE, "UTF-8"), "\n";
 
 ?>
---EXPECT--
+--EXPECTF--
 String: ß
 Lower: ß
 Lower Simple: ß
@@ -96,6 +96,8 @@ Title Simple: Ä°
 Fold: i̇
 Fold Simple: Ä°
 
+
+Deprecated: Function mb_internal_encoding() is deprecated since 8.5, use internal_encoding INI settings instead in %s on line %d
 69
 69
 69

ext/mbstring/tests/gh13815.phpt

@@ -13,8 +13,10 @@ $strSjis = mb_convert_encoding($strUtf8, 'Shift_JIS', 'UTF-8');
 var_dump(mb_strlen(mb_trim($strSjis)));
 var_dump(mb_strlen(mb_trim($strSjis, encoding: 'Shift_JIS')));
 ?>
---EXPECT--
+--EXPECTF--
 int(1)
 int(1)
+
+Deprecated: Function mb_internal_encoding() is deprecated since 8.5, use internal_encoding INI settings instead in %s on line %d
 int(1)
 int(1)

ext/mbstring/tests/gh9535.phpt

@@ -2,6 +2,8 @@
 GH-9535 (mb_strcut(): The behavior of mb_strcut in mbstring has been changed in PHP8.1)
 --EXTENSIONS--
 mbstring
+--INI--
+error_reporting=E_ALL & ~E_DEPRECATED
 --FILE--
 <?php
 $encodings = [

ext/mbstring/tests/gh9535b.phpt

@@ -2,6 +2,8 @@
 Test output of mb_strcut for text encodings which use escape sequences
 --EXTENSIONS--
 mbstring
+--INI--
+error_reporting=E_ALL & ~E_DEPRECATED
 --FILE--
 <?php
 

ext/mbstring/tests/ini_language.phpt

@@ -10,6 +10,8 @@ mbstring.language=Japanese
 var_dump(ini_get('internal_encoding'));
 var_dump(mb_internal_encoding());
 ?>
---EXPECT--
+--EXPECTF--
 string(9) "Shift_JIS"
+
+Deprecated: Function mb_internal_encoding() is deprecated since 8.5, use internal_encoding INI settings instead in %s on line %d
 string(4) "SJIS"

ext/mbstring/tests/internal_encoding.phpt

@@ -46,23 +46,38 @@ var_dump(mb_strlen("\xc3\xb6"));
 
 ?>
 --EXPECTF--
+Deprecated: Function mb_internal_encoding() is deprecated since 8.5, use internal_encoding INI settings instead in %s on line %d
 string(10) "ISO-8859-1"
 int(2)
 
 Deprecated: ini_set(): Use of mbstring.internal_encoding is deprecated in %s on line %d
+
+Deprecated: Function mb_internal_encoding() is deprecated since 8.5, use internal_encoding INI settings instead in %s on line %d
 string(5) "UTF-8"
 int(1)
+
+Deprecated: Function mb_internal_encoding() is deprecated since 8.5, use internal_encoding INI settings instead in %s on line %d
 string(5) "UTF-8"
 int(1)
 
 Deprecated: ini_set(): Use of mbstring.internal_encoding is deprecated in %s on line %d
+
+Deprecated: Function mb_internal_encoding() is deprecated since 8.5, use internal_encoding INI settings instead in %s on line %d
 string(10) "ISO-8859-2"
 int(2)
+
+Deprecated: Function mb_internal_encoding() is deprecated since 8.5, use internal_encoding INI settings instead in %s on line %d
+
+Deprecated: Function mb_internal_encoding() is deprecated since 8.5, use internal_encoding INI settings instead in %s on line %d
 string(5) "UTF-8"
 int(1)
+
+Deprecated: Function mb_internal_encoding() is deprecated since 8.5, use internal_encoding INI settings instead in %s on line %d
 string(5) "UTF-8"
 int(1)
 
