bin/143351: [request] update flex(1) to at least 2.5.33
austinenglish at gmail.com
Sat Feb 6 04:05:24 UTC 2010
On Mon, Feb 1, 2010 at 11:57 PM, Garrett Cooper <yanefbsd at gmail.com> wrote:
> On Mon, Feb 1, 2010 at 2:33 AM, Austin English <austinenglish at gmail.com> wrote:
>> On Fri, Jan 29, 2010 at 6:48 PM, <linimon at freebsd.org> wrote:
>>> Old Synopsis: update flex to at least 2.5.33
>>> New Synopsis: [request] update flex(1) to at least 2.5.33
>>> State-Changed-From-To: open->suspended
>>> State-Changed-By: linimon
>>> State-Changed-When: Sat Jan 30 00:45:41 UTC 2010
>>> Mark suspended awaiting someone to take up the (hard) task of deciding
>>> if this will create regressions. See:
>> Indeed, it does. I did a build world to make sure it worked (using the
>> freebsd9 snapshot), then installed flex 2.5.35 and rebuilt.
>> /usr/src/cddl failed to build, and possible others (ran make -k, but
>> haven't checked the log yet). Should I file bugs for each program that
>> doesn't build in newer flex? or sit back and wait for someone to fix
>> flex for BSD and then include it?
> File bugs for any and all issues you find, point them back to this
> PR, and as long as everything points in the right direction and is
> done in proper order, everything should work just fine (TM).
> The version bump definitely deserves a heads up though in the
> current@ and stable@ community though so people can fix prepare to fix
> their potentially buggy / backwards incompatible yacc code, or bite
> the bullet and multislot version their tools [via ports].
Actually, I was mistaken. I used flex from
http://flex.sourceforge.net/, rather than the flex in ports. I just
did a make clean ; make world with flex from ports, and everything
builds fine (this is with FreeBSD-9.0-CURRENT-201001-i386). I also
compared make's output, no new compiler warnings appear.
Of course, this isn't a complete analysis, but things look much better
than it initially appeared.
More information about the freebsd-bugs