Newsgroups: php.internals Path: news.php.net Xref: news.php.net php.internals:12249 Return-Path: Mailing-List: contact internals-help@lists.php.net; run by ezmlm Delivered-To: mailing list internals@lists.php.net Received: (qmail 61051 invoked by uid 1010); 19 Aug 2004 05:41:38 -0000 Delivered-To: ezmlm-scan-internals@lists.php.net Delivered-To: ezmlm-internals@lists.php.net Received: (qmail 60993 invoked by uid 1007); 19 Aug 2004 05:41:38 -0000 Message-ID: <20040819054138.60992.qmail@pb1.pair.com> To: internals@lists.php.net Date: Wed, 18 Aug 2004 22:41:37 -0700 Lines: 24 X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 6.00.2800.1437 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1441 X-Posted-By: 64.142.6.231 Subject: Re: From: pollita@php.net ("Sara Golemon") > > > - compile time hooking (no dynamic goto targets) > > > > > > > > That kind of makes them useless for writing parsers, which was one of > > the justifications of them. > > Hmmm you're right. I forgot that in C when doing parsers it's > commonplace to map the label addresses to some lookup indexes and then > jump to the address with goto. Although it's also quite common to do > parsers without dynamic jumping, but yeah that's a dandy handy feature > :) At this point I'd like to retract my agreement with static goto > targets. Now I wonder if it is possible to have the compiler recognize a > static jump from a dynamic jump and do the hooking accordingly to > maximize efficiency in the static case? > Implementation Two covered all the points listed including staic hooking. The dynamic hooks could still be done, but it'd take some reworking. I kinda done with it for now though, there have been too many karmicly strong negative votes for me to believe it's going anywhere near acceptance. The patch will stay where it is if anyone wants it as is though. -Sara