tr portability fixes
Subsystems (2)
PATCH 1/16 tr portability fixes
In reply to a message not in the archiveSpecifying character ranges in tr differs between System V and POSIX. In System V, brackets are required (e.g., '[A-Z]'), whereas in POSIX they are not.
We can mostly get around this by just using the bracket form for both sets, as in:
tr '[A-Z] '[a-z]'
in which case POSIX interpets this as "'[' becomes '['", which is OK.
However, this doesn't work with multiple sequences, like:
# rot13 tr '[A-Z][a-z]' '[N-Z][A-M][n-z][a-m]'
where the POSIX version does not behave the same as the System V version. In this case, we must simply enumerate the sequence.
This patch fixes problematic uses of tr in git scripts and test scripts in one of three ways:
- if a single sequence, make sure it uses brackets
- if multiple sequences, enumerate
- if extra brackets (e.g., tr '[A]' 'a'), eliminate brackets
---
This was posted earlier, but mid-thread.
git-bisect.sh | 4 ++-- git-filter-branch.sh | 4 ++-- t/t4022-diff-rewrite.sh | 5 ++++- t/t7003-filter-branch.sh | 2 +- 4 files changed, 9 insertions(+), 6 deletions(-)
git-bisect.sh +2 -2
@@ -293,14 +293,14 @@ bisect_next() { bisect_next_check good skip=$(git for-each-ref --format='%(objectname)' \ - "refs/bisect/skip-*" | tr '[\012]' ' ') || exit + "refs/bisect/skip-*" | tr '\012' ' ') || exit BISECT_OPT='' test -n "$skip" && BISECT_OPT='--bisect-all' bad=$(git rev-parse --verify refs/bisect/bad) && good=$(git for-each-ref --format='^%(objectname)' \ - "refs/bisect/good-*" | tr '[\012]' ' ') && + "refs/bisect/good-*" | tr '\012' ' ') && eval="git rev-list --bisect-vars $BISECT_OPT $good $bad --" && eval="$eval $(cat "$GIT_DIR/BISECT_NAMES")" && eval=$(filter_skipped "$eval" "$skip") &&
git-filter-branch.sh +2 -2
@@ -58,8 +58,8 @@ eval "$functions" # "author" or "committer set_ident () { - lid="$(echo "$1" | tr "A-Z" "a-z")" - uid="$(echo "$1" | tr "a-z" "A-Z")" + lid="$(echo "$1" | tr "[A-Z]" "[a-z]")" + uid="$(echo "$1" | tr "[a-z]" "[A-Z]")" pick_id_script=' /^'$lid' /{ s/'\''/'\''\\'\'\''/g
t/t4022-diff-rewrite.sh +4 -1
@@ -8,7 +8,10 @@ test_expect_success setup ' cat ../../COPYING >test && git add test && - tr 'a-zA-Z' 'n-za-mN-ZA-M' <../../COPYING >test + tr \ + "abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ" \ + "nopqrstuvwxyzabcdefghijklmNOPQRSTUVWXYZABCDEFGHIJKLM" \ + <../../COPYING >test '
t/t7003-filter-branch.sh +1 -1
@@ -4,7 +4,7 @@ test_description='git-filter-branch' . ./test-lib.sh make_commit () { - lower=$(echo $1 | tr A-Z a-z) + lower=$(echo $1 | tr '[A-Z]' '[a-z]') echo $lower > $lower git add $lower test_tick
Signature · 1 line
1.5.4.4.543.g30fdd.dirty
Quoted 5 lines · Jeff King
We can mostly get around this by just using the bracket form for both sets, as in:
tr '[A-Z] '[a-z]'
Quoted 7 lines
in which case POSIX interpets this as "'[' becomes '['", which is OK.
However, this doesn't work with multiple sequences, like:
# rot13 tr '[A-Z][a-z]' '[N-Z][A-M][n-z][a-m]'
Not that it matters a lot, but I wonder whether
tr '[A-M][N-Z][a-m][n-z]' '[N-Z][A-M][n-z][a-m]'
would have done the trick.
-- Hannes
Quoted 10 lines · Johannes Sixt
Quoted 4 lines
However, this doesn't work with multiple sequences, like:
# rot13 tr '[A-Z][a-z]' '[N-Z][A-M][n-z][a-m]'Not that it matters a lot, but I wonder whether
tr '[A-M][N-Z][a-m][n-z]' '[N-Z][A-M][n-z][a-m]'would have done the trick.
For the record, it does on Solaris (and I really can't imagine it _not_ working anywhere else, but then I couldn't imagine my example not working, either. ;) ).
-Peff
PATCH 2/16 t0050: perl portability fix
In reply to a message not in the archiveOlder versions of perl (such as 5.005) don't understand -CO, nor do they understand the "U" pack specifier. Instead of using perl, let's just printf the binary bytes we are interested in.
--- t/t0050-filesystem.sh | 4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-)
t/t0050-filesystem.sh +2 -2
@@ -4,8 +4,8 @@ test_description='Various filesystem issues' . ./test-lib.sh -auml=`perl -CO -e 'print pack("U",0x00E4)'` -aumlcdiar=`perl -CO -e 'print pack("U",0x0061).pack("U",0x0308)'` +auml=`printf '\xc3\xa4'` +aumlcdiar=`printf '\x61\xcc\x88'` test_expect_success 'see if we expect ' '
Signature · 1 line
1.5.4.4.543.g30fdd.dirty
Re: [PATCH 02/16] t0050: perl portability fix
Quoted 5 lines · Jeff King
Older versions of perl (such as 5.005) don't understand -CO, nor do they understand the "U" pack specifier. Instead of using perl, let's just printf the binary bytes we are interested in.
Signed-off-by Jeff King <[email protected]>
Yay! We need this on Windows with MSYS's perl, too. Works great:
-- Hannes
PATCH 3/16 more tr portability test script fixes
In reply to a message not in the archiveDealing with NULs is not always safe with tr. On Solaris, incoming NULs are silently deleted by both the System V and UCB versions of tr. When converting to NULs, the System V version works fine, but the UCB version silently ignores the request to convert the character.
This patch changes all instances of tr using NULs to use "perl -pe 'y///'" instead.
---
This dramatically increases the amount of perl in the testsuite. However, it is a portable subset of perl, and it isn't the first perl in the testsuite.
t/diff-lib.sh | 4 ++-- t/t0020-crlf.sh | 2 +- t/t1300-repo-config.sh | 4 ++-- t/t3300-funny-names.sh | 6 +++--- t/t4020-diff-external.sh | 2 +- t/t4103-apply-binary.sh | 4 ++-- t/t4116-apply-reverse.sh | 4 ++-- t/t4200-rerere.sh | 2 +- t/t5300-pack-object.sh | 2 +- test-sha1.sh | 4 ++-- 10 files changed, 17 insertions(+), 17 deletions(-)
t/diff-lib.sh +2 -2
@@ -21,8 +21,8 @@ compare_diff_raw_z () { # Also we do not check SHA1 hash generation in this test, which # is a job for t0000-basic.sh - tr '\000' '\012' <"$1" | sed -e "$sanitize_diff_raw_z" >.tmp-1 - tr '\000' '\012' <"$2" | sed -e "$sanitize_diff_raw_z" >.tmp-2 + perl -pe 'y/\000/\012/' <"$1" | sed -e "$sanitize_diff_raw_z" >.tmp-1 + perl -pe 'y/\000/\012/' <"$2" | sed -e "$sanitize_diff_raw_z" >.tmp-2 git diff .tmp-1 .tmp-2 && rm -f .tmp-1 .tmp-2 }
t/t0020-crlf.sh +1 -1
@@ -5,7 +5,7 @@ test_description='CRLF conversion' . ./test-lib.sh q_to_nul () { - tr Q '\000' + perl -pe 'y/Q/\000/' } q_to_cr () {
t/t1300-repo-config.sh +2 -2
@@ -657,12 +657,12 @@ Qsection.sub=section.val4 Qsection.sub=section.val5Q EOF -git config --null --list | tr '\000' 'Q' > result +git config --null --list | perl -pe 'y/\000/Q/' > result echo >>result test_expect_success '--null --list' 'cmp result expect' -git config --null --get-regexp 'val[0-9]' | tr '\000' 'Q' > result +git config --null --get-regexp 'val[0-9]' | perl -pe 'y/\000/Q/' > result echo >>result test_expect_success '--null --get-regexp' 'cmp result expect'
t/t3300-funny-names.sh +3 -3
@@ -54,7 +54,7 @@ echo 'just space no-funny tabs ," (dq) and spaces' >expected test_expect_success 'git ls-files -z with-funny' \ - 'git ls-files -z | tr \\000 \\012 >current && + 'git ls-files -z | perl -pe y/\\000/\\012/ >current && git diff expected current' t1=`git write-tree`
@@ -83,11 +83,11 @@ test_expect_success 'git diff-tree with-funny' \ echo 'A tabs ," (dq) and spaces' >expected test_expect_success 'git diff-index -z with-funny' \ - 'git diff-index -z --name-status $t0 | tr \\000 \\012 >current && + 'git diff-index -z --name-status $t0 | perl -pe y/\\000/\\012/ >current && git diff expected current' test_expect_success 'git diff-tree -z with-funny' \ - 'git diff-tree -z --name-status $t0 $t1 | tr \\000 \\012 >current && + 'git diff-tree -z --name-status $t0 $t1 | perl -pe y/\\000/\\012/ >current && git diff expected current' cat > expected <<\EOF
t/t4020-diff-external.sh +1 -1
@@ -99,7 +99,7 @@ test_expect_success 'no diff with -diff' ' git diff | grep Binary ' -echo NULZbetweenZwords | tr Z '\000' > file +echo NULZbetweenZwords | perl -pe 'y/Z/\000/' > file test_expect_success 'force diff with "diff"' ' echo >.gitattributes "file diff" &&
t/t4103-apply-binary.sh +2 -2
@@ -24,10 +24,10 @@ git update-index --add --remove file1 file2 file4 git-commit -m 'Initial Version' 2>/dev/null git-checkout -b binary -tr 'x' '\000' <file1 >file3 +perl -pe 'y/x/\000/' <file1 >file3 cat file3 >file4 git add file2 -tr '\000' 'v' <file3 >file1 +perl -pe 'y/\000/v/' <file3 >file1 rm -f file2 git update-index --add --remove file1 file2 file3 file4 git-commit -m 'Second Version'
t/t4116-apply-reverse.sh +2 -2
@@ -12,14 +12,14 @@ test_description='git apply in reverse test_expect_success setup ' for i in a b c d e f g h i j k l m n; do echo $i; done >file1 && - tr "ijk" '\''\000\001\002'\'' <file1 >file2 && + perl -pe "y/ijk/\\000\\001\\002/" <file1 >file2 && git add file1 file2 && git commit -m initial && git tag initial && for i in a b c g h i J K L m o n p q; do echo $i; done >file1 && - tr "mon" '\''\000\001\002'\'' <file1 >file2 && + perl -pe "y/mon/\\000\\001\\002/" <file1 >file2 && git commit -a -m second && git tag second &&
t/t4200-rerere.sh +1 -1
@@ -129,7 +129,7 @@ test_expect_success 'rerere kicked in' "! grep ======= a1" test_expect_success 'rerere prefers first change' 'git diff a1 expect' rm $rr/postimage -echo "$sha1 a1" | tr '\012' '\000' > .git/rr-cache/MERGE_RR +echo "$sha1 a1" | perl -pe 'y/\012/\000/' > .git/rr-cache/MERGE_RR test_expect_success 'rerere clear' 'git rerere clear'
t/t5300-pack-object.sh +1 -1
@@ -15,7 +15,7 @@ test_expect_success \ 'rm -f .git/index* for i in a b c do - dd if=/dev/zero bs=4k count=1 | tr "\\000" $i >$i && + dd if=/dev/zero bs=4k count=1 | perl -pe "y/\\000/$i/" >$i && git update-index --add $i || return 1 done && cat c >d && echo foo >>d && git update-index --add d &&
test-sha1.sh +2 -2
@@ -10,7 +10,7 @@ do { test -z "$pfx" || echo "$pfx" dd if=/dev/zero bs=1048576 count=$cnt 2>/dev/null | - tr '\000' 'g' + perl -pe 'y/\000/g/' } | ./test-sha1 $cnt ` if test "$expect" = "$actual"
@@ -55,7 +55,7 @@ do { test -z "$pfx" || echo "$pfx" dd if=/dev/zero bs=1048576 count=$cnt 2>/dev/null | - tr '\000' 'g' + perl -pe 'y/\000/g/' } | sha1sum | sed -e 's/ .*//' `
Signature · 1 line
1.5.4.4.543.g30fdd.dirty
Re: [PATCH 03/16] more tr portability test script fixes
t/diff-lib.sh compare_diff_raw_z () { · line 25
Quoted 10 lines · Jeff King
t/diff-lib.sh +2
-2@@ -21,8 +21,8 @@ compare_diff_raw_z () { # Also we do not check SHA1 hash generation in this test, which # is a job for t0000-basic.sh - tr '\000' '\012' <"$1" | sed -e "$sanitize_diff_raw_z" >.tmp-1 - tr '\000' '\012' <"$2" | sed -e "$sanitize_diff_raw_z" >.tmp-2 + perl -pe 'y/\000/\012/' <"$1" | sed -e "$sanitize_diff_raw_z" >.tmp-1 + perl -pe 'y/\000/\012/' <"$2" | sed -e "$sanitize_diff_raw_z" >.tmp-2
It might make sense performance-wise to integrate the job of the sed call into the perl call here. Haven't tested it, though.
Gruesse,
Signature · 2 lines
Frank Lichtenheld <[email protected]> www: http://www.djpig.de/
Re: [PATCH 03/16] more tr portability test script fixes
Quoted 7 lines · Frank Lichtenheld
t/diff-lib.sh compare_diff_raw_z () { · line 25
Quoted 4 lines
t/diff-lib.sh +2
-2- tr '\000' '\012' <"$1" | sed -e "$sanitize_diff_raw_z" >.tmp-1 - tr '\000' '\012' <"$2" | sed -e "$sanitize_diff_raw_z" >.tmp-2 + perl -pe 'y/\000/\012/' <"$1" | sed -e "$sanitize_diff_raw_z" >.tmp-1 + perl -pe 'y/\000/\012/' <"$2" | sed -e "$sanitize_diff_raw_z" >.tmp-2It might make sense performance-wise to integrate the job of the sed call into the perl call here. Haven't tested it, though.
Probably. My changes were pretty mechanical, and I tried to keep them as small as possible.
I'm not sure we care too much about an extra fork in a test script for performance, though it would probably end up more readable.
-Peff
PATCH 4/16 grep portability fix: don't use "-e" or "-q"
In reply to a message not in the archiveSystem V versions of grep (such as Solaris /usr/bin/grep) don't understand either of these options. git's usage of "grep -e pattern" fell into one of two categories:
1. equivalent to "grep pattern". -e is only useful here if the pattern begins with a "-", but all of the patterns are hardcoded and do not begin with a dash.
2. stripping comments and blank lines with
grep -v -e "^$" -e "^#"
We can fortunately do this in the affirmative as
grep '^[^#]'Uses of "-q" can be replaced with redirection to /dev/null. In many tests, however, "grep -q" is used as "if this string is in the expected output, we are OK". In this case, it is fine to just remove the "-q" entirely; it simply makes the "verbose" mode of the test slightly more verbose.
