[pacman-dev] [PATCH] makepkg: use tput for terminal-safe colored and bold text

newer
[pacman-dev] [PATCH] Fix a small...

Cedric Staniewski

22 Oct 2009 22 Oct '09

7:08 p.m.

Suggested-by: Dan McGee Signed-off-by: Cedric Staniewski --- I do not know if this patch is usefull at all, because I do not notice any change. Just grabbed this from Dan's TODO list and wanted to play a bit with tput. scripts/makepkg.sh.in | 10 +++++----- 1 files changed, 5 insertions(+), 5 deletions(-) diff --git a/scripts/makepkg.sh.in b/scripts/makepkg.sh.in index 40367ae..a64a867 100644 --- a/scripts/makepkg.sh.in +++ b/scripts/makepkg.sh.in @@ -85,7 +85,7 @@ PACMAN_OPTS= plain() { local mesg=$1; shift if [ $COLORMSG -eq 1 ]; then - printf "\033[1;1m ${mesg}\033[1;0m\n" "$@" >&2 + printf "$(tput bold) ${mesg}$(tput sgr0)\n" "$@" >&2 else printf " ${mesg}\n" "$@" >&2 fi @@ -94,7 +94,7 @@ plain() { msg() { local mesg=$1; shift if [ $COLORMSG -eq 1 ]; then - printf "\033[1;32m==>\033[1;0m\033[1;1m ${mesg}\033[1;0m\n" "$@" >&2 + printf "$(echo -e "bold\nsetaf 2" | tput -S)==>$(echo -e "sgr0\nbold" | tput -S) ${mesg}$(tput sgr0)\n" "$@" >&2 else printf "==> ${mesg}\n" "$@" >&2 fi @@ -103,7 +103,7 @@ msg() { msg2() { local mesg=$1; shift if [ $COLORMSG -eq 1 ]; then - printf "\033[1;34m ->\033[1;0m\033[1;1m ${mesg}\033[1;0m\n" "$@" >&2 + printf "$(echo -e "bold\nsetaf 4" | tput -S) ->$(echo -e "sgr0\nbold" | tput -S) ${mesg}$(tput sgr0)\n" "$@" >&2 else printf " -> ${mesg}\n" "$@" >&2 fi @@ -112,7 +112,7 @@ msg2() { warning() { local mesg=$1; shift if [ $COLORMSG -eq 1 ]; then - printf "\033[1;33m==> $(gettext "WARNING:")\033[1;0m\033[1;1m ${mesg}\033[1;0m\n" "$@" >&2 + printf "$(echo -e "bold\nsetaf 3" | tput -S)==> $(gettext "WARNING:")$(echo -e "sgr0\nbold" | tput -S) ${mesg}$(tput sgr0)\n" "$@" >&2 else printf "==> $(gettext "WARNING:") ${mesg}\n" "$@" >&2 fi @@ -121,7 +121,7 @@ warning() { error() { local mesg=$1; shift if [ $COLORMSG -eq 1 ]; then - printf "\033[1;31m==> $(gettext "ERROR:")\033[1;0m\033[1;1m ${mesg}\033[1;0m\n" "$@" >&2 + printf "$(echo -e "bold\nsetaf 1" | tput -S)==> $(gettext "ERROR:")$(echo -e "sgr0\nbold" | tput -S) ${mesg}$(tput sgr0)\n" "$@" >&2 else printf "==> $(gettext "ERROR:") ${mesg}\n" "$@" >&2 fi -- 1.6.5.1

Show replies by date

Allan McRae

23 Oct 23 Oct

12:41 a.m.

New subject: [pacman-dev] [PATCH] makepkg: use tput for terminal-safe colored and bold text

Cedric Staniewski wrote:

...

Suggested-by: Dan McGee Signed-off-by: Cedric Staniewski ---

I do not know if this patch is usefull at all, because I do not notice any change. Just grabbed this from Dan's TODO list and wanted to play a bit with tput.

