Re: cvs commit: ports/sysutils/screen Makefile pkg-plist ports/sysutils/screen/ (fwd)
Cy Schubert writes:
[...]
> +.if defined(WITH_KOI8U)
> +EXTRA_PATCHES+= `${ECHO} ${FILESDIR}/opt-koi8u-*`
> +PLIST_FILES+=%%DATADIR%%/utf8encodings/a3
> +.endif
${ECHO} is "true" when make(1) is invoked in -s (silent) mode.
That's why PM recommends using ${ECHO_CMD}/${ECHO_MSG} instead.
___
freebsd-ports@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-ports
To unsubscribe, send any mail to "freebsd-ports-unsubscr...@freebsd.org"
Re: cvs commit: ports/sysutils/screen Makefile pkg-plist ports/sysutils/screen/ (fwd)
Wed, May 04, 2011 at 01:37:23, amdmi3 wrote about "Re: cvs commit: ports/sysutils/screen Makefile pkg-plist ports/sysutils/screen/ (fwd)": > > > Unfortunately I will not commit any patched version of this patch without > > > having it fully tested. As I do not use UTF8 character sets, I cannot > > > tell > > > if the patch will work or not. > > > > OK. Dmitry and other volunteers, please test the following. At least > > it correctly works on my system within the following modes: > > 1) utf-8 both outside and inside > > 2) koi8-u both outside and inside > > 3) koi8-r both outside and inside > > 4) koi8-u outside, utf-8 inside > > 5) utf-8 outside, koi8-u inside > > Works for me with utf-8 outside, utf-8 and koi8-r inside. Thanks, that's good news. -netch- ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to "freebsd-ports-unsubscr...@freebsd.org"
Re: cvs commit: ports/sysutils/screen Makefile pkg-plist ports/sysutils/screen/ (fwd)
Tue, May 03, 2011 at 20:30:11, Cy.Schubert wrote about "Re: cvs commit: ports/sysutils/screen Makefile pkg-plist ports/sysutils/screen/ (fwd)": > > I noticed that this patch is being discussed on screen-devel. It would be > > ideal if the screen upline would adopt this patch. Hmmm, the only "discussion" is that I proposed it simultaneously to upstream. This of course would be better but I know that the guys there are much more... hmmm... beholdive:) and it is rather long to wait for next version. > > On another thought, I'd like to offer a compromise. Inclusion of this patch > > could be controlled through an option. Agreed, thanks. > A more perfect patch... > > + SHOWENC "Show encoding on the status line" OFF \ > + KOI8U "Add support for koi8-u encooding" OFF ~ typo -netch- ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to "freebsd-ports-unsubscr...@freebsd.org"
Re: cvs commit: ports/sysutils/screen Makefile pkg-plist ports/sysutils/screen/ (fwd)
In message <201105040312.p443c7wg091...@cwsys.cwsent.com>, Cy Schubert
writes:
> In message <20110503213723.GA16152@hades.panopticon>, Dmitry Marakasov
> writes:
> > * Valentin Nechayev (ne...@netch.kiev.ua) wrote:
> >
> > > > Unfortunately I will not commit any patched version of this patch witho
> ut
> >
> > > > having it fully tested. As I do not use UTF8 character sets, I cannot t
> el
> > l
> > > > if the patch will work or not.
> > >
> > > OK. Dmitry and other volunteers, please test the following. At least
> > > it correctly works on my system within the following modes:
> > > 1) utf-8 both outside and inside
> > > 2) koi8-u both outside and inside
> > > 3) koi8-r both outside and inside
> > > 4) koi8-u outside, utf-8 inside
> > > 5) utf-8 outside, koi8-u inside
> >
> > Works for me with utf-8 outside, utf-8 and koi8-r inside.
>
> I noticed that this patch is being discussed on screen-devel. It would be
> ideal if the screen upline would adopt this patch.
