Re: cvs: php-src /ext/spl spl_directory.c spl_directory.h /ext/standard file.c file.h /ext/standard/tests/file bug40501.csv bug40501.phpt
fgetcsv_error.phpt

Hello David, can you MFH this to 5.3. marcus Wednesday, October 3, 2007, 12:31:04 PM, you wrote: > > http://cvs.php.net/viewvc.cgi/php-src/ext/spl/spl_directory.c?r1=1.136&r2=1.137&diff_format=u�����0� > �HL$i�� > Index: php-src/ext/spl/spl_directory.c > diff -u php-src/ext/spl/spl_directory.c:1.136 > php-src/ext/spl/spl_directory.c:1.137 > --- php-src/ext/spl/spl_directory.c:1.136 Thu Sep 27 18:28:42 2007 > +++ php-src/ext/spl/spl_directory.c Wed Oct 3 10:31:03 2007 > @@ -16,7 +16,7 @@ > +----------------------------------------------------------------------+ > */ > > -/* $Id: spl_directory.c,v 1.136 2007/09/27 18:28:42 dmitry Exp $ */ > +/* $Id: spl_directory.c,v 1.137 2007/10/03 10:31:03 dsp Exp $ */ > > #ifdef HAVE_CONFIG_H > # include "config.h" > @@ -1771,7 +1771,7 @@ > spl_filesystem_file_call(intern, func_ptr, pass_num_args, return_value, arg2 > TSRMLS_CC); \ > } > > -static int spl_filesystem_file_read_csv(spl_filesystem_object *intern, char delimiter, char > enclosure, zval *return_value TSRMLS_DC) /* {{{ */ > +static int spl_filesystem_file_read_csv(spl_filesystem_object *intern, char delimiter, char > enclosure, char escape, zval *return_value TSRMLS_DC) /* {{{ */ > { > int ret = SUCCESS; > > @@ -1788,7 +1788,7 @@ > } > ALLOC_INIT_ZVAL(intern->u.file.current_zval); > > - php_fgetcsv(intern->u.file.stream, delimiter, enclosure, buf_len, buf, > intern->u.file.current_zval TSRMLS_CC); > + php_fgetcsv(intern->u.file.stream, delimiter, enclosure, escape, buf_len, > buf, > intern->u.file.current_zval TSRMLS_CC); > if (return_value) { > if (Z_TYPE_P(return_value) != IS_NULL) { > zval_dtor(return_value); > @@ -1814,7 +1814,7 @@ > return FAILURE; > } > if (intern->flags & SPL_FILE_OBJECT_READ_CSV) { > - return spl_filesystem_file_read_csv(intern, > intern->u.file.delimiter, > intern->u.file.enclosure, NULL TSRMLS_CC); > + return spl_filesystem_file_read_csv(intern, > intern->u.file.delimiter, > intern->u.file.enclosure, intern->u.file.escape, NULL TSRMLS_CC); > } else { > zend_call_method_with_0_params(&this_ptr, Z_OBJCE_P(getThis()), > &intern->u.file.func_getCurr, "getCurrentLine", &retval); > } > @@ -2149,13 +2149,20 @@ > SPL_METHOD(SplFileObject, fgetcsv) > { > spl_filesystem_object *intern = > (spl_filesystem_object*)zend_object_store_get_object(getThis() TSRMLS_CC); > - char delimiter = intern->u.file.delimiter, enclosure = intern->u.file.enclosure; > - char *delim, *enclo; > - int d_len, e_len; > + char delimiter = intern->u.file.delimiter, enclosure = intern->u.file.enclosure, > escape > = intern->u.file.escape; > + char *delim, *enclo, *esc; > + int d_len, e_len, esc_len; > > - if (zend_parse_parameters(ZEND_NUM_ARGS() TSRMLS_CC, "|ss", &delim, > &d_len, &enclo, > &e_len) == SUCCESS) { > + if (zend_parse_parameters(ZEND_NUM_ARGS() TSRMLS_CC, "|sss", &delim, > &d_len, &enclo, > &e_len, &esc, &esc_len) == SUCCESS) { > switch(ZEND_NUM_ARGS()) > { > + case 3: > + if (esc_len != 1) { > + php_error_docref(NULL TSRMLS_CC, E_WARNING, > "escape > must be a character"); > + RETURN_FALSE; > + } > + escape = esc[0]; > + > case 2: > if (e_len != 1) { > php_error_docref(NULL TSRMLS_CC, E_WARNING, > "enclosure > must be a character"); > @@ -2173,23 +2180,30 @@ > case 0: > break; > } > - spl_filesystem_file_read_csv(intern, delimiter, enclosure, return_value > TSRMLS_CC); > + spl_filesystem_file_read_csv(intern, delimiter, enclosure, escape, return_value > TSRMLS_CC); > } > } > /* }}} */ > > -/* {{{ proto void SplFileObject::setCsvControl([string delimiter = ',' [, string > enclosure = > '"']]) > +/* {{{ proto void SplFileObject::setCsvControl([string delimiter = ',' [, string > enclosure = > '"' [, string escape = '\\']]]) > Set the delimiter and enclosure character used in fgetcsv */ > SPL_METHOD(SplFileObject, setCsvControl) > { > spl_filesystem_object *intern = > (spl_filesystem_object*)zend_object_store_get_object(getThis() TSRMLS_CC); > - char delimiter = ',', enclosure = '"'; > - char *delim, *enclo; > - int d_len, e_len; > + char delimiter = ',', enclosure = '"', escape='\\'; > + char *delim, *enclo, *esc; > + int d_len, e_len, esc_len; > > - if (zend_parse_parameters(ZEND_NUM_ARGS() TSRMLS_CC, "|ss", &delim, > &d_len, &enclo, > &e_len) == SUCCESS) { > + if (zend_parse_parameters(ZEND_NUM_ARGS() TSRMLS_CC, "|sss", &delim, > &d_len, &enclo, > &e_len, &esc, &esc_len) == SUCCESS) { > switch(ZEND_NUM_ARGS()) > { > + case 3: > + if (esc_len != 1) { > + php_error_docref(NULL TSRMLS_CC, E_WARNING, > "escape > must be a character"); > + RETURN_FALSE; > + } > + escape = esc[0]; > + /* no break */ > case 2: > if (e_len != 1) { > php_error_docref(NULL TSRMLS_CC, E_WARNING, > "enclosure > must be a character"); > @@ -2209,6 +2223,7 @@ > } > intern->u.file.delimiter = delimiter; > intern->u.file.enclosure = enclosure; > + intern->u.file.escape = escape; > } > } > /* }}} */ > > http://cvs.php.net/viewvc.cgi/php-src/ext/spl/spl_directory.h?r1=1.34&r2=1.35&diff_format=u > Index: php-src/ext/spl/spl_directory.h > diff -u php-src/ext/spl/spl_directory.h:1.34 php-src/ext/spl/spl_directory.h:1.35 > --- php-src/ext/spl/spl_directory.h:1.34 Sun Mar 4 00:42:30 2007 > +++ php-src/ext/spl/spl_directory.h Wed Oct 3 10:31:03 2007 > @@ -16,7 +16,7 @@ > +----------------------------------------------------------------------+ > */ > > -/* $Id: spl_directory.h,v 1.34 2007/03/04 00:42:30 helly Exp $ */ > +/* $Id: spl_directory.h,v 1.35 2007/10/03 10:31:03 dsp Exp $ */ > > #ifndef SPL_DIRECTORY_H > #define SPL_DIRECTORY_H > @@ -93,6 +93,7 @@ > zend_function *func_getCurr; > char delimiter; > char enclosure; > + char escape; > } file; > } u; > spl_filesystem_iterator it; > > http://cvs.php.net/viewvc.cgi/php-src/ext/standard/file.c?r1=1.500&r2=1.501&diff_format=u > Index: php-src/ext/standard/file.c > diff -u php-src/ext/standard/file.c:1.500 php-src/ext/standard/file.c:1.501 > --- php-src/ext/standard/file.c:1.500 Wed Sep 5 12:55:36 2007 > +++ php-src/ext/standard/file.c Wed Oct 3 10:31:04 2007 > @@ -21,7 +21,7 @@ > +----------------------------------------------------------------------+ > */ > > -/* $Id: file.c,v 1.500 2007/09/05 12:55:36 iliaa Exp $ */ > +/* $Id: file.c,v 1.501 2007/10/03 10:31:04 dsp Exp $ */ > > /* Synced with php 3.0 revision 1.218 1999-06-16 [ssb] */ > > @@ -2260,12 +2260,12 @@ > /* }}} */ > > PHPAPI void php_fgetcsv(php_stream *stream, /* {{{ */ > - char delimiter, char enclosure, > - size_t buf_len, char *buf, > - zval *return_value TSRMLS_DC) > + char delimiter, char enclosure, char escape, > + size_t buf_len, char *buf, > + zval *return_value TSRMLS_DC) > { > - char *delim = &delimiter, *enc = &enclosure, *buffer = buf; > - int delim_len = 1, enc_len = 1, buffer_len = buf_len; > + char *delim = &delimiter, *enc = &enclosure, *buffer = buf, *esc; > + int delim_len = 1, enc_len = 1, esc_len = 1, buffer_len = buf_len; > zend_uchar type = IS_STRING; > > if (stream) { > @@ -2285,22 +2285,30 @@ > INIT_PZVAL(return_value); > return; > } > + if (FAILURE == > zend_string_to_unicode(ZEND_U_CONVERTER(UG(runtime_encoding_conv)), (UChar**)&esc, > &esc_len, > &escape, 1 TSRMLS_CC)) { > + efree(delim); > + efree(enc); > + INIT_PZVAL(return_value); > + return; > + } > if (FAILURE == > zend_string_to_unicode(ZEND_U_CONVERTER(UG(runtime_encoding_conv)), (UChar**)&buffer, > &buffer_len, buf, buf_len TSRMLS_CC)) { > efree(delim); > efree(enc); > + efree(esc); > INIT_PZVAL(return_value); > return; > } > > - php_u_fgetcsv(stream, (UChar*)delim, delim_len, (UChar*)enc, enc_len, &esc, > 1, > + php_u_fgetcsv(stream, (UChar*)delim, delim_len, (UChar*)enc, enc_len, > (UChar*)esc, esc_len, > (UChar*)buffer, buffer_len, return_value TSRMLS_CC); > > /* Types converted, free storage */ > efree(delim); > efree(enc); > + efree(esc); > } else { > /* Binary stream with binary delimiter/enclosures/prefetch */ > - php_fgetcsv_ex(stream, delim, delim_len, enc, enc_len, "\\", 1, > buffer, > buffer_len, return_value TSRMLS_CC); > + php_fgetcsv_ex(stream, delim, delim_len, enc, enc_len, esc, esc_len, buffer, > buffer_len, return_value TSRMLS_CC); > } > } > > @@ -2365,8 +2373,12 @@ > } > > /* Is it an escape character? */ > - if (PHP_FGETCSV_BIN_CHECK(p, e, escape, escape_len)) { > - /* Skip escape sequence and let next char be treated as > literal */ > + if ((PHP_FGETCSV_BIN_CHECK(p, e, escape, escape_len) && > escape > != enclosure) > + || (PHP_FGETCSV_BIN_CHECK(p, e, escape, escape_len) > + && > PHP_FGETCSV_BIN_CHECK(p+1, e, escape, > escape_len) && escape == enclosure)) { > + /* Skip escape sequence and let next char be treated as > literal > + If enclosure is the same character as esacpe, it is > considered as esacped > + if it appears twice */ > p += escape_len; > /* FALL THROUGH */ > } > @@ -2569,8 +2581,12 @@ > } > > /* Is it an escape character? */ > - if (PHP_FGETCSV_UNI_CHECK(p, e, escape, escape_len)) { > - /* Skip escape sequence and let next char be treated as > literal */ > + if ((PHP_FGETCSV_UNI_CHECK(p, e, escape, escape_len) && > escape > != enclosure) > + || (PHP_FGETCSV_UNI_CHECK(p, e, escape, escape_len) > + && > PHP_FGETCSV_UNI_CHECK(p+1, e, escape, > escape_len) && escape == enclosure)) { > + /* Skip escape sequence and let next char be treated as > literal > + If enclosure is the same character as esacpe, it is > considered as esacped > + if it appears twice */ > p += escape_len; > /* FALL THROUGH */ > } > > http://cvs.php.net/viewvc.cgi/php-src/ext/standard/file.h?r1=1.104&r2=1.105&diff_format=u��M]|�m�K� > ���� > Index: php-src/ext/standard/file.h > diff -u php-src/ext/standard/file.h:1.104 php-src/ext/standard/file.h:1.105 > --- php-src/ext/standard/file.h:1.104 Wed Jan 10 15:06:58 2007 > +++ php-src/ext/standard/file.h Wed Oct 3 10:31:04 2007 > @@ -16,7 +16,7 @@ > +----------------------------------------------------------------------+ > */ > > -/* $Id: file.h,v 1.104 2007/01/10 15:06:58 bjori Exp $ */ > +/* $Id: file.h,v 1.105 2007/10/03 10:31:04 dsp Exp $ */ > > /* Synced with php 3.0 revision 1.30 1999-06-16 [ssb] */ > > @@ -78,7 +78,7 @@ > PHPAPI int php_copy_file_ex(char *src, char *dest, int src_chk TSRMLS_DC); > PHPAPI int php_mkdir_ex(char *dir, long mode, int options TSRMLS_DC); > PHPAPI int php_mkdir(char *dir, long mode TSRMLS_DC); > -PHPAPI void php_fgetcsv(php_stream *stream, char delimiter, char enclosure, size_t buf_len, > char *buf, zval *return_value TSRMLS_DC); > +PHPAPI void php_fgetcsv(php_stream *stream, char delimiter, char enclosure, char escape, > size_t buf_len, char *buf, zval *return_value TSRMLS_DC); > PHPAPI void php_fgetcsv_ex(php_stream *stream, char *delimiter, int delimiter_len, char > *enclosure, int enclosure_len, char *escape, int escape_len, > char *buffer, int buffer_len, zval *return_value TSRMLS_DC); > PHPAPI void php_u_fgetcsv(php_stream *stream, UChar *delimiter, int delimiter_len, UChar > *enclosure, int enclosure_len, UChar *escape, int escape_len, > > http://cvs.php.net/viewvc.cgi/php-src/ext/standard/tests/file/fgetcsv_error.phpt?r1=1.2&r2=1.3&diff_format=u > Index: php-src/ext/standard/tests/file/fgetcsv_error.phpt > diff -u php-src/ext/standard/tests/file/fgetcsv_error.phpt:1.2 > php-src/ext/standard/tests/file/fgetcsv_error.phpt:1.3 > --- php-src/ext/standard/tests/file/fgetcsv_error.phpt:1.2 Wed Aug 8 13:15:01 2007 > +++ php-src/ext/standard/tests/file/fgetcsv_error.phpt Wed Oct 3 10:31:04 2007 > @@ -36,7 +36,7 @@ > for($loop_counter = 1; $loop_counter <= count($invalid_args); $loop_counter++) { > echo "-- Iteration $loop_counter --\n"; > var_dump( fgetcsv($invalid_args[$loop_counter - 1]) ); // with default args > - var_dump( fgetcsv($invalid_args[$loop_counter - 1], $len, $delim, $enclosure) ); // all args > specified > + var_dump( fgetcsv($invalid_args[$loop_counter - 1], $len, $delim, $enclosure, $escape) ); // > all args specified > } > > echo "Done\n"; > > http://cvs.php.net/viewvc.cgi/php-src/ext/standard/tests/file/bug40501.csv?view=markup&rev=1.1 > Index: php-src/ext/standard/tests/file/bug40501.csv > +++ php-src/ext/standard/tests/file/bug40501.csv > "this element contains the delimiter, and ends with an odd number of > backslashes (ex: 1)\",and it isn't the last element$ > > http://cvs.php.net/viewvc.cgi/php-src/ext/standard/tests/file/bug40501.phpt?view=markup&rev=1.1 > Index: php-src/ext/standard/tests/file/bug40501.phpt > +++ php-src/ext/standard/tests/file/bug40501.phpt > --TEST-- > Bug #40501 (fgetcsv() can't handle trailing odd number of backslashes) > --FILE-- > <?php > $file = dirname(__FILE__).'/bug40501.csv'; > $h = fopen($file, 'r'); > $data = fgetcsv($h, NULL, ',', '"', '"'); > fclose($h); > var_dump($data); ?>> > --EXPECT-- > array(2) { > [0]=> > string(88) "this element contains the delimiter, and ends with an odd number of > backslashes (ex: 1)\" > [1]=> > string(30) "and it isn't the last element$" > } Best regards, Marcus