From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 9207 invoked from network); 13 Mar 2004 22:41:23 -0000 Received: from sunsite.dk (130.225.247.90) by ns1.primenet.com.au with SMTP; 13 Mar 2004 22:41:23 -0000 Received: (qmail 23542 invoked by alias); 13 Mar 2004 22:41:16 -0000 Mailing-List: contact zsh-workers-help@sunsite.dk; run by ezmlm Precedence: bulk X-No-Archive: yes X-Seq: 19619 Received: (qmail 23468 invoked from network); 13 Mar 2004 22:41:16 -0000 Received: from localhost (HELO sunsite.dk) (127.0.0.1) by localhost with SMTP; 13 Mar 2004 22:41:16 -0000 X-MessageWall-Score: 0 (sunsite.dk) Received: from [193.109.254.211] by sunsite.dk (MessageWall 1.0.8) with SMTP; 13 Mar 2004 22:41:15 -0000 X-VirusChecked: Checked X-Env-Sender: okiddle@yahoo.co.uk X-Msg-Ref: server-19.tower-36.messagelabs.com!1079217674!4623217 X-StarScan-Version: 5.2.5; banners=-,-,- X-Originating-IP: [158.234.9.163] Received: (qmail 14989 invoked from network); 13 Mar 2004 22:41:14 -0000 Received: from iris.logica.co.uk (158.234.9.163) by server-19.tower-36.messagelabs.com with SMTP; 13 Mar 2004 22:41:14 -0000 Received: from trentino.logica.co.uk ([158.234.142.61]) by iris.logica.co.uk (8.12.3/8.12.3/Debian -4) with ESMTP id i2DMfECk021196 for ; Sat, 13 Mar 2004 22:41:14 GMT Received: from trentino.logica.co.uk (localhost [127.0.0.1]) by trentino.logica.co.uk (Postfix) with ESMTP id 8290579721C1 for ; Sat, 13 Mar 2004 23:40:29 +0100 (CET) X-VirusChecked: Checked X-StarScan-Version: 5.1.13; banners=.,-,- From: Oliver Kiddle To: Zsh workers Subject: PATCH: compiling on MacOS X 10.1 Date: Sat, 13 Mar 2004 23:40:29 +0100 Message-ID: <14248.1079217629@trentino.logica.co.uk> Sourceforge actually has both a MacOS X 10.1 and 10.2 system. Before I had tried the 10.2 system but I just tried 10.1 and it had a problem. Where configure used the preprocessor to get a list of signals, it gets lines of this form: #1 "/usr/include/sys/signal.h" 1 We expect a space or tab after the `#' in the sed expression. This patch allows a number. I've checked the output on a few other systems and the only other lines beginning with `#' I ever saw are `#pragma' so this should be safe. The patch also includes a section to avoid a large number of ugly warnings on MacOS like the following: errtmp.c:3: undefined type, found `XXNAMES' errtmp.c:3: illegal external declaration, missing `;' after `XXEPERM' They don't matter because at the end it does this: cpp-precomp: warning: errors during smart preprocessing, retrying in basic mode But they are easily avoided by using -traditional-cpp. zpty doesn't entirely work on Mac OS X: test Y03 hangs. Theres also a test failure in C02 that I haven't yet understood. Oliver Index: configure.ac =================================================================== RCS file: /cvsroot/zsh/zsh/configure.ac,v retrieving revision 1.14 diff -u -r1.14 configure.ac --- configure.ac 12 Mar 2004 16:29:30 -0000 1.14 +++ configure.ac 13 Mar 2004 21:48:27 -0000 @@ -412,6 +412,11 @@ AC_PROG_GCC_TRADITIONAL dnl Do we need -traditional flag for gcc. AC_C_CONST dnl Does compiler support `const'. +dnl Default preprocessing on Mac OS X produces warnings +case "$host_os" in + darwin*) CPP="$CPP -traditional-cpp" ;; +esac + fp_PROG_CC_STDC AC_MSG_CHECKING([whether to use prototypes]) if test ."$ansi2knr" = .yes || test ."$ansi2knr" = .no; then @@ -1232,7 +1237,7 @@ echo "#include " > nametmp.c errfile_list="`$CPP nametmp.c | sed -n -e 's/^#line[ ].*\"\(.*\)\"/\1/p' \ - -e 's/^#[ ].*\"\(.*\)\"/\1/p' | + -e 's/^#[ 0-9].*\"\(.*\)\"/\1/p' | sed 's/\\\\\\\\/\//g' | $AWK '{ if (\$1 ~ \"err\") files[[\$1]] = \$1 } END { for (var in files) print var }'`"