Re: -STABLE (was: ncurses.h and #define trace _nc_trace)

2000-06-13 Thread Steve Kargl
Trevor Johnson wrote: > Gerald Pfeifer wrote: > > > Is there a way to > [...] > > b) mark a port to notify users not to try using it on -STABLE and > > -CURRENT? > > After a look at > http://www.freebsd.org/porters-handbook/porting-versions.html and at > src/sys/sys/param.h,v in the CVS tre

Re: -STABLE (was: ncurses.h and #define trace _nc_trace)

2000-06-13 Thread Trevor Johnson
Gerald Pfeifer wrote: > Is there a way to [...] > b) mark a port to notify users not to try using it on -STABLE and > -CURRENT? After a look at http://www.freebsd.org/porters-handbook/porting-versions.html and at src/sys/sys/param.h,v in the CVS tree, I made the appended patch. -- Trevor Jo

Re: -STABLE (was: ncurses.h and #define trace _nc_trace)

2000-06-13 Thread Gerald Pfeifer
[ Jordan, I'm Cc:ing you as release engineer as -STABLE has been partially broken for nearly three weeks now. ] On Sat, 3 Jun 2000, Gerald Pfeifer wrote: > This causes clear breakage in -STABLE and lots of wasted time for users > of the Wine port and myself (answering support requests), for exa

Re: -STABLE (was: ncurses.h and #define trace _nc_trace)

2000-06-06 Thread Gerald Pfeifer
On Sun, 4 Jun 2000, Andrey A. Chernov wrote: > Ask Peter to commit this patch at least... Thanks! Note that Peter has been Cc:ed on all messages concerning this issue, but didn't reply yet, so I just opened PR misc/19077 and included your patch there. Gerald -- Gerald "Jerry" [EMAIL PROTECTED]

Re: -STABLE (was: ncurses.h and #define trace _nc_trace)

2000-06-04 Thread Daniel C. Sobral
"Andrey A. Chernov" wrote: > > On Mon, Jun 05, 2000 at 08:42:47AM +0900, Daniel C. Sobral wrote: > > Huh? I'm obviously missing something here... Why would the branches be > > damaged by this? > > Because they forked. It cause merge troubles on the next import into the > vendor branch forever.

Re: -STABLE (was: ncurses.h and #define trace _nc_trace)

2000-06-04 Thread Andrey A. Chernov
On Mon, Jun 05, 2000 at 08:42:47AM +0900, Daniel C. Sobral wrote: > Huh? I'm obviously missing something here... Why would the branches be > damaged by this? Because they forked. It cause merge troubles on the next import into the vendor branch forever. -- Andrey A. Chernov <[EMAIL PROTECTED]>

Re: -STABLE (was: ncurses.h and #define trace _nc_trace)

2000-06-04 Thread Daniel C. Sobral
"Andrey A. Chernov" wrote: > > On Sun, Jun 04, 2000 at 11:36:47PM +0900, Daniel C. Sobral wrote: > > You revert to the original by applying the reverse patch of your changes > > and committing that. This can be easily done with cvs. And then you > > Are you joking? CVS branches will be damaged b

Re: -STABLE (was: ncurses.h and #define trace _nc_trace)

2000-06-04 Thread Andrey A. Chernov
On Sat, Jun 03, 2000 at 07:10:56PM +0200, Gerald Pfeifer wrote: > However you do it, please fix this ASAP, and be it by reverting the Ask Peter to commit this patch at least... --- include/curses.h.in.bak Wed May 24 14:44:45 2000 +++ include/curses.h.in Mon Jun 5 02:28:33 2000 @@ -1313,8 +1

Re: -STABLE (was: ncurses.h and #define trace _nc_trace)

2000-06-04 Thread Andrey A. Chernov
On Sun, Jun 04, 2000 at 11:36:47PM +0900, Daniel C. Sobral wrote: > You revert to the original by applying the reverse patch of your changes > and committing that. This can be easily done with cvs. And then you Are you joking? CVS branches will be damaged by this again. > changes in the past fir

Re: -STABLE (was: ncurses.h and #define trace _nc_trace)

2000-06-04 Thread Daniel C. Sobral
"Andrey A. Chernov" wrote: > > > However you do it, please fix this ASAP, and be it by reverting the > > original, broken patch! > > I wish but I can't. If I'll touch those misterious CVS branches structure > again, everyone shoot at me as in previous time. I think it is enough for > me. Nowd

Re: -STABLE (was: ncurses.h and #define trace _nc_trace)

2000-06-03 Thread Andrey A. Chernov
On Sat, Jun 03, 2000 at 07:10:56PM +0200, Gerald Pfeifer wrote: > On Fri, 26 May 2000, Andrey A. Chernov wrote: > > Yes, #define was a quick solution. > > We need either to add a stub or to import ncurses author fix (removing > > all trace function from production library). > > > > Latest is bet

-STABLE (was: ncurses.h and #define trace _nc_trace)

2000-06-03 Thread Gerald Pfeifer
On Fri, 26 May 2000, Andrey A. Chernov wrote: > Yes, #define was a quick solution. > We need either to add a stub or to import ncurses author fix (removing > all trace function from production library). > > Latest is better IMHO, but I left to choose best way to Peter. However you do it, please