Fixed multibyte related issues

php · Mar 14, 2011 · bbc879b · bbc879b
1 parent 0be5ca5
commit bbc879b
Show file tree

Hide file tree

Showing 11 changed files with 42 additions and 41 deletions.
diff --git a/Zend/tests/multibyte/multibyte_encoding_002.phpt b/Zend/tests/multibyte/multibyte_encoding_002.phpt
@@ -11,7 +11,7 @@ if (!extension_loaded("mbstring")) {
 ?>
 --INI--
 zend.multibyte=1
-zend.internal_encoding=iso-8859-1
+mbstring.internal_encoding=iso-8859-1
 --FILE--
 <?php
 print "Hello World\n";

diff --git a/Zend/zend_language_scanner.l b/Zend/zend_language_scanner.l
@@ -346,9 +346,7 @@ static const zend_encoding* zend_multibyte_detect_unicode(TSRMLS_D)
 
 	if (script_encoding) {
 		/* remove BOM */
-		script = (unsigned char*)emalloc(LANG_SCNG(script_org_size)+1-bom_size);
-		memcpy(script, LANG_SCNG(script_org)+bom_size, LANG_SCNG(script_org_size)+1-bom_size);
-		LANG_SCNG(script_org) = script;
+		LANG_SCNG(script_org) += bom_size;
 		LANG_SCNG(script_org_size) -= bom_size;
 
 		return script_encoding;

diff --git a/ext/exif/exif.c b/ext/exif/exif.c
@@ -2664,13 +2664,13 @@ static int exif_process_user_comment(image_info_type *ImageInfo, char **pszInfoP
 				decode = ImageInfo->decode_unicode_le;
 			}
 			if (zend_multibyte_encoding_converter(
-					pszInfoPtr, 
+					(unsigned char**)pszInfoPtr, 
 					&len, 
-					szValuePtr,
+					(unsigned char*)szValuePtr,
 					ByteCount,
-					ImageInfo->encode_unicode,
-					decode
-					TSRMLS_CC) != 0) {
+					zend_multibyte_fetch_encoding(ImageInfo->encode_unicode TSRMLS_CC),
+					zend_multibyte_fetch_encoding(decode TSRMLS_CC)
+					TSRMLS_CC) < 0) {
 				len = exif_process_string_raw(pszInfoPtr, szValuePtr, ByteCount);
 			}
 			return len;
@@ -2684,13 +2684,13 @@ static int exif_process_user_comment(image_info_type *ImageInfo, char **pszInfoP
 			szValuePtr = szValuePtr+8;
 			ByteCount -= 8;
 			if (zend_multibyte_encoding_converter(
-					pszInfoPtr, 
+					(unsigned char**)pszInfoPtr, 
 					&len, 
-					szValuePtr,
+					(unsigned char*)szValuePtr,
 					ByteCount,
-					ImageInfo->encode_jis,
-					ImageInfo->motorola_intel ? ImageInfo->decode_jis_be : ImageInfo->decode_jis_le
-					TSRMLS_CC) != 0) {
+					zend_multibyte_fetch_encoding(ImageInfo->encode_jis TSRMLS_CC),
+					zend_multibyte_fetch_encoding(ImageInfo->motorola_intel ? ImageInfo->decode_jis_be : ImageInfo->decode_jis_le TSRMLS_CC)
+					TSRMLS_CC) < 0) {
 				len = exif_process_string_raw(pszInfoPtr, szValuePtr, ByteCount);
 			}
 			return len;
@@ -2723,13 +2723,13 @@ static int exif_process_unicode(image_info_type *ImageInfo, xp_field_type *xp_fi
 
 	/* Copy the comment */
 	if (zend_multibyte_encoding_converter(
-			&xp_field->value, 
+			(unsigned char**)&xp_field->value, 
 			&xp_field->size, 
-			szValuePtr,
+			(unsigned char*)szValuePtr,
 			ByteCount,
-			ImageInfo->encode_unicode,
-			ImageInfo->motorola_intel ? ImageInfo->decode_unicode_be : ImageInfo->decode_unicode_le
-			TSRMLS_CC) != 0) {
+			zend_multibyte_fetch_encoding(ImageInfo->encode_unicode TSRMLS_CC),
+			zend_multibyte_fetch_encoding(ImageInfo->motorola_intel ? ImageInfo->decode_unicode_be : ImageInfo->decode_unicode_le TSRMLS_CC)
+			TSRMLS_CC) < 0) {
 		xp_field->size = exif_process_string_raw(&xp_field->value, szValuePtr, ByteCount);
 	}
 	return xp_field->size;

diff --git a/ext/mbstring/libmbfl/filters/mbfilter_ascii.c b/ext/mbstring/libmbfl/filters/mbfilter_ascii.c
@@ -37,7 +37,7 @@
 
 static int mbfl_filt_ident_ascii(int c, mbfl_identify_filter *filter);
 
