make test for 4.4.3RC1（続き）- mb_substr編
あー、忙しくて現実逃避する時間も無い。。＿|￣|○
先日の件ですが。
「mbstring.func_overload」の機能を使って文字列関数をオーバーロードするとmb_substr関数のテストが失敗するようになりました（コレの３番目）。
何故、失敗するようになったかを調べてみました。まずはテスト。
<?php declare( encoding="EUC-JP" ); echo PHP_VERSION."\n"; var_dump( mb_language( ) ); var_dump( mb_internal_encoding( ) ); $euc_jp = "0123この文字列は日本語です。"; var_dump( substr( $euc_jp, 100, 10 ) ); var_dump( mb_substr( $euc_jp, 100, 10 ) ); ?> 結果 ---- % ./php-4.3.3-cli ./mb_substr.php 4.3.3 string(8) "Japanese" string(9) "eucJP-win" bool(false) string(0) "" % ./php-4.3.9-cli ./mb_substr.php 4.3.9 string(8) "Japanese" string(9) "eucJP-win" bool(false) string(0) "" % ./php-4.3.11-cli ./mb_substr.php 4.3.11 string(8) "Japanese" string(9) "eucJP-win" bool(false) string(0) "" % ./php-4.4.1-cli ./mb_substr.php 4.4.1 string(8) "Japanese" string(9) "eucJP-win" bool(false) string(0) "" % ./php-4.4.2-cli ./mb_substr.php 4.4.2 string(8) "Japanese" string(9) "eucJP-win" bool(false) string(0) "" % ./php-4.4.3RC1-cli ./mb_substr.php 4.4.3RC1 string(8) "Japanese" string(9) "eucJP-win" bool(false) string(0) "" % ./php-4.3.3-cli -d mbstring.func_overload=2 ./mb_substr.php 4.3.3 string(8) "Japanese" string(9) "eucJP-win" string(0) "" string(0) "" % ./php-4.3.9-cli -d mbstring.func_overload=2 ./mb_substr.php 4.3.9 string(8) "Japanese" string(9) "eucJP-win" string(0) "" string(0) "" % ./php-4.3.11-cli -d mbstring.func_overload=2 ./mb_substr.php 4.3.11 string(8) "Japanese" string(9) "eucJP-win" string(0) "" string(0) "" % ./php-4.4.1-cli -d mbstring.func_overload=2 ./mb_substr.php 4.4.1 string(8) "Japanese" string(9) "eucJP-win" string(0) "" string(0) "" % ./php-4.4.2-cli -d mbstring.func_overload=2 ./mb_substr.php 4.4.2 string(8) "Japanese" string(9) "eucJP-win" bool(false) bool(false) % ./php-4.4.3RC1-cli -d mbstring.func_overload=2 ./mb_substr.php 4.4.3RC1 string(8) "Japanese" string(9) "eucJP-win" bool(false) bool(false)
テストの結果から、4.4.1〜4.4.2間の変更で挙動が変わった事が分かります。
カンが鋭い人は、もう分かったかもしれませんが。（ｗ
でもって、4.4.1〜4.4.2間の変更点は以下の通りです。長いけど。
--- mbstring.c 2005/09/21 13:19:19 1.142.2.47.2.1
+++ mbstring.c 2005/12/30 05:25:54 1.142.2.47.2.7
@@ -17,7 +17,7 @@
+----------------------------------------------------------------------+
*/
-/* $Id: mbstring.c,v 1.142.2.47.2.1 2005/09/21 13:19:19 iliaa Exp $ */
+/* $Id: mbstring.c,v 1.142.2.47.2.7 2005/12/30 05:25:54 hirokawa Exp $ */
/*
* PHP4 Multibyte String module "mbstring"
@@ -1555,7 +1555,7 @@
from_encoding = mbfl_no_encoding_invalid;
identd = mbfl_encoding_detector_new(
(enum mbfl_no_encoding *)info->from_encodings,
- info->num_from_encodings);
+ info->num_from_encodings, 0);
if (identd) {
n = 0;
while (n < num) {
@@ -1902,7 +1902,7 @@
} else {
/* auto detect */
from_encoding = mbfl_no_encoding_invalid;
- identd = mbfl_encoding_detector_new(elist, elistsz);
+ identd = mbfl_encoding_detector_new(elist, elistsz, 0);
if (identd != NULL) {
n = 0;
while (n < num) {
@@ -2404,6 +2404,11 @@
}
}
+ if (((MBSTRG(func_overload) & MB_OVERLOAD_STRING) == MB_OVERLOAD_STRING)
+ && (from >= mbfl_strlen(&string))) {
+ RETURN_FALSE;
+ }
+
ret = mbfl_substr(&string, &result, from, len);
if (ret != NULL) {
RETVAL_STRINGL((char *)ret->val, ret->len, 0); /* the string is already strdup()'ed */
@@ -2485,6 +2490,13 @@
}
}
+ if (from > Z_STRLEN_PP(arg1)) {
+ RETURN_FALSE;
+ }
+ if (((unsigned) from + (unsigned) len) > Z_STRLEN_PP(arg1)) {
+ len = Z_STRLEN_PP(arg1) - from;
+ }
+
ret = mbfl_strcut(&string, &result, from, len);
if (ret != NULL) {
RETVAL_STRINGL(ret->val, ret->len, 0); /* the string is already strdup()'ed */
@@ -2656,7 +2668,7 @@