 Deprecated: ini_set(): Use of mbstring.internal_encoding is deprecated in %s on line %d
+
+Deprecated: Function mb_internal_encoding() is deprecated since 8.5, use internal_encoding INI settings instead in %s on line %d
 string(10) "ISO-8859-3"
 int(2)

ext/mbstring/tests/mb_chr.phpt

@@ -46,7 +46,7 @@ try {
 }
 
 ?>
---EXPECT--
+--EXPECTF--
 bool(true)
 bool(true)
 bool(true)
@@ -56,4 +56,6 @@ mb_chr(): Argument #2 ($encoding) must be a valid encoding, "pass" given
 mb_chr() does not support the "JIS" encoding
 mb_chr() does not support the "CP50222" encoding
 mb_chr() does not support the "UTF-7" encoding
+
+Deprecated: Function mb_internal_encoding() is deprecated since 8.5, use internal_encoding INI settings instead in %s on line %d
 mb_chr() does not support the "UTF-7" encoding

ext/mbstring/tests/mb_decode_mimeheader_variation4.phpt

@@ -8,89 +8,89 @@ mbstring
 // We convert runs of whitespace, including CR, LF, tab, and space, to a single space...
 // but ONLY when that run of whitespace does not occur right in the middle between two
 // valid MIME encoded words
-mb_internal_encoding('UCS-2');
+ini_set('internal_encoding', 'UCS-2');
 var_dump(bin2hex(mb_decode_mimeheader("2,\rGCG\xb3GS")));
 
 // We DO convert a run of whitespace to a single space at the very beginning of the input string,
 // as long as it is followed by a non-whitespace character
-mb_internal_encoding('ASCII');
+ini_set('internal_encoding', 'ASCII');
 var_dump(bin2hex(mb_decode_mimeheader("\n8i")));
 
 // But not if it is just a CR or LF at the end of the string
-mb_internal_encoding('ASCII');
+ini_set('internal_encoding', 'ASCII');
 var_dump(bin2hex(mb_decode_mimeheader("\r")));
 
 // Not if it is a run of whitespace going right up to the end of the string
-mb_internal_encoding('ASCII');
+ini_set('internal_encoding', 'ASCII');
 var_dump(bin2hex(mb_decode_mimeheader("\n ")));
 
 // Handle = which doesn't initiate a valid encoded word
-mb_internal_encoding('ASCII');
+ini_set('internal_encoding', 'ASCII');
 var_dump(bin2hex(mb_decode_mimeheader(",\x13@=,")));
 
 // Encoded word which should not be accepted
-mb_internal_encoding('ASCII');
+ini_set('internal_encoding', 'ASCII');
 var_dump(bin2hex(mb_decode_mimeheader("=?I?B??=")));
 
 // Encoded word with invalid charset name, and input string ends with whitespace
 // The old implementation of mb_decode_mimeheader would get 'stuck' on the invalid encoding name
 // and could never get out of that state; it would not even try to interpret any other encoded words
 // up to the end of the input string
-mb_internal_encoding('ISO-8859-7');
+ini_set('internal_encoding', 'ISO-8859-7');
 var_dump(bin2hex(mb_decode_mimeheader("=?=\x20?=?R\xb7=?=\x20?\x8b\x00====?===??=\xc5UC-R\xb7=?=\x20?=?=====?=\x20?======?======?=\x20?======?===??=\xc5UC-KR\xb7=?=\x20?=?===?==?\x0a")));
 
 // While the old implementation would generally trim off whitespace at the end of the input string,
 // but there was a bug whereby it would not do this when the whitespace character was the 100th
 // byte of an invalid charset name
-mb_internal_encoding('UCS-2');
+ini_set('internal_encoding', 'UCS-2');
 var_dump(bin2hex(mb_decode_mimeheader("=?\xc2\x86tf7,U\x01\x00`@\x00\x04|\xf1D\x18\x00\x00\x00v\xcb\xcb\xcb\xcb\xcb\xcb\xcb\xcb\xcb\xcb\xcb\xff\x13f7,U&\x00S\x01\x00\x17,D\xcb\xcb\xcb\xcb\xcb\xcb\xcb\x01\x00\x00\x14\xcb\xcb\xcb\xcb\xcb\xcb\xcb\xcb\xcb\xcb\xcb\xcb\xcb\xcb\xcb\xcb\xcb\xcb\xcb\xcb\xcb\xcb\xcb\xcb\xcb\xcb\xcb\xcb\xcb\xcb\xcb\xcb\xcb\x00\x11\x00\x00\x00\x00\x00\x00\x0a")));
 
 // Empty encoded word
-mb_internal_encoding('ASCII');
+ini_set('internal_encoding', 'ASCII');
 var_dump(bin2hex(mb_decode_mimeheader("=?us?B?")));
 
 // Encoded word with just one invalid Base64 byte
-mb_internal_encoding('ASCII');
+ini_set('internal_encoding', 'ASCII');
 var_dump(bin2hex(mb_decode_mimeheader("=?us?B?-")));
 
