This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
[Bug testsuite/41166] [4.5 Regression] Syntax error: Unterminated quoted string
- From: "rwild at gcc dot gnu dot org" <gcc-bugzilla at gcc dot gnu dot org>
- To: gcc-bugs at gcc dot gnu dot org
- Date: 25 Aug 2009 16:32:23 -0000
- Subject: [Bug testsuite/41166] [4.5 Regression] Syntax error: Unterminated quoted string
- References: <bug-41166-682@http.gcc.gnu.org/bugzilla/>
- Reply-to: gcc-bugzilla at gcc dot gnu dot org
------- Comment #3 from rwild at gcc dot gnu dot org 2009-08-25 16:32 -------
(In reply to comment #2)
> > value_of_substed=`echo @substed@ | ./config.status --file=-`
>
> It's a bad idea for the script to depend at all on either config.status or
> a build tree
I don't understand this remark. If the script shouldn't depend on
config.status, well, then why does it read it? There should be no
additional limitation between reading the script and running it,
at least how I understand it.
> - right now it's necessary to fake a config.status file when
> doing installed-compiler testing.
Well, it would certainly be easier to fake
echo @TOPLEVEL_CONFIGURE_ARGUMENTS@ | ./config.status --file=-
with a two-line script of the form:
#! /bin/sh
echo args...
than it is to fake internal details and remove the complex quoting
done inside config.status.
> The gcc testsuite should arrange to put the "Configured with:" line from
> gcc -v output in the .sum file (it's already in some .log files, and in
> acats.sum), and the script should use that line to extract configure
> options.
Fine with me, but not my area of expertise.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=41166