It looks fine to me... I just have no idea on the advantage we are achieving with this change apart from the apparent terminal safeness. Was the old version not safe? Dan: comments?

Dan McGee

12:46 a.m.

New subject: [pacman-dev] [PATCH] makepkg: use tput for terminal-safe colored and bold text

On Thu, Oct 22, 2009 at 7:41 PM, Allan McRae wrote:

...

Cedric Staniewski wrote:

...
Suggested-by: Dan McGee Signed-off-by: Cedric Staniewski ---

I do not know if this patch is usefull at all, because I do not notice any change. Just grabbed this from Dan's TODO list and wanted to play a bit with tput.

It looks fine to me... I just have no idea on the advantage we are achieving with this change apart from the apparent terminal safeness. Was the old version not safe?

Dan: comments?

First comment- that TODO list is still huge, wow. :) Anyway, this seems pretty reasonable to me, but not sure it is worth it. At the least, we should capture these sequences once on script startup, and then use the global variable in each function. And does $(tput offbold) make more sense for the reset? -Dan

Allan McRae

1:50 a.m.

New subject: [pacman-dev] [PATCH] makepkg: use tput for terminal-safe colored and bold text

Dan McGee wrote:

...

On Thu, Oct 22, 2009 at 7:41 PM, Allan McRae wrote:

...
Cedric Staniewski wrote:

...
Suggested-by: Dan McGee Signed-off-by: Cedric Staniewski ---

I do not know if this patch is usefull at all, because I do not notice any change. Just grabbed this from Dan's TODO list and wanted to play a bit with tput.

It looks fine to me... I just have no idea on the advantage we are achieving with this change apart from the apparent terminal safeness. Was the old version not safe?

Dan: comments?

First comment- that TODO list is still huge, wow. :)

Anyway, this seems pretty reasonable to me, but not sure it is worth it. At the least, we should capture these sequences once on script startup, and then use the global variable in each function. And does $(tput offbold) make more sense for the reset?

There does not appear to be such a thing as "tput offbold". Anyway, I like the idea of setting these all at the start. if [ $COLORMSG -eq 1 ]; then BOLD_ON=$(tput bold) BOLD_OFF=${tput srg0) GREEN_ON... ... fi then the messages can just be plain() { local mesg=$1; shift printf "${BOLD_ON} ${mesg}${BOLD_OFF}\n" "$@" >&2 fi } with none of the tests needed. Allan

Dan McGee

4:07 a.m.

New subject: [pacman-dev] [PATCH] makepkg: use tput for terminal-safe colored and bold text

On Thu, Oct 22, 2009 at 8:50 PM, Allan McRae wrote:

...

Dan McGee wrote:

...
On Thu, Oct 22, 2009 at 7:41 PM, Allan McRae wrote:

...
Cedric Staniewski wrote:

...
Suggested-by: Dan McGee Signed-off-by: Cedric Staniewski ---

I do not know if this patch is usefull at all, because I do not notice any change. Just grabbed this from Dan's TODO list and wanted to play a bit with tput.

It looks fine to me... I just have no idea on the advantage we are achieving with this change apart from the apparent terminal safeness. Was the old version not safe?

Dan: comments?

First comment- that TODO list is still huge, wow. :)

Anyway, this seems pretty reasonable to me, but not sure it is worth it. At the least, we should capture these sequences once on script startup, and then use the global variable in each function. And does $(tput offbold) make more sense for the reset?

There does not appear to be such a thing as "tput offbold".

You are quite right, I just read the manpage wrong: bold=`tput smso` offbold=`@TPUT@ rmso` Set the shell variables bold, to begin stand-out mode sequence, and offbold, to end standout mode sequence, for the current terminal. This might be followed by a prompt: echo "${bold}Please type in your name: ${offbold}\c"

Cedric Staniewski

10:59 a.m.

New subject: [pacman-dev] [PATCH] makepkg: use tput for terminal-safe colored and bold text

Dan McGee wrote:

...