---
One might disagree with my "grep without -q actually shows useful verbose output in tests" statement. In that case, we can >/dev/null all of those instances.
git-rebase--interactive.sh | 6 +++--- git-submodule.sh | 6 +++--- t/t0030-stripspace.sh | 34 +++++++++++++++++----------------- t/t3404-rebase-interactive.sh | 3 +-- t/t3800-mktag.sh | 2 +- t/t5400-send-pack.sh | 2 +- t/t7502-status.sh | 2 +- t/t7600-merge.sh | 4 ++-- t/t9400-git-cvsserver-server.sh | 26 +++++++++++++------------- 9 files changed, 42 insertions(+), 43 deletions(-)
git-rebase--interactive.sh +3 -3
@@ -78,8 +78,8 @@ mark_action_done () { sed -e 1q < "$TODO" >> "$DONE" sed -e 1d < "$TODO" >> "$TODO".new mv -f "$TODO".new "$TODO" - count=$(($(grep -ve '^$' -e '^#' < "$DONE" | wc -l))) - total=$(($count+$(grep -ve '^$' -e '^#' < "$TODO" | wc -l))) + count=$(grep -c '^[^#]' < "$DONE") + total=$(($count+$(grep -c '^[^#]' < "$TODO"))) if test "$last_count" != "$count" then last_count=$count
@@ -110,7 +110,7 @@ die_abort () { } has_action () { - grep -vqe '^$' -e '^#' "$1" + grep '^[^#]' "$1" >/dev/null } pick_one () {
git-submodule.sh +3 -3
@@ -230,7 +230,7 @@ cmd_init() shift done - git ls-files --stage -- "$@" | grep -e '^160000 ' | + git ls-files --stage -- "$@" | grep '^160000 ' | while read mode sha1 stage path do # Skip already registered paths
@@ -284,7 +284,7 @@ cmd_update() shift done - git ls-files --stage -- "$@" | grep -e '^160000 ' | + git ls-files --stage -- "$@" | grep '^160000 ' | while read mode sha1 stage path do name=$(module_name "$path") || exit
@@ -367,7 +367,7 @@ cmd_status() shift done - git ls-files --stage -- "$@" | grep -e '^160000 ' | + git ls-files --stage -- "$@" | grep '^160000 ' | while read mode sha1 stage path do name=$(module_name "$path") || exit
t/t0030-stripspace.sh +17 -17
@@ -245,12 +245,12 @@ test_expect_success \ test_expect_success \ 'text plus spaces without newline at end should not show spaces' ' - ! (printf "$ttt$sss" | git stripspace | grep -q " ") && - ! (printf "$ttt$ttt$sss" | git stripspace | grep -q " ") && - ! (printf "$ttt$ttt$ttt$sss" | git stripspace | grep -q " ") && - ! (printf "$ttt$sss$sss" | git stripspace | grep -q " ") && - ! (printf "$ttt$ttt$sss$sss" | git stripspace | grep -q " ") && - ! (printf "$ttt$sss$sss$sss" | git stripspace | grep -q " ") + ! (printf "$ttt$sss" | git stripspace | grep " " >/dev/null) && + ! (printf "$ttt$ttt$sss" | git stripspace | grep " " >/dev/null) && + ! (printf "$ttt$ttt$ttt$sss" | git stripspace | grep " " >/dev/null) && + ! (printf "$ttt$sss$sss" | git stripspace | grep " " >/dev/null) && + ! (printf "$ttt$ttt$sss$sss" | git stripspace | grep " " >/dev/null) && + ! (printf "$ttt$sss$sss$sss" | git stripspace | grep " " >/dev/null) ' test_expect_success \
@@ -282,12 +282,12 @@ test_expect_success \ test_expect_success \ 'text plus spaces at end should not show spaces' ' - ! (echo "$ttt$sss" | git stripspace | grep -q " ") && - ! (echo "$ttt$ttt$sss" | git stripspace | grep -q " ") && - ! (echo "$ttt$ttt$ttt$sss" | git stripspace | grep -q " ") && - ! (echo "$ttt$sss$sss" | git stripspace | grep -q " ") && - ! (echo "$ttt$ttt$sss$sss" | git stripspace | grep -q " ") && - ! (echo "$ttt$sss$sss$sss" | git stripspace | grep -q " ") + ! (echo "$ttt$sss" | git stripspace | grep " " >/dev/null) && + ! (echo "$ttt$ttt$sss" | git stripspace | grep " " >/dev/null) && + ! (echo "$ttt$ttt$ttt$sss" | git stripspace | grep " " >/dev/null) && + ! (echo "$ttt$sss$sss" | git stripspace | grep " " >/dev/null) && + ! (echo "$ttt$ttt$sss$sss" | git stripspace | grep " " >/dev/null) && + ! (echo "$ttt$sss$sss$sss" | git stripspace | grep " " >/dev/null) ' test_expect_success \
@@ -341,11 +341,11 @@ test_expect_success \ test_expect_success \ 'spaces without newline at end should not show spaces' ' - ! (printf "" | git stripspace | grep -q " ") && - ! (printf "$sss" | git stripspace | grep -q " ") && - ! (printf "$sss$sss" | git stripspace | grep -q " ") && - ! (printf "$sss$sss$sss" | git stripspace | grep -q " ") && - ! (printf "$sss$sss$sss$sss" | git stripspace | grep -q " ") + ! (printf "" | git stripspace | grep " " >/dev/null) && + ! (printf "$sss" | git stripspace | grep " " >/dev/null) && + ! (printf "$sss$sss" | git stripspace | grep " " >/dev/null) && + ! (printf "$sss$sss$sss" | git stripspace | grep " " >/dev/null) && + ! (printf "$sss$sss$sss$sss" | git stripspace | grep " " >/dev/null) ' test_expect_success \
t/t3404-rebase-interactive.sh +1 -2
@@ -149,8 +149,7 @@ test_expect_success 'stop on conflicting pick' ' diff -u expect .git/.dotest-merge/patch && diff -u expect2 file1 && test 4 = $(grep -v "^#" < .git/.dotest-merge/done | wc -l) && - test 0 = $(grep -ve "^#" -e "^$" < .git/.dotest-merge/git-rebase-todo | - wc -l) + test 0 = $(grep -c "^[^#]" < .git/.dotest-merge/git-rebase-todo) ' test_expect_success 'abort' '
t/t3800-mktag.sh +1 -1
@@ -15,7 +15,7 @@ check_verify_failure () { expect="$2" test_expect_success "$1" ' ( ! git-mktag <tag.sig 2>message ) && - grep -q "$expect" message + grep "$expect" message ' }
t/t5400-send-pack.sh +1 -1
@@ -120,7 +120,7 @@ test_expect_success \ cd .. && git-clone parent child && cd child && git-push --all && cd ../parent && - git-branch -a >branches && ! grep -q origin/master branches + git-branch -a >branches && ! grep origin/master branches ' rewound_push_setup() {
t/t7502-status.sh +1 -1
@@ -33,7 +33,7 @@ test_expect_success 'setup' ' test_expect_success 'status (1)' ' - grep -e "use \"git rm --cached <file>\.\.\.\" to unstage" output + grep "use \"git rm --cached <file>\.\.\.\" to unstage" output '
t/t7600-merge.sh +2 -2
@@ -371,7 +371,7 @@ test_expect_success 'override config option -n' ' git merge --summary c2 >diffstat.txt && verify_merge file result.1-5 msg.1-5 && verify_parents $c1 $c2 && - if ! grep -e "^ file | *2 +-$" diffstat.txt + if ! grep "^ file | *2 +-$" diffstat.txt then echo "[OOPS] diffstat was not generated" fi
@@ -386,7 +386,7 @@ test_expect_success 'override config option --summary' ' git merge -n c2 >diffstat.txt && verify_merge file result.1-5 msg.1-5 && verify_parents $c1 $c2 && - if grep -e "^ file | *2 +-$" diffstat.txt + if grep "^ file | *2 +-$" diffstat.txt then echo "[OOPS] diffstat was generated" false
t/t9400-git-cvsserver-server.sh +13 -13
@@ -94,7 +94,7 @@ EOF test_expect_success 'pserver authentication' \ 'cat request-anonymous | git-cvsserver pserver >log 2>&1 && - tail -n1 log | grep -q "^I LOVE YOU$"' + tail -n1 log | grep "^I LOVE YOU$"' test_expect_success 'pserver authentication failure (non-anonymous user)' \ 'if cat request-git | git-cvsserver pserver >log 2>&1
@@ -103,11 +103,11 @@ test_expect_success 'pserver authentication failure (non-anonymous user)' \ else true fi && - tail -n1 log | grep -q "^I HATE YOU$"' + tail -n1 log | grep "^I HATE YOU$"' test_expect_success 'pserver authentication (login)' \ 'cat login-anonymous | git-cvsserver pserver >log 2>&1 && - tail -n1 log | grep -q "^I LOVE YOU$"' + tail -n1 log | grep "^I LOVE YOU$"' test_expect_success 'pserver authentication failure (login/non-anonymous user)' \ 'if cat login-git | git-cvsserver pserver >log 2>&1
@@ -116,7 +116,7 @@ test_expect_success 'pserver authentication failure (login/non-anonymous user)' else true fi && - tail -n1 log | grep -q "^I HATE YOU$"' + tail -n1 log | grep "^I HATE YOU$"' # misuse pserver authentication for testing of req_Root
@@ -146,15 +146,15 @@ test_expect_success 'req_Root failure (relative pathname)' \ else true fi && - tail log | grep -q "^error 1 Root must be an absolute pathname$"' + tail log | grep "^error 1 Root must be an absolute pathname$"' test_expect_success 'req_Root failure (conflicting roots)' \ 'cat request-conflict | git-cvsserver pserver >log 2>&1 && - tail log | grep -q "^error 1 Conflicting roots specified$"' + tail log | grep "^error 1 Conflicting roots specified$"' test_expect_success 'req_Root (strict paths)' \ 'cat request-anonymous | git-cvsserver --strict-paths pserver $SERVERDIR >log 2>&1 && - tail -n1 log | grep -q "^I LOVE YOU$"' + tail -n1 log | grep "^I LOVE YOU$"' test_expect_success 'req_Root failure (strict-paths)' ' ! cat request-anonymous |
@@ -163,7 +163,7 @@ test_expect_success 'req_Root failure (strict-paths)' ' test_expect_success 'req_Root (w/o strict-paths)' \ 'cat request-anonymous | git-cvsserver pserver $WORKDIR/ >log 2>&1 && - tail -n1 log | grep -q "^I LOVE YOU$"' + tail -n1 log | grep "^I LOVE YOU$"' test_expect_success 'req_Root failure (w/o strict-paths)' ' ! cat request-anonymous |
@@ -181,7 +181,7 @@ EOF test_expect_success 'req_Root (base-path)' \ 'cat request-base | git-cvsserver --strict-paths --base-path $WORKDIR/ pserver $SERVERDIR >log 2>&1 && - tail -n1 log | grep -q "^I LOVE YOU$"' + tail -n1 log | grep "^I LOVE YOU$"' test_expect_success 'req_Root failure (base-path)' ' ! cat request-anonymous |
@@ -192,14 +192,14 @@ GIT_DIR="$SERVERDIR" git config --bool gitcvs.enabled false || exit 1 test_expect_success 'req_Root (export-all)' \ 'cat request-anonymous | git-cvsserver --export-all pserver $WORKDIR >log 2>&1 && - tail -n1 log | grep -q "^I LOVE YOU$"' + tail -n1 log | grep "^I LOVE YOU$"' test_expect_success 'req_Root failure (export-all w/o whitelist)' \ '! (cat request-anonymous | git-cvsserver --export-all pserver >log 2>&1 || false)' test_expect_success 'req_Root (everything together)' \ 'cat request-base | git-cvsserver --export-all --strict-paths --base-path $WORKDIR/ pserver $SERVERDIR >log 2>&1 && - tail -n1 log | grep -q "^I LOVE YOU$"' + tail -n1 log | grep "^I LOVE YOU$"' GIT_DIR="$SERVERDIR" git config --bool gitcvs.enabled true || exit 1
@@ -216,7 +216,7 @@ test_expect_success 'gitcvs.enabled = false' \ else true fi && - cat cvs.log | grep -q "GITCVS emulation disabled" && + grep "GITCVS emulation disabled" cvs.log && test ! -d cvswork2' rm -fr cvswork2
@@ -237,7 +237,7 @@ test_expect_success 'gitcvs.ext.enabled = false' \ else true fi && - cat cvs.log | grep -q "GITCVS emulation disabled" && + grep "GITCVS emulation disabled" cvs.log && test ! -d cvswork2' rm -fr cvswork2
Signature · 1 line
1.5.4.4.543.g30fdd.dirty
Re: [PATCH 04/16] grep portability fix: don't use "-e" or "-q"
Quoted 2 lines · Jeff King
System V versions of grep (such as Solaris
/usr/bin/grep) don't understand either of these options. git's usage of
It might be fair for other System V people to qualify the above statement with "Historic System V" (personally I felt that Solaris without xpg4 was unusable, and I wish you luck tackling it).
More seriously, you would need to disable "when grepping work-tree, running the native grep is faster and just as capable" optimization in builtin-grep.c::grep_cache(). Treat these problematic System V platforms as if they are not __unix__.
I am surprised that you did not have issues with "tail -n$number". Historic way to spell it was "tail -$number" wasn't it?
Re: [PATCH 04/16] grep portability fix: don't use "-e" or "-q"
Quoted 9 lines · Junio C Hamano
Quoted 2 lines · Jeff King
System V versions of grep (such as Solaris
/usr/bin/grep) don't understand either of these options. git's usage ofIt might be fair for other System V people to qualify the above statement with "Historic System V" (personally I felt that Solaris without xpg4 was unusable, and I wish you luck tackling it).
Quoted 7 lines
More seriously, you would need to disable "when grepping work-tree, running the native grep is faster and just as capable" optimization in builtin-grep.c::grep_cache(). Treat these problematic System V platforms as if they are not
__unix__.I am surprised that you did not have issues with "tail -n$number". Historic way to spell it was "tail -$number" wasn't it?
Heh, I notice you had to deal with these issues in the later patches ;-)
PATCH 5/16 remove use of "tail -n 1" and "tail -1"
In reply to a message not in the archiveThe "-n" syntax is not supported by System V versions of tail (which prefer "tail -1"). Unfortunately "tail -1" is not actually POSIX. We had some of both forms in our scripts.
Since neither form works everywhere, this patch replaces both with the equivalent sed invocation:
sed -ne '$p'
---
Actually, "tail -1" _does_ seem to be accepted everywhere, even though it isn't POSIX. I remember the GNU utils complaining about it (but maybe just with POSIXLY_CORRECT?) a while back, but that seems to have been reverted. But certainly this sed invocation should work everywhere.
git-am.sh | 2 +- git-rebase--interactive.sh | 2 +- t/t3404-rebase-interactive.sh | 4 ++-- t/t5302-pack-index.sh | 4 ++-- t/t6030-bisect-porcelain.sh | 6 +++--- t/t7600-merge.sh | 2 +- t/t9400-git-cvsserver-server.sh | 20 ++++++++++---------- 7 files changed, 20 insertions(+), 20 deletions(-)
git-am.sh +1 -1
@@ -357,7 +357,7 @@ do LAST_SIGNED_OFF_BY=` sed -ne '/^Signed-off-by: /p' \ "$dotest/msg-clean" | - tail -n 1 + sed -ne '$p' ` ADD_SIGNOFF=` test "$LAST_SIGNED_OFF_BY" = "$SIGNOFF" || {
git-rebase--interactive.sh +1 -1
@@ -218,7 +218,7 @@ nth_string () { make_squash_message () { if test -f "$SQUASH_MSG"; then COUNT=$(($(sed -n "s/^# This is [^0-9]*\([1-9][0-9]*\).*/\1/p" \ - < "$SQUASH_MSG" | tail -n 1)+1)) + < "$SQUASH_MSG" | sed -ne '$p')+1)) echo "# This is a combination of $COUNT commits." sed -e 1d -e '2,/^./{ /^$/d
t/t3404-rebase-interactive.sh +2 -2
@@ -122,8 +122,8 @@ test_expect_success 'reflog for the branch shows state before rebase' ' test_expect_success 'exchange two commits' ' FAKE_LINES="2 1" git rebase -i HEAD~2 && - test H = $(git cat-file commit HEAD^ | tail -n 1) && - test G = $(git cat-file commit HEAD | tail -n 1) + test H = $(git cat-file commit HEAD^ | sed -ne \$p) && + test G = $(git cat-file commit HEAD | sed -ne \$p) ' cat > expect << EOF
t/t5302-pack-index.sh +2 -2
@@ -103,7 +103,7 @@ test_expect_success \ test_expect_success \ '[index v1] 2) create a stealth corruption in a delta base reference' \ '# this test assumes a delta smaller than 16 bytes at the end of the pack - git show-index <1.idx | sort -n | tail -n 1 | ( + git show-index <1.idx | sort -n | sed -ne \$p | ( read delta_offs delta_sha1 && git cat-file blob "$delta_sha1" > blob_1 && chmod +w ".git/objects/pack/pack-${pack1}.pack" &&
@@ -141,7 +141,7 @@ test_expect_success \ test_expect_success \ '[index v2] 2) create a stealth corruption in a delta base reference' \ '# this test assumes a delta smaller than 16 bytes at the end of the pack - git show-index <1.idx | sort -n | tail -n 1 | ( + git show-index <1.idx | sort -n | sed -ne \$p | ( read delta_offs delta_sha1 delta_crc && git cat-file blob "$delta_sha1" > blob_3 && chmod +w ".git/objects/pack/pack-${pack1}.pack" &&
t/t6030-bisect-porcelain.sh +3 -3
@@ -219,7 +219,7 @@ test_expect_success 'bisect run & skip: cannot tell between 2' ' add_line_into_file "6: Yet a line." hello && HASH6=$(git rev-parse --verify HEAD) && echo "#"\!"/bin/sh" > test_script.sh && - echo "tail -1 hello | grep Ciao > /dev/null && exit 125" >> test_script.sh && + echo "sed -ne \\\$p hello | grep Ciao > /dev/null && exit 125" >> test_script.sh && echo "grep line hello > /dev/null" >> test_script.sh && echo "test \$? -ne 0" >> test_script.sh && chmod +x test_script.sh &&
@@ -244,8 +244,8 @@ test_expect_success 'bisect run & skip: find first bad' ' add_line_into_file "7: Should be the last line." hello && HASH7=$(git rev-parse --verify HEAD) && echo "#"\!"/bin/sh" > test_script.sh && - echo "tail -1 hello | grep Ciao > /dev/null && exit 125" >> test_script.sh && - echo "tail -1 hello | grep day > /dev/null && exit 125" >> test_script.sh && + echo "sed -ne \\\$p hello | grep Ciao > /dev/null && exit 125" >> test_script.sh && + echo "sed -ne \\\$p hello | grep day > /dev/null && exit 125" >> test_script.sh && echo "grep Yet hello > /dev/null" >> test_script.sh && echo "test \$? -ne 0" >> test_script.sh && chmod +x test_script.sh &&
t/t7600-merge.sh +1 -1
@@ -165,7 +165,7 @@ verify_mergeheads() { fi && while test $# -gt 0 do - head=$(head -n $i .git/MERGE_HEAD | tail -n 1) + head=$(head -n $i .git/MERGE_HEAD | sed -ne \$p) if test "$1" != "$head" then echo "[OOPS] MERGE_HEAD $i != $1"
t/t9400-git-cvsserver-server.sh +10 -10
@@ -54,7 +54,7 @@ test_expect_success 'setup' ' test_expect_success 'basic checkout' \ 'GIT_CONFIG="$git_config" cvs -Q co -d cvswork master && test "$(echo $(grep -v ^D cvswork/CVS/Entries|cut -d/ -f2,3,5 | head -n 1))" = "empty/1.1/" - test "$(echo $(grep -v ^D cvswork/CVS/Entries|cut -d/ -f2,3,5 | tail -n 1))" = "secondrootfile/1.1/"' + test "$(echo $(grep -v ^D cvswork/CVS/Entries|cut -d/ -f2,3,5 | sed -ne \$p))" = "secondrootfile/1.1/"' #------------------------ # PSERVER AUTHENTICATION
@@ -94,7 +94,7 @@ EOF test_expect_success 'pserver authentication' \ 'cat request-anonymous | git-cvsserver pserver >log 2>&1 && - tail -n1 log | grep "^I LOVE YOU$"' + sed -ne \$p log | grep "^I LOVE YOU$"' test_expect_success 'pserver authentication failure (non-anonymous user)' \ 'if cat request-git | git-cvsserver pserver >log 2>&1
@@ -103,11 +103,11 @@ test_expect_success 'pserver authentication failure (non-anonymous user)' \ else true fi && - tail -n1 log | grep "^I HATE YOU$"' + sed -ne \$p log | grep "^I HATE YOU$"' test_expect_success 'pserver authentication (login)' \ 'cat login-anonymous | git-cvsserver pserver >log 2>&1 && - tail -n1 log | grep "^I LOVE YOU$"' + sed -ne \$p log | grep "^I LOVE YOU$"' test_expect_success 'pserver authentication failure (login/non-anonymous user)' \ 'if cat login-git | git-cvsserver pserver >log 2>&1
@@ -116,7 +116,7 @@ test_expect_success 'pserver authentication failure (login/non-anonymous user)' else true fi && - tail -n1 log | grep "^I HATE YOU$"' + sed -ne \$p log | grep "^I HATE YOU$"' # misuse pserver authentication for testing of req_Root
@@ -154,7 +154,7 @@ test_expect_success 'req_Root failure (conflicting roots)' \ test_expect_success 'req_Root (strict paths)' \ 'cat request-anonymous | git-cvsserver --strict-paths pserver $SERVERDIR >log 2>&1 && - tail -n1 log | grep "^I LOVE YOU$"' + sed -ne \$p log | grep "^I LOVE YOU$"' test_expect_success 'req_Root failure (strict-paths)' ' ! cat request-anonymous |
@@ -163,7 +163,7 @@ test_expect_success 'req_Root failure (strict-paths)' ' test_expect_success 'req_Root (w/o strict-paths)' \ 'cat request-anonymous | git-cvsserver pserver $WORKDIR/ >log 2>&1 && - tail -n1 log | grep "^I LOVE YOU$"' + sed -ne \$p log | grep "^I LOVE YOU$"' test_expect_success 'req_Root failure (w/o strict-paths)' ' ! cat request-anonymous |
@@ -181,7 +181,7 @@ EOF test_expect_success 'req_Root (base-path)' \ 'cat request-base | git-cvsserver --strict-paths --base-path $WORKDIR/ pserver $SERVERDIR >log 2>&1 && - tail -n1 log | grep "^I LOVE YOU$"' + sed -ne \$p log | grep "^I LOVE YOU$"' test_expect_success 'req_Root failure (base-path)' ' ! cat request-anonymous |
@@ -192,14 +192,14 @@ GIT_DIR="$SERVERDIR" git config --bool gitcvs.enabled false || exit 1 test_expect_success 'req_Root (export-all)' \ 'cat request-anonymous | git-cvsserver --export-all pserver $WORKDIR >log 2>&1 && - tail -n1 log | grep "^I LOVE YOU$"' + sed -ne \$p log | grep "^I LOVE YOU$"' test_expect_success 'req_Root failure (export-all w/o whitelist)' \ '! (cat request-anonymous | git-cvsserver --export-all pserver >log 2>&1 || false)' test_expect_success 'req_Root (everything together)' \ 'cat request-base | git-cvsserver --export-all --strict-paths --base-path $WORKDIR/ pserver $SERVERDIR >log 2>&1 && - tail -n1 log | grep "^I LOVE YOU$"' + sed -ne \$p log | grep "^I LOVE YOU$"' GIT_DIR="$SERVERDIR" git config --bool gitcvs.enabled true || exit 1
Signature · 1 line
1.5.4.4.543.g30fdd.dirty
PATCH 6/16 add test_cmp function for test scripts
In reply to a message not in the archiveMany scripts compare actual and expected output using "diff -u". This is nicer than "cmp" because the output shows how the two differ. However, not all versions of diff understand -u, leading to unnecessary test failure.