>
> On another thought, I'd like to offer a compromise. Inclusion of this patch
> could be controlled through an option. Ifdefs would normally work however
> conditionally patching man pages should also be considered. What I have in
> mind is naming the patches with a koi8u prefix, e.g. koi8u-patch-*, and
> setting $EXTRA_PATCHES to include those patches. If the patches cause any
> other gas not discovered during testing any resulting problems are limited
> to only those who specified the option, making fixing such a problem less
> critical than having to immediately reverse the commit at a future date to
> satisfy most users' requirements for a stable package. To this end
> following is a patch which satisfies this requirement.
A more perfect patch...
Index: Makefile
===
RCS file: /home/pcvs/ports/sysutils/screen/Makefile,v
retrieving revision 1.84
diff -u -r1.84 Makefile
--- Makefile3 May 2011 19:34:48 - 1.84
+++ Makefile4 May 2011 03:28:26 -
@@ -7,7 +7,7 @@
PORTNAME= screen
PORTVERSION= 4.0.3
-PORTREVISION= 12
+PORTREVISION= 13
CATEGORIES=sysutils
MASTER_SITES= ftp://ftp.uni-erlangen.de/pub/utilities/screen/ \
http://komquats.com/distfiles/ \
@@ -24,7 +24,8 @@
NETHACK "Enable nethack-style messages" ON \
XTERM_256 "Enable support for 256 colour xterm" OFF \
HOSTINLOCKED "Print user@host in locked message" OFF \
- SHOWENC "Show encoding on the status line" OFF
+ SHOWENC "Show encoding on the status line" OFF \
+ KOI8U "Add support for koi8-u encooding" OFF
.include
@@ -71,9 +72,17 @@
EXTRA_PATCHES+=${FILESDIR}/opt-showencoding
.endif
+.if defined(WITH_KOI8U)
+EXTRA_PATCHES+= `${ECHO} ${FILESDIR}/opt-koi8u-*`
+PLIST_FILES+= %%DATADIR%%/utf8encodings/a3
+.endif
+
post-patch:
@${RM} ${WRKSRC}/doc/screen.info*
@${REINPLACE_CMD} -e 's|/dev/ptmx|/nonexistent|' ${WRKSRC}/configure
+.if defined(WITH_KOI8U)
+ @uudecode -o ${WRKSRC}/utf8encodings/a3 ${FILESDIR}/uue-utf8encodings_a3
+.endif
post-install:
@${MKDIR} ${EXAMPLESDIR}
Index: files/opt-koi8u-ansi.h
===
RCS file: files/opt-koi8u-ansi.h
diff -N files/opt-koi8u-ansi.h
--- /dev/null 1 Jan 1970 00:00:00 -
+++ files/opt-koi8u-ansi.h 4 May 2011 03:28:26 -
@@ -0,0 +1,10 @@
+--- ansi.h
ansi.h
+@@ -128,6 +128,7 @@
+ #define KOI8R 6
+ #define CP12517
+ #define GBK 20
++#define KOI8U 21
+
+ #define EUC EUC_JP
+
Index: files/opt-koi8u-doc_screen.1
===
RCS file: files/opt-koi8u-doc_screen.1
diff -N files/opt-koi8u-doc_screen.1
--- /dev/null 1 Jan 1970 00:00:00 -
+++ files/opt-koi8u-doc_screen.14 May 2011 03:28:26 -
@@ -0,0 +1,15 @@
+diff -u -rNu doc/screen.1 doc/screen.1
+--- doc/screen.1 2003-12-05 15:51:57.0 +0200
doc/screen.1 2011-03-29 11:22:55.739089194 +0300
+@@ -1688,8 +1688,9 @@
+ There is also a way to select a terminal encoding depending on
+ the terminal type by using the \*QKJ\*U termcap entry.
