Newsgroups: php.internals Path: news.php.net Xref: news.php.net php.internals:12265 Return-Path: Mailing-List: contact internals-help@lists.php.net; run by ezmlm Delivered-To: mailing list internals@lists.php.net Received: (qmail 22648 invoked by uid 1010); 19 Aug 2004 13:21:52 -0000 Delivered-To: ezmlm-scan-internals@lists.php.net Delivered-To: ezmlm-internals@lists.php.net Received: (qmail 17829 invoked from network); 19 Aug 2004 13:20:03 -0000 Received: from unknown (HELO amber.vis-av.com) (66.92.75.243) by pb1.pair.com with SMTP; 19 Aug 2004 13:20:03 -0000 Received: (qmail 29549 invoked from network); 19 Aug 2004 13:20:03 -0000 Received: from unknown (HELO random.?none?) (192.168.1.9) by amber.internal with SMTP; 19 Aug 2004 13:20:03 -0000 Received: (nullmailer pid 16214 invoked by uid 0); Thu, 19 Aug 2004 13:20:02 -0000 To: Derick Rethans Cc: internals References: Reply-To: Derrell.Lipman@UnwiredUniverse.com Date: Thu, 19 Aug 2004 09:20:02 -0400 In-Reply-To: (Derick Rethans's message of "Thu, 19 Aug 2004 09:21:02 +0200 (CEST)") Message-ID: Lines: 26 User-Agent: Gnus/5.090006 (Oort Gnus v0.06) XEmacs/21.4 (Common Lisp, i386-debian-linux) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Subject: Re: [PHP-DEV] flex.skl problem? From: Derrell.Lipman@UnwiredUniverse.com Derick Rethans writes: > On Wed, 18 Aug 2004 Derrell.Lipman@UnwiredUniverse.com wrote: > >> I'm trying to compile PHP from cvs, but it barfs while trying to generate >> zend_language_scanner.c from zend_language_scanner.l. The problem is in the >> specified skeleton file: flex.skl, and it seems to be dying near the beginning >> on the first '%-' line. I can't find any reference in the flex grammar >> documentation that even discusses such a line, so I don't know what it's >> supposed to do. >> >> I'm using flex version 2.5.31 > > Don't do that, we only support 2.5.4 (the flex people modified some > internal stuff which is incompatible with our skeleton file). Yuck! That seems kind of nasty to have to downgrade an application. (I'm tracking Debian unstable.) Any chance that the flex-generated C files could be distributed in cvs and the Makefile modified to not try to generate them from the flex input? Better yet, what's involved in modifying the skeleton file to work with an up-to-date flex? I might give it a go if it's not something requiring a total rewrite... Derrell