This adds a test_cmp function to the test scripts and switches all "diff -u" invocations to use it. The function uses the contents of "$GIT_TEST_CMP" to compare its arguments; the default is "diff -u".
On systems with a less-capable diff, you can do:
GIT_TEST_CMP=cmp make test
---
The replacements were done mechanically with:
perl -pi -e 's/(?<!git.)diff -u/test_cmp/' t/*.sh
Perhaps this should be a command-line option instead of an environment variable? Maybe it should be auto-detected or set up in the Makefile?
t/t0003-attributes.sh | 2 +- t/t0022-crlf-rename.sh | 2 +- t/t1005-read-tree-reset.sh | 2 +- t/t2200-add-update.sh | 2 +- t/t3001-ls-files-others-exclude.sh | 2 +- t/t3050-subprojects-fetch.sh | 4 ++-- t/t3060-ls-files-with-tree.sh | 2 +- t/t3201-branch-contains.sh | 6 +++--- t/t3404-rebase-interactive.sh | 4 ++-- t/t3405-rebase-malformed.sh | 4 ++-- t/t3406-rebase-message.sh | 2 +- t/t3701-add-interactive.sh | 4 ++-- t/t3902-quoted.sh | 16 ++++++++-------- t/t3903-stash.sh | 2 +- t/t4023-diff-rename-typechange.sh | 6 +++--- t/t4024-diff-optimize-common.sh | 2 +- t/t4025-hunk-header.sh | 2 +- t/t4027-diff-submodule.sh | 6 +++--- t/t4105-apply-fuzz.sh | 2 +- t/t4125-apply-ws-fuzz.sh | 8 ++++---- t/t4150-am-subdir.sh | 10 +++++----- t/t4201-shortlog.sh | 2 +- t/t5505-remote.sh | 6 +++--- t/t5510-fetch.sh | 2 +- t/t5512-ls-remote.sh | 8 ++++---- t/t6004-rev-list-path-optim.sh | 2 +- t/t6009-rev-list-parent.sh | 2 +- t/t6027-merge-binary.sh | 4 ++-- t/t6029-merge-subtree.sh | 2 +- t/t7010-setup.sh | 18 +++++++++--------- t/t7201-co.sh | 18 +++++++++--------- t/t7501-commit.sh | 14 +++++++------- t/t7502-commit.sh | 14 +++++++------- t/t7502-status.sh | 2 +- t/t7600-merge.sh | 2 +- t/t8003-blame.sh | 4 ++-- t/t9001-send-email.sh | 2 +- t/t9116-git-svn-log.sh | 24 ++++++++++++------------ t/t9200-git-cvsexportcommit.sh | 14 +++++++------- t/test-lib.sh | 18 ++++++++++++++++++ 40 files changed, 133 insertions(+), 115 deletions(-)
t/t0003-attributes.sh +1 -1
@@ -11,7 +11,7 @@ attr_check () { git check-attr test -- "$path" >actual && echo "$path: test: $2" >expect && - diff -u expect actual + test_cmp expect actual }
t/t0022-crlf-rename.sh +1 -1
@@ -26,7 +26,7 @@ test_expect_success 'diff -M' ' git diff-tree -M -r --name-status HEAD^ HEAD | sed -e "s/R[0-9]*/RNUM/" >actual && echo "RNUM sample elpmas" >expect && - diff -u expect actual + test_cmp expect actual '
t/t1005-read-tree-reset.sh +1 -1
@@ -24,7 +24,7 @@ test_expect_success 'setup' ' test_expect_success 'reset should work' ' git read-tree -u --reset HEAD^ && git ls-files >actual && - diff -u expect actual + test_cmp expect actual ' test_done
t/t2200-add-update.sh +1 -1
@@ -62,7 +62,7 @@ test_expect_success 'cache tree has not been corrupted' ' sed -e "s/ 0 / /" >expect && git ls-tree -r $(git write-tree) | sed -e "s/ blob / /" >current && - diff -u expect current + test_cmp expect current '
t/t3001-ls-files-others-exclude.sh +1 -1
@@ -97,7 +97,7 @@ cat > expect << EOF EOF test_expect_success 'git-status honours core.excludesfile' \ - 'diff -u expect output' + 'test_cmp expect output' test_expect_success 'trailing slash in exclude allows directory match(1)' '
t/t3050-subprojects-fetch.sh +2 -2
@@ -26,7 +26,7 @@ test_expect_success clone ' cd cloned && (git rev-parse HEAD; git ls-files -s) >../actual ) && - diff -u expected actual + test_cmp expected actual ' test_expect_success advance '
@@ -46,7 +46,7 @@ test_expect_success fetch ' git pull && (git rev-parse HEAD; git ls-files -s) >../actual ) && - diff -u expected actual + test_cmp expected actual ' test_done
t/t3060-ls-files-with-tree.sh +1 -1
@@ -66,6 +66,6 @@ test_expect_success 'git -ls-files --with-tree should succeed from subdir' ' cd .. test_expect_success \ 'git -ls-files --with-tree should add entries from named tree.' \ - 'diff -u expected output' + 'test_cmp expected output' test_done
t/t3201-branch-contains.sh +3 -3
@@ -31,7 +31,7 @@ test_expect_success 'branch --contains=master' ' { echo " master" && echo "* side" } >expect && - diff -u expect actual + test_cmp expect actual '
@@ -41,7 +41,7 @@ test_expect_success 'branch --contains master' ' { echo " master" && echo "* side" } >expect && - diff -u expect actual + test_cmp expect actual '
@@ -51,7 +51,7 @@ test_expect_success 'branch --contains=side' ' { echo "* side" } >expect && - diff -u expect actual + test_cmp expect actual '
t/t3404-rebase-interactive.sh +2 -2
@@ -146,8 +146,8 @@ EOF test_expect_success 'stop on conflicting pick' ' git tag new-branch1 && ! git rebase -i master && - diff -u expect .git/.dotest-merge/patch && - diff -u expect2 file1 && + test_cmp expect .git/.dotest-merge/patch && + test_cmp expect2 file1 && test 4 = $(grep -v "^#" < .git/.dotest-merge/done | wc -l) && test 0 = $(grep -c "^[^#]" < .git/.dotest-merge/git-rebase-todo) '
t/t3405-rebase-malformed.sh +2 -2
@@ -41,8 +41,8 @@ test_expect_success rebase ' git rebase master side && git cat-file commit HEAD | sed -e "1,/^\$/d" >F1 && - diff -u F0 F1 && - diff -u F F0 + test_cmp F0 F1 && + test_cmp F F0 ' test_done
t/t3406-rebase-message.sh +1 -1
@@ -37,7 +37,7 @@ test_expect_success 'rebase -m' ' git rebase -m master >report && sed -n -e "/^Already applied: /p" \ -e "/^Committed: /p" report >actual && - diff -u expect actual + test_cmp expect actual '
t/t3701-add-interactive.sh +2 -2
@@ -24,7 +24,7 @@ EOF test_expect_success 'diff works (initial)' ' (echo d; echo 1) | git add -i >output && sed -ne "/new file/,/content/p" <output >diff && - diff -u expected diff + test_cmp expected diff ' test_expect_success 'revert works (initial)' ' git add file &&
@@ -57,7 +57,7 @@ EOF test_expect_success 'diff works (commit)' ' (echo d; echo 1) | git add -i >output && sed -ne "/^index/,/content/p" <output >diff && - diff -u expected diff + test_cmp expected diff ' test_expect_success 'revert works (commit)' ' git add file &&
t/t3902-quoted.sh +8 -8
@@ -78,28 +78,28 @@ EOF test_expect_success 'check fully quoted output from ls-files' ' - git ls-files >current && diff -u expect.quoted current + git ls-files >current && test_cmp expect.quoted current ' test_expect_success 'check fully quoted output from diff-files' ' git diff --name-only >current && - diff -u expect.quoted current + test_cmp expect.quoted current ' test_expect_success 'check fully quoted output from diff-index' ' git diff --name-only HEAD >current && - diff -u expect.quoted current + test_cmp expect.quoted current ' test_expect_success 'check fully quoted output from diff-tree' ' git diff --name-only HEAD^ HEAD >current && - diff -u expect.quoted current + test_cmp expect.quoted current '
@@ -111,28 +111,28 @@ test_expect_success 'setting core.quotepath' ' test_expect_success 'check fully quoted output from ls-files' ' - git ls-files >current && diff -u expect.raw current + git ls-files >current && test_cmp expect.raw current ' test_expect_success 'check fully quoted output from diff-files' ' git diff --name-only >current && - diff -u expect.raw current + test_cmp expect.raw current ' test_expect_success 'check fully quoted output from diff-index' ' git diff --name-only HEAD >current && - diff -u expect.raw current + test_cmp expect.raw current ' test_expect_success 'check fully quoted output from diff-tree' ' git diff --name-only HEAD^ HEAD >current && - diff -u expect.raw current + test_cmp expect.raw current '
t/t3903-stash.sh +1 -1
@@ -34,7 +34,7 @@ EOF test_expect_success 'parents of stash' ' test $(git rev-parse stash^) = $(git rev-parse HEAD) && git diff stash^2..stash > output && - diff -u output expect + test_cmp output expect ' test_expect_success 'apply needs clean working directory' '
t/t4023-diff-rename-typechange.sh +3 -3
@@ -57,7 +57,7 @@ test_expect_success 'cross renames to be detected for regular files' ' echo "R100 foo bar" echo "R100 bar foo" } | sort >expect && - diff -u expect actual + test_cmp expect actual '
@@ -68,7 +68,7 @@ test_expect_success 'cross renames to be detected for typechange' ' echo "R100 foo bar" echo "R100 bar foo" } | sort >expect && - diff -u expect actual + test_cmp expect actual '
@@ -79,7 +79,7 @@ test_expect_success 'moves and renames' ' echo "R100 foo bar" echo "T100 foo" } | sort >expect && - diff -u expect actual + test_cmp expect actual '
t/t4024-diff-optimize-common.sh +1 -1
@@ -150,7 +150,7 @@ test_expect_success 'diff -U0' ' do git diff -U0 file-?$n done | zc >actual && - diff -u expect actual + test_cmp expect actual '
t/t4025-hunk-header.sh +1 -1
@@ -37,7 +37,7 @@ test_expect_success 'hunk header truncation with an overly long line' ' echo " A $N$N$N$N$N$N$N$N$N2" echo " L $N$N$N$N$N$N$N$N$N1" ) >expected && - diff -u actual expected + test_cmp actual expected '
t/t4027-diff-submodule.sh +3 -3
@@ -37,17 +37,17 @@ test_expect_success setup ' test_expect_success 'git diff --raw HEAD' ' git diff --raw --abbrev=40 HEAD >actual && - diff -u expect actual + test_cmp expect actual ' test_expect_success 'git diff-index --raw HEAD' ' git diff-index --raw HEAD >actual.index && - diff -u expect actual.index + test_cmp expect actual.index ' test_expect_success 'git diff-files --raw' ' git diff-files --raw >actual.files && - diff -u expect actual.files + test_cmp expect actual.files ' test_done
t/t4105-apply-fuzz.sh +1 -1
@@ -9,7 +9,7 @@ dotest () { test_expect_success "$name" " git checkout-index -f -q -u file && git apply $* && - diff -u expect file + test_cmp expect file " }
t/t4125-apply-ws-fuzz.sh +4 -4
@@ -56,7 +56,7 @@ test_expect_success nofix ' git apply --whitespace=nowarn patch-1 && # The result should obviously match. - diff -u file-1 file + test_cmp file-1 file ' test_expect_success 'withfix (forward)' '
@@ -70,7 +70,7 @@ test_expect_success 'withfix (forward)' ' git apply --whitespace=fix patch-0 && git apply --whitespace=fix patch-1 && - diff -u file-fixed file + test_cmp file-fixed file ' test_expect_success 'withfix (backward)' '
@@ -91,12 +91,12 @@ test_expect_success 'withfix (backward)' ' sed -e /h/d file-fixed >fixed-head && sed -e /h/d file >file-head && - diff -u fixed-head file-head && + test_cmp fixed-head file-head && sed -n -e /h/p file-fixed >fixed-tail && sed -n -e /h/p file >file-tail && - ! diff -u fixed-tail file-tail + ! test_cmp fixed-tail file-tail '
t/t4150-am-subdir.sh +5 -5
@@ -22,14 +22,14 @@ test_expect_success 'am regularly from stdin' ' git checkout initial && git am <patchfile && git diff master >actual && - diff -u expect actual + test_cmp expect actual ' test_expect_success 'am regularly from file' ' git checkout initial && git am patchfile && git diff master >actual && - diff -u expect actual + test_cmp expect actual ' test_expect_success 'am regularly from stdin in subdirectory' '
@@ -41,7 +41,7 @@ test_expect_success 'am regularly from stdin in subdirectory' ' git am <../patchfile ) && git diff master>actual && - diff -u expect actual + test_cmp expect actual ' test_expect_success 'am regularly from file in subdirectory' '
@@ -53,7 +53,7 @@ test_expect_success 'am regularly from file in subdirectory' ' git am ../patchfile ) && git diff master >actual && - diff -u expect actual + test_cmp expect actual ' test_expect_success 'am regularly from file in subdirectory with full path' '
@@ -66,7 +66,7 @@ test_expect_success 'am regularly from file in subdirectory with full path' ' git am "$P/patchfile" ) && git diff master >actual && - diff -u expect actual + test_cmp expect actual ' test_done
t/t4201-shortlog.sh +1 -1
@@ -45,6 +45,6 @@ A U Thor (5): EOF -test_expect_success 'shortlog wrapping' 'diff -u expect out' +test_expect_success 'shortlog wrapping' 'test_cmp expect out' test_done
t/t5505-remote.sh +3 -3
@@ -24,7 +24,7 @@ setup_repository () { tokens_match () { echo "$1" | tr ' ' '\012' | sort | sed -e '/^$/d' >expect && echo "$2" | tr ' ' '\012' | sort | sed -e '/^$/d' >actual && - diff -u expect actual + test_cmp expect actual } check_remote_track () {
@@ -73,7 +73,7 @@ test_expect_success 'add another remote' ' sed -e "/^refs\/remotes\/origin\//d" \ -e "/^refs\/remotes\/second\//d" >actual && >expect && - diff -u expect actual + test_cmp expect actual ) '
@@ -93,7 +93,7 @@ test_expect_success 'remove remote' ' git for-each-ref "--format=%(refname)" refs/remotes | sed -e "/^refs\/remotes\/origin\//d" >actual && >expect && - diff -u expect actual + test_cmp expect actual ) '
t/t5510-fetch.sh +1 -1
@@ -249,7 +249,7 @@ test_expect_success 'bundle should record HEAD correctly' ' do echo "$(git rev-parse --verify $h) $h" done >expect && - diff -u expect actual + test_cmp expect actual '
t/t5512-ls-remote.sh +4 -4
@@ -24,28 +24,28 @@ test_expect_success setup ' test_expect_success 'ls-remote --tags .git' ' git ls-remote --tags .git >actual && - diff -u expected.tag actual + test_cmp expected.tag actual ' test_expect_success 'ls-remote .git' ' git ls-remote .git >actual && - diff -u expected.all actual + test_cmp expected.all actual ' test_expect_success 'ls-remote --tags self' ' git ls-remote --tags self >actual && - diff -u expected.tag actual + test_cmp expected.tag actual ' test_expect_success 'ls-remote self' ' git ls-remote self >actual && - diff -u expected.all actual + test_cmp expected.all actual '
t/t6004-rev-list-path-optim.sh +1 -1
@@ -45,7 +45,7 @@ test_expect_success 'further setup' ' test_expect_success 'path optimization 2' ' ( echo "$side"; echo "$initial" ) >expected && git rev-list HEAD -- a >actual && - diff -u expected actual + test_cmp expected actual ' test_done
t/t6009-rev-list-parent.sh +1 -1
@@ -31,7 +31,7 @@ test_expect_failure 'one is ancestor of others and should not be shown' ' git rev-list one --not four >result && >expect && - diff -u expect result + test_cmp expect result '
t/t6027-merge-binary.sh +2 -2
@@ -45,7 +45,7 @@ test_expect_success resolve ' false else git ls-files -s >current - diff -u current expect + test_cmp current expect fi '
@@ -60,7 +60,7 @@ test_expect_success recursive ' false else git ls-files -s >current - diff -u current expect + test_cmp current expect fi '
t/t6029-merge-subtree.sh +1 -1
@@ -25,7 +25,7 @@ test_expect_success 'subtree available and works like recursive' ' git merge -s subtree side && for i in mundo $s world; do echo $i; done >expect && - diff -u expect hello + test_cmp expect hello '
t/t7010-setup.sh +9 -9