On Thu, Oct 22, 2009 at 8:50 PM, Allan McRae wrote:

...
Dan McGee wrote:

...
On Thu, Oct 22, 2009 at 7:41 PM, Allan McRae wrote:

...
Cedric Staniewski wrote:

...
Suggested-by: Dan McGee Signed-off-by: Cedric Staniewski ---

I do not know if this patch is usefull at all, because I do not notice any change. Just grabbed this from Dan's TODO list and wanted to play a bit with tput.

It looks fine to me... I just have no idea on the advantage we are achieving with this change apart from the apparent terminal safeness. Was the old version not safe?

Dan: comments?

First comment- that TODO list is still huge, wow. :)

Anyway, this seems pretty reasonable to me, but not sure it is worth it. At the least, we should capture these sequences once on script startup, and then use the global variable in each function. And does $(tput offbold) make more sense for the reset?

There does not appear to be such a thing as "tput offbold".

You are quite right, I just read the manpage wrong: bold=`tput smso` offbold=`@TPUT@ rmso` Set the shell variables bold, to begin stand-out mode sequence, and offbold, to end standout mode sequence, for the current terminal. This might be followed by a prompt: echo "${bold}Please type in your name: ${offbold}\c"

Just for the record, the capability names can be found at the terminfo manpage. The tput manpage was not that helpful in this context. By the way, the example you quoted does not work as I expected for me; it does not bold the text, but changes the background color to grey.

Allan McRae

11:17 a.m.

New subject: [pacman-dev] [PATCH] makepkg: use tput for terminal-safe colored and bold text

Cedric Staniewski wrote:

...

Dan McGee wrote:

...
On Thu, Oct 22, 2009 at 8:50 PM, Allan McRae wrote:

...
Dan McGee wrote:

...
On Thu, Oct 22, 2009 at 7:41 PM, Allan McRae wrote:

...
Cedric Staniewski wrote:

...
Suggested-by: Dan McGee Signed-off-by: Cedric Staniewski ---

I do not know if this patch is usefull at all, because I do not notice any change. Just grabbed this from Dan's TODO list and wanted to play a bit with tput.

It looks fine to me... I just have no idea on the advantage we are achieving with this change apart from the apparent terminal safeness. Was the old version not safe?

Dan: comments?

First comment- that TODO list is still huge, wow. :)

Anyway, this seems pretty reasonable to me, but not sure it is worth it. At the least, we should capture these sequences once on script startup, and then use the global variable in each function. And does $(tput offbold) make more sense for the reset?

There does not appear to be such a thing as "tput offbold".

You are quite right, I just read the manpage wrong: bold=`tput smso` offbold=`@TPUT@ rmso` Set the shell variables bold, to begin stand-out mode sequence, and offbold, to end standout mode sequence, for the current terminal. This might be followed by a prompt: echo "${bold}Please type in your name: ${offbold}\c"

Just for the record, the capability names can be found at the terminfo manpage. The tput manpage was not that helpful in this context. By the way, the example you quoted does not work as I expected for me; it does not bold the text, but changes the background color to grey.

It does the same for me in XFCE terminal. Aparently urxvt works with the example... Allan

Cedric Staniewski

10:51 a.m.

New subject: [pacman-dev] [PATCH] makepkg: use tput for terminal-safe colored and bold text

Allan McRae wrote:

...

Dan McGee wrote:

...
On Thu, Oct 22, 2009 at 7:41 PM, Allan McRae wrote:

...
Cedric Staniewski wrote:

...
Suggested-by: Dan McGee Signed-off-by: Cedric Staniewski ---

I do not know if this patch is usefull at all, because I do not notice any change. Just grabbed this from Dan's TODO list and wanted to play a bit with tput.

It looks fine to me... I just have no idea on the advantage we are achieving with this change apart from the apparent terminal safeness. Was the old version not safe?

Dan: comments?

