From: Alex Dowad <alexinbeijing@gmail.com>
Date: Mon, 9 Nov 2020 20:59:07 +0000 (+0200)
Subject: Test EUC-JP and Shift-JIS more thoroughly
X-Git-Url: https://granicus.if.org/sourcecode?a=commitdiff_plain;h=d1d50c2b7acc7e48331c46bf9ed2c3817ccd110a;p=php

Test EUC-JP and Shift-JIS more thoroughly

Previously, the unit tests for these text encodings covered all mappings
from legacy -> Unicode, and all _reversible_ mappings from Unicode -> legacy.
However, we should also test the few Unicode -> legacy mappings which
are not reversible.
---

diff --git a/ext/mbstring/tests/armscii8_encoding.phpt b/ext/mbstring/tests/armscii8_encoding.phpt
index 960836fe6b..b5bea077d0 100644
--- a/ext/mbstring/tests/armscii8_encoding.phpt
+++ b/ext/mbstring/tests/armscii8_encoding.phpt
@@ -23,7 +23,7 @@ testAllInvalidChars($invalid, $toUnicode, 'ARMSCII-8', 'UTF-16BE', "\x00%");
 testTruncatedChars($truncated, 'ARMSCII-8', 'UTF-16BE', "\x00%");
 echo "Tested ARMSCII-8 -> UTF-16BE\n";
 
-findInvalidChars($fromUnicode, $invalid, $unused, map(range(0,0xFF), 2));
+findInvalidChars($fromUnicode, $invalid, $unused, array_fill_keys(range(0,0xFF), 2));
 convertAllInvalidChars($invalid, $fromUnicode, 'UTF-16BE', 'ARMSCII-8', '%');
 echo "Tested UTF-16BE -> ARMSCII-8\n";
 ?>
diff --git a/ext/mbstring/tests/encoding_tests.inc b/ext/mbstring/tests/encoding_tests.inc
index 9a374b6137..e95bc8fd09 100644
--- a/ext/mbstring/tests/encoding_tests.inc
+++ b/ext/mbstring/tests/encoding_tests.inc
@@ -207,13 +207,6 @@ function findInvalidChars($valid, &$invalid, &$truncated, $startBytes = array())
     }
 }
 
-// Helper for building `$startBytes` map for above function
-function map($keys, $value, $array = array()) {
-    foreach ($keys as $key)
-        $array[$key] = $value;
-    return $array;
-}
-
 function testEncodingFromUTF16ConversionTable($path, $encoding, $replacement = '%', $startBytes = array()) {
     srand(1000); // Make results consistent
     mb_substitute_character(0x25); // '%'
@@ -225,7 +218,7 @@ function testEncodingFromUTF16ConversionTable($path, $encoding, $replacement = '
     testTruncatedChars($truncated, $encoding, 'UTF-16BE', "\x00%");
     echo "Tested $encoding -> UTF-16BE\n";
 
-    findInvalidChars($fromUnicode, $invalid, $unused, map(range(0,0xFF), 2));
+    findInvalidChars($fromUnicode, $invalid, $unused, array_fill_keys(range(0,0xFF), 2));
     convertAllInvalidChars($invalid, $fromUnicode, 'UTF-16BE', $encoding, $replacement);
     echo "Tested UTF-16BE -> $encoding\n";
 }
diff --git a/ext/mbstring/tests/eucjp_encoding.phpt b/ext/mbstring/tests/eucjp_encoding.phpt
index a05097cd84..dc321d3bae 100644
--- a/ext/mbstring/tests/eucjp_encoding.phpt
+++ b/ext/mbstring/tests/eucjp_encoding.phpt
@@ -43,7 +43,7 @@ $fromUnicode["\x00\x00\x00\x7E"] = "\x7E";
 /* Likewise with 0x005C */
 $fromUnicode["\x00\x00\x00\x5C"] = "\x5C";
 
-findInvalidChars($validChars, $invalidChars, $truncated, map(range(0xA1, 0xFE), 2, array(0x8E => 2, 0x8F => 3)));
+findInvalidChars($validChars, $invalidChars, $truncated, array_fill_keys(range(0xA1, 0xFE), 2) + array(0x8E => 2, 0x8F => 3));
 
 /* In the JIS X 0212 character set, kuten code 0x2237 (EUC-JP 0x8FA2B7)
  * is an ordinary tilde character
@@ -61,6 +61,9 @@ echo "Encoding verification and conversion work for all invalid characters\n";
 testValidString("\x8F\xA2\xB7", "\x00\x00\x00~", 'EUC-JP', 'UTF-32BE', false);
 echo "Irreversible mapping of 0x8FA2B7 follows JIS X 0212 correctly\n";
 
+testAllValidChars($fromUnicode, 'UTF-32BE', 'EUC-JP', false);
+echo "Unicode -> EUC-JP conversion works on all valid characters\n";
+
 $invalidChars = array();
 for ($cp = 0; $cp <= 0xFFFF; $cp++) {
   $char = pack('N', $cp);
@@ -74,4 +77,5 @@ echo "Unicode -> EUC-JP conversion works on all invalid characters\n";
 Encoding verification and conversion work for all valid characters
 Encoding verification and conversion work for all invalid characters
 Irreversible mapping of 0x8FA2B7 follows JIS X 0212 correctly
+Unicode -> EUC-JP conversion works on all valid characters
 Unicode -> EUC-JP conversion works on all invalid characters
diff --git a/ext/mbstring/tests/sjis_encoding.phpt b/ext/mbstring/tests/sjis_encoding.phpt
index 5445a90fbd..a75ce85a4f 100644
--- a/ext/mbstring/tests/sjis_encoding.phpt
+++ b/ext/mbstring/tests/sjis_encoding.phpt
@@ -46,16 +46,20 @@ testAllValidChars($validChars, 'Shift-JIS', 'UTF-16BE');
 echo "SJIS verification and conversion works on all valid characters\n";
 
 findInvalidChars($validChars, $invalidChars, $truncated,
-  map(range(0x81, 0x9F), 2, map(range(0xE0, 0xEF), 2)));
+  array_fill_keys(range(0x81, 0x9F), 2) + array_fill_keys(range(0xE0, 0xEF), 2));
 testAllInvalidChars($invalidChars, $validChars, 'Shift-JIS', 'UTF-16BE', "\x00%");
 testTruncatedChars($truncated, 'Shift-JIS', 'UTF-16BE', "\x00%");
 echo "SJIS verification and conversion works on all invalid characters\n";
 
-findInvalidChars($fromUnicode, $invalidChars, $unused, map(range(0, 0xFF), 2));
+testAllValidChars($fromUnicode, 'UTF-16BE', 'Shift-JIS', false);
+echo "Unicode -> SJIS conversion works on all valid characters\n";
+
+findInvalidChars($fromUnicode, $invalidChars, $unused, array_fill_keys(range(0, 0xFF), 2));
 convertAllInvalidChars($invalidChars, $fromUnicode, 'UTF-16BE', 'Shift-JIS', '%');
 echo "Unicode -> SJIS conversion works on all invalid characters\n";
 ?>
 --EXPECT--
 SJIS verification and conversion works on all valid characters
 SJIS verification and conversion works on all invalid characters
+Unicode -> SJIS conversion works on all valid characters
 Unicode -> SJIS conversion works on all invalid characters