@@ -18,7 +18,7 @@ test_expect_success 'git add (absolute)' ' git add "$D/a/b/c/d" && git ls-files >current && echo a/b/c/d >expect && - diff -u expect current + test_cmp expect current '
@@ -32,7 +32,7 @@ test_expect_success 'git add (funny relative)' ' ) && git ls-files >current && echo a/e/f >expect && - diff -u expect current + test_cmp expect current '
@@ -43,7 +43,7 @@ test_expect_success 'git rm (absolute)' ' git rm -f --cached "$D/a/b/c/d" && git ls-files >current && echo a/e/f >expect && - diff -u expect current + test_cmp expect current '
@@ -57,7 +57,7 @@ test_expect_success 'git rm (funny relative)' ' ) && git ls-files >current && echo a/b/c/d >expect && - diff -u expect current + test_cmp expect current '
@@ -67,7 +67,7 @@ test_expect_success 'git ls-files (absolute)' ' git add a && git ls-files "$D/a/e/../b" >current && echo a/b/c/d >expect && - diff -u expect current + test_cmp expect current '
@@ -80,7 +80,7 @@ test_expect_success 'git ls-files (relative #1)' ' git ls-files "../b/c" ) >current && echo c/d >expect && - diff -u expect current + test_cmp expect current '
@@ -93,7 +93,7 @@ test_expect_success 'git ls-files (relative #2)' ' git ls-files --full-name "../e/f" ) >current && echo a/e/f >expect && - diff -u expect current + test_cmp expect current '
@@ -126,13 +126,13 @@ test_expect_success 'log using absolute path names' ' git log a/b/c/d >f1.txt && git log "$(pwd)/a/b/c/d" >f2.txt && - diff -u f1.txt f2.txt + test_cmp f1.txt f2.txt ' test_expect_success 'blame using absolute path names' ' git blame a/b/c/d >f1.txt && git blame "$(pwd)/a/b/c/d" >f2.txt && - diff -u f1.txt f2.txt + test_cmp f1.txt f2.txt ' test_expect_success 'setup deeper work tree' '
t/t7201-co.sh +9 -9
@@ -83,13 +83,13 @@ test_expect_success "checkout with unrelated dirty tree without -m" ' fill 0 1 2 3 4 5 6 7 8 >same && cp same kept git checkout side >messages && - diff -u same kept + test_cmp same kept (cat > messages.expect <<EOF M same EOF ) && touch messages.expect && - diff -u messages.expect messages + test_cmp messages.expect messages ' test_expect_success "checkout -m with dirty tree" '
@@ -106,19 +106,19 @@ test_expect_success "checkout -m with dirty tree" ' M one EOF ) && - diff -u expect.messages messages && + test_cmp expect.messages messages && fill "M one" "A three" "D two" >expect.master && git diff --name-status master >current.master && - diff -u expect.master current.master && + test_cmp expect.master current.master && fill "M one" >expect.side && git diff --name-status side >current.side && - diff -u expect.side current.side && + test_cmp expect.side current.side && : >expect.index && git diff --cached >current.index && - diff -u expect.index current.index + test_cmp expect.index current.index ' test_expect_success "checkout -m with dirty tree, renamed" '
@@ -136,7 +136,7 @@ test_expect_success "checkout -m with dirty tree, renamed" ' git checkout -m renamer && fill 1 3 4 5 7 8 >expect && - diff -u expect uno && + test_cmp expect uno && ! test -f one && git diff --cached >current && ! test -s current
@@ -161,7 +161,7 @@ test_expect_success 'checkout -m with merge conflict' ' git diff master:one :3:uno | sed -e "1,/^@@/d" -e "/^ /d" -e "s/^-/d/" -e "s/^+/a/" >current && fill d2 aT d7 aS >expect && - diff -u current expect && + test_cmp current expect && git diff --cached two >current && ! test -s current '
@@ -178,7 +178,7 @@ If you want to create a new branch from this checkout, you may do so HEAD is now at 7329388... Initial A one, A two EOF ) && - diff -u messages.expect messages && + test_cmp messages.expect messages && H=$(git rev-parse --verify HEAD) && M=$(git show-ref -s --verify refs/heads/master) && test "z$H" = "z$M" &&
t/t7501-commit.sh +7 -7
@@ -203,7 +203,7 @@ test_expect_success 'sign off (1)' ' git var GIT_COMMITTER_IDENT | sed -e "s/>.*/>/" -e "s/^/Signed-off-by: /" ) >expected && - diff -u expected actual + test_cmp expected actual '
@@ -223,7 +223,7 @@ $existing" && git var GIT_COMMITTER_IDENT | sed -e "s/>.*/>/" -e "s/^/Signed-off-by: /" ) >expected && - diff -u expected actual + test_cmp expected actual '
@@ -240,7 +240,7 @@ test_expect_success 'multiple -m' ' echo echo three ) >expected && - diff -u expected actual + test_cmp expected actual '
@@ -301,12 +301,12 @@ test_expect_success 'same tree (merge and amend merge)' ' git merge -s ours side -m "empty ok" && git diff HEAD^ HEAD >actual && : >expected && - diff -u expected actual && + test_cmp expected actual && git commit --amend -m "empty really ok" && git diff HEAD^ HEAD >actual && : >expected && - diff -u expected actual + test_cmp expected actual '
@@ -323,7 +323,7 @@ test_expect_success 'amend using the message from another commit' ' git commit --allow-empty --amend -C "$old" && git show --pretty="format:%ad %s" "$old" >expected && git show --pretty="format:%ad %s" HEAD >actual && - diff -u expected actual + test_cmp expected actual '
@@ -341,7 +341,7 @@ test_expect_success 'amend using the message from a commit named with tag' ' git commit --allow-empty --amend -C tagged-old && git show --pretty="format:%ad %s" "$old" >expected && git show --pretty="format:%ad %s" HEAD >actual && - diff -u expected actual + test_cmp expected actual '
t/t7502-commit.sh +7 -7
@@ -85,7 +85,7 @@ test_expect_success 'verbose' ' git add negative && git status -v | sed -ne "/^diff --git /p" >actual && echo "diff --git a/negative b/negative" >expect && - diff -u expect actual + test_cmp expect actual '
@@ -95,7 +95,7 @@ test_expect_success 'cleanup commit messages (verbatim,-t)' ' { echo;echo "# text";echo; } >expect && git commit --cleanup=verbatim -t expect -a && git cat-file -p HEAD |sed -e "1,/^\$/d" |head -n 3 >actual && - diff -u expect actual + test_cmp expect actual '
@@ -104,7 +104,7 @@ test_expect_success 'cleanup commit messages (verbatim,-F)' ' echo >>negative && git commit --cleanup=verbatim -F expect -a && git cat-file -p HEAD |sed -e "1,/^\$/d">actual && - diff -u expect actual + test_cmp expect actual '
@@ -113,7 +113,7 @@ test_expect_success 'cleanup commit messages (verbatim,-m)' ' echo >>negative && git commit --cleanup=verbatim -m "$(cat expect)" -a && git cat-file -p HEAD |sed -e "1,/^\$/d">actual && - diff -u expect actual + test_cmp expect actual '
@@ -124,7 +124,7 @@ test_expect_success 'cleanup commit messages (whitespace,-F)' ' echo "# text" >expect && git commit --cleanup=whitespace -F text -a && git cat-file -p HEAD |sed -e "1,/^\$/d">actual && - diff -u expect actual + test_cmp expect actual '
@@ -135,7 +135,7 @@ test_expect_success 'cleanup commit messages (strip,-F)' ' echo sample >expect && git commit --cleanup=strip -F text -a && git cat-file -p HEAD |sed -e "1,/^\$/d">actual && - diff -u expect actual + test_cmp expect actual '
@@ -150,7 +150,7 @@ test_expect_success 'cleanup commit messages (strip,-F,-e)' ' { echo;echo sample;echo; } >text && git commit -e -F text -a && head -n 4 .git/COMMIT_EDITMSG >actual && - diff -u expect actual + test_cmp expect actual '
t/t7502-status.sh +1 -1
@@ -146,7 +146,7 @@ cat <<EOF >expect EOF test_expect_success 'status of partial commit excluding new file in index' ' git status dir1/modified >output && - diff -u expect output + test_cmp expect output ' test_done
t/t7600-merge.sh +1 -1
@@ -108,7 +108,7 @@ create_merge_msgs() { } verify_diff() { - if ! diff -u "$1" "$2" + if ! test_cmp "$1" "$2" then echo "$3" false
t/t8003-blame.sh +2 -2
@@ -112,7 +112,7 @@ test_expect_success 'blame wholesale copy' ' echo mouse-Second echo mouse-Third } >expected && - diff -u expected current + test_cmp expected current '
@@ -125,7 +125,7 @@ test_expect_success 'blame wholesale copy and more' ' echo cow-Fifth echo mouse-Third } >expected && - diff -u expected current + test_cmp expected current '
t/t9001-send-email.sh +1 -1
@@ -81,7 +81,7 @@ test_expect_success 'Show all headers' ' -e "s/^\(Message-Id:\).*/\1 MESSAGE-ID-STRING/" \ -e "s/^\(X-Mailer:\).*/\1 X-MAILER-STRING/" \ >actual-show-all-headers && - diff -u expected-show-all-headers actual-show-all-headers + test_cmp expected-show-all-headers actual-show-all-headers ' z8=zzzzzzzz
t/t9116-git-svn-log.sh +12 -12
@@ -55,74 +55,74 @@ printf 'r1 \nr2 \nr4 \n' > expected-range-r1-r2-r4 test_expect_success 'test ascending revision range' " git reset --hard trunk && - git svn log -r 1:4 | grep '^r[0-9]' | cut -d'|' -f1 | diff -u expected-range-r1-r2-r4 - + git svn log -r 1:4 | grep '^r[0-9]' | cut -d'|' -f1 | test_cmp expected-range-r1-r2-r4 - " printf 'r4 \nr2 \nr1 \n' > expected-range-r4-r2-r1 test_expect_success 'test descending revision range' " git reset --hard trunk && - git svn log -r 4:1 | grep '^r[0-9]' | cut -d'|' -f1 | diff -u expected-range-r4-r2-r1 - + git svn log -r 4:1 | grep '^r[0-9]' | cut -d'|' -f1 | test_cmp expected-range-r4-r2-r1 - " printf 'r1 \nr2 \n' > expected-range-r1-r2 test_expect_success 'test ascending revision range with unreachable revision' " git reset --hard trunk && - git svn log -r 1:3 | grep '^r[0-9]' | cut -d'|' -f1 | diff -u expected-range-r1-r2 - + git svn log -r 1:3 | grep '^r[0-9]' | cut -d'|' -f1 | test_cmp expected-range-r1-r2 - " printf 'r2 \nr1 \n' > expected-range-r2-r1 test_expect_success 'test descending revision range with unreachable revision' " git reset --hard trunk && - git svn log -r 3:1 | grep '^r[0-9]' | cut -d'|' -f1 | diff -u expected-range-r2-r1 - + git svn log -r 3:1 | grep '^r[0-9]' | cut -d'|' -f1 | test_cmp expected-range-r2-r1 - " printf 'r2 \n' > expected-range-r2 test_expect_success 'test ascending revision range with unreachable upper boundary revision and 1 commit' " git reset --hard trunk && - git svn log -r 2:3 | grep '^r[0-9]' | cut -d'|' -f1 | diff -u expected-range-r2 - + git svn log -r 2:3 | grep '^r[0-9]' | cut -d'|' -f1 | test_cmp expected-range-r2 - " test_expect_success 'test descending revision range with unreachable upper boundary revision and 1 commit' " git reset --hard trunk && - git svn log -r 3:2 | grep '^r[0-9]' | cut -d'|' -f1 | diff -u expected-range-r2 - + git svn log -r 3:2 | grep '^r[0-9]' | cut -d'|' -f1 | test_cmp expected-range-r2 - " printf 'r4 \n' > expected-range-r4 test_expect_success 'test ascending revision range with unreachable lower boundary revision and 1 commit' " git reset --hard trunk && - git svn log -r 3:4 | grep '^r[0-9]' | cut -d'|' -f1 | diff -u expected-range-r4 - + git svn log -r 3:4 | grep '^r[0-9]' | cut -d'|' -f1 | test_cmp expected-range-r4 - " test_expect_success 'test descending revision range with unreachable lower boundary revision and 1 commit' " git reset --hard trunk && - git svn log -r 4:3 | grep '^r[0-9]' | cut -d'|' -f1 | diff -u expected-range-r4 - + git svn log -r 4:3 | grep '^r[0-9]' | cut -d'|' -f1 | test_cmp expected-range-r4 - " printf -- '------------------------------------------------------------------------\n' > expected-separator test_expect_success 'test ascending revision range with unreachable boundary revisions and no commits' " git reset --hard trunk && - git svn log -r 5:6 | diff -u expected-separator - + git svn log -r 5:6 | test_cmp expected-separator - " test_expect_success 'test descending revision range with unreachable boundary revisions and no commits' " git reset --hard trunk && - git svn log -r 6:5 | diff -u expected-separator - + git svn log -r 6:5 | test_cmp expected-separator - " test_expect_success 'test ascending revision range with unreachable boundary revisions and 1 commit' " git reset --hard trunk && - git svn log -r 3:5 | grep '^r[0-9]' | cut -d'|' -f1 | diff -u expected-range-r4 - + git svn log -r 3:5 | grep '^r[0-9]' | cut -d'|' -f1 | test_cmp expected-range-r4 - " test_expect_success 'test descending revision range with unreachable boundary revisions and 1 commit' " git reset --hard trunk && - git svn log -r 5:3 | grep '^r[0-9]' | cut -d'|' -f1 | diff -u expected-range-r4 - + git svn log -r 5:3 | grep '^r[0-9]' | cut -d'|' -f1 | test_cmp expected-range-r4 - " test_done
t/t9200-git-cvsexportcommit.sh +7 -7
@@ -37,7 +37,7 @@ check_entries () { else printf '%s\n' "$2" | tr '|' '\012' >expected fi - diff -u expected actual + test_cmp expected actual } test_expect_success \
@@ -257,8 +257,8 @@ test_expect_success '-w option should work with relative GIT_DIR' ' (cd "$GIT_DIR" && GIT_DIR=. git cvsexportcommit -w "$CVSWORK" -c $id && check_entries "$CVSWORK/W" "file1.txt/1.1/|file2.txt/1.1/" && - diff -u "$CVSWORK/W/file1.txt" ../W/file1.txt && - diff -u "$CVSWORK/W/file2.txt" ../W/file2.txt + test_cmp "$CVSWORK/W/file1.txt" ../W/file1.txt && + test_cmp "$CVSWORK/W/file2.txt" ../W/file2.txt ) '
@@ -279,9 +279,9 @@ test_expect_success 'check files before directories' ' git cvsexportcommit -w "$CVSWORK" -c $id && check_entries "$CVSWORK/E" "DS/1.1/|newfile5.txt/1.1/" && check_entries "$CVSWORK" "DS/1.1/|release-notes/1.2/" && - diff -u "$CVSWORK/DS" DS && - diff -u "$CVSWORK/E/DS" E/DS && - diff -u "$CVSWORK/release-notes" release-notes + test_cmp "$CVSWORK/DS" DS && + test_cmp "$CVSWORK/E/DS" E/DS && + test_cmp "$CVSWORK/release-notes" release-notes '
@@ -293,7 +293,7 @@ test_expect_success 'commit a file with leading spaces in the name' ' id=$(git rev-parse HEAD) && git cvsexportcommit -w "$CVSWORK" -c $id && check_entries "$CVSWORK" " space/1.1/|DS/1.1/|release-notes/1.2/" && - diff -u "$CVSWORK/ space" " space" + test_cmp "$CVSWORK/ space" " space" '
t/test-lib.sh +18 -0
@@ -42,6 +42,7 @@ export GIT_MERGE_VERBOSITY export GIT_AUTHOR_EMAIL GIT_AUTHOR_NAME export GIT_COMMITTER_EMAIL GIT_COMMITTER_NAME export EDITOR VISUAL +GIT_TEST_CMP=${GIT_TEST_CMP:-diff -u} # Protect ourselves from common misconfiguration to export # CDPATH into the environment
@@ -302,6 +303,23 @@ test_must_fail () { test $? -gt 0 -a $? -le 128 } +# test_cmp is a helper function to compare actual and expected output. +# You can use it like: +# +# test_expect_success 'foo works' ' +# echo expected >expected && +# foo >actual && +# test_cmp expected actual +# ' +# +# This could be written as either "cmp" or "diff -u", but: +# - cmp's output is not nearly as easy to read as diff -u +# - not all diff versions understand "-u" + +test_cmp() { + $GIT_TEST_CMP "$@" +} + # Most tests can use the created repository, but some may need to create more. # Usage: test_create_repo <directory> test_create_repo () {
Signature · 1 line
1.5.4.4.543.g30fdd.dirty
Re: [PATCH 06/16] add test_cmp function for test scripts
I think I have an earlier round of this in 'pu'.