First comment- that TODO list is still huge, wow. :)

Hehe. :)

...

...
Anyway, this seems pretty reasonable to me, but not sure it is worth it. At the least, we should capture these sequences once on script startup, and then use the global variable in each function. And does $(tput offbold) make more sense for the reset?

There does not appear to be such a thing as "tput offbold".

Anyway, I like the idea of setting these all at the start.

if [ $COLORMSG -eq 1 ]; then BOLD_ON=$(tput bold) BOLD_OFF=${tput srg0) GREEN_ON... ... fi

then the messages can just be

plain() { local mesg=$1; shift printf "${BOLD_ON} ${mesg}${BOLD_OFF}\n" "$@" >&2 fi }

with none of the tests needed.

Allan

That definitely makes sense. If you think the attached patch is fine, I plan to change the order of these two ones and resubmit them.

...

From 50901c551ddefb9a5eca500ac63ab0a0c7eef167 Mon Sep 17 00:00:00 2001 From: Cedric Staniewski Date: Fri, 23 Oct 2009 12:27:54 +0200 -- 8< -- Subject: [PATCH] makepkg: define the required escape sequences globally

In doing so, it is possible to get rid of all the tests for colored messages except for one global one. Signed-off-by: Cedric Staniewski --- scripts/makepkg.sh.in | 42 +++++++++++++++--------------------------- 1 files changed, 15 insertions(+), 27 deletions(-) diff --git a/scripts/makepkg.sh.in b/scripts/makepkg.sh.in index a64a867..b89acdb 100644 --- a/scripts/makepkg.sh.in +++ b/scripts/makepkg.sh.in @@ -72,7 +72,6 @@ IGNOREARCH=0 HOLDVER=0 PKGFUNC=0 SPLITPKG=0 -COLORMSG=0 # Forces the pkgver of the current PKGBUILD. Used by the fakeroot call # when dealing with svn/cvs/etc PKGBUILDs. @@ -84,47 +83,27 @@ PACMAN_OPTS= plain() { local mesg=$1; shift - if [ $COLORMSG -eq 1 ]; then - printf "$(tput bold) ${mesg}$(tput sgr0)\n" "$@" >&2 - else - printf " ${mesg}\n" "$@" >&2 - fi + printf "${BOLD} ${mesg}${ALL_OFF}\n" "$@" >&2 } msg() { local mesg=$1; shift - if [ $COLORMSG -eq 1 ]; then - printf "$(echo -e "bold\nsetaf 2" | tput -S)==>$(echo -e "sgr0\nbold" | tput -S) ${mesg}$(tput sgr0)\n" "$@" >&2 - else - printf "==> ${mesg}\n" "$@" >&2 - fi + printf "${GREEN}==>${ALL_OFF_BOLD} ${mesg}${ALL_OFF}\n" "$@" >&2 } msg2() { local mesg=$1; shift - if [ $COLORMSG -eq 1 ]; then - printf "$(echo -e "bold\nsetaf 4" | tput -S) ->$(echo -e "sgr0\nbold" | tput -S) ${mesg}$(tput sgr0)\n" "$@" >&2 - else - printf " -> ${mesg}\n" "$@" >&2 - fi + printf "${BLUE} ->${ALL_OFF_BOLD} ${mesg}${ALL_OFF}\n" "$@" >&2 } warning() { local mesg=$1; shift - if [ $COLORMSG -eq 1 ]; then - printf "$(echo -e "bold\nsetaf 3" | tput -S)==> $(gettext "WARNING:")$(echo -e "sgr0\nbold" | tput -S) ${mesg}$(tput sgr0)\n" "$@" >&2 - else - printf "==> $(gettext "WARNING:") ${mesg}\n" "$@" >&2 - fi + printf "${YELLOW}==> $(gettext "WARNING:")${ALL_OFF_BOLD} ${mesg}${ALL_OFF}\n" "$@" >&2 } error() { local mesg=$1; shift - if [ $COLORMSG -eq 1 ]; then - printf "$(echo -e "bold\nsetaf 1" | tput -S)==> $(gettext "ERROR:")$(echo -e "sgr0\nbold" | tput -S) ${mesg}$(tput sgr0)\n" "$@" >&2 - else - printf "==> $(gettext "ERROR:") ${mesg}\n" "$@" >&2 - fi + printf "${RED}==> $(gettext "ERROR:")${ALL_OFF_BOLD} ${mesg}${ALL_OFF}\n" "$@" >&2 } @@ -1576,7 +1555,16 @@ fi # check if messages are to be printed using color if [ -t 2 -a ! "$USE_COLOR" = "n" -a "$(check_buildenv color)" = "y" ]; then - COLORMSG=1 + readonly ALL_OFF="$(tput sgr0)" \ + BOLD="$(tput bold)" + readonly ALL_OFF_BOLD="${ALL_OFF}${BOLD}" \ + RED="${BOLD}$(tput setaf 1)" \ + BLUE="${BOLD}$(tput setaf 4)" \ + GREEN="${BOLD}$(tput setaf 2)" \ + YELLOW="${BOLD}$(tput setaf 3)" +else + unset ALL_OFF ALL_OFF_BOLD BOLD RED BLUE GREEN YELLOW + readonly ALL_OFF ALL_OFF_BOLD BOLD RED BLUE GREEN YELLOW fi # override settings with an environment variable for batch processing -- 1.6.5.1

