| Message ID | 20200719110033.78844-1-adilger@dilger.ca |
|---|---|
| State | Accepted |
| Headers | show |
| Series | tests: replace perl usage with shell built-in | expand |
On Sun, Jul 19, 2020 at 05:00:33AM -0600, Andreas Dilger wrote: > A couple of tests use perl only for generating a string of > N characters long. Instead of requiring perl to run a few > tests, use shell built-in commands and don't repeatedly run > a separate subshell just to get a string of characters. > > Signed-off-by: Andreas Dilger <adilger@dilger.ca> Thanks, applied. - Ted
On Thu, Oct 01, 2020 at 10:58:09AM -0400, Theodore Y. Ts'o wrote: > On Sun, Jul 19, 2020 at 05:00:33AM -0600, Andreas Dilger wrote: > > A couple of tests use perl only for generating a string of > > N characters long. Instead of requiring perl to run a few > > tests, use shell built-in commands and don't repeatedly run > > a separate subshell just to get a string of characters. > > h > > Signed-off-by: Andreas Dilger <adilger@dilger.ca> > > Thanks, applied. .... and I have to revert the patch. It's using a non-standard shell construction which fails on strict POSIX compliant shells, such as dash. - Ted
On Oct 1, 2020, at 11:08 AM, Theodore Y. Ts'o <tytso@MIT.EDU> wrote: > > On Thu, Oct 01, 2020 at 10:58:09AM -0400, Theodore Y. Ts'o wrote: >> On Sun, Jul 19, 2020 at 05:00:33AM -0600, Andreas Dilger wrote: >>> A couple of tests use perl only for generating a string of >>> N characters long. Instead of requiring perl to run a few >>> tests, use shell built-in commands and don't repeatedly run >>> a separate subshell just to get a string of characters. >>> h >>> Signed-off-by: Andreas Dilger <adilger@dilger.ca> >> >> Thanks, applied. > > .... and I have to revert the patch. It's using a non-standard shell > construction which fails on strict POSIX compliant shells, such as > dash. Could you please point out which constructs it is unhappy with? I had an earlier version of the patch using a different mechanism, but then updated it after feedback from Lukas. Cheers, Andreas
diff --git a/tests/d_xattr_sorting/script b/tests/d_xattr_sorting/script index 866611502012..65c74840f517 100644 --- a/tests/d_xattr_sorting/script +++ b/tests/d_xattr_sorting/script @@ -22,7 +22,7 @@ echo Exit status is $status >> $OUT.new B=$(mktemp ${TMPDIR:-/tmp}/b.XXXXXX) -perl -e 'print "x" x 256;' > $B +printf 'x%.0s' {1..256} > $B echo "ea_set -f /tmp/b / security.SMEG64" >> $OUT.new $DEBUGFS -w -R "ea_set -f $B / security.SMEG64" $TMPFILE >> $OUT.new 2>&1 diff --git a/tests/f_badsymlinks2/mkimage.sh b/tests/f_badsymlinks2/mkimage.sh index 6bbf020de0d7..297633790bee 100755 --- a/tests/f_badsymlinks2/mkimage.sh +++ b/tests/f_badsymlinks2/mkimage.sh @@ -18,10 +18,12 @@ do_tune2fs() { mount image mnt } +A=$(printf 'A%.0s' {1..4096}) symlink() { local len=$1 local src=$2 - local target=$(perl -e 'print "A" x '$len) + local target=${A:0:$len} + ln -s $target $src stat -c %i $src } diff --git a/tests/f_create_symlinks/script b/tests/f_create_symlinks/script index 169f58dbed2f..8906f3d65653 100644 --- a/tests/f_create_symlinks/script +++ b/tests/f_create_symlinks/script @@ -15,22 +15,22 @@ fi dd if=/dev/zero of=$TMPFILE bs=1k count=512 > /dev/null 2>&1 echo mke2fs -q -F -o Linux -b 1024 -g 256 -O inline_data,extents -I 256 test.img 1024 > $OUT.new -$MKE2FS -q -F -o Linux -b 1024 -g 256 -O inline_data,extents -I 256 $TMPFILE 1024 >> $OUT 2>&1 +$MKE2FS -q -F -o Linux -b 1024 -g 256 -O inline_data,extents -I 256 $TMPFILE 1024 >> $OUT.new 2>&1 $FSCK $FSCK_OPT -N test_filesys $TMPFILE >> $OUT.new 2>&1 status=$? echo Exit status is $status >> $OUT.new +B=$(printf 'x%.0s' {1..1500}) for i in 30 60 70 500 1023 1024 1500; do - echo "debugfs -R \"symlink /l_$i $(perl -e "print 'x' x $i;")\" test.img" >> $OUT.new - $DEBUGFS -w -R "symlink /l_$i $(perl -e "print 'x' x $i;")" $TMPFILE \ - 2>&1 | sed -f $cmd_dir/filter.sed >> $OUT.new + echo "debugfs -R \"symlink /l_$i ${B:0:i}\" test.img" >> $OUT.new + $DEBUGFS -w -R "symlink /l_$i ${B:0:i}" $TMPFILE >> $OUT.new 2>&1 done +unset B for i in 30 60 70 500 1023 1024 1500; do echo "debugfs -R \"stat /l_$i\" test.img" >> $OUT.new - $DEBUGFS -R "stat /l_$i" $TMPFILE 2>&1 | \ - grep -v "time: " >> $OUT.new + $DEBUGFS -R "stat /l_$i" $TMPFILE 2>&1 | grep -v "time: " >> $OUT.new done $FSCK $FSCK_OPT -N test_filesys $TMPFILE >> $OUT.new 2>&1 diff --git a/util/gen-sample-fs b/util/gen-sample-fs index 8e139160fd01..290da33f51c3 100755 --- a/util/gen-sample-fs +++ b/util/gen-sample-fs @@ -7,8 +7,10 @@ cp /dev/null $FS mke2fs -q -t ext4 -O inline_data,^has_journal -I 256 -b 4096 -N 64 $FS 256 mount -t ext4 $FS $MNT ln -s symlink_data $MNT/symlink + +L=$(printf 'x%.0s' {1..1024}) for i in 30 70 500 1023 1024; do - ln -s /$(perl -e "print 'x' x $i;") $MNT/l_$i + ln -s /${L:0:$i} $MNT/l_$i done touch $MNT/acl setfacl -m u:daemon:r $MNT/acl
A couple of tests use perl only for generating a string of N characters long. Instead of requiring perl to run a few tests, use shell built-in commands and don't repeatedly run a separate subshell just to get a string of characters. Signed-off-by: Andreas Dilger <adilger@dilger.ca> --- tests/d_xattr_sorting/script | 2 +- tests/f_badsymlinks2/mkimage.sh | 4 +++- tests/f_create_symlinks/script | 12 ++++++------ util/gen-sample-fs | 4 +++- 4 files changed, 13 insertions(+), 9 deletions(-)