NetBSD Problem Report #37173

From dholland@eecs.harvard.edu  Mon Oct 22 20:30:21 2007
Return-Path: <dholland@eecs.harvard.edu>
Received: from mail.netbsd.org (mail.netbsd.org [204.152.190.11])
	by narn.NetBSD.org (Postfix) with ESMTP id D2D0863B8FE
	for <gnats-bugs@gnats.NetBSD.org>; Mon, 22 Oct 2007 20:30:20 +0000 (UTC)
Message-Id: <20071022191637.31AD8FBD0@tanaqui.eecs.harvard.edu>
Date: Mon, 22 Oct 2007 15:16:37 -0400 (EDT)
From: dholland@eecs.harvard.edu
Reply-To: dholland@eecs.harvard.edu
To: gnats-bugs@NetBSD.org
Subject: PKG_OPTIONS.tin and curses
X-Send-Pr-Version: 3.95

>Number:         37173
>Category:       pkg
>Synopsis:       PKG_OPTIONS.tin and curses
>Confidential:   no
>Severity:       serious
>Priority:       medium
>Responsible:    wiz
>State:          closed
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Mon Oct 22 20:35:00 +0000 2007
>Closed-Date:    Sun Nov 25 12:43:57 +0000 2007
>Last-Modified:  Sat Dec 08 18:40:01 +0000 2007
>Originator:     David A. Holland <dholland@eecs.harvard.edu>
>Release:        NetBSD 2.1.0_STABLE (20070915) (current pkgsrc)
>Organization:
   EECS
>Environment:
System: NetBSD bantha 2.1.0_STABLE NetBSD 2.1.0_STABLE (GENERIC) #3: Fri Dec  1 14:58:15 EST 2006  root@bantha:/usr/src/sys/arch/i386/compile/GENERIC i386
Architecture: i386
Machine: i386
>Description:

Recently the PKG_OPTIONS for tin were rearranged to allow choosing
curses, ncurses, or ncursesw. None of these, however, is the same as
the behavior prior to the rearrangement; the choice apparently has
broad user-visible implications, and consequently the users complained.

>How-To-Repeat:

Recompile tin with the default PKG_OPTIONS. The reports I got included
changes in color display and arrow key behavior, and maybe other stuff.

>Fix:

This adds an option to select the old behavior. It might be best to
make it the default as well, so as to create the least surprise.

Index: options.mk
===================================================================
RCS file: /cvsroot/pkgsrc/news/tin/options.mk,v
retrieving revision 1.8
diff -u -r1.8 options.mk
--- options.mk	19 Sep 2007 05:55:05 -0000	1.8
+++ options.mk	22 Oct 2007 19:11:55 -0000
@@ -3,7 +3,7 @@
 PKG_OPTIONS_VAR=		PKG_OPTIONS.tin
 PKG_SUPPORTED_OPTIONS=		icu inet6 tin-use-inn-spool
 PKG_OPTIONS_REQUIRED_GROUPS=	display
-PKG_OPTIONS_GROUP.display=	curses ncurses ncursesw
+PKG_OPTIONS_GROUP.display=	nocurses curses ncurses ncursesw
 PKG_SUGGESTED_OPTIONS=		curses
 # untested
 #PKG_SUPPORTED_OPTIONS+=	socks

>Release-Note:

>Audit-Trail:

Responsible-Changed-From-To: pkg-manager->wiz
Responsible-Changed-By: obache@netbsd.org
Responsible-Changed-When: Tue, 23 Oct 2007 00:11:38 +0000
Responsible-Changed-Why:
Over to maintainer.


From: "Valeriy E. Ushakov" <uwe@stderr.spb.ru>
To: gnats-bugs@netbsd.org
Cc: 
Subject: Re: pkg/37173
Date: Sat, 17 Nov 2007 02:56:27 +0300

 Suggested hack makes keypad keys (arrows &co) in tin work for me again.

 Our curses has no chance of working with tin, b/c tin does something
 in its curses code that is not strictly legal.  In tcurses.c:ReadCh
 it does ungetch() on KEY_* values while our curses doesn't support
 that (as per SUS).  To quote the SUS (emphasis mine):

   The ungetch() function pushes the SINGLE-BYTE character ch onto the
   head of the input queue.

 and KEY_* values are from 0x100 up, i.e. not single-byte.  So later,
 when get_arrow_key calls getch() again it receives truncated value
 (KEY_* & 0xff) instead of the KEY_* value.

 -uwe

State-Changed-From-To: open->closed
State-Changed-By: wiz@netbsd.org
State-Changed-When: Sun, 25 Nov 2007 12:43:57 +0000
State-Changed-Why:
I made the curses support completely optional -- just add '-curses' to your
tin options.
Thanks for the PR!


From: Thomas Klausner <wiz@netbsd.org>
To: gnats-bugs@NetBSD.org
Cc: 
Subject: PR/37173 CVS commit: pkgsrc/news/tin
Date: Sun, 25 Nov 2007 12:44:07 +0000 (UTC)

 Module Name:	pkgsrc
 Committed By:	wiz
 Date:		Sun Nov 25 12:44:07 UTC 2007

 Modified Files:
 	pkgsrc/news/tin: options.mk

 Log Message:
 Make curses support completely optional, in response to PR 37173
 by David A. Holland.


 To generate a diff of this commit:
 cvs rdiff -r1.8 -r1.9 pkgsrc/news/tin/options.mk

 Please note that diffs are not public domain; they are subject to the
 copyright notices on the relevant files.

From: Julian Coleman <jdc@netbsd.org>
To: gnats-bugs@NetBSD.org
Cc: 
Subject: PR/37173 CVS commit: src/lib/libcurses
Date: Sat,  8 Dec 2007 18:38:12 +0000 (UTC)

 Module Name:	src
 Committed By:	jdc
 Date:		Sat Dec  8 18:38:11 UTC 2007

 Modified Files:
 	src/lib/libcurses: curses_private.h get_wch.c getch.c screen.c

 Log Message:
 Keep pushed-back characters locally.  Fixes problems where KEY_* symbols
 are pushed back.  Should fix the arrow keys part of PR pkg/37173.

 While we are here, make getch() and get_wch() check for resize immediately,
 instead of reading a key, checking for resize and then having to push-back
 the just read key.


 To generate a diff of this commit:
 cvs rdiff -r1.40 -r1.41 src/lib/libcurses/curses_private.h
 cvs rdiff -r1.4 -r1.5 src/lib/libcurses/get_wch.c
 cvs rdiff -r1.49 -r1.50 src/lib/libcurses/getch.c
 cvs rdiff -r1.20 -r1.21 src/lib/libcurses/screen.c

 Please note that diffs are not public domain; they are subject to the
 copyright notices on the relevant files.

>Unformatted:

Home	PR Database Search