Allan McRae

11:17 a.m.

New subject: [pacman-dev] [PATCH] makepkg: use tput for terminal-safe colored and bold text

Cedric Staniewski wrote:

...

Allan McRae wrote:

...
Dan McGee wrote:

...
On Thu, Oct 22, 2009 at 7:41 PM, Allan McRae wrote:

...
Cedric Staniewski wrote:

...
Suggested-by: Dan McGee Signed-off-by: Cedric Staniewski ---

I do not know if this patch is usefull at all, because I do not notice any change. Just grabbed this from Dan's TODO list and wanted to play a bit with tput.

It looks fine to me... I just have no idea on the advantage we are achieving with this change apart from the apparent terminal safeness. Was the old version not safe?

Dan: comments?

First comment- that TODO list is still huge, wow. :)

Hehe. :)

...
...
Anyway, this seems pretty reasonable to me, but not sure it is worth it. At the least, we should capture these sequences once on script startup, and then use the global variable in each function. And does $(tput offbold) make more sense for the reset?

There does not appear to be such a thing as "tput offbold".

Anyway, I like the idea of setting these all at the start.

if [ $COLORMSG -eq 1 ]; then BOLD_ON=$(tput bold) BOLD_OFF=${tput srg0) GREEN_ON... ... fi

then the messages can just be

plain() { local mesg=$1; shift printf "${BOLD_ON} ${mesg}${BOLD_OFF}\n" "$@" >&2 fi }

with none of the tests needed.

Allan

That definitely makes sense. If you think the attached patch is fine, I plan to change the order of these two ones and resubmit them.

...
From 50901c551ddefb9a5eca500ac63ab0a0c7eef167 Mon Sep 17 00:00:00 2001 From: Cedric Staniewski Date: Fri, 23 Oct 2009 12:27:54 +0200 -- 8< -- Subject: [PATCH] makepkg: define the required escape sequences globally

In doing so, it is possible to get rid of all the tests for colored messages except for one global one.

