RE: [PHP-DEV] Give the Language a Rest motion

From: Date: Fri, 10 Mar 2006 07:10:15 +0000
Subject: RE: [PHP-DEV] Give the Language a Rest motion
References: 1  Groups: php.internals 
Request: Send a blank email to internals+get-22322@lists.php.net to get a copy of this message
Just for info: GCC-4.1 now uses faster hand-written recursive-descent parser (instead of bison generated). Thanks. Dmitry. > -----Original Message----- > From: Zeev Suraski [mailto:zeev@zend.com] > Sent: Friday, March 10, 2006 2:26 AM > To: Marcus Boerger > Cc: Sara Golemon; internals@lists.php.net > Subject: Re: [PHP-DEV] Give the Language a Rest motion > > > No speed boost with opcode caches, which will be bundled in PHP 6 :) > > Zeev > > At 01:15 10/03/2006, Marcus Boerger wrote: > >Hello Sara, > > > > but if we were moving from flex to re2c for that > tokenizing scripts > >we'd get a nice speed boost, too. Typically re2c based > scanners are 2 > >to 3 times faster than lex based ones. And oh-re2c allows unicode > >scanning (2 byte > >input) and you can use the same .re to generate two .c files > if necessary. > > > >best regards > >matcus > > > >Thursday, March 9, 2006, 6:44:45 PM, you wrote: > > > > >>> The inability to inject tokens and expressions into the > lexer and > > >>> parser is another limitation on what can be done from > extensions > > >>> in terms of syntax level features. Yes, I know this is > more of a > > >>> problem with bison and flex than with the design of ZE, > but that > > >>> doesn't make it any less bothersome. > > >> > > >> Do other compiler tools allow this? > > >> > > > I've heard second hand that lemon does. But I'm quite > certain that > > > rewriting the lexer/parser is not at the top of the list of > > productive ways > > > to advance the language. > > > > > -Sara > > > > > > > > > >Best regards, > > Marcus > > > >-- > >PHP Internals - PHP Runtime Development Mailing List > >To unsubscribe, visit: http://www.php.net/unsub.php > > -- > PHP Internals - PHP Runtime Development Mailing List > To unsubscribe, visit: http://www.php.net/unsub.php > > >

« previous php.internals (#22322) next »