Johannes Schindelin [off-list ref] writes:
At least on one of my machines, Term::ReadLine tries to be clever and
sends \x1b\x5b1034h at the end of the script when TERM is set to 'xterm'.
To prevent that, force TERM=dumb just for the test.
We try to set up a pretty vanilla environment for test repeatability in
t/test-lib.sh, and I suspect we would simply want to do this over there.
Wouldn't this make more sense?
diff --git a/t/test-lib.sh b/t/test-lib.sh
index 87a5ea4..7f6331a 100644
--- a/t/test-lib.sh
+++ b/t/test-lib.sh
@@ -7,8 +7,9 @@
LANG=C
LC_ALL=C
PAGER=cat
+TERM=dumb
TZ=UTC
-export LANG LC_ALL PAGER TZ
+export LANG LC_ALL PAGER TERM TZ
EDITOR=:
VISUAL=:
unset GIT_EDITOR
Hi,
On Thu, 6 Mar 2008, Junio C Hamano wrote:
Johannes Schindelin [off-list ref] writes:
quoted
At least on one of my machines, Term::ReadLine tries to be clever and
sends \x1b\x5b1034h at the end of the script when TERM is set to 'xterm'.
To prevent that, force TERM=dumb just for the test.
We try to set up a pretty vanilla environment for test repeatability in
t/test-lib.sh, and I suspect we would simply want to do this over there.
Wouldn't this make more sense?
Sure!
Ciao,
Dscho
Le vendredi 7 mars 2008, Junio C Hamano a écrit :
quoted hunk
Johannes Schindelin [off-list ref] writes:
quoted
At least on one of my machines, Term::ReadLine tries to be clever and
sends \x1b\x5b1034h at the end of the script when TERM is set to
'xterm'.
To prevent that, force TERM=dumb just for the test.
We try to set up a pretty vanilla environment for test repeatability in
t/test-lib.sh, and I suspect we would simply want to do this over there.
Wouldn't this make more sense?
diff --git a/t/test-lib.sh b/t/test-lib.sh
index 87a5ea4..7f6331a 100644
--- a/t/test-lib.sh
+++ b/t/test-lib.sh
@@ -7,8 +7,9 @@
LANG=C
LC_ALL=C
PAGER=cat
+TERM=dumb
TZ=UTC
-export LANG LC_ALL PAGER TZ
+export LANG LC_ALL PAGER TERM TZ
EDITOR=:
VISUAL=:
unset GIT_EDITOR
This may not work well with colored output.
In test-lib.sh line 62 there is :
[ "x$TERM" != "xdumb" ] &&
[ -t 1 ] &&
tput bold >/dev/null 2>&1 &&
tput setaf 1 >/dev/null 2>&1 &&
tput sgr0 >/dev/null 2>&1 &&
color=t
Thanks,
Christian.