-static const char *mbfl_encoding_ascii_aliases[] = {"ANSI_X3.4-1968", "iso-ir-6", "ANSI_X3.4-1986", "ISO_646.irv:1991", "US-ASCII", "ISO646-US", "us", "IBM367", "cp367", "csASCII", NULL};
+static const char *mbfl_encoding_ascii_aliases[] = {"ANSI_X3.4-1968", "iso-ir-6", "ANSI_X3.4-1986", "ISO_646.irv:1991", "US-ASCII", "ISO646-US", "us", "IBM367", "IBM-367", "cp367", "csASCII", NULL};
 
 const mbfl_encoding mbfl_encoding_ascii = {
 	mbfl_no_encoding_ascii,

diff --git a/ext/mbstring/libmbfl/filters/mbfilter_cp5022x.c b/ext/mbstring/libmbfl/filters/mbfilter_cp5022x.c
@@ -462,7 +462,7 @@ mbfl_filt_conv_wchar_jis_ms(int c, mbfl_convert_filter *filter)
 			s = 0x224c;
 		}
 	}
-	if (s <= 0 || s >= 0x8080 && s < 0x10000) {
+	if (s <= 0 || (s >= 0x8080 && s < 0x10000)) {
 		int i;
 		s = -1;
 
@@ -693,7 +693,7 @@ mbfl_filt_conv_wchar_cp50221(int c, mbfl_convert_filter *filter)
 			s = 0x224c;
 		}
 	}
-	if (s <= 0 || s >= 0x8080 && s < 0x10000) {
+	if (s <= 0 || (s >= 0x8080 && s < 0x10000)) {
 		int i;
 		s = -1;
 
@@ -841,7 +841,7 @@ mbfl_filt_conv_wchar_cp50222(int c, mbfl_convert_filter *filter)
 			s = 0x224c;
 		}
 	}
-	if (s <= 0 || s >= 0x8080 && s < 0x10000) {
+	if (s <= 0 || (s >= 0x8080 && s < 0x10000)) {
 		int i;
 		s = -1;
 

diff --git a/ext/mbstring/libmbfl/filters/mbfilter_cp850.c b/ext/mbstring/libmbfl/filters/mbfilter_cp850.c
@@ -33,7 +33,7 @@
 
 static int mbfl_filt_ident_cp850(int c, mbfl_identify_filter *filter);
 
-static const char *mbfl_encoding_cp850_aliases[] = {"CP850", "CP-850", "IBM-850", NULL};
+static const char *mbfl_encoding_cp850_aliases[] = {"CP850", "CP-850", "IBM850", "IBM-850", NULL};
 
 const mbfl_encoding mbfl_encoding_cp850 = {
 	mbfl_no_encoding_cp850,

diff --git a/ext/mbstring/libmbfl/filters/mbfilter_cp866.c b/ext/mbstring/libmbfl/filters/mbfilter_cp866.c
@@ -37,7 +37,7 @@
 
 static int mbfl_filt_ident_cp866(int c, mbfl_identify_filter *filter);
 
-static const char *mbfl_encoding_cp866_aliases[] = {"CP866", "CP-866", "IBM-866", NULL};
+static const char *mbfl_encoding_cp866_aliases[] = {"CP866", "CP-866", "IBM866", "IBM-866", NULL};
 
 const mbfl_encoding mbfl_encoding_cp866 = {
 	mbfl_no_encoding_cp866,

diff --git a/ext/mbstring/libmbfl/mbfl/mbfl_memory_device.c b/ext/mbstring/libmbfl/mbfl/mbfl_memory_device.c
@@ -218,7 +218,7 @@ mbfl_memory_device_strcat(mbfl_memory_device *device, const char *psrc)
 	const unsigned char *p;
 
 	len = 0;
-	p = psrc;
+	p = (const unsigned char*)psrc;
 	while (*p) {
 		p++;
 		len++;
@@ -235,7 +235,7 @@ mbfl_memory_device_strcat(mbfl_memory_device *device, const char *psrc)
 		device->buffer = tmp;
 	}
 