string.no_encoding = from_encoding;
} else if (size > 1) {
/* auto detect */
- from_encoding = mbfl_identify_encoding_no(&string, list, size);
+ from_encoding = mbfl_identify_encoding_no(&string, list, size, 0);
if (from_encoding != mbfl_no_encoding_invalid) {
string.no_encoding = from_encoding;
} else {
@@ -2924,16 +2936,17 @@
}
/* }}} */
-/* {{{ proto string mb_encode_mimeheader(string str [, string charset [, string transfer-encoding [, string linefeed]]])
+/* {{{ proto string mb_encode_mimeheader(string str [, string charset [, string transfer-encoding [, string linefeed [, int indent]]]])
Converts the string to MIME "encoded-word" in the format of =?charset?(B|Q)?encoded_string?= */
PHP_FUNCTION(mb_encode_mimeheader)
{
- pval **argv[4];
+ pval **argv[5];
enum mbfl_no_encoding charset, transenc;
mbfl_string string, result, *ret;
char *p, *linefeed;
+ int indent;
- if (ZEND_NUM_ARGS() < 1 || ZEND_NUM_ARGS() > 4 || zend_get_parameters_array_ex(ZEND_NUM_ARGS(), argv) == FAILURE) {
+ if (ZEND_NUM_ARGS() < 1 || ZEND_NUM_ARGS() > 5 || zend_get_parameters_array_ex(ZEND_NUM_ARGS(), argv) == FAILURE) {
WRONG_PARAM_COUNT;
}
@@ -2970,6 +2983,12 @@
linefeed = Z_STRVAL_PP(argv[3]);
}
+ indent = 0;
+ if (ZEND_NUM_ARGS() >= 5) {
+ convert_to_long_ex(argv[4]);
+ indent = Z_LVAL_PP(argv[4]);
+ }
+
convert_to_string_ex(argv[0]);
mbfl_string_init(&string);
mbfl_string_init(&result);
@@ -2977,7 +2996,7 @@
string.no_encoding = MBSTRG(current_internal_encoding);
string.val = Z_STRVAL_PP(argv[0]);
string.len = Z_STRLEN_PP(argv[0]);
- ret = mbfl_mime_header_encode(&string, &result, charset, transenc, linefeed, 0);
+ ret = mbfl_mime_header_encode(&string, &result, charset, transenc, linefeed, indent);
if (ret != NULL) {
RETVAL_STRINGL((char *)ret->val, ret->len, 0) /* the string is already strdup()'ed */
} else {
@@ -3195,7 +3214,7 @@
stack = (pval ***)safe_emalloc(stack_max, sizeof(pval **), 0);
if (stack != NULL) {
stack_level = 0;
- identd = mbfl_encoding_detector_new(elist, elistsz);
+ identd = mbfl_encoding_detector_new(elist, elistsz, 0);
if (identd != NULL) {
n = 2;
while (n < argc || stack_level > 0) {
@@ -3460,6 +3479,22 @@
* Sends an email message with MIME scheme
*/
#if HAVE_SENDMAIL
+#define SKIP_LONG_HEADER_SEP_MBSTRING(str, pos) \
+ if (str[pos] == '\r' && str[pos + 1] == '\n' && (str[pos + 2] == ' ' || str[pos + 2] == '\t')) { \
+ pos += 3; \
+ while (str[pos] == ' ' || str[pos] == '\t') { \
+ pos++; \
+ } \
+ continue; \
+ } \
+ else if (str[pos] == '\n' && (str[pos + 1] == ' ' || str[pos + 1] == '\t')) { \
+ pos += 2; \
+ while (str[pos] == ' ' || str[pos] == '\t') { \
+ pos++; \
+ } \
+ continue; \
+ } \
+
PHP_FUNCTION(mb_send_mail)
{
int argc, n;
@@ -3474,8 +3509,9 @@
body_enc; /* body transfar encoding */
mbfl_memory_device device; /* automatic allocateable buffer for additional header */
const mbfl_language *lang;
- char *force_extra_parameters = INI_STR("mail.force_extra_parameters");
int err = 0;
+ char *to_r;
+ int to_len, i;
/* initialize */
mbfl_memory_device_init(&device, 0, 0);
@@ -3502,6 +3538,32 @@
convert_to_string_ex(argv[0]);
if (Z_STRVAL_PP(argv[0])) {
to = Z_STRVAL_PP(argv[0]);
+ to_len = Z_STRLEN_PP(argv[0]);
+ if (to_len > 0) {
+ to_r = estrndup(to, to_len);
+ for (; to_len; to_len--) {
+ if (!isspace((unsigned char) to_r[to_len - 1])) {
+ break;
+ }
+ to_r[to_len - 1] = '\0';
+ }
+ for (i = 0; to_r[i]; i++) {
+ if (iscntrl((unsigned char) to_r[i])) {
+ /* According to RFC 822, section 3.1.1 long headers may be
+separated into
+ * parts using CRLF followed at least one linear-white-space
+character ('\t' or ' ').