Re: [PATCH 06/16] add test_cmp function for test scripts
Quoted 1 line · Junio C Hamano
I think I have an earlier round of this in 'pu'.
Oops, so you do. I remember discussing it but didn't recall a patch coming out of it.
I see you have queued mine now in pu over yours. I actually think yours looks a little nicer (you sanity-check the comparator, and you are more careful with stdin (something that I considered, but figured we could add if something actually broke)).
Any reason not to keep your existing one over mine?
-Peff
Re: [PATCH 06/16] add test_cmp function for test scripts
Quoted 7 lines · Jeff King
Quoted 1 line · Junio C Hamano
I think I have an earlier round of this in 'pu'.
Oops, so you do. I remember discussing it but didn't recall a patch coming out of it.
Quoted 6 lines
I see you have queued mine now in pu over yours. I actually think yours looks a little nicer (you sanity-check the comparator, and you are more careful with stdin (something that I considered, but figured we could add if something actually broke)).
Any reason not to keep your existing one over mine?
Yours is much simpler.
And it is tested on the field, so if it ever breaks I have somebody else to blame ;-)
PATCH 7/16 t4020: don't use grep -a
In reply to a message not in the archiveSolaris /usr/bin/grep doesn't understand "-a". In this case we can just include the expected output with the test, which is a better test anyway.
---
Ooh, we get to use binary patches.
t/t4020-diff-external.sh | 3 ++- t/t4020/diff.NUL | Bin 0 -> 116 bytes 2 files changed, 2 insertions(+), 1 deletions(-) create mode 100644 t/t4020/diff.NUL
t/t4020-diff-external.sh +2 -1
@@ -103,7 +103,8 @@ echo NULZbetweenZwords | perl -pe 'y/Z/\000/' > file test_expect_success 'force diff with "diff"' ' echo >.gitattributes "file diff" && - git diff | grep -a second + git diff >actual && + test_cmp ../t4020/diff.NUL actual ' test_done
t/t4020/diff.NUL new file binary +0 -0
Signature · 1 line
1.5.4.4.543.g30fdd.dirty
PATCH 8/16 t4200: use cut instead of sed
In reply to a message not in the archiveSome versions of sed (like the one on Solaris) don't like to match literal tabs, and simply print nothing. Instead, let's use cut.
--- t/t4200-rerere.sh | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-)
t/t4200-rerere.sh +1 -1
@@ -53,7 +53,7 @@ test_expect_success 'conflicting merge' ' ! git merge first ' -sha1=$(sed -e 's/ .*//' .git/rr-cache/MERGE_RR) +sha1=$(cut -d' ' -f1 .git/rr-cache/MERGE_RR) rr=.git/rr-cache/$sha1 test_expect_success 'recorded preimage' "grep ======= $rr/preimage"
Signature · 1 line
1.5.4.4.543.g30fdd.dirty
Re: [PATCH 08/16] t4200: use cut instead of sed
Quoted 5 lines · Jeff King
Some versions of sed (like the one on Solaris) don't like to match literal tabs, and simply print nothing. Instead, let's use cut.
-sha1=$(sed -e 's/ .*//' .git/rr-cache/MERGE_RR)
This is a bit hard to believe. On one of my ancient Sun box:
$ uname -a SunOS sic.twinsun.com 5.8 Generic_117350-45 sun4u sparc SUNW,UltraSPARC-IIi-Engine $ ls -l /bin/sed -r-xr-xr-x 1 root bin 28748 Aug 2 2005 /bin/sed
the above "sed" does not misbehave (/bin/sh does, of course, on $(...), but that is a different story).
Re: [PATCH 08/16] t4200: use cut instead of sed
Quoted 7 lines · Junio C Hamano
Quoted 5 lines
Some versions of sed (like the one on Solaris) don't like to match literal tabs, and simply print nothing. Instead, let's use cut.
-sha1=$(sed -e 's/ .*//' .git/rr-cache/MERGE_RR)
This is a bit hard to believe. On one of my ancient Sun box:
Ah, sorry. I tested this line by hand, found it didn't work, and stupidly jumped to the assumption that it was the literal tab (that being the only interesting thing in the input).
But the actual problem is that MERGE_RR lacks a trailing newline. I don't see any code to add newlines, even though it seems possible that we will write out several paths. So I think we need a newline here:
builtin-rerere.c +2 -1
@@ -58,7 +58,8 @@ static int write_rr(struct path_list *rr, int out_fd)int length =strlen (path) +1 ;if (write_in_full (out_fd, rr->items[i].util,40 ) !=40 ||write_in_full (out_fd,"\t" ,1 ) !=1 || -write_in_full (out_fd, path, length) != length) +write_in_full (out_fd, path, length) != length || +write_in_full (out_fd,"\n" ,1 ) !=1 )die ("unable to write rerere record" ); }if (commit_lock_file (&write_lock) !=0 )
And unless I am missing something, rerere on multiple paths is very broken (but that seems weird, since this code is so old).
-Peff
Re: [PATCH 08/16] t4200: use cut instead of sed
Quoted 9 lines · Jeff King
But the actual problem is that
MERGE_RRlacks a trailing newline. I don't see any code to add newlines, even though it seems possible that we will write out several paths. So I think we need a newline here:builtin-rerere.c +0
-0
builtin-rerere.c static int write_rr(struct path_list *rr, int out_fd) · line 65
Quoted 9 lines
builtin-rerere.c +2
-1@@ -58,7 +58,8 @@ static int write_rr(struct path_list *rr, int out_fd)int length =strlen (path) +1 ;if (write_in_full (out_fd, rr->items[i].util,40 ) !=40 ||write_in_full (out_fd,"\t" ,1 ) !=1 || -write_in_full (out_fd, path, length) != length) +write_in_full (out_fd, path, length) != length || +write_in_full (out_fd,"\n" ,1 ) !=1 )die ("unable to write rerere record" ); }if (commit_lock_file (&write_lock) !=0 )
No, check 3f43d72392b6c0477debd7edbd49bae9b7f41e60^:git-rerere.perl; the records in this file are supposed to be NUL terminated (the paths are allowed to have LF in them).
Quoted 2 lines
And unless I am missing something, rerere on multiple paths is very broken (but that seems weird, since this code is so old).
I do not recall offhand an example of multi-path rerere working or not working correctly, but I am not surprised if the C re-implementation is broken. Will need to check.
PATCH 9/16 t6000lib: tr portability fix
In reply to a message not in the archiveSome versions of tr complain if the number of characters in both sets isn't the same. So here we must manually expand the dashes in set2.
---
This almost makes me want to just use sed instead. But quoting that line noise would probably make it less readable.
t/t6000lib.sh | 5 ++++- 1 files changed, 4 insertions(+), 1 deletions(-)
t/t6000lib.sh +4 -1
@@ -97,7 +97,10 @@ check_output() # from front and back. name_from_description() { - tr "'" '-' | tr '~`!@#$%^&*()_+={}[]|\;:"<>,/? ' '-' | tr -s '-' | tr '[A-Z]' '[a-z]' | sed "s/^-*//;s/-*\$//" + tr "'" '-' | + tr '~`!@#$%^&*()_+={}[]|\;:"<>,/? ' \ + '------------------------------' | + tr -s '-' | tr '[A-Z]' '[a-z]' | sed "s/^-*//;s/-*\$//" }
Signature · 1 line
1.5.4.4.543.g30fdd.dirty
PATCH 10/16 add NO_EXTERNAL_GREP build option
In reply to a message not in the archivePreviously, we just chose whether to allow external grep based on the __unix__ define. However, there are systems which define this macro but which have an inferior group (e.g., one that does not support all options used by t7002). This allows users to accept the potential speed penalty to get a more consistent grep experience (and to pass the testsuite).
---
This might have fallouts for msysgit (i.e., they need to define NO_EXTERNAL_GREP instead of relying on __unix__ not being defined).
Makefile | 7 +++++++ builtin-grep.c | 4 ++-- 2 files changed, 9 insertions(+), 2 deletions(-)
Makefile +7 -0
@@ -148,6 +148,9 @@ all:: # is a simplified version of the merge sort used in glibc. This is # recommended if Git triggers O(n^2) behavior in your platform's qsort(). # +# Define NO_EXTERNAL_GREP if you don't want "git grep" to ever call +# your external grep (e.g., if your system lacks grep, or if its grep is +# not very featureful). GIT-VERSION-FILE: .FORCE-GIT-VERSION-FILE @$(SHELL_PATH) ./GIT-VERSION-GEN
@@ -467,6 +470,7 @@ ifeq ($(uname_O),Cygwin) NO_STRCASESTR = YesPlease NO_MEMMEM = YesPlease NO_SYMLINK_HEAD = YesPlease + NO_EXTERNAL_GREP = YesPlease NEEDS_LIBICONV = YesPlease NO_FAST_WORKING_DIRECTORY = UnfortunatelyYes NO_TRUSTABLE_FILEMODE = UnfortunatelyYes
@@ -760,6 +764,9 @@ endif ifdef DIR_HAS_BSD_GROUP_SEMANTICS COMPAT_CFLAGS += -DDIR_HAS_BSD_GROUP_SEMANTICS endif +ifdef NO_EXTERNAL_GREP + BASIC_CFLAGS += -DNO_EXTERNAL_GREP +endif ifeq ($(TCLTK_PATH),) NO_TCLTK=NoThanks
builtin-grep.c +2 -2
@@ -153,7 +153,7 @@ static int grep_file(struct grep_opt *opt, const char *filename)return i; } -#ifdef __unix__ +#ifndef NO_EXTERNAL_GREP static int exec_grep (int argc,const char **argv) {pid_t pid;
@@ -372,7 +372,7 @@ static int grep_cache(struct grep_opt *opt, const char **paths, int cached)int nr;read_cache (); -#ifdef __unix__ +#ifndef NO_EXTERNAL_GREP /* * Use the external "grep" command for the case where * we grep through the checked-out files. It tends to
Signature · 1 line
1.5.4.4.543.g30fdd.dirty
Re: [PATCH 10/16] add NO_EXTERNAL_GREP build option
Quoted 8 lines · Jeff King
Previously, we just chose whether to allow external grep based on the
__unix__define. However, there are systems which define this macro but which have an inferior group (e.g., one that does not support all options used by t7002). This allows users to accept the potential speed penalty to get a more consistent grep experience (and to pass the testsuite).
Quoted 4 lines
Signed-off-by Jeff King <[email protected]>---
This might have fallouts for msysgit (i.e., they need to defineNO_EXTERNAL_GREPinstead of relying on__unix__not being defined).
You name it. Would you mind converting exec_grep() to use run_command(), too? Or better inline it since it won't do a lot more than run_command()? That way we at least won't get a broken git when I merge git.git that has this patch.
Thanks,
-- Hannes
Re: [PATCH 10/16] add NO_EXTERNAL_GREP build option
Quoted 7 lines · Johannes Sixt
Quoted 2 lines
This might have fallouts for msysgit (i.e., they need to define
NO_EXTERNAL_GREPinstead of relying on__unix__not being defined).You name it. Would you mind converting
exec_grep()to userun_command(), too? Or better inline it since it won't do a lot more thanrun_command()? That way we at least won't get a broken git when I merge git.git that has this patch.
Junio's fixups should restore the automagic behavior, so you shouldn't see any problems now, I think. But the run_command cleanup is sensible.
This is on top of what Junio has in pu.
-- >8 --
use run_command for external grep
The behavior should be identical, but there's no good reason not to use our abstraction library.
--- builtin-grep.c | 32 ++++---------------------------- 1 files changed, 4 insertions(+), 28 deletions(-)
builtin-grep.c +4 -28
@@ -11,6 +11,7 @@#include "tree-walk.h" #include "builtin.h" #include "grep.h" +#include "run-command.h" #ifndef NO_EXTERNAL_GREP #ifdef __unix__
@@ -162,32 +163,6 @@ static int grep_file(struct grep_opt *opt, const char *filename) }#if !NO_EXTERNAL_GREP -static int exec_grep (int argc,const char **argv) -{ -pid_t pid; -int status; - - argv[argc] =NULL ; - pid =fork (); -if (pid <0 ) -return pid; -if (!pid) { -execvp ("grep" , (char **) argv); -exit (255 ); - } -while (waitpid (pid, &status,0 ) <0 ) { -if (errno ==EINTR ) -continue ; -return -1 ; - } -if (WIFEXITED (status)) { -if (!WEXITSTATUS (status)) -return 1 ; -return 0 ; - } -return -1 ; -} -#define MAXARGS 1000#define ARGBUF 4096#define push_arg (a) do { \
@@ -253,7 +228,8 @@ static int flush_grep(struct grep_opt *opt, argc -=2 ; } - status =exec_grep (argc, argv); + argv[argc] =NULL ; + status =run_command_v_opt (argv,0 );if (kept_0) {/*
@@ -264,7 +240,7 @@ static int flush_grep(struct grep_opt *opt, argv[arg0++] = kept_0; argv[arg0] = argv[argc+1 ]; } -return status; +return status ==0 ?1 :-1 ; }static int external_grep (struct grep_opt *opt,const char **paths,int cached)
Signature · 1 line
1.5.4.4.553.g83e84.dirty
Re: [PATCH 10/16] add NO_EXTERNAL_GREP build option
Quoted 2 lines · Jeff King
Junio's fixups should restore the automagic behavior, so you shouldn't see any problems now, I think. But the
run_commandcleanup is sensible.
Thanks. But it turns out that things are not /that/ trivial. We better live with Junio's fixup.
Quoted 13 lines
builtin-grep.c +0
-13-while (waitpid (pid, &status,0 ) <0 ) { -if (errno ==EINTR ) -continue ; -return -1 ; - } -if (WIFEXITED (status)) { -if (!WEXITSTATUS (status)) -return 1 ; -return 0 ; - } -return -1 ; -} -
...
Quoted 1 line
builtin-grep.c +1
-0+ status =run_command_v_opt (argv,0 );
...
builtin-grep.c static int flush_grep(struct grep_opt *opt, · line 243
Quoted 1 line
builtin-grep.c +1
-0+return status ==0 ?1 :-1 ;
grep can return 0 (success, something found), 1 (nothing found), and other values for "real" failures like usage errors or crashes. This conditional throws the latter two into the same pot, which makes git-grep unable to distinguish "nothing found" from failure; cf. the call sites of flush_grep(), where want to set the flag 'hit'.
-- Hannes
Re: [PATCH 10/16] add NO_EXTERNAL_GREP build option
Quoted 9 lines · Johannes Sixt
Quoted 1 line
builtin-grep.c +1
-0+ status =run_command_v_opt (argv,0 );...
builtin-grep.c static int flush_grep(struct grep_opt *opt, · line 243
Quoted 1 line
builtin-grep.c +1
-0+return status ==0 ?1 :-1 ;grep can return 0 (success, something found), 1 (nothing found), and other values for "real" failures like usage errors or crashes. This conditional throws the latter two into the same pot, which makes git-grep unable to distinguish "nothing found" from failure; cf. the call sites of
flush_grep(), where want to set the flag 'hit'.
