Re: -Werror considered harmful

2013-02-27 Thread Colin Walters
On Tue, 2013-02-26 at 15:29 -0500, Behdad Esfahbod wrote: I'm not questioning that. I think the right question would be: how many bugs got fixed in GLib because of the warnings being enabled by default Initally? A lot, some quite bad. We were leaking internal symbols for example. But

Re: -Werror considered harmful

2013-02-26 Thread Dan Winship
On 02/26/2013 02:39 AM, Behdad Esfahbod wrote: In file included from ../../../glib/libcharset/localcharset.c:28: /usr/lib/gcc/i586-mingw32msvc/4.2.1-sjlj/../../../../i586-mingw32msvc/include/stdio.h:372: error: no previous prototype for 'getc' This is a bug in whatever version of mingw you are

Re: -Werror considered harmful

2013-02-26 Thread Tristan Van Berkom
Hi Behdad. I'll be the first to agree that -Werror is evil stuff, we even fell into this catch 22 only around a year ago where one group thought it was a good idea to infect our builds with -Werror, while another group thought it was a good idea to add deprecation warnings, ... I'm not even sure

Re: -Werror considered harmful

2013-02-26 Thread Maciej Piechotka
Wouldn't it be better to selectively enable the -Wall -Werror rather then selectively disabled it? People compiling the library potentially care less about warning/errors then developers and know less about build system (was it --disable-fatal-warnings? --disable-Werror? etc.) Enabling warnings

Re: -Werror considered harmful

2013-02-26 Thread Colin Walters
On Tue, 2013-02-26 at 15:03 +0100, Dieter Verfaillie wrote: On 2013-02-26 14:20, Maciej Piechotka wrote: I would be for not including -Werror inside configure.ac at all and trust people with commit rights to enable it (either by explicit option or by CFLAGS variable). Agreed,

Re: -Werror considered harmful

2013-02-26 Thread Dieter Verfaillie
On 2013-02-26 17:55, Colin Walters wrote: Note we're not talking about -Werror. Ah, but it was mostly that which I was reacting to... mvg, Dieter ___ desktop-devel-list mailing list desktop-devel-list@gnome.org

Re: -Werror considered harmful

2013-02-26 Thread Behdad Esfahbod
[Replying to all four messages below.] On 13-02-26 03:19 AM, Dan Winship wrote: On 02/26/2013 02:39 AM, Behdad Esfahbod wrote: In file included from ../../../glib/libcharset/localcharset.c:28: /usr/lib/gcc/i586-mingw32msvc/4.2.1-sjlj/../../../../i586-mingw32msvc/include/stdio.h:372: error: no

Re: -Werror considered harmful

2013-02-26 Thread Martin Pitt
Colin Walters [2013-02-26 11:55 -0500]: I don't believe the baseline approach is craziness Neither do I. I find some selective -Werror= highly useful, and I'd even go as far as to say that _not_ having any of them is craziness. Errors triggered by things like

Re: -Werror considered harmful

2013-02-26 Thread Pascal Terjan
On Wed, Feb 27, 2013 at 5:57 AM, Martin Pitt martin.p...@ubuntu.com wrote: Colin Walters [2013-02-26 11:55 -0500]: I don't believe the baseline approach is craziness Neither do I. I find some selective -Werror= highly useful, and I'd even go as far as to say that _not_ having any of them is

-Werror considered harmful

2013-02-25 Thread Behdad Esfahbod
[I think I'm not on d-d-l, CC is appreciated.] rant Hi, I'm sending this email instead of filing a bug because I think I need to say this as widely as I can: modules adding -Werror (or -Werror=... variants thereof) are harmful Background: A few years ago Javier opened this bug against