From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 4142 invoked by alias); 19 Dec 2002 08:37:58 -0000 Mailing-List: contact insight-help@sources.redhat.com; run by ezmlm Precedence: bulk List-Subscribe: List-Archive: List-Post: List-Help: , Sender: insight-owner@sources.redhat.com Received: (qmail 4133 invoked from network); 19 Dec 2002 08:37:57 -0000 Received: from unknown (HELO mta6.snfc21.pbi.net) (206.13.28.240) by 209.249.29.67 with SMTP; 19 Dec 2002 08:37:57 -0000 Received: from modrick ([66.120.11.181]) by mta6.snfc21.pbi.net (iPlanet Messaging Server 5.1 HotFix 1.6 (built Oct 18 2002)) with SMTP id <0H7C00AW9YMYSK@mta6.snfc21.pbi.net> for insight@sources.redhat.com; Thu, 19 Dec 2002 00:37:46 -0800 (PST) Date: Thu, 19 Dec 2002 00:37:00 -0000 From: Mo DeJong Subject: Re: [RFA] Find init.tcl on cygwin In-reply-to: <20021219070416.GA21543@redhat.com> To: insight@sources.redhat.com Message-id: <20021219003946.24d69eec.mdejong@uncounted.org> MIME-version: 1.0 Content-type: text/plain; charset=US-ASCII Content-transfer-encoding: 7bit References: <20021219070416.GA21543@redhat.com> X-SW-Source: 2002-q4/txt/msg00225.txt.bz2 On Thu, 19 Dec 2002 02:04:16 -0500 Christopher Faylor wrote: > This has been sitting in my sandbox for a long time. > > Ok to check in? > > cgf ... > /* CYGNUS LOCAL */ > - sprintf(installLib, "share/tcl%s", TCL_VERSION); > + sprintf(installLib, "usr/share/tcl%s", TCL_VERSION); > /* END CYGNUS LOCAL */ What are the configure flags that you are building with under Cygwin? Is it something like `.../configure --prefix=/ --datadir=/usr` ? The reason I am concerned is that a configure like the following: .../configure --prefix=/usr/mydir ; make ; make install will install the init.tcl file into /usr/mydir/share/tcl8.3/init.tcl. With the change you suggest, it seems like the `make install` rule will still copy the file to the same location but now the TclpInitLibraryPath function will be looking in /usr/mydir/usr/share/tcl8.3/init.tcl. Also, this change will make the windows version of `make install` behave differently from the unix version, and I don't quite follow why you would want to do that. What is the use case this patch is intended to fix? Mo