-	p = psrc;
+	p = (const unsigned char*)psrc;
 	w = &device->buffer[device->pos];
 	device->pos += len;
 	while (len > 0) {

diff --git a/ext/mbstring/mb_gpc.c b/ext/mbstring/mb_gpc.c
@@ -282,7 +282,7 @@ const mbfl_encoding *_php_mb_encoding_handler_ex(const php_mb_encoding_handler_i
 			if (info->report_errors) {
 				php_error_docref(NULL TSRMLS_CC, E_WARNING, "Unable to detect encoding");
 			}
-			from_encoding = mbfl_no_encoding_pass;
+			from_encoding = &mbfl_encoding_pass;
 		}
 	}
 

diff --git a/ext/mbstring/mbstring.c b/ext/mbstring/mbstring.c
@@ -2958,7 +2958,7 @@ MBSTRING_API char * php_mb_convert_encoding(const char *input, size_t length, co
 				string.no_encoding = from_encoding->no_encoding;
 			} else {
 				php_error_docref(NULL TSRMLS_CC, E_WARNING, "Unable to detect character encoding");
-				from_encoding = mbfl_no_encoding_pass;
+				from_encoding = &mbfl_encoding_pass;
 				to_encoding = from_encoding;
 				string.no_encoding = from_encoding->no_encoding;
 			}
@@ -3496,7 +3496,7 @@ PHP_FUNCTION(mb_convert_variables)
 		break;
 	}
 	if (elistsz <= 0) {
-		from_encoding = mbfl_no_encoding_pass;
+		from_encoding = &mbfl_encoding_pass;
 	} else if (elistsz == 1) {
 		from_encoding = *elist;
 	} else {
@@ -3565,15 +3565,15 @@ PHP_FUNCTION(mb_convert_variables)
 
 		if (!from_encoding) {
 			php_error_docref(NULL TSRMLS_CC, E_WARNING, "Unable to detect encoding");
-			from_encoding = mbfl_no_encoding_pass;
+			from_encoding = &mbfl_encoding_pass;
 		}
 	}
 	if (elist != NULL) {
 		efree((void *)elist);
 	}
 	/* create converter */
 	convd = NULL;
-	if (from_encoding != mbfl_no_encoding_pass) {
+	if (from_encoding != &mbfl_encoding_pass) {
 		convd = mbfl_buffer_converter_new2(from_encoding, to_encoding, 0);
 		if (convd == NULL) {
 			php_error_docref(NULL TSRMLS_CC, E_WARNING, "Unable to create converter");
@@ -4418,6 +4418,7 @@ PHP_FUNCTION(mb_get_info)
 			array_init(return_value);
 			for (i = 0; i < n; i++) {
 				add_next_index_string(return_value, (*entry)->name, 1);
+				entry++;
 			}
 		}
 	} else if (!strcasecmp("substitute_character", typ)) {
@@ -4693,7 +4694,7 @@ static void php_mb_gpc_get_detect_order(const zend_encoding ***list, size_t *lis
 
 static void php_mb_gpc_set_input_encoding(const zend_encoding *encoding TSRMLS_DC) /* {{{ */
 {
-	MBSTRG(http_input_identify) = encoding;
+	MBSTRG(http_input_identify) = (const mbfl_encoding*)encoding;
 }
 /* }}} */
 

diff --git a/ext/mbstring/tests/mb_encoding_aliases.phpt b/ext/mbstring/tests/mb_encoding_aliases.phpt
@@ -13,26 +13,28 @@ var_dump(mb_encoding_aliases("8bit"));
 ?>
 --EXPECTF--
 Warning: mb_encoding_aliases() expects exactly 1 parameter, 0 given in %s on line 2
-array(10) {
+array(11) {
   [0]=>
   string(14) "ANSI_X3.4-1968"
   [1]=>
   string(14) "ANSI_X3.4-1986"
   [2]=>
-  string(6) "IBM367"
+  string(7) "IBM-367"
   [3]=>
-  string(9) "ISO646-US"
+  string(6) "IBM367"
   [4]=>
-  string(16) "ISO_646.irv:1991"
+  string(9) "ISO646-US"
   [5]=>
-  string(8) "US-ASCII"
+  string(16) "ISO_646.irv:1991"
   [6]=>
-  string(5) "cp367"
+  string(8) "US-ASCII"
   [7]=>
-  string(7) "csASCII"
+  string(5) "cp367"
   [8]=>
-  string(8) "iso-ir-6"
+  string(7) "csASCII"
   [9]=>
+  string(8) "iso-ir-6"
+  [10]=>
   string(2) "us"
 }
 array(0) {