+ * To prevent these separators from being replaced with a space,
+we use the
+ * SKIP_LONG_HEADER_SEP_MBSTRING to skip over them.
+ */
+ SKIP_LONG_HEADER_SEP_MBSTRING(to_r, i);
+ to_r[i] = ' ';
+ }
+ }
+ } else {
+ to_r = to;
+ }
} else {
php_error_docref(NULL TSRMLS_CC, E_WARNING, "Missing To: field");
err = 1;
@@ -3516,7 +3578,7 @@
orig_str.no_encoding = MBSTRG(current_internal_encoding);
if (orig_str.no_encoding == mbfl_no_encoding_invalid
|| orig_str.no_encoding == mbfl_no_encoding_pass) {
- orig_str.no_encoding = mbfl_identify_encoding_no(&orig_str, MBSTRG(current_detect_order_list), MBSTRG(current_detect_order_list_size));
+ orig_str.no_encoding = mbfl_identify_encoding_no(&orig_str, MBSTRG(current_detect_order_list), MBSTRG(current_detect_order_list_size), 0);
}
pstr = mbfl_mime_header_encode(&orig_str, &conv_str, tran_cs, head_enc, "\n", sizeof("Subject: [PHP-jp nnnnnnnn]"));
if (pstr != NULL) {
@@ -3539,7 +3601,7 @@
if (orig_str.no_encoding == mbfl_no_encoding_invalid
|| orig_str.no_encoding == mbfl_no_encoding_pass) {
- orig_str.no_encoding = mbfl_identify_encoding_no(&orig_str, MBSTRG(current_detect_order_list), MBSTRG(current_detect_order_list_size));
+ orig_str.no_encoding = mbfl_identify_encoding_no(&orig_str, MBSTRG(current_detect_order_list), MBSTRG(current_detect_order_list_size), 0);
}
pstr = NULL;
@@ -3596,18 +3658,19 @@
extra_cmd = Z_STRVAL_PP(argv[4]);
}
- if (force_extra_parameters) {
- extra_cmd = estrdup(force_extra_parameters);
- } else if (extra_cmd) {
+ if (extra_cmd) {
extra_cmd = php_escape_shell_cmd(extra_cmd);
}
- if (!err && php_mail(to, subject, message, headers, extra_cmd TSRMLS_CC)) {
+ if (!err && php_mail(to_r, subject, message, headers, extra_cmd TSRMLS_CC)) {
RETVAL_TRUE;
} else {
RETVAL_FALSE;
}
+ if (to_r != to) {
+ efree(to_r);
+ }
if (extra_cmd) {
efree(extra_cmd);
}
@@ -3901,7 +3964,7 @@
mbfl_string_init(&string);
string.no_language = MBSTRG(current_language);
- identd = mbfl_encoding_detector_new(elist, size);
+ identd = mbfl_encoding_detector_new(elist, size, 0);
if (identd) {
int n = 0;でもって、原因となった変更点は以下の通り。
--- mbstring.c 2005/12/15 03:37:22 1.142.2.47.2.6
+++ mbstring.c 2005/12/30 05:25:54 1.142.2.47.2.7
@@ -17,7 +17,7 @@
+----------------------------------------------------------------------+
*/
-/* $Id: mbstring.c,v 1.142.2.47.2.6 2005/12/15 03:37:22 iliaa Exp $ */
+/* $Id: mbstring.c,v 1.142.2.47.2.7 2005/12/30 05:25:54 hirokawa Exp $ */
/*
* PHP4 Multibyte String module "mbstring"
@@ -1555,7 +1555,7 @@
from_encoding = mbfl_no_encoding_invalid;
identd = mbfl_encoding_detector_new(
(enum mbfl_no_encoding *)info->from_encodings,
- info->num_from_encodings);
+ info->num_from_encodings, 0);
if (identd) {
n = 0;
while (n < num) {
@@ -1902,7 +1902,7 @@
} else {
/* auto detect */
