Re: cvs: php-src /ext/pdo_mysql mysql_statement.c
| From: | Wez Furlong | Date: | Thu, 14 Jul 2005 12:29:39 +0000 |
| Subject: | Re: cvs: php-src /ext/pdo_mysql mysql_statement.c | ||
| References: | 1 2 | Groups: | php.cvs |
| Request: | Send a blank email to php-cvs+get-32922@lists.php.net to get a copy of this message | ||
Then explain why half the test cases fail silently with the code as it was, and why all tests pass and run perfectly now?
--Wez.
Andrey Hristov wrote:
Hi Wez, I do never commit code which makes test case fail. Here not tests failed therefore I did commit the code. In addition, (my_ulonglong) -1 is not 0. The first value is returned from mysql_stmt_affected_rows() and mysql_affected_rows() to indicate a failure. Both functions return unsigned long and therefore to indicate an error they return a number of rows so far thought that will not appear in a result set (or number of affected rows). Therefore, I do think, if my mind is not borked :), that your fix is not correct. Andrey Quoting Wez Furlong <wez@php.net>:wez Wed Jul 13 22:03:55 2005 EDTModified files:/php-src/ext/pdo_mysql mysql_statement.cLog: it is not an error for a statement to return 0 rows. Andrey: please run the test suite before each commit! This broke beta 3 and was clearly visible as a result of running the tests. Running further tests now. http://cvs.php.net/diff.php/php-src/ext/pdo_mysql/mysql_statement.c?r1=1.41&r2=1.42&ty=u Index: php-src/ext/pdo_mysql/mysql_statement.c diff -u php-src/ext/pdo_mysql/mysql_statement.c:1.41 php-src/ext/pdo_mysql/mysql_statement.c:1.42--- php-src/ext/pdo_mysql/mysql_statement.c:1.41 Tue Jul 12 15:22:05 2005 +++ php-src/ext/pdo_mysql/mysql_statement.c Wed Jul 13 22:03:54 2005@@ -17,7 +17,7 @@ +----------------------------------------------------------------------+ */ -/* $Id: mysql_statement.c,v 1.41 2005/07/12 19:22:05 andrey Exp $ */ +/* $Id: mysql_statement.c,v 1.42 2005/07/14 02:03:54 wez Exp $ */ #ifdef HAVE_CONFIG_H #include "config.h" @@ -96,11 +96,9 @@my_bool on = 1; /* if we have bound the buffers don't set the attribute again */ if (!S->result) { - /* - FIXME: using directly internal structs - but for now cleaner - then calling 2 times result_metadata. - */ - for (i= 0; i < S->stmt->field_count; i++) { + /* FIXME: using directly internal structs - but for now cleaner + * then calling 2 times result_metadata. */ + for (i = 0; i < S->stmt->field_count; i++) { switch (S->fields[i].type) { case MYSQL_TYPE_MEDIUM_BLOB: case MYSQL_TYPE_LONG_BLOB:@@ -124,12 +122,12 @@if (S->bound_result) { int i; - for (i = 0; i < stmt->column_count; i++) { - efree(S->bound_result[i].buffer); + for (i = 0; i < stmt->column_count; i++) { + efree(S->bound_result[i].buffer); } - efree(S->bound_result); - efree(S->out_null); - efree(S->out_length); + efree(S->bound_result); + efree(S->out_null); + efree(S->out_length); }stmt->column_count = (int)mysql_num_fields(S->result);@@ -140,8 +138,8 @@/* summon memory to hold the row */ for (i = 0; i < stmt->column_count; i++) { S->bound_result[i].buffer_length = - S->fields[i].max_length? S->fields[i].max_length: - S->fields[i].length; + S->fields[i].max_length? S->fields[i].max_length: + S->fields[i].length; S->bound_result[i].buffer = emalloc(S->bound_result[i].buffer_length); S->bound_result[i].is_null = &S->out_null[i]; S->bound_result[i].length = &S->out_length[i];@@ -156,11 +154,11 @@}; - if ((row_count= mysql_stmt_affected_rows(S->stmt)) != (my_ulonglong)-1) { + row_count = mysql_stmt_affected_rows(S->stmt); + if (row_cont != (my_ulonglong)-1) { stmt->row_count = row_count; - return 1; } - return 0; + return 1; }#endif/* ensure that we free any previous unfetched results */-- PHP CVS Mailing List (http://www.php.net/) To unsubscribe, visit: http://www.php.net/unsub.php