I noticed that, as well, while writing it. But if you look at external_grep, it just lumps the two cases together anyway. It only ever compares "0 < status", so I think the behavior should be identical.
Which isn't to say that the code couldn't be _improved_ by noting grep failures and crashes, but I don't think it is any worse.
-Peff
Re: [PATCH 10/16] add NO_EXTERNAL_GREP build option
Quoted 11 lines · Jeff King
Quoted 8 lines · Johannes Sixt
Quoted 1 line
builtin-grep.c +1
-0+ status =run_command_v_opt (argv,0 );...
builtin-grep.c static int flush_grep(struct grep_opt *opt, · line 243
Quoted 1 line
builtin-grep.c +1
-0+return status ==0 ?1 :-1 ;grep can return 0 (success, something found), 1 (nothing found), and other values for "real" failures like usage errors or crashes. This conditional throws the latter two into the same pot, which makes git-grep unable to distinguish "nothing found" from failure; cf. the call sites of
flush_grep(), where want to set the flag 'hit'.
Quoted 3 lines
I noticed that, as well, while writing it. But if you look at
external_grep, it just lumps the two cases together anyway. It only ever compares "0 < status", so I think the behavior should be identical.
Yes, indeed. I missed that status < 0 is not checked for.
-- Hannes
Re: [PATCH 10/16] add NO_EXTERNAL_GREP build option
In reply to Jeff KingQuoted 18 lines · Jeff King
Previously, we just chose whether to allow external grep based on the
__unix__define. However, there are systems which define this macro but which have an inferior group (e.g., one that does not support all options used by t7002). This allows users to accept the potential speed penalty to get a more consistent grep experience (and to pass the testsuite).Signed-off-by Jeff King <[email protected]>---
This might have fallouts for msysgit (i.e., they need to defineNO_EXTERNAL_GREPinstead of relying on__unix__not being defined).
...builtin-grep.c +0
-0
builtin-grep.c static int grep_file(struct grep_opt *opt, const char *filename) · line 156
Quoted 5 lines
builtin-grep.c +1
-1@@ -153,7 +153,7 @@ static int grep_file(struct grep_opt *opt, const char *filename)return i; } -#ifdef __unix__ +#ifndef NO_EXTERNAL_GREP
Perhaps place
#ifndef NO_EXTERNAL_GREP
#ifndef __unix__
#define NO_EXTERNAL_GREP 1
#else
#define NO_EXTERNAL_GREP 0
#endif
#endifin git-compat-util.h, and make the in-code reference to
#if NO_EXTERNAL_GREP
... optimization using external grep ...
#endifRe: [PATCH 10/16] add NO_EXTERNAL_GREP build option
Quoted 10 lines · Junio C Hamano
Perhaps place
#ifndef NO_EXTERNAL_GREP #ifndef __unix__ #define NO_EXTERNAL_GREP 1 #else #define NO_EXTERNAL_GREP 0 #endif #endif
Quoted 5 lines
in
git-compat-util.h, and make the in-code reference to#if NO_EXTERNAL_GREP ... optimization using external grep ... #endif
I agree that is much nicer. Looks like you have already marked it up in pu; what is there looks sane.
-Peff
Re: [PATCH 10/16] add NO_EXTERNAL_GREP build option
In reply to Jeff KingThere is one more reason other than you said in the Makefile patch one might want to avoid external grep. Cygwin may have perfectly well working grep, but the reason they avoid external grep is because forking is too slow there.
---
Makefile | 5 ++--- builtin-grep.c | 12 ++++++++++-- 2 files changed, 12 insertions(+), 5 deletions(-)
Makefile +2 -3
@@ -149,8 +149,8 @@ all:: # recommended if Git triggers O(n^2) behavior in your platform's qsort(). # # Define NO_EXTERNAL_GREP if you don't want "git grep" to ever call -# your external grep (e.g., if your system lacks grep, or if its grep is -# not very featureful). +# your external grep (e.g., if your system lacks grep, if its grep is +# broken, or spawning external process is slower than built-in grep git has). GIT-VERSION-FILE: .FORCE-GIT-VERSION-FILE @$(SHELL_PATH) ./GIT-VERSION-GEN
@@ -470,7 +470,6 @@ ifeq ($(uname_O),Cygwin) NO_STRCASESTR = YesPlease NO_MEMMEM = YesPlease NO_SYMLINK_HEAD = YesPlease - NO_EXTERNAL_GREP = YesPlease NEEDS_LIBICONV = YesPlease NO_FAST_WORKING_DIRECTORY = UnfortunatelyYes NO_TRUSTABLE_FILEMODE = UnfortunatelyYes
builtin-grep.c +10 -2
@@ -12,6 +12,14 @@#include "builtin.h" #include "grep.h" +#ifndef NO_EXTERNAL_GREP +#ifdef __unix__ +#define NO_EXTERNAL_GREP 0 +#else +#define NO_EXTERNAL_GREP 1 +#endif +#endif +/* * git grep pathspecs are somewhat different from diff-tree pathspecs; * pathname wildcards are allowed.
@@ -153,7 +161,7 @@ static int grep_file(struct grep_opt *opt, const char *filename)return i; } -#ifndef NO_EXTERNAL_GREP +#if !NO_EXTERNAL_GREP static int exec_grep (int argc,const char **argv) {pid_t pid;
@@ -372,7 +380,7 @@ static int grep_cache(struct grep_opt *opt, const char **paths, int cached)int nr;read_cache (); -#ifndef NO_EXTERNAL_GREP +#if !NO_EXTERNAL_GREP /* * Use the external "grep" command for the case where * we grep through the checked-out files. It tends to
PATCH 11/16 config: add --literal-match option
In reply to a message not in the archiveWhen limiting the values to be set (or returned), the user previously had the option of specifying a regex. In some cases, however, they may want to find a literal value. The option --literal-match converts any matching regex into a literal string comparison.
---
I think this is a nice addition regardless, but it is used by the next patch.
The patch is about twice as long as it needs to be since getting and setting in builtin-config follow two almost-the-same parallel codepaths. I suspect this could be cleaned up, but I didn't look too closely.
Documentation/git-config.txt | 7 ++++ builtin-config.c | 50 +++++++++++++++++++++++-------- builtin-remote.c | 2 +- cache.h | 2 +- config.c | 67 ++++++++++++++++++++++++++++-------------- t/t1300-repo-config.sh | 13 ++++++++ 6 files changed, 104 insertions(+), 37 deletions(-)
Documentation/git-config.txt +7 -0
@@ -152,6 +152,13 @@ See also <<FILES>>. output. The optional `default` parameter is used instead, if there is no color configured for `name`. +--literal-match:: + + Some invocations of git-config will limit their actions based on + matching a config value to a regular expression. If this option + is used, then any such matches are done as a string comparison + rather than as a regular expression match. + [[FILES]] FILES -----
builtin-config.c +37 -13
120 lines not on this page · raw message
builtin-remote.c +1 -1
@@ -107,7 +107,7 @@ static int add(int argc, const char **argv)else strbuf_addf (&buf2,"refs/heads/%s:refs/remotes/%s/%s" , item->path, name, item->path); -if (git_config_set_multivar (buf.buf, buf2.buf,"^$" ,0 )) +if (git_config_set_multivar (buf.buf, buf2.buf,"^$" ,0 ,0 ))return 1 ; }
cache.h +1 -1
@@ -695,7 +695,7 @@ extern unsigned long git_config_ulong(const char *, const char *);extern int git_config_bool (const char *,const char *);extern int git_config_string (const char **,const char *,const char *);extern int git_config_set (const char *,const char *); -extern int git_config_set_multivar (const char *,const char *,const char *,int ); +extern int git_config_set_multivar (const char *,const char *,const char *,int ,int );extern int git_config_rename_section (const char *,const char *);extern const char *git_etc_gitconfig (void );extern int check_repository_format_version (const char *var,const char *value);
config.c +45 -22
120 lines not on this page · raw message
t/t1300-repo-config.sh +13 -0
@@ -682,4 +682,17 @@ test_expect_success 'symlinked configuration' ' ' +test_expect_success 'literal matching works' ' + + git config literal.key1 value && + git config literal.key2 va..e && + git config --get-regexp "literal..*" va..e >output && + grep key1 output && + grep key2 output && + git config --literal-match --get-regexp "literal..*" va..e >output && + ! grep key1 output && + grep key2 output + +' + test_done
Signature · 1 line
1.5.4.4.543.g30fdd.dirty
Re: [PATCH 11/16] config: add --literal-match option
Quoted 6 lines · Jeff King
When limiting the values to be set (or returned), the user previously had the option of specifying a regex. In some cases, however, they may want to find a literal value. The option --literal-match converts any matching regex into a literal string comparison.
Quoted 8 lines
Signed-off-by Jeff King <[email protected]>---
I think this is a nice addition regardless, but it is used by the next patch.The patch is about twice as long as it needs to be since getting and setting in builtin-config follow two almost-the-same parallel codepaths. I suspect this could be cleaned up, but I didn't look too closely.
I think that is a good new feature to propose.
Historically, the config_set_multivar() function has been one of the most buggy part of the then-current codebase. It might be a good idea to clean-up first and then enhance.
But in either case I am quite reluctant to touch this part of the code right now before 1.5.5, especially without extra sets of eyeballs.
Re: [PATCH 11/16] config: add --literal-match option
Quoted 4 lines · Junio C Hamano
Quoted 3 lines
The patch is about twice as long as it needs to be since getting and setting in builtin-config follow two almost-the-same parallel codepaths. I suspect this could be cleaned up, but I didn't look too closely.
Quoted 8 lines
I think that is a good new feature to propose.
Historically, the
config_set_multivar()function has been one of the most buggy part of the then-current codebase. It might be a good idea to clean-up first and then enhance.But in either case I am quite reluctant to touch this part of the code right now before 1.5.5, especially without extra sets of eyeballs.
That sounds sensible. I will throw this on the backburner until post-1.5.5, then, and try to respin it with some config cleanups.
-Peff
Re: [PATCH 11/16] config: add --literal-match option
In reply to Jeff KingDocumentation/git-config.txt See also <<FILES>>. · line 161
Quoted 7 lines · Jeff King
+--literal-match::
Documentation/git-config.txt +6
-0+ + Some invocations of git-config will limit their actions based on + matching a config value to a regular expression. If this option + is used, then any such matches are done as a string comparison + rather than as a regular expression match. +
Why this option is not named --fixed-strings, as everywhere else, then?
Signature · 3 lines
Jakub Narebski Warsaw, Poland ShadeHawk on #git
Re: [PATCH 11/16] config: add --literal-match option
Quoted 9 lines · Jakub Narebski
Documentation/git-config.txt See also <<FILES>>. · line 161
Quoted 7 lines
+--literal-match::
Documentation/git-config.txt +6
-0+ + Some invocations of git-config will limit their actions based on + matching a config value to a regular expression. If this option + is used, then any such matches are done as a string comparison + rather than as a regular expression match. +Why this option is not named --fixed-strings, as everywhere else, then?
Because I never use --fixed-strings anywhere else and didn't think of it? :)
I think, though, that there is a key difference, and using that name would lead to confusion. In grep, --fixed-strings means "find this fixed _substring_ within the input" whereas this option means "find the value that is byte-for-byte equal to this."
-Peff
PATCH 12/16 git-submodule: avoid sed input with no newline
In reply to a message not in the archiveSome versions of sed don't like this, and give no output at all. Instead, we can use git-config to pare down the matches for us.
The content of the last three lines of the patch aren't changed at all; they merely fix a bogus 7-space indentation.
---
And this is safe because of the --literal-match from the last patch.
git-submodule.sh | 13 ++++++------- 1 files changed, 6 insertions(+), 7 deletions(-)
git-submodule.sh +6 -7
@@ -71,13 +71,12 @@ resolve_relative_url () module_name() { # Do we have "submodule.<something>.path = $1" defined in .gitmodules file? - re=$(printf '%s' "$1" | sed -e 's/[].[^$\\*]/\\&/g') - name=$( GIT_CONFIG=.gitmodules \ - git config --get-regexp '^submodule\..*\.path$' | - sed -n -e 's|^submodule\.\(.*\)\.path '"$re"'$|\1|p' ) - test -z "$name" && - die "No submodule mapping found in .gitmodules for path '$path'" - echo "$name" + name=$(git config --literal-match -f .gitmodules \ + --get-regexp 'submodule\..*\.path$' "$1" | + sed -e 's/submodule\.//' -e 's/\.path.*//') + test -z "$name" && + die "No submodule mapping found in .gitmodules for path '$path'" + echo "$name" } #
Signature · 1 line
1.5.4.4.543.g30fdd.dirty
Re: [PATCH 12/16] git-submodule: avoid sed input with no newline
Quoted 3 lines · Jeff King
Some versions of sed don't like this, and give no output at all. Instead, we can use git-config to pare down the matches for us.
Good use of --literal.
But doesn't this make you wonder if "--literal-match --get-regexp" is quite a strange combination? "Literal" covers the value part but the key is still regexp (and we do want it to behave that way). However, maybe we would want to also allow "give entries whose key is this literal key and whose value matches this regexp"?
Quoted 7 lines
git-submodule.sh +6
-0+ name=$(git config --literal-match -f .gitmodules \ + --get-regexp 'submodule\..*\.path$' "$1" | + sed -e 's/submodule\.//' -e 's/\.path.*//') + test -z "$name" && + die "No submodule mapping found in .gitmodules for path '$path'" + echo "$name" }
Re: [PATCH 12/16] git-submodule: avoid sed input with no newline
Quoted 7 lines · Junio C Hamano
Good use of --literal.
But doesn't this make you wonder if "--literal-match --get-regexp" is quite a strange combination? "Literal" covers the value part but the key is still regexp (and we do want it to behave that way). However, maybe we would want to also allow "give entries whose key is this literal key and whose value matches this regexp"?
git config --get?
The problem is that we are matching on two elements: key and value. The matching style of the key element is chosen by --get versus --get-regexp. However, the value_regex is selected implicitly by its presence, so there is no room to modify the flag in the same way. Perhaps --literal-value-match or just --literal-value would be a better name.
-Peff
PATCH 13/16 filter-branch: don't use xargs -0
In reply to a message not in the archiveSome versions of xargs don't understand "-0"; fortunately in this case we can get the same effect by using "git clean".
--- git-filter-branch.sh | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-)
git-filter-branch.sh +1 -1
@@ -281,7 +281,7 @@ while read commit parents; do die "Could not checkout the index" # files that $commit removed are now still in the working tree; # remove them, else they would be added again - git ls-files -z --others | xargs -0 rm -f + git clean -q -f -x eval "$filter_tree" < /dev/null || die "tree filter failed: $filter_tree"
Signature · 1 line
1.5.4.4.543.g30fdd.dirty
PATCH 14/16 filter-branch: use $SHELL_PATH instead of 'sh'
In reply to a message not in the archiveOn some systems, 'sh' isn't very friendly. In particular, t7003 fails on Solaris because it doesn't understand $(). Instead, use the specified SHELL_PATH to run shell code.
---
I think this makes sense regardless of the Solaris shell. We should be consistent about which SHELL_PATH we use in scripts.
Makefile | 1 + git-filter-branch.sh | 2 +- 2 files changed, 2 insertions(+), 1 deletions(-)
Makefile +1 -0
@@ -874,6 +874,7 @@ common-cmds.h: $(wildcard Documentation/git-*.txt) $(patsubst %.sh,%,$(SCRIPT_SH)) : % : %.sh $(QUIET_GEN)$(RM) $@ $@+ && \ sed -e '1s|#!.*/sh|#!$(SHELL_PATH_SQ)|' \ + -e 's|@SHELL_PATH@|$(SHELL_PATH_SQ)|' \ -e 's|@@PERL@@|$(PERL_PATH_SQ)|g' \ -e 's/@@GIT_VERSION@@/$(GIT_VERSION)/g' \ -e 's/@@NO_CURL@@/$(NO_CURL)/g' \
git-filter-branch.sh +1 -1
@@ -309,7 +309,7 @@ while read commit parents; do sed -e '1,/^$/d' <../commit | \ eval "$filter_msg" > ../message || die "msg filter failed: $filter_msg" - sh -c "$filter_commit" "git commit-tree" \ + @SHELL_PATH@ -c "$filter_commit" "git commit-tree" \ $(git write-tree) $parentstr < ../message > ../map/$commit done <../revs
Signature · 1 line
1.5.4.4.543.g30fdd.dirty
PATCH 15/16 t9112: add missing #!/bin/sh header
In reply to a message not in the archive---
This causes the test to barf under Solaris, even though the "correct" behavior for me was to just start it and skip the tests.
t/t9112-git-svn-md5less-file.sh | 2 ++ 1 files changed, 2 insertions(+), 0 deletions(-)
t/t9112-git-svn-md5less-file.sh +2 -0
@@ -1,3 +1,5 @@ +#!/bin/sh + test_description='test that git handles an svn repository with missing md5sums' . ./lib-git-svn.sh
Signature · 1 line
1.5.4.4.543.g30fdd.dirty
PATCH 16/16 t7505: use SHELL_PATH in hook
In reply to a message not in the archiveThe hook doesn't run properly under Solaris /bin/sh. Let's use the SHELL_PATH the user told us about already instead.