from_encoding = mbfl_no_encoding_invalid;
- identd = mbfl_encoding_detector_new(elist, elistsz);
+ identd = mbfl_encoding_detector_new(elist, elistsz, 0);
if (identd != NULL) {
n = 0;
while (n < num) {
@@ -2404,6 +2404,11 @@
}
}
+ if (((MBSTRG(func_overload) & MB_OVERLOAD_STRING) == MB_OVERLOAD_STRING)
+ && (from >= mbfl_strlen(&string))) {
+ RETURN_FALSE;
+ }
+
ret = mbfl_substr(&string, &result, from, len);
if (ret != NULL) {
RETVAL_STRINGL((char *)ret->val, ret->len, 0); /* the string is already strdup()'ed */
@@ -2663,7 +2668,7 @@
string.no_encoding = from_encoding;
} else if (size > 1) {
/* auto detect */
- from_encoding = mbfl_identify_encoding_no(&string, list, size);
+ from_encoding = mbfl_identify_encoding_no(&string, list, size, 0);
if (from_encoding != mbfl_no_encoding_invalid) {
string.no_encoding = from_encoding;
} else {
@@ -3209,7 +3214,7 @@
stack = (pval ***)safe_emalloc(stack_max, sizeof(pval **), 0);
if (stack != NULL) {
stack_level = 0;
- identd = mbfl_encoding_detector_new(elist, elistsz);
+ identd = mbfl_encoding_detector_new(elist, elistsz, 0);
if (identd != NULL) {
n = 2;
while (n < argc || stack_level > 0) {
@@ -3573,7 +3578,7 @@
orig_str.no_encoding = MBSTRG(current_internal_encoding);
if (orig_str.no_encoding == mbfl_no_encoding_invalid
|| orig_str.no_encoding == mbfl_no_encoding_pass) {
- orig_str.no_encoding = mbfl_identify_encoding_no(&orig_str, MBSTRG(current_detect_order_list), MBSTRG(current_detect_order_list_size));
+ orig_str.no_encoding = mbfl_identify_encoding_no(&orig_str, MBSTRG(current_detect_order_list), MBSTRG(current_detect_order_list_size), 0);
}
pstr = mbfl_mime_header_encode(&orig_str, &conv_str, tran_cs, head_enc, "\n", sizeof("Subject: [PHP-jp nnnnnnnn]"));
if (pstr != NULL) {
@@ -3596,7 +3601,7 @@
if (orig_str.no_encoding == mbfl_no_encoding_invalid
|| orig_str.no_encoding == mbfl_no_encoding_pass) {
- orig_str.no_encoding = mbfl_identify_encoding_no(&orig_str, MBSTRG(current_detect_order_list), MBSTRG(current_detect_order_list_size));
+ orig_str.no_encoding = mbfl_identify_encoding_no(&orig_str, MBSTRG(current_detect_order_list), MBSTRG(current_detect_order_list_size), 0);
}
pstr = NULL;
@@ -3959,7 +3964,7 @@
mbfl_string_init(&string);
string.no_language = MBSTRG(current_language);
- identd = mbfl_encoding_detector_new(elist, size);
+ identd = mbfl_encoding_detector_new(elist, size, 0);
if (identd) {
int n = 0;抜粋すると以下の部分です。思いっきり「RETURN_FALSE」してた。（ｗ
+ if (((MBSTRG(func_overload) & MB_OVERLOAD_STRING) == MB_OVERLOAD_STRING)
+ && (from >= mbfl_strlen(&string))) {
+ RETURN_FALSE;
+ }
+（テスト結果から察すると）普段substr関数を使っていた人から「挙動が違う！」って文句がきたから関数をオーバーロードした時*だけ*は仕様をsubstr関数に併せたのでは。。。という風に推測できますね。その割には（長い事）放置されてたなぁ。
テスト・プログラムは修正しないとな。。。一応、確認してみよっと。
っていうか、長さで判断しろよな。。。とか思ってしまいますが。
「""」や「false」とか使い分けるの激しく面倒な人なので。