 // Encoded word with an invalid Base64 byte followed by a valid Base64 byte
-mb_internal_encoding('ASCII');
+ini_set('internal_encoding', 'ASCII');
 var_dump(bin2hex(mb_decode_mimeheader("=?us?B?-s")));
 
 // Empty encoded word with a ? which looks like it should be terminator, but = is missing
-mb_internal_encoding('ASCII');
+ini_set('internal_encoding', 'ASCII');
 var_dump(bin2hex(mb_decode_mimeheader("=?us?B??")));
 
 // Encoded word with just one invalid Base64 byte, but this time properly terminated
-mb_internal_encoding('ASCII');
+ini_set('internal_encoding', 'ASCII');
 var_dump(bin2hex(mb_decode_mimeheader("=?us?B?\x00?=")));
 
 // Invalid encoded word, followed immediately by valid encoded word
-mb_internal_encoding('ASCII');
+ini_set('internal_encoding', 'ASCII');
 var_dump(bin2hex(mb_decode_mimeheader("=?=?hz?b?")));
 
 // Another example of invalid encoded word followed immediately by valid encoded word
-mb_internal_encoding('ASCII');
+ini_set('internal_encoding', 'ASCII');
 var_dump(bin2hex(mb_decode_mimeheader("=?==?hz?b?")));
 
 // Yet another example
-mb_internal_encoding('ASCII');
+ini_set('internal_encoding', 'ASCII');
 var_dump(bin2hex(mb_decode_mimeheader("=?,=?hz?b?")));
 
 // In PHP 8.0-8.1 this would cause a crash
-mb_internal_encoding('UUENCODE');
+ini_set('internal_encoding', 'UUENCODE');
 var_dump(bin2hex(mb_decode_mimeheader("")));
 
 // The conversion filter for SJIS-Mobile#SOFTBANK did not work correctly when it was
 // passed the last buffer of wchars without passing 'end' flag, then called one more
 // time with an empty buffer and 'end' flag to finish up
-mb_internal_encoding('SJIS-Mobile#SOFTBANK');
+ini_set('internal_encoding', 'SJIS-Mobile#SOFTBANK');
 var_dump(bin2hex(mb_decode_mimeheader("6")));
 
 // Same for SJIS-Mobile#KDDI and SJIS-Mobile#DOCOMO
-mb_internal_encoding('SJIS-Mobile#KDDI');
+ini_set('internal_encoding', 'SJIS-Mobile#SOFTBANK');
 var_dump(bin2hex(mb_decode_mimeheader("6")));
-mb_internal_encoding('SJIS-Mobile#DOCOMO');
+ini_set('internal_encoding', 'SJIS-Mobile#SOFTBANK');
 var_dump(bin2hex(mb_decode_mimeheader("6")));
 
 ?>

ext/mbstring/tests/mb_encode_mimeheader_basic.phpt

@@ -24,7 +24,7 @@ foreach ($english as $lang => $input) {
     var_dump(mb_encode_mimeheader($input, 'UTF-8', 'Q'));
 }
 
-mb_internal_encoding('utf-8');
+ini_set('internal_encoding', 'utf-8');
 
 foreach ($nonEnglish as $lang => $input) {
     echo "\nLanguage: $lang\n";

ext/mbstring/tests/mb_encode_mimeheader_basic2.phpt

@@ -22,7 +22,7 @@ $inputs = array('SJIS' => $sjis_string,
 foreach ($inputs as $lang => $input) {
     echo "\nLanguage: $lang\n";
     echo "-- Base 64: --\n";
-    mb_internal_encoding($lang);
+    ini_set('internal_encoding', $lang);
     $outEncoding = "UTF-8";
     var_dump(mb_encode_mimeheader($input, $outEncoding, 'B'));
     echo "-- Quoted-Printable --\n";

ext/mbstring/tests/mb_encode_mimeheader_basic3.phpt

@@ -22,7 +22,7 @@ $inputs = array('SJIS' => $sjis_string,
 foreach ($inputs as $lang => $input) {
     echo "\nLanguage: $lang\n";
     echo "-- Base 64: --\n";
-    mb_internal_encoding($lang);
+    ini_set('internal_encoding', $lang);
     $outEncoding = $lang;
     var_dump(mb_encode_mimeheader($input, $outEncoding, 'B'));
     echo "-- Quoted-Printable --\n";