---
I discussed this before, but I never followed up with a patch. So here it is.
t/t7505-prepare-commit-msg-hook.sh | 3 ++- 1 files changed, 2 insertions(+), 1 deletions(-)
t/t7505-prepare-commit-msg-hook.sh +2 -1
@@ -25,7 +25,8 @@ export FAKE_EDITOR HOOKDIR="$(git rev-parse --git-dir)/hooks" HOOK="$HOOKDIR/prepare-commit-msg" mkdir -p "$HOOKDIR" -cat > "$HOOK" <<'EOF' +echo "#!$SHELL_PATH" > "$HOOK" +cat >> "$HOOK" <<'EOF' #!/bin/sh if test "$2" = commit; then source=$(git-rev-parse "$3")
Signature · 1 line
1.5.4.4.543.g30fdd.dirty
Re: [PATCH 16/16] t7505: use SHELL_PATH in hook
Hi!
Quoted 2 lines · Jeff King
The hook doesn't run properly under Solaris
/bin/sh. Let's use theSHELL_PATHthe user told us about already instead.
[...]
t/t7505-prepare-commit-msg-hook.sh export FAKE_EDITOR · line 30
Quoted 10 lines
t/t7505-prepare-commit-msg-hook.sh +2
-1@@ -25,7 +25,8 @@ export FAKE_EDITOR HOOKDIR="$(git rev-parse --git-dir)/hooks" HOOK="$HOOKDIR/prepare-commit-msg" mkdir -p "$HOOKDIR" -cat > "$HOOK" <<'EOF' +echo "#!$SHELL_PATH" > "$HOOK" +cat >> "$HOOK" <<'EOF' #!/bin/sh
^^^^^^^^^
The above line should be removed in my humble opinion.
BR,
/Adam
Signature · 2 lines
.:. Adam Piatyszek (ediap) .:.....................................:. .:. [email protected] .:................................:.
Re: [PATCH 16/16] t7505: use SHELL_PATH in hook
Quoted 6 lines · Adam Piatyszek
t/t7505-prepare-commit-msg-hook.sh export FAKE_EDITOR · line 30
Quoted 4 lines
t/t7505-prepare-commit-msg-hook.sh +2
-1-cat > "$HOOK" <<'EOF' +echo "#!$SHELL_PATH" > "$HOOK" +cat >> "$HOOK" <<'EOF' #!/bin/sh^^^^^^^^^
The above line should be removed in my humble opinion.
Oops, yes. I did delete it, but then accidentally editor-typo'd it back into existence. Good eyes.
-Peff
Re: [PATCH 08/16] t4200: use cut instead of sed
In reply to Junio C HamanoQuoted 9 lines · Junio C Hamano
Quoted 9 lines
index c607aad..e4a1dc1 100644
builtin-rerere.c +1
-1@@ -58,7 +58,8 @@ static int write_rr(struct path_list *rr, int out_fd)int length =strlen (path) +1 ; if (write_in_full(out_fd, rr->items[i].util, 40) != 40 || write_in_full(out_fd, "\t", 1) != 1 || -write_in_full (out_fd, path, length) != length) + write_in_full(out_fd, path, length) != length ||
Quoted 8 lines
builtin-rerere.c static int write_rr(struct path_list *rr, int out_fd) · line 65
Quoted 4 lines
builtin-rerere.c +1
-0+ write_in_full(out_fd, "\n", 1) != 1)die ("unable to write rerere record" ); }if (commit_lock_file (&write_lock) !=0 )No, check 3f43d72392b6c0477debd7edbd49bae9b7f41e60^:git-rerere.perl; the records in this file are supposed to be NUL terminated (the paths are allowed to have LF in them).
Ah, OK. And it actually does do that correctly ("length = strlen(path) + 1"). So I think there isn't a bug there in rerere.
But if it's NUL-terminated, then using sed _definitely_ isn't portable here. cut does work on Solaris in this case, but that might or might not be portable to other systems with similar NUL problems. I'm not sure what is the best route. We really just want to grab the sha1 from the beginning of the first line. dd count=40? :)
-Peff
PATCH 1/1 t/t6000lib.sh: tr portability fix fix
In reply to Jeff KingSome versions of tr have a problem with character sets which begin with multiple dashes and attempt to interpret them as long options. Use the '--' notation to signal the end of command line options.
---
Quoted 8 lines · Jeff King
Some versions of tr complain if the number of characters in both sets isn't the same. So here we must manually expand the dashes in set2.
Signed-off-by Jeff King <[email protected]>---
This almost makes me want to just use sed instead. But quoting that line noise would probably make it less readable.
I get the following error on t6002-rev-list-bisect.sh:
* ok 31: bisection diff --bisect u3 ^U <= 0 * ok 32: bisection diff --bisect u4 ^U <= 0 * ok 33: bisection diff --bisect u5 ^U <= 0
tr: unrecognized option ------------------------------' Try tr --help' for more information.
* FAIL 34: --bisect l5 ^root
check_output "git rev-list $_bisect_option l5 ^root"
tr: unrecognized option ------------------------------' Try tr --help' for more information.
* FAIL 35: --bisect l5 ^root ^c3
...
This is tr version 5.2.1.
This patch fixes things. If the dashdash notation is not portable, then backslashing each dash also works. i.e. '\-\-\-\-.. etc. but as you mentioned something like that is less readable, but possibly not as bad as a sed version.
-brandon
t/t6000lib.sh | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-)
t/t6000lib.sh +1 -1
@@ -98,7 +98,7 @@ check_output() name_from_description() { tr "'" '-' | - tr '~`!@#$%^&*()_+={}[]|\;:"<>,/? ' \ + tr -- '~`!@#$%^&*()_+={}[]|\;:"<>,/? ' \ '------------------------------' | tr -s '-' | tr '[A-Z]' '[a-z]' | sed "s/^-*//;s/-*\$//" }
Signature · 1 line
1.5.4.4.481.g5075
Re: [PATCH] t/t6000lib.sh: tr portability fix fix
Quoted 4 lines · Brandon Casey
This patch fixes things. If the dashdash notation is not portable, then backslashing each dash also works. i.e. '\-\-\-\-.. etc. but as you mentioned something like that is less readable, but possibly not as bad as a sed version.
It seems to work fine on Solaris with all versions of tr. I did just blindly extend the '-' without thinking, though...I wonder if there are systems that will get confused about it being a range. It might be safer to just use sed anyway.
-Peff
Re: [PATCH] t/t6000lib.sh: tr portability fix fix
Quoted 6 lines · Jeff King
Quoted 2 lines · Brandon Casey
If the dashdash notation is not portable, then backslashing each dash also works. i.e. '\-\-\-\-..
I wonder if there are
systems that will get confused about it being a range.
Oh, now I understand _why_ backslashing the dashes worked. When your email arrived I was still trying to figure out why
echo hello | tr aeiou '\-\-\-\-\-'
correctly converted the e and o into dashes. Because tr must have a way for the user to escape the range notation. I don't use tr very often.
-brandon
PATCH 1/1 t/t6000lib.sh: tr portability fix fix
In reply to Jeff KingSome versions of tr have a problem with character sets which begin with multiple dashes and attempt to interpret them as long options. Escape each dash to avoid this confusion and also prevent a possible interpretation of the dashes as a range.
---
Quoted 11 lines · Jeff King
Quoted 4 lines · Brandon Casey
This patch fixes things. If the dashdash notation is not portable, then backslashing each dash also works. i.e. '\-\-\-\-.. etc. but as you mentioned something like that is less readable, but possibly not as bad as a sed version.
It seems to work fine on Solaris with all versions of tr. I did just blindly extend the '-' without thinking, though...I wonder if there are systems that will get confused about it being a range. It might be safer to just use sed anyway.
Here's the version with escaped dashes. If you do the sed version, it's something to compare to for readability.
-brandon
t/t6000lib.sh | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-)
t/t6000lib.sh +1 -1
@@ -99,7 +99,7 @@ name_from_description() { tr "'" '-' | tr '~`!@#$%^&*()_+={}[]|\;:"<>,/? ' \ - '------------------------------' | + '\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-' | tr -s '-' | tr '[A-Z]' '[a-z]' | sed "s/^-*//;s/-*\$//" }
Signature · 1 line
1.5.4.4.481.g5075
Re: [PATCH] t/t6000lib.sh: tr portability fix fix
t/t6000lib.sh name_from_description() · line 102
Quoted 7 lines · Brandon Casey
Here's the version with escaped dashes. If you do the sed version, it's something to compare to for readability.
[...]
t/t6000lib.sh +1
-1- '------------------------------' | + '\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-' |
Ugh. How about:
sed 'yA~`!@#$%^&*()_+={}[]|\;:"<>,/? A------------------------------A'
The 'A' delimiter is because in my test, Solaris sed didn't seem to understand \/ to include the literal '/'. And all of the other punctuation is already used in the pattern. ;)
-Peff
Re: [PATCH] t/t6000lib.sh: tr portability fix fix
Quoted 5 lines · Jeff King
Quoted 3 lines · Brandon Casey
Here's the version with escaped dashes. If you do the sed version, it's something to compare to for readability.
Quoted 8 lines
t/t6000lib.sh name_from_description() · line 102
Quoted 4 lines
[...]
t/t6000lib.sh +1
-1- '------------------------------' | + '\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-' |Ugh. How about:
sed 'yA~`!@#$%^&*()_+={}[]|\;:"<>,/? A------------------------------A'
Not working. I get:
* ok 33: bisection diff --bisect u5 ^U <= 0
sed: -e expression #1, char 64: unterminated `y' command
* FAIL 34: --bisect l5 ^root
check_output "git rev-list $_bisect_option l5 ^root"
sed: -e expression #1, char 64: unterminated `y' command
* FAIL 35: --bisect l5 ^root ^c3
check_output "git rev-list $_bisect_option l5 ^root ^c3"
sed: -e expression #1, char 64: unterminated `y' command
* FAIL 36: --bisect l5 ^root ^c3 ^b4
But this does:
sed 'yA~`!@#$%^&*()_+={}\[]|\\;:"<>,/? A------------------------------A'
I have to escape open bracket and backslash on my end (linux). :(
Have to leave now, so if that doesn't work for you (which I'm thinking it won't), I won't be able to test on my end for a while.
-brandon
Re: [PATCH 03/16] more tr portability test script fixes
In reply to Jeff KingJeff King, Wed, Mar 12, 2008 22:31:06 +0100:
t/diff-lib.sh compare_diff_raw_z () { · line 25
Quoted 4 lines
t/diff-lib.sh +2
-2- tr '\000' '\012' <"$1" | sed -e "$sanitize_diff_raw_z" >.tmp-1 - tr '\000' '\012' <"$2" | sed -e "$sanitize_diff_raw_z" >.tmp-2 + perl -pe 'y/\000/\012/' <"$1" | sed -e "$sanitize_diff_raw_z" >.tmp-1 + perl -pe 'y/\000/\012/' <"$2" | sed -e "$sanitize_diff_raw_z" >.tmp-2
These break in presence of ActiveState Perl on Windows.
I suggest replacing such simple construction with a simplified, in-tree, version of tr.
Patches follow
PATCH 1/1 Add test-tr: poor-man tr for the test suite
It offers a limited set of POSIX tr, in particular: no character class support and no [n*m] operators. Only 8bit. C-escapes supported, and character ranges. Deletion and squeezing should work, but -s does not match what the GNU tr from coreutils (which, in turn, does not match SuSv2).
---
Makefile | 2 +- t/t0000-test-progs.sh | 72 +++++++++++++++ test-tr.c | 236 +++++++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 309 insertions(+), 1 deletions(-) create mode 100755 t/t0000-test-progs.sh create mode 100644 test-tr.c
Makefile +1 -1
@@ -1177,7 +1177,7 @@ endif ### Testing rules -TEST_PROGRAMS = test-chmtime$X test-genrandom$X test-date$X test-delta$X test-sha1$X test-match-trees$X test-absolute-path$X test-parse-options$X +TEST_PROGRAMS = test-chmtime$X test-genrandom$X test-date$X test-delta$X test-sha1$X test-match-trees$X test-absolute-path$X test-parse-options$X test-tr$X all:: $(TEST_PROGRAMS)
t/t0000-test-progs.sh new file +72 -0
72 lines not on this page · raw message
test-tr.c new file +236 -0
236 lines not on this page · raw message
Signature · 1 line
1.5.5.rc0.53.g97734
PATCH 1/1 Use test-tr in the test suite
--- t/annotate-tests.sh | 4 ++-- t/diff-lib.sh | 4 ++-- t/t0000-basic.sh | 2 +- t/t0020-crlf.sh | 10 +++++----- t/t0021-conversion.sh | 2 +- t/t1300-repo-config.sh | 4 ++-- t/t3300-funny-names.sh | 6 +++--- t/t3402-rebase-merge.sh | 4 ++-- t/t4004-diff-rename-symlink.sh | 2 +- t/t4015-diff-whitespace.sh | 6 +++--- t/t4019-diff-wserror.sh | 2 +- t/t4020-diff-external.sh | 2 +- t/t4022-diff-rewrite.sh | 2 +- t/t4101-apply-nonl.sh | 4 ++-- t/t4103-apply-binary.sh | 4 ++-- t/t4116-apply-reverse.sh | 4 ++-- t/t4118-apply-empty-context.sh | 4 ++-- t/t4200-rerere.sh | 2 +- t/t4201-shortlog.sh | 4 ++-- t/t5300-pack-object.sh | 2 +- t/t5500-fetch-pack.sh | 2 +- t/t5505-remote.sh | 4 ++-- t/t6003-rev-list-topo-order.sh | 2 +- t/t7003-filter-branch.sh | 2 +- t/t9200-git-cvsexportcommit.sh | 2 +- t/test-lib.sh | 12 +++++++----- 26 files changed, 50 insertions(+), 48 deletions(-)
t/annotate-tests.sh +2 -2
@@ -104,7 +104,7 @@ test_expect_success \ test_expect_success \ 'an incomplete line added' \ - 'echo "incomplete" | tr -d "\\012" >>file && + 'echo "incomplete" | test-tr -d "\\012" >>file && GIT_AUTHOR_NAME="C" git commit -a -m "Incomplete"' test_expect_success \
@@ -115,7 +115,7 @@ test_expect_success \ 'some edit' \ 'mv file file.orig && sed -e "s/^3A/99/" -e "/^1A/d" -e "/^incomplete/d" < file.orig > file && - echo "incomplete" | tr -d "\\012" >>file && + echo "incomplete" | test-tr -d "\\012" >>file && GIT_AUTHOR_NAME="D" git commit -a -m "edit"' test_expect_success \
t/diff-lib.sh +2 -2
@@ -21,8 +21,8 @@ compare_diff_raw_z () { # Also we do not check SHA1 hash generation in this test, which # is a job for t0000-basic.sh - perl -pe 'y/\000/\012/' <"$1" | sed -e "$sanitize_diff_raw_z" >.tmp-1 - perl -pe 'y/\000/\012/' <"$2" | sed -e "$sanitize_diff_raw_z" >.tmp-2 + test-tr '\000' '\012' <"$1" | sed -e "$sanitize_diff_raw_z" >.tmp-1 + test-tr '\000' '\012' <"$2" | sed -e "$sanitize_diff_raw_z" >.tmp-2 git diff .tmp-1 .tmp-2 && rm -f .tmp-1 .tmp-2 }
t/t0000-basic.sh +1 -1
@@ -324,7 +324,7 @@ test_expect_success 'very long name in the index handled sanely' ' ( git ls-files -s path4 | sed -e "s/ .*/ /" | - tr -d "\012" + test-tr -d "\012" echo "$a" ) | git update-index --index-info && len=$(git ls-files "a*" | wc -c) &&
t/t0020-crlf.sh +5 -5
@@ -5,20 +5,20 @@ test_description='CRLF conversion' . ./test-lib.sh q_to_nul () { - perl -pe 'y/Q/\000/' + test-tr Q '\000' } q_to_cr () { - tr Q '\015' + test-tr Q '\015' } append_cr () { - sed -e 's/$/Q/' | tr Q '\015' + sed -e 's/$/Q/' | test-tr Q '\015' } remove_cr () { - tr '\015' Q <"$1" | grep Q >/dev/null && - tr '\015' Q <"$1" | sed -ne 's/Q$//p' + test-tr '\015' Q <"$1" | grep Q >/dev/null && + test-tr '\015' Q <"$1" | sed -ne 's/Q$//p' } test_expect_success setup '
t/t0021-conversion.sh +1 -1
@@ -5,7 +5,7 @@ test_description='blob conversion via gitattributes' . ./test-lib.sh cat <<\EOF >rot13.sh -tr \ +test-tr \ 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ' \ 'nopqrstuvwxyzabcdefghijklmNOPQRSTUVWXYZABCDEFGHIJKLM' EOF
t/t1300-repo-config.sh +2 -2
@@ -657,12 +657,12 @@ Qsection.sub=section.val4 Qsection.sub=section.val5Q EOF -git config --null --list | perl -pe 'y/\000/Q/' > result +git config --null --list | test-tr '\000' Q > result echo >>result test_expect_success '--null --list' 'cmp result expect' -git config --null --get-regexp 'val[0-9]' | perl -pe 'y/\000/Q/' > result +git config --null --get-regexp 'val[0-9]' | test-tr '\000' Q > result echo >>result test_expect_success '--null --get-regexp' 'cmp result expect'