+
+-Supported encodings are eucJP, SJIS, eucKR, eucCN, Big5, GBK, KOI8-R,
+-CP1251, UTF-8, ISO8859-2, ISO8859-3, ISO8859-4, ISO8859-5, ISO8859-6,
++Supported encodings are eucJP, SJIS, eucKR, eucCN, Big5, GBK,
++KOI8-R, KOI8-U, CP1251, UTF-8,
++ISO8859-2, ISO8859-3, ISO8859-4, ISO8859-5, ISO8859-6,
+ ISO8859-7, ISO8859-8, ISO8859-9, ISO8859-10, ISO8859-15, jis.
+
+ See also \*Qdefencoding\*U, which changes the default setting of a new
Index: files/opt-koi8u-doc_screen.texinfo
===
RCS file: files/opt-koi8u-doc_screen.texinfo
diff -N files/opt-koi8u-doc_screen.texinfo
--- /dev/null 1 Jan 1970 00:00:00 -
+++ files/opt-koi8u-doc_screen.texinfo 4 May 2011 03:2
Re: cvs commit: ports/sysutils/screen Makefile pkg-plist ports/sysutils/screen/ (fwd)
In message <20110503213723.GA16152@hades.panopticon>, Dmitry Marakasov
writes:
> * Valentin Nechayev (ne...@netch.kiev.ua) wrote:
>
> > > Unfortunately I will not commit any patched version of this patch without
>
> > > having it fully tested. As I do not use UTF8 character sets, I cannot tel
> l
> > > if the patch will work or not.
> >
> > OK. Dmitry and other volunteers, please test the following. At least
> > it correctly works on my system within the following modes:
> > 1) utf-8 both outside and inside
> > 2) koi8-u both outside and inside
> > 3) koi8-r both outside and inside
> > 4) koi8-u outside, utf-8 inside
> > 5) utf-8 outside, koi8-u inside
>
> Works for me with utf-8 outside, utf-8 and koi8-r inside.
I noticed that this patch is being discussed on screen-devel. It would be
ideal if the screen upline would adopt this patch.
On another thought, I'd like to offer a compromise. Inclusion of this patch
could be controlled through an option. Ifdefs would normally work however
conditionally patching man pages should also be considered. What I have in
mind is naming the patches with a koi8u prefix, e.g. koi8u-patch-*, and
setting $EXTRA_PATCHES to include those patches. If the patches cause any
other gas not discovered during testing any resulting problems are limited
to only those who specified the option, making fixing such a problem less
critical than having to immediately reverse the commit at a future date to
satisfy most users' requirements for a stable package. To this end
following is a patch which satisfies this requirement.
Index: Makefile
===
RCS file: /home/pcvs/ports/sysutils/screen/Makefile,v
retrieving revision 1.84
diff -u -r1.84 Makefile
--- Makefile3 May 2011 19:34:48 - 1.84
+++ Makefile4 May 2011 03:09:00 -
@@ -7,7 +7,7 @@
PORTNAME= screen
PORTVERSION= 4.0.3
-PORTREVISION= 12
+PORTREVISION= 13
CATEGORIES=sysutils
MASTER_SITES= ftp://ftp.uni-erlangen.de/pub/utilities/screen/ \
http://komquats.com/distfiles/ \
@@ -24,7 +24,8 @@
NETHACK "Enable nethack-style messages" ON \
XTERM_256 "Enable support for 256 colour xterm" OFF \
HOSTINLOCKED "Print user@host in locked message" OFF \
- SHOWENC "Show encoding on the status line" OFF
+ SHOWENC "Show encoding on the status line" OFF \
+ KOI8U "Add support for koi8-u encooding" OFF
.include
@@ -71,9 +72,15 @@
EXTRA_PATCHES+=${FILESDIR}/opt-showencoding
.endif
+.if defined(WITH_KOI8U)
+EXTRA_PATCHES+= `${ECHO} ${FILESDIR}/opt-koi8u-*`
+PLIST_FILES+= %%DATADIR%%/utf8encodings/a3
+.endif
+
post-patch:
@${RM} ${WRKSRC}/doc/screen.info*
@${REINPLACE_CMD} -e 's|/dev/ptmx|/nonexistent|' ${WRKSRC}/configure
+ @uudecode -o ${WRKSRC}/utf8encodings/a3 ${FILESDIR}/uue-utf8encodings_a3
post-install:
@${MKDIR} ${EXAMPLESDIR}
Index: files/opt-koi8u-ansi.h
===
RCS file: files/opt-koi8u-ansi.h
diff -N files/opt-koi8u-ansi.h
--- /dev/null 1 Jan 1970 00:00:00 -
+++ files/opt-koi8u-ansi.h 4 May 2011 03:09:00 -
@@ -0,0 +1,10 @@
+--- ansi.h
ansi.h
+@@ -128,6 +128,7 @@
+ #define KOI8R 6
+ #define CP12517
+ #define GBK 20
++#define KOI8U 21
+
+ #define EUC EUC_JP
+
Index: files/opt-koi8u-doc_screen.1
===
RCS file: files/opt-koi8u-doc_screen.1
diff -N files/opt-koi8u-doc_screen.1
--- /dev/null 1 Jan 1970 00:00:00 -
+++ files/opt-koi8u-doc_screen.14 May 2011 03:09:00 -
@@ -0,0 +1,15 @@
+diff -u -rNu doc/screen.1 doc/screen.1
+--- doc/screen.1 2003-12-05 15:51:57.0 +0200
doc/screen.1 2011-03-29 11:22:55.739089194 +0300
+@@ -1688,8 +1688,9 @@
+ There is also a way to select a terminal encoding depending on
+ the terminal type by using the \*QKJ\*U termcap entry.
+
+-Supported encodings are eucJP, SJIS, eucKR, eucCN, Big5, GBK, KOI8-R,
+-CP1251, UTF-8, ISO8859-2, ISO8859-3, ISO8859-4, ISO8859-5, ISO8859-6,
++Supported encodings are eucJP, SJIS, eucKR, eucCN, Big5, GBK,
++KOI8-R, KOI8-U, CP1251, UTF-8,
++ISO8859-2, ISO8859-3, ISO8859-4, ISO8859-5, ISO8859-6,
+ ISO8859-7, ISO8859-8, ISO8859-9, ISO8859-10, ISO8859-15, jis.
+
+ See also \*Qdefencoding\*U, which changes the default setting of a new
Index: files/opt-koi8u-doc_screen.texinfo
===
RCS file: files/opt-koi8u-doc_screen.texinfo
diff -N files/opt-koi8u-doc_screen.texinfo
--- /dev/null 1 Jan 1970 00:00:00 -
+++ files/opt-koi8u-doc_screen.texinfo 4 May 2011 03:09:00 -
@@ -0,0 +1,13 @@
+diff -u -rNu doc/screen.texinfo doc/screen.texinfo
+--- doc/screen.texinfo 2003-12-05 15:51:46.0 +0200
doc/screen.texinfo 2011-03-29 11:22:30.536269863 +0300
+@@ -2726,
Re: cvs commit: ports/sysutils/screen Makefile pkg-plist ports/sysutils/screen/ (fwd)
* Valentin Nechayev (ne...@netch.kiev.ua) wrote: > > Unfortunately I will not commit any patched version of this patch without > > having it fully tested. As I do not use UTF8 character sets, I cannot tell > > if the patch will work or not. > > OK. Dmitry and other volunteers, please test the following. At least > it correctly works on my system within the following modes: > 1) utf-8 both outside and inside > 2) koi8-u both outside and inside > 3) koi8-r both outside and inside > 4) koi8-u outside, utf-8 inside > 5) utf-8 outside, koi8-u inside Works for me with utf-8 outside, utf-8 and koi8-r inside. -- Dmitry Marakasov . 55B5 0596 FF1E 8D84 5F56 9510 D35A 80DD F9D2 F77D amd...@amdmi3.ru ..: jabber: amd...@jabber.ruhttp://www.amdmi3.ru ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to "freebsd-ports-unsubscr...@freebsd.org"
Re: cvs commit: ports/sysutils/screen Makefile pkg-plist ports/sysutils/screen/ (fwd)
Tue, May 03, 2011 at 12:49:28, Cy.Schubert wrote about "Re: cvs commit:
ports/sysutils/screen Makefile pkg-plist ports/sysutils/screen/ (fwd)":
> Unfortunately I will not commit any patched version of this patch without
> having it fully tested. As I do not use UTF8 character sets, I cannot tell
> if the patch will work or not.
