This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
Re: CVS Problem: java/parse.c and java/parse-scan.c deleted
- To: "John David Anglin" <dave at hiauly1 dot hia dot nrc dot ca>
- Subject: Re: CVS Problem: java/parse.c and java/parse-scan.c deleted
- From: Alexandre Oliva <aoliva at redhat dot com>
- Date: 12 Sep 2001 19:17:32 -0300
- Cc: jsm28 at cam dot ac dot uk, David dot Billinghurst at riotinto dot com, gcc-bugs at gcc dot gnu dot org
- Organization: GCC Team, Red Hat
- References: <200109100311.f8A3B9Sw009150@hiauly1.hia.nrc.ca>
On Sep 10, 2001, "John David Anglin" <dave@hiauly1.hia.nrc.ca> wrote:
>> On Sep 6, 2001, "John David Anglin" <dave@hiauly1.hia.nrc.ca> wrote:
>>
>> > Here is an updated patch for gcc_update which doesn't touch missing files.
>> > It just prints a warning. OK?
>>
>> Err... I'm not sure I understand why we need this. We just shouldn't
>> be touching files that are not in CVS. Approving the patch that
> No problem. However, missing files can occur in other ways as well.
But the cvs update before touch would have fetched them, wouldn't it?
Hmm, but if you're doing just --touch, it wouldn't.
> This patch changes the script so that it doesn't touch missing files.
> It gives a little bit of extra security and a warning when it happens.
What I don't like much is the warning. I mean, if I don't check out
say the Java sub-directory I wouldn't like to be warned about it every
time I run gcc_update. We might just ignore the file in this case.
But this is moot right now, since we don't have files in
sub-directories depending on files in parent directories. We'd bail
out in case any of the dependencies are missing, which covers the
same-directory case.
> The point is as you state above to not touch missing files, irrespective
> of whether there was a script error, a download problem or a user error.
> From my personal perspective, I wouldn't have wasted time with a build
> and researching the problem if I had seen the warning when I did the CVS
> update. The problem would have been immediately obvious. That's why
> I think the patch should be installed.
I agree. But I'm not sure I"m entitled to approve the change. Gerry,
as the original author, has the final word, IIRC. Gerry?
--
Alexandre Oliva Enjoy Guarana', see http://www.ic.unicamp.br/~oliva/
Red Hat GCC Developer aoliva@{cygnus.com, redhat.com}
CS PhD student at IC-Unicamp oliva@{lsd.ic.unicamp.br, gnu.org}
Free Software Evangelist *Please* write to mailing lists, not to me