t/t3300-funny-names.sh +3 -3
@@ -54,7 +54,7 @@ echo 'just space no-funny tabs ," (dq) and spaces' >expected test_expect_success 'git ls-files -z with-funny' \ - 'git ls-files -z | perl -pe y/\\000/\\012/ >current && + 'git ls-files -z | test-tr "\000" "\012" >current && git diff expected current' t1=`git write-tree`
@@ -83,11 +83,11 @@ test_expect_success 'git diff-tree with-funny' \ echo 'A tabs ," (dq) and spaces' >expected test_expect_success 'git diff-index -z with-funny' \ - 'git diff-index -z --name-status $t0 | perl -pe y/\\000/\\012/ >current && + 'git diff-index -z --name-status $t0 | test-tr "\000" "\012" >current && git diff expected current' test_expect_success 'git diff-tree -z with-funny' \ - 'git diff-tree -z --name-status $t0 $t1 | perl -pe y/\\000/\\012/ >current && + 'git diff-tree -z --name-status $t0 $t1 | test-tr "\000" "\012" >current && git diff expected current' cat > expected <<\EOF
t/t3402-rebase-merge.sh +2 -2
@@ -30,11 +30,11 @@ test_expect_success setup ' git update-index --force-remove original && git commit -a -m"side renames and edits." && - tr "[a-z]" "[A-Z]" <original >newfile && + test-tr a-z A-Z <original >newfile && git add newfile && git commit -a -m"side edits further." && - tr "[a-m]" "[A-M]" <original >newfile && + test-tr a-m A-M <original >newfile && rm -f original && git commit -a -m"side edits once again." &&
t/t4004-diff-rename-symlink.sh +1 -1
@@ -14,7 +14,7 @@ by an edit for them. test_expect_success \ 'prepare reference tree' \ - 'echo xyzzy | tr -d '\\\\'012 >yomin && + 'echo xyzzy | test-tr -d "\012" >yomin && ln -s xyzzy frotz && git update-index --add frotz yomin && tree=$(git write-tree) &&
t/t4015-diff-whitespace.sh +3 -3
@@ -51,7 +51,7 @@ test_expect_success "Ray's example with -w" 'git diff expect out' git diff -b > out test_expect_success "Ray's example with -b" 'git diff expect out' -tr 'Q' '\015' << EOF > x +test-tr 'Q' '\015' << EOF > x whitespace at beginning whitespace change whitespace in the middle
@@ -71,7 +71,7 @@ unchanged line CR at end EOF -tr 'Q' '\015' << EOF > expect +test-tr 'Q' '\015' << EOF > expect diff --git a/x b/x index d99af23..8b32fb5 100644 --- a/x
@@ -99,7 +99,7 @@ EOF git diff -w > out test_expect_success 'another test, with -w' 'git diff expect out' -tr 'Q' '\015' << EOF > expect +test-tr 'Q' '\015' << EOF > expect diff --git a/x b/x index d99af23..8b32fb5 100644 --- a/x
t/t4019-diff-wserror.sh +1 -1
@@ -12,7 +12,7 @@ test_expect_success setup ' echo " Eight SP indent" >>F && echo " HT and SP indent" >>F && echo "With trailing SP " >>F && - echo "Carriage ReturnQ" | tr Q "\015" >>F && + echo "Carriage ReturnQ" | test-tr Q "\015" >>F && echo "No problem" >>F '
t/t4020-diff-external.sh +1 -1
@@ -99,7 +99,7 @@ test_expect_success 'no diff with -diff' ' git diff | grep Binary ' -echo NULZbetweenZwords | perl -pe 'y/Z/\000/' > file +echo NULZbetweenZwords | test-tr 'Z' '\000' > file test_expect_success 'force diff with "diff"' ' echo >.gitattributes "file diff" &&
t/t4022-diff-rewrite.sh +1 -1
@@ -8,7 +8,7 @@ test_expect_success setup ' cat ../../COPYING >test && git add test && - tr \ + test-tr \ "abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ" \ "nopqrstuvwxyzabcdefghijklmNOPQRSTUVWXYZABCDEFGHIJKLM" \ <../../COPYING >test
t/t4101-apply-nonl.sh +2 -2
@@ -12,8 +12,8 @@ test_description='git apply should handle files with incomplete lines. (echo a; echo b) >frotz.0 (echo a; echo b; echo c) >frotz.1 -(echo a; echo b | tr -d '\012') >frotz.2 -(echo a; echo c; echo b | tr -d '\012') >frotz.3 +(echo a; echo b | test-tr -d '\012') >frotz.2 +(echo a; echo c; echo b | test-tr -d '\012') >frotz.3 for i in 0 1 2 3 do
t/t4103-apply-binary.sh +2 -2
@@ -24,10 +24,10 @@ git update-index --add --remove file1 file2 file4 git-commit -m 'Initial Version' 2>/dev/null git-checkout -b binary -perl -pe 'y/x/\000/' <file1 >file3 +test-tr x '\000' <file1 >file3 cat file3 >file4 git add file2 -perl -pe 'y/\000/v/' <file3 >file1 +test-tr '\000' v <file3 >file1 rm -f file2 git update-index --add --remove file1 file2 file3 file4 git-commit -m 'Second Version'
t/t4116-apply-reverse.sh +2 -2
@@ -12,14 +12,14 @@ test_description='git apply in reverse test_expect_success setup ' for i in a b c d e f g h i j k l m n; do echo $i; done >file1 && - perl -pe "y/ijk/\\000\\001\\002/" <file1 >file2 && + test-tr "ijk" "\000\001\002" <file1 >file2 && git add file1 file2 && git commit -m initial && git tag initial && for i in a b c g h i J K L m o n p q; do echo $i; done >file1 && - perl -pe "y/mon/\\000\\001\\002/" <file1 >file2 && + test-tr "mon" "\000\001\002" <file1 >file2 && git commit -a -m second && git tag second &&
t/t4118-apply-empty-context.sh +2 -2
@@ -18,13 +18,13 @@ test_expect_success setup ' cat file1 >file1.orig && { cat file1 && - echo Q | tr -d "\\012" + echo Q | test-tr -d "\012" } >file2 && cat file2 >file2.orig git add file1 file2 && sed -e "/^B/d" <file1.orig >file1 && sed -e "/^[BQ]/d" <file2.orig >file2 && - echo Q | tr -d "\\012" >>file2 && + echo Q | test-tr -d "\012" >>file2 && cat file1 >file1.mods && cat file2 >file2.mods && git diff |
t/t4200-rerere.sh +1 -1
@@ -129,7 +129,7 @@ test_expect_success 'rerere kicked in' "! grep ======= a1" test_expect_success 'rerere prefers first change' 'git diff a1 expect' rm $rr/postimage -echo "$sha1 a1" | perl -pe 'y/\012/\000/' > .git/rr-cache/MERGE_RR +echo "$sha1 a1" | test-tr '\012' '\000' > .git/rr-cache/MERGE_RR test_expect_success 'rerere clear' 'git rerere clear'
t/t4201-shortlog.sh +2 -2
@@ -19,12 +19,12 @@ git commit --quiet -m "This is a very, very long first line for the commit messa # test if the wrapping is still valid when replacing all i's by treble clefs. echo 3 > a1 -git commit --quiet -m "$(echo "This is a very, very long first line for the commit message to see if it is wrapped correctly" | sed "s/i/1234/g" | tr 1234 '\360\235\204\236')" a1 +git commit --quiet -m "$(echo "This is a very, very long first line for the commit message to see if it is wrapped correctly" | sed "s/i/1234/g" | test-tr 1234 '\360\235\204\236')" a1 # now fsck up the utf8 git config i18n.commitencoding non-utf-8 echo 4 > a1 -git commit --quiet -m "$(echo "This is a very, very long first line for the commit message to see if it is wrapped correctly" | sed "s/i/1234/g" | tr 1234 '\370\235\204\236')" a1 +git commit --quiet -m "$(echo "This is a very, very long first line for the commit message to see if it is wrapped correctly" | sed "s/i/1234/g" | test-tr 1234 '\370\235\204\236')" a1 echo 5 > a1 git commit --quiet -m "a 12 34 56 78" a1
t/t5300-pack-object.sh +1 -1
@@ -15,7 +15,7 @@ test_expect_success \ 'rm -f .git/index* for i in a b c do - dd if=/dev/zero bs=4k count=1 | perl -pe "y/\\000/$i/" >$i && + dd if=/dev/zero bs=4k count=1 | test-tr "\000" "$i" >$i && git update-index --add $i || return 1 done && cat c >d && echo foo >>d && git update-index --add d &&
t/t5500-fetch-pack.sh +1 -1
@@ -37,7 +37,7 @@ add () { } count_objects () { - ls .git/objects/??/* 2>>log2.txt | wc -l | tr -d " " + ls .git/objects/??/* 2>>log2.txt | wc -l | test-tr -d " " } test_expect_object_count () {
t/t5505-remote.sh +2 -2
@@ -22,8 +22,8 @@ setup_repository () { } tokens_match () { - echo "$1" | tr ' ' '\012' | sort | sed -e '/^$/d' >expect && - echo "$2" | tr ' ' '\012' | sort | sed -e '/^$/d' >actual && + echo "$1" | test-tr ' ' '\012' | sort | sed -e '/^$/d' >expect && + echo "$2" | test-tr ' ' '\012' | sort | sed -e '/^$/d' >actual && test_cmp expect actual }
t/t6003-rev-list-topo-order.sh +1 -1
@@ -79,7 +79,7 @@ save_tag g4 unique_commit g6 tree -p g3 -p h2 git update-ref HEAD $(tag l5) -test_output_expect_success 'rev-list has correct number of entries' 'git rev-list HEAD | wc -l | tr -d \" \"' <<EOF +test_output_expect_success 'rev-list has correct number of entries' 'git rev-list HEAD | wc -l | test-tr -d \" \"' <<EOF 19 EOF
t/t7003-filter-branch.sh +1 -1
@@ -4,7 +4,7 @@ test_description='git-filter-branch' . ./test-lib.sh make_commit () { - lower=$(echo $1 | tr '[A-Z]' '[a-z]') + lower=$(echo $1 | test-tr 'A-Z' 'a-z') echo $lower > $lower git add $lower test_tick
t/t9200-git-cvsexportcommit.sh +1 -1
@@ -35,7 +35,7 @@ check_entries () { then >expected else - printf '%s\n' "$2" | tr '|' '\012' >expected + printf '%s\n' "$2" | test-tr '|' '\012' >expected fi test_cmp expected actual }
t/test-lib.sh +7 -5
@@ -44,11 +44,16 @@ export GIT_COMMITTER_EMAIL GIT_COMMITTER_NAME export EDITOR VISUAL GIT_TEST_CMP=${GIT_TEST_CMP:-diff -u} +# Test the binaries we have just built. The tests are kept in +# t/ subdirectory and are run in trash subdirectory. +PATH=$(pwd)/..:$PATH +export PATH + # Protect ourselves from common misconfiguration to export # CDPATH into the environment unset CDPATH -case $(echo $GIT_TRACE |tr "[A-Z]" "[a-z]") in +case $(echo $GIT_TRACE |test-tr A-Z a-z) in 1|2|true) echo "* warning: Some tests will not work if GIT_TRACE" \ "is set as to trace on STDERR ! *"
@@ -369,16 +374,13 @@ test_done () { esac } -# Test the binaries we have just built. The tests are kept in -# t/ subdirectory and are run in trash subdirectory. -PATH=$(pwd)/..:$PATH GIT_EXEC_PATH=$(pwd)/.. GIT_TEMPLATE_DIR=$(pwd)/../templates/blt unset GIT_CONFIG unset GIT_CONFIG_LOCAL GIT_CONFIG_NOSYSTEM=1 GIT_CONFIG_NOGLOBAL=1 -export PATH GIT_EXEC_PATH GIT_TEMPLATE_DIR GIT_CONFIG_NOSYSTEM GIT_CONFIG_NOGLOBAL +export GIT_EXEC_PATH GIT_TEMPLATE_DIR GIT_CONFIG_NOSYSTEM GIT_CONFIG_NOGLOBAL GITPERLLIB=$(pwd)/../perl/blib/lib:$(pwd)/../perl/blib/arch/auto/Git export GITPERLLIB
Signature · 1 line
1.5.5.rc0.53.g97734
Re: [PATCH 03/16] more tr portability test script fixes
In reply to Alex RiesenQuoted 10 lines · Alex Riesen
Jeff King, Wed, Mar 12, 2008 22:31:06 +0100:
t/diff-lib.sh compare_diff_raw_z () { · line 25
Quoted 4 lines
t/diff-lib.sh +2
-2- tr '\000' '\012' <"$1" | sed -e "$sanitize_diff_raw_z" >.tmp-1 - tr '\000' '\012' <"$2" | sed -e "$sanitize_diff_raw_z" >.tmp-2 + perl -pe 'y/\000/\012/' <"$1" | sed -e "$sanitize_diff_raw_z" >.tmp-1 + perl -pe 'y/\000/\012/' <"$2" | sed -e "$sanitize_diff_raw_z" >.tmp-2These break in presence of ActiveState Perl on Windows.
I suggest replacing such simple construction with a simplified, in-tree, version of tr.
<sigh> It's sad that it must come to that, but your test-tr patches seem like the only sane choice. They seem to work fine on my Solaris box.
Note that there are still a few uses of 'tr' in actual git scripts. However, they are pretty tame, so I think they should work everywhere. Otherwise, test-tr must become "git tr". :)
-Peff
Re: [PATCH 03/16] more tr portability test script fixes
Jeff King, Tue, Mar 18, 2008 23:44:37 +0100:
Quoted 10 lines · Alex Riesen
Quoted 9 lines
Jeff King, Wed, Mar 12, 2008 22:31:06 +0100:
t/diff-lib.sh compare_diff_raw_z () { · line 25
Quoted 4 lines
t/diff-lib.sh +2
-2- tr '\000' '\012' <"$1" | sed -e "$sanitize_diff_raw_z" >.tmp-1 - tr '\000' '\012' <"$2" | sed -e "$sanitize_diff_raw_z" >.tmp-2 + perl -pe 'y/\000/\012/' <"$1" | sed -e "$sanitize_diff_raw_z" >.tmp-1 + perl -pe 'y/\000/\012/' <"$2" | sed -e "$sanitize_diff_raw_z" >.tmp-2These break in presence of ActiveState Perl on Windows.
Quoted 8 lines
Quoted 2 lines
I suggest replacing such simple construction with a simplified, in-tree, version of tr.
<sigh> It's sad that it must come to that, but your test-tr patches seem like the only sane choice. They seem to work fine on my Solaris box.
Note that there are still a few uses of 'tr' in actual git scripts. However, they are pretty tame, so I think they should work everywhere.
Yes, I thought about them too and hope for the same (they'll do).
Quoted 1 line
Otherwise, test-tr must become "git tr". :)
We already use "git diff" and "git apply" as reliable "diff" and "patch" :)
Re: [PATCH 03/16] more tr portability test script fixes
In reply to Jeff KingQuoted 8 lines · Jeff King
Quoted 6 lines · Alex Riesen
Jeff King, Wed, Mar 12, 2008 22:31:06 +0100:
Quoted 4 lines
t/diff-lib.sh +2
-2- tr '\000' '\012' <"$1" | sed -e "$sanitize_diff_raw_z" >.tmp-1 - tr '\000' '\012' <"$2" | sed -e "$sanitize_diff_raw_z" >.tmp-2 + perl -pe 'y/\000/\012/' <"$1" | sed -e "$sanitize_diff_raw_z" >.tmp-1 + perl -pe 'y/\000/\012/' <"$2" | sed -e "$sanitize_diff_raw_z" >.tmp-2
Quoted 7 lines
Quoted 4 lines
These break in presence of ActiveState Perl on Windows.
I suggest replacing such simple construction with a simplified, in-tree, version of tr.
<sigh> It's sad that it must come to that, but your test-tr patches seem like the only sane choice. They seem to work fine on my Solaris box.
I am very tempted to say that it might make more sense to declare ActiveState unsupported. How many times have we suffered from it, and notice it was only from Alex every time, nobody else? Are there silent majorities involved here?
Re: [PATCH 03/16] more tr portability test script fixes
Junio C Hamano, Wed, Mar 19, 2008 22:40:57 +0100:
Quoted 12 lines
Quoted 10 lines · Jeff King
Quoted 8 lines · Alex Riesen
Jeff King, Wed, Mar 12, 2008 22:31:06 +0100:
t/diff-lib.sh compare_diff_raw_z () { · line 25
Quoted 4 lines
t/diff-lib.sh +2
-2- tr '\000' '\012' <"$1" | sed -e "$sanitize_diff_raw_z" >.tmp-1 - tr '\000' '\012' <"$2" | sed -e "$sanitize_diff_raw_z" >.tmp-2 + perl -pe 'y/\000/\012/' <"$1" | sed -e "$sanitize_diff_raw_z" >.tmp-1 + perl -pe 'y/\000/\012/' <"$2" | sed -e "$sanitize_diff_raw_z" >.tmp-2These break in presence of ActiveState Perl on Windows.
Quoted 8 lines
Quoted 5 lines
Quoted 2 lines
I suggest replacing such simple construction with a simplified, in-tree, version of tr.
<sigh> It's sad that it must come to that, but your test-tr patches seem like the only sane choice. They seem to work fine on my Solaris box.
I am very tempted to say that it might make more sense to declare ActiveState unsupported.
I am not asking to really support it, it is too offending a thought. I am just asking to consider the patch, which removes another dependency on local system. Wasn't there issues with tr already?
Quoted 2 lines
How many times have we suffered from it, and notice it was only from Alex every time, nobody else?
Well, I say that every time. I think by know everyone, who seen a patch from me, noticed how very specially stupid my kind of setup is.
BTW, who suffered from what?
Quoted 1 line
Are there silent majorities involved here?
Yeah, the windows-people. They are used to silently workaround their system.