OK. Dmitry and other volunteers, please test the following. At least
it correctly works on my system within the following modes:
1) utf-8 both outside and inside
2) koi8-u both outside and inside
3) koi8-r both outside and inside
4) koi8-u outside, utf-8 inside
5) utf-8 outside, koi8-u inside
The factor I missed in previous patch is that coding number in ansi.h
shall be equal to offset in encodings[].
diff -u -rNu 0/ports/sysutils/screen/Makefile 1/ports/sysutils/screen/Makefile
--- 0/ports/sysutils/screen/Makefile2011-01-04 05:28:30.0 +0200
+++ 1/ports/sysutils/screen/Makefile2011-03-29 12:12:08.557420238 +0300
@@ -7,7 +7,7 @@
PORTNAME= screen
PORTVERSION= 4.0.3
-PORTREVISION= 10
+PORTREVISION= 12
CATEGORIES=sysutils
MASTER_SITES= ftp://ftp.uni-erlangen.de/pub/utilities/screen/ \
http://komquats.com/distfiles/ \
@@ -74,6 +74,7 @@
post-patch:
@${RM} ${WRKSRC}/doc/screen.info*
@${REINPLACE_CMD} -e 's|/dev/ptmx|/nonexistent|' ${WRKSRC}/configure
+ uudecode -o ${WRKSRC}/utf8encodings/a3 ${FILESDIR}/uue-utf8encodings_a3
post-install:
@${MKDIR} ${EXAMPLESDIR}
diff -u -rNu 0/ports/sysutils/screen/files/patch-ansi.h
1/ports/sysutils/screen/files/patch-ansi.h
--- 0/ports/sysutils/screen/files/patch-ansi.h 1970-01-01 03:00:00.0
+0300
+++ 1/ports/sysutils/screen/files/patch-ansi.h 2011-05-03 23:19:08.376097400
+0300
@@ -0,0 +1,10 @@
+--- ansi.h
ansi.h
+@@ -128,6 +128,7 @@
+ #define KOI8R 6
+ #define CP12517
+ #define GBK 20
++#define KOI8U 21
+
+ #define EUC EUC_JP
+
diff -u -rNu 0/ports/sysutils/screen/files/patch-doc_screen.1
1/ports/sysutils/screen/files/patch-doc_screen.1
--- 0/ports/sysutils/screen/files/patch-doc_screen.11970-01-01
03:00:00.0 +0300
+++ 1/ports/sysutils/screen/files/patch-doc_screen.12011-03-29
12:04:42.049544451 +0300
@@ -0,0 +1,15 @@
+diff -u -rNu doc/screen.1 doc/screen.1
+--- doc/screen.1 2003-12-05 15:51:57.0 +0200
doc/screen.1 2011-03-29 11:22:55.739089194 +0300
+@@ -1688,8 +1688,9 @@
+ There is also a way to select a terminal encoding depending on
+ the terminal type by using the \*QKJ\*U termcap entry.
+
+-Supported encodings are eucJP, SJIS, eucKR, eucCN, Big5, GBK, KOI8-R,
+-CP1251, UTF-8, ISO8859-2, ISO8859-3, ISO8859-4, ISO8859-5, ISO8859-6,
++Supported encodings are eucJP, SJIS, eucKR, eucCN, Big5, GBK,
++KOI8-R, KOI8-U, CP1251, UTF-8,
++ISO8859-2, ISO8859-3, ISO8859-4, ISO8859-5, ISO8859-6,
+ ISO8859-7, ISO8859-8, ISO8859-9, ISO8859-10, ISO8859-15, jis.
