Re: cvs: php-src /ext/unicode config.w32 unicode.c

The basic idea is that we always return a result, whether we raise an error or not. The type of error we raise also differs: in case of U_ILLEGAR_CHAR_FOUND, we definitely want to stop all processing, return what we processed so far, and raise a more noticeable error. For U_INVALID_CHAR_FOUND/U_IRREGULAR_CHAR_FOUND we can raise an error of lower severity. But yes, we do need a wrapper for conversion failure processing - something generic enough to be used by all converters. -Andrei On Aug 13, 2005, at 6:43 AM, Wez Furlong wrote:

wez        Sat Aug 13 09:43:32 2005 EDT

Modified files:

    /php-src/ext/unicode    config.w32 unicode.c

Log: Fixup config.w32. Implement unicode_decode() and unicode_encode(), as described in README.UNICODE. Still need to decide how to handle errors here, since there is no error return, and a conversion error is not necessarily fatal. http://cvs.php.net/diff.php/php-src/ext/unicode/config.w32?r1=1.2&r2=1.3&ty=u Index: php-src/ext/unicode/config.w32 diff -u php-src/ext/unicode/config.w32:1.2 php-src/ext/unicode/config.w32:1.3

--- php-src/ext/unicode/config.w32:1.2    Fri Aug 12 05:10:04 2005
+++ php-src/ext/unicode/config.w32    Sat Aug 13 09:43:31 2005

@@ -1,9 +1,8 @@ -// $Id: config.w32,v 1.2 2005/08/12 09:10:04 sniper Exp $ +// $Id: config.w32,v 1.3 2005/08/13 13:43:31 wez Exp $ // vim:ft=javascript -ARG_ENABLE('unicode' , 'ICU API extension', 'no'); -if (PHP_UNICODE) { - +ARG_ENABLE('unicode' , 'ICU API extension', 'yes'); +if (PHP_UNICODE != 'no') { EXTENSION("unicode", "unicode.c unicode_filter.c locale.c"); AC_DEFINE('HAVE_UNICODE', 1, 'ICU API extension'); } http://cvs.php.net/diff.php/php-src/ext/unicode/unicode.c?r1=1.2&r2=1.3&ty=u Index: php-src/ext/unicode/unicode.c diff -u php-src/ext/unicode/unicode.c:1.2 php-src/ext/unicode/unicode.c:1.3

--- php-src/ext/unicode/unicode.c:1.2    Fri Aug 12 05:10:04 2005
+++ php-src/ext/unicode/unicode.c    Sat Aug 13 09:43:31 2005

@@ -11,19 +11,101 @@ | license@php.net so we can mail you a copy

immediately.               |
   +----------------------------------------------------------------------+
   | Authors: Andrei Zmievski <andrei@php.net>                            |
+  |          Wez Furlong <wez@php.net>                                   |
   +----------------------------------------------------------------------+

*/ -/* $Id: unicode.c,v 1.2 2005/08/12 09:10:04 sniper Exp $ */ +/* $Id: unicode.c,v 1.3 2005/08/13 13:43:31 wez Exp $ */ #include "php_unicode.h" - #if HAVE_UNICODE +#include "zend_unicode.h" + +/* {{{ proto unicode unicode_decode(string $input, string $encoding) + Takes a string in the souce encoding and converts it to a UTF-16 unicode string, returning the result */ +static PHP_FUNCTION(unicode_decode) +{

+    union {
+        void *vptr;
+        char *bin;
+    } input;
+    zend_uchar type;
+    int len;
+    char *encoding;
+    int enclen;
+    UErrorCode status;
+    UConverter *conv = NULL;
+    UChar *target;
+    int32_t targetlen;

+

+    if (FAILURE == zend_parse_parameters(ZEND_NUM_ARGS() TSRMLS_CC, "ts", &input.vptr, &len, &type, &encoding, &enclen)) {
+        return;
+    }

+

+    if (type == IS_UNICODE) {
+        php_error_docref(NULL TSRMLS_CC, E_WARNING, "input string is already unicode");
+        RETURN_FALSE;
+    }

+

+    status = U_ZERO_ERROR;
+    conv = ucnv_open(encoding, &status);
+    if (!conv) {
+        php_error_docref(NULL TSRMLS_CC, E_WARNING, "could not locate converter for %s", encoding);
+        RETURN_FALSE;
+    }

+    status = U_ZERO_ERROR;
+    zend_convert_to_unicode(conv, &target, &targetlen, input.bin, len, &status);
+    if (U_FAILURE(status)) {
+        /* TODO: error handling semantics ? */
+        php_error_docref(NULL TSRMLS_CC, E_WARNING, "conversion was not entirely successful: %d", status);
+    }
+    RETVAL_UNICODEL(target, targetlen, 0);

+

+    ucnv_close(conv);

+} +/* }}} */ + +/* {{{ proto string unicode_encode(unicode $input, string $encoding) + Takes a unicode string and converts it to a string in the specified encoding */ +static PHP_FUNCTION(unicode_encode) +{

+    UChar *uni;
+    int len;
+    char *encoding;
+    int enclen;
+    UErrorCode status;
+    UConverter *conv = NULL;
+    char *target;
+    int32_t targetlen;

+

+    if (FAILURE == zend_parse_parameters(ZEND_NUM_ARGS() TSRMLS_CC, "us", &uni, &len, &encoding, &enclen)) {
+        return;
+    }

+

+    status = U_ZERO_ERROR;
+    conv = ucnv_open(encoding, &status);
+    if (!conv) {
+        php_error_docref(NULL TSRMLS_CC, E_WARNING, "could not locate converter for %s", encoding);
+        RETURN_FALSE;
+    }

+

+    status = U_ZERO_ERROR;
+    zend_convert_from_unicode(conv, &target, &targetlen, uni, len, &status);
+    if (U_FAILURE(status)) {
+        /* TODO: error handling semantics ? */
+        php_error_docref(NULL TSRMLS_CC, E_WARNING, "conversion was not entirely successful: %d", status);
+    }
+    RETVAL_STRINGL(target, targetlen, 0);

+

+    ucnv_close(conv);

+} /* {{{ unicode_functions[] */ function_entry unicode_functions[] = {

     PHP_FE(icu_loc_get_default, NULL)
     PHP_FE(icu_loc_set_default, NULL)
+    PHP_FE(unicode_decode, NULL)
+    PHP_FE(unicode_encode, NULL)
     { NULL, NULL, NULL }

}; /* }}} */ --PHP CVS Mailing List (http://www.php.net/) To unsubscribe, visit: http://www.php.net/unsub.php