Re: cvs: php-src /main output.c /tests/output bug46900.phpt bug46903.phpt ob_014.phpt ob_015.phpt ob_start_basic_002.phpt ob_start_basic_004.phpt

2009/2/15 Michael Wallner <mike@php.net>: > Robin Fernandes wrote: >> >> robinf Sun Dec 28 19:50:58 2008 UTC >> >> Added files: /php-src/tests/output bug46903.phpt >> bug46900.phpt >> Modified files: /php-src/tests/output >> ob_start_basic_004.phpt >> ob_start_basic_002.phpt ob_015.phpt >> ob_014.phpt /php-src/main output.c Log: >> Fix bugs #46900 and #46903. >> > >> >> >> http://cvs.php.net/viewvc.cgi/php-src/main/output.c?r1=1.214&r2=1.215&diff_format=u >> Index: php-src/main/output.c >> diff -u php-src/main/output.c:1.214 php-src/main/output.c:1.215 >> --- php-src/main/output.c:1.214 Mon Aug 18 07:45:59 2008 >> +++ php-src/main/output.c Sun Dec 28 19:50:58 2008 >> @@ -19,7 +19,7 @@ >> >> +----------------------------------------------------------------------+ >> */ >> -/* $Id: output.c,v 1.214 2008/08/18 07:45:59 tony2001 Exp $ */ >> +/* $Id: output.c,v 1.215 2008/12/28 19:50:58 robinf Exp $ */ >> #ifndef PHP_OUTPUT_DEBUG >> # define PHP_OUTPUT_DEBUG 0 >> @@ -1342,6 +1342,8 @@ >> } >> if (chunk_size < 0) { >> chunk_size = 0; >> + } else if (chunk_size == 1) { >> + chunk_size = 4096; >> } >> >> if (SUCCESS != php_output_start_user(output_handler, chunk_size, >> flags TSRMLS_CC)) { > > The documentation stated IMHO silly magic behavior of the old output layer. > Why start sprinkling the new code with useless magic meanings? Hi Mike, Fair point regarding bug 46903. In this case, I can see how documenting a change in behaviour could be better than implementing the currently documented behaviour - either approach is fine by me. I'm happy to revert this and raise a doc bug instead, if that's the right way to go. Cheers, Robin