+
+ See also \*Qdefencoding\*U, which changes the default setting of a new
diff -u -rNu 0/ports/sysutils/screen/files/patch-doc_screen.texinfo
1/ports/sysutils/screen/files/patch-doc_screen.texinfo
--- 0/ports/sysutils/screen/files/patch-doc_screen.texinfo 1970-01-01
03:00:00.0 +0300
+++ 1/ports/sysutils/screen/files/patch-doc_screen.texinfo 2011-03-29
12:05:34.814438254 +0300
@@ -0,0 +1,13 @@
+diff -u -rNu doc/screen.texinfo doc/screen.texinfo
+--- doc/screen.texinfo 2003-12-05 15:51:46.0 +0200
doc/screen.texinfo 2011-03-29 11:22:30.536269863 +0300
+@@ -2726,7 +2726,8 @@
+
+ Supported encodings are
+ @code{eucJP}, @code{SJIS}, @code{eucKR},
+-@code{eucCN}, @code{Big5}, @code{GBK}, @code{KOI8-R}, @code{CP1251},
++@code{eucCN}, @code{Big5}, @code{GBK},
++@code{KOI8-R}, @code{KOI8-U}, @code{CP1251},
+ @code{UTF-8}, @code{ISO8859-2}, @code{ISO8859-3},
+ @code{ISO8859-4}, @code{ISO8859-5}, @code{ISO8859-6},
+ @code{ISO8859-7}, @code{ISO8859-8}, @code{ISO8859-9},
diff -u -rNu 0/ports/sysutils/screen/files/patch-encoding.c
1/ports/sysutils/screen/files/patch-encoding.c
--- 0/ports/sysutils/screen/files/patch-encoding.c 1970-01-01
03:00:00.0 +0300
+++ 1/ports/sysutils/screen/files/patch-encoding.c 2011-05-03
23:20:03.466558553 +0300
@@ -0,0 +1,20 @@
+--- encoding.c
encoding.c
+@@ -57,6 +57,7 @@
+
+ /* big5 font: ^X */
+ /* KOI8-R font: 96 ! */
++/* KOI8-U font: 96 # */
+ /* CP1251 font: 96 ? */
+
+ struct encoding encodings[] = {
+@@ -80,7 +81,8 @@
+ { "ISO8859-10", 0, 0x80|'V', 0, 0, 0 },
+ { "ISO8859-15", 0, 0x80|'b', 0, 0, 0 },
+ { "jis",0, 0, 0, 0, "\002\004I" },
+- { "GBK","B\031BB01",0x80|'b', 1, 1, "\031" }
++ { "GBK","B\031BB01",0x80|'b', 1, 1, "\031" },
++ { "KOI8-
Re: cvs commit: ports/sysutils/screen Makefile pkg-plist ports/sysutils/screen/ (fwd)
In message <20110503194409.gc65...@netch.kiev.ua>, Valentin Nechayev writes: > Tue, May 03, 2011 at 12:25:51, Cy.Schubert wrote about "Re: cvs commit: port > s/sysutils/screen Makefile pkg-plist ports/sysutils/screen/ (fwd)": > > > I need to back this out. Sorry. You may want to post the patch on ports@ > > and have ppl test it for a while. > > Please wait for an hour - seems I see the problem Unfortunately I will not commit any patched version of this patch without having it fully tested. As I do not use UTF8 character sets, I cannot tell if the patch will work or not. It is suggested you publish the patch to freebsd-ports and have a few people test it first. As soon is there is a group of people on the mailing list who have vetted the patch I will recommit it. Otherwise I cannot take the risk. -- Cheers, Cy Schubert FreeBSD UNIX: Web: http://www.FreeBSD.org Q: How did the regular expression cross the road? A: ^.*$ ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to "freebsd-ports-unsubscr...@freebsd.org"