Signed-off-by: Cedric Staniewski --- scripts/makepkg.sh.in | 42 +++++++++++++++--------------------------- 1 files changed, 15 insertions(+), 27 deletions(-)

diff --git a/scripts/makepkg.sh.in b/scripts/makepkg.sh.in index a64a867..b89acdb 100644 --- a/scripts/makepkg.sh.in +++ b/scripts/makepkg.sh.in @@ -72,7 +72,6 @@ IGNOREARCH=0 HOLDVER=0 PKGFUNC=0 SPLITPKG=0 -COLORMSG=0

# Forces the pkgver of the current PKGBUILD. Used by the fakeroot call # when dealing with svn/cvs/etc PKGBUILDs. @@ -84,47 +83,27 @@ PACMAN_OPTS=

plain() { local mesg=$1; shift - if [ $COLORMSG -eq 1 ]; then - printf "$(tput bold) ${mesg}$(tput sgr0)\n" "$@" >&2 - else - printf " ${mesg}\n" "$@" >&2 - fi + printf "${BOLD} ${mesg}${ALL_OFF}\n" "$@" >&2 }

msg() { local mesg=$1; shift - if [ $COLORMSG -eq 1 ]; then - printf "$(echo -e "bold\nsetaf 2" | tput -S)==>$(echo -e "sgr0\nbold" | tput -S) ${mesg}$(tput sgr0)\n" "$@" >&2 - else - printf "==> ${mesg}\n" "$@" >&2 - fi + printf "${GREEN}==>${ALL_OFF_BOLD} ${mesg}${ALL_OFF}\n" "$@" >&2 }

msg2() { local mesg=$1; shift - if [ $COLORMSG -eq 1 ]; then - printf "$(echo -e "bold\nsetaf 4" | tput -S) ->$(echo -e "sgr0\nbold" | tput -S) ${mesg}$(tput sgr0)\n" "$@" >&2 - else - printf " -> ${mesg}\n" "$@" >&2 - fi + printf "${BLUE} ->${ALL_OFF_BOLD} ${mesg}${ALL_OFF}\n" "$@" >&2 }

warning() { local mesg=$1; shift - if [ $COLORMSG -eq 1 ]; then - printf "$(echo -e "bold\nsetaf 3" | tput -S)==> $(gettext "WARNING:")$(echo -e "sgr0\nbold" | tput -S) ${mesg}$(tput sgr0)\n" "$@" >&2 - else - printf "==> $(gettext "WARNING:") ${mesg}\n" "$@" >&2 - fi + printf "${YELLOW}==> $(gettext "WARNING:")${ALL_OFF_BOLD} ${mesg}${ALL_OFF}\n" "$@" >&2 }

error() { local mesg=$1; shift - if [ $COLORMSG -eq 1 ]; then - printf "$(echo -e "bold\nsetaf 1" | tput -S)==> $(gettext "ERROR:")$(echo -e "sgr0\nbold" | tput -S) ${mesg}$(tput sgr0)\n" "$@" >&2 - else - printf "==> $(gettext "ERROR:") ${mesg}\n" "$@" >&2 - fi + printf "${RED}==> $(gettext "ERROR:")${ALL_OFF_BOLD} ${mesg}${ALL_OFF}\n" "$@" >&2 }

@@ -1576,7 +1555,16 @@ fi

# check if messages are to be printed using color if [ -t 2 -a ! "$USE_COLOR" = "n" -a "$(check_buildenv color)" = "y" ]; then - COLORMSG=1 + readonly ALL_OFF="$(tput sgr0)" \ + BOLD="$(tput bold)" + readonly ALL_OFF_BOLD="${ALL_OFF}${BOLD}" \ + RED="${BOLD}$(tput setaf 1)" \ + BLUE="${BOLD}$(tput setaf 4)" \ + GREEN="${BOLD}$(tput setaf 2)" \ + YELLOW="${BOLD}$(tput setaf 3)" +else + unset ALL_OFF ALL_OFF_BOLD BOLD RED BLUE GREEN YELLOW + readonly ALL_OFF ALL_OFF_BOLD BOLD RED BLUE GREEN YELLOW fi

# override settings with an environment variable for batch processing

That looks good to me. Feel free to tidy it as you see fit and forward the patch. Allan

5306

Age (days ago)

5307

Last active (days ago)

List overview

Download

8 comments

3 participants

participants (3)

Allan McRae
Cedric Staniewski
Dan McGee