Mailing-List: contact cygwin-apps-help AT sourceware DOT cygnus DOT com; run by ezmlm Sender: cygwin-apps-owner AT sourceware DOT cygnus DOT com List-Subscribe: List-Archive: List-Post: List-Help: , Delivered-To: mailing list cygwin-apps AT sources DOT redhat DOT com Message-ID: <3C05E139.2070701@ece.gatech.edu> Date: Thu, 29 Nov 2001 02:18:17 -0500 From: Charles Wilson User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:0.9.4) Gecko/20011019 Netscape6/6.2 X-Accept-Language: en-us MIME-Version: 1.0 To: cygwin-apps AT cygwin DOT com Subject: Re: Recent setup.ini corruption References: <20011128193300 DOT GA6598 AT redhat DOT com> <3C056C9D DOT 4C486264 AT ece DOT gatech DOT edu> <20011129011113 DOT GC9605 AT redhat DOT com> Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Weirdness. The setup.ini currently on mirrors.rcn.net doesn't have extraneous " marks. However, when I run setup.exe, it crashes. But, it DOES save an uncorrupted copy of the brand new setup.ini to my local disk. So, I ran setup.exe and selected "install from local directory" -- and it was able to parse the setup.ini FINE. No crash. Huh? This is repeatable. What does the lex parser have to do with the net-io/localdisk-io code? --Chuck Christopher Faylor wrote: > On Wed, Nov 28, 2001 at 06:00:45PM -0500, Charles Wilson wrote: > >>Christopher Faylor wrote: >> >>>The recent reports of setup.ini corruption were due, I believe, to the >>>use of embedded "'s in some recent ldesc additions. >>> >>This was my fault. Sorry.... >> > > Hey, I didn't say nothin. I wasn't blamin no one. > > >>I've also sent Corinna an updated set of setup.hints that use >>texinfo-proper `foo' internal quotation marks, instead of "foo". >> > > I was thinking about modifying upset to use the `' style quoting, but > I figured it wouldn't actually get used all that often, so it wasn't > worth my effort. > > >>>I've modified upset to convert these types of quotes to apostrophes and >>>this seems to have fixed the problem. >>> >>Thanks. I'm sure this type of screwup will happen again (maybe someone >>OTHER than me, next time) so it's good if upset can handle this >>gracefully. >> > > Yes. I guess we really need an upset -c or an upset-lint, or something. > > cgf >