git-svn test suite failures due to Subversion race

PATCH Michael Spang 4 participants 7 messages Jun 15, 2016

Share & tools

Michael Spang

author

2016-06-15 22:42 UTC #0 lore raw

Some of the git-svn tests can fail on fast machines due to a race in
Subversion: if a file is modified in the same second it was checked out (or in for that matter), Subversion will not consider it modified. Apparently there is also a "racy Subversion" problem parallel to the "racy-git" problem. The machine is an Athlon 64 X2 3800+.

For example, test #3 of t9106-git-svn-commit-diff-clobber.sh will fail if Subversion happens to fail to make any commit in test #2 of the same file. Test #2 will silently pass if no commit was made, as it is not an error to commit nothing. The commit in #3 is supposed to conflict with the one from #2, but obviously won't if that commit wasn't made. So in this case test #3's commit succeeds when failure is expected, and the test fails. The [annotated] output of a test run where this happens is attached. A couple of other tests have the same problem.

This may be a known issue, but my search of the archives was fruitless and it doesn't appear to be documented anywhere. It's obviously one that would need to ultimately be fixed in Subversion, although a workaround in the test suite might help those whose builds depend on a passing test suite. It's problematic for me to have the git test suite fail because the git package for Debian runs the test suite while building, and will abort the build if there are failures.

Until this race is fixed in Subversion I guess I'm stuck either skipping the git-svn tests or inserting sleep 1 in a few places to work around it. A patch that works around this problem in all of the tests that failed for me is attached in case its useful to someone. Even faster machines may reveal more instances of the problem. I did not attempt to "fix" any tests that will still pass if commits are lost.

Cheers,
Michael Spang

Quoted 1 line

From 97e90fcd7cf600726ec468016eb37d1e1de38dde Mon Sep 17 00:00:00 2001

From: Michael Spang <[email protected]>
Date: Sun, 11 Feb 2007 20:56:22 -0500
Subject: [PATCH] Work around Subversion race in git-svn tests.

Signed-off-by Michael Spang <[email protected]>

---
 t/t9103-git-svn-graft-branches.sh      |    1 +
 t/t9106-git-svn-commit-diff-clobber.sh |    2 ++
 2 files changed, 3 insertions(+), 0 deletions(-)

t/t9103-git-svn-graft-branches.sh +1 -0

index 4e55778..f90f3a8 100755
@@ -14,6 +14,7 @@ test_expect_success 'initialize repo' "
 	svn cp -m 'tag a' $svnrepo/trunk $svnrepo/tags/a &&
 	svn cp -m 'branch a' $svnrepo/trunk $svnrepo/branches/a &&
 	svn co $svnrepo wc &&
+	sleep 1 &&
 	cd wc &&
 	echo feedme >> branches/a/readme &&
 	svn commit -m hungry &&

t/t9106-git-svn-commit-diff-clobber.sh +2 -0

index 59b6425..70c9f58 100755
@@ -16,6 +16,7 @@ test_expect_success 'initialize repo' "
 	"
 test_expect_success 'commit change from svn side' "
 	svn co $svnrepo t.svn &&
+	sleep 1 &&
 	cd t.svn &&
 	echo second line from svn >> file &&
 	svn commit -m 'second line from svn' &&

@@ -43,6 +44,7 @@ test_expect_failure 'dcommit fails to commit because of conflict' "
 	git-svn fetch &&
 	git reset --hard refs/remotes/git-svn &&
 	svn co $svnrepo t.svn &&
+	sleep 1 &&
 	cd t.svn &&
 	echo fourth line from svn >> file &&
 	svn commit -m 'fourth line from svn' &&

Signature · 1 line

1.5.0.rc4.gcc46a

001 $ sh t9106-git-svn-commit-diff-clobber.sh -v
002 * expecting success:

003            mkdir import &&
004            cd import &&
005            echo initial > file &&
006            svn import -m 'initial' . file:///users/mspang/src/git/git/t/trash/svnrepo &&
007            cd .. &&
008            echo initial > file &&
009            git update-index --add file &&
010            git commit -a -m 'initial'

011
012 Adding file
013
014 Committed revision 1.
015 Created initial commit 39d53013acc58c2675a4d856da562282c21f5235
016 1 files changed, 1 insertions(+), 0 deletions(-)
017 create mode 100644 file
018 * ok 1: initialize repo
019
020 * expecting success:

021            svn co file:///users/mspang/src/git/git/t/trash/svnrepo t.svn &&
022            cd t.svn &&
023            echo second line from svn >> file &&
024            svn commit -m 'second line from svn' &&
025            cd .. &&
026            rm -rf t.svn

027
028 A t.svn/file
029 Checked out revision 1.

[missing]   Sending        file
[missing]   Transmitting file data .
[missing]   Committed revision 2.

030 * ok 2: commit change from svn side
031
032 * expecting failure:

033            echo second line from git >> file &&
034            git commit -a -m 'second line from git' &&
035            git-svn commit-diff -r1 HEAD~1 HEAD file:///users/mspang/src/git/git/t/trash/svnrepo

036
037 Created commit eafa51760d91bfcfa1ce7e4f18bd0854cc723333
038 1 files changed, 1 insertions(+), 0 deletions(-)
039 diff-tree HEAD~1 HEAD
040 M file
041 Committed 2
042 * FAIL 3: commit conflicting change from git

[fails due to last missing commit]

043

044                    echo second line from git >> file &&
045                    git commit -a -m 'second line from git' &&
046                    git-svn commit-diff -r1 HEAD~1 HEAD file:///users/mspang/src/git/git/t/trash/svnrepo
047
048
049    * expecting success:
050            git reset --hard HEAD~1 &&
051            echo second line from svn >> file &&
052            git commit -a -m 'second line from svn' &&
053            echo third line from git >> file &&
054            git commit -a -m 'third line from git' &&
055            git-svn commit-diff -r2 HEAD~1 HEAD file:///users/mspang/src/git/git/t/trash/svnrepo

056
057 HEAD is now at 39d5301... initial
058 Created commit c8ce91056ec7f8082bc1f86f5233b3c73f952ff5
059 1 files changed, 1 insertions(+), 0 deletions(-)
060 Created commit c6cfbf71727661a3ca3d25144fe0c579ff8e02fb
061 1 files changed, 1 insertions(+), 0 deletions(-)
062 diff-tree HEAD~1 HEAD
063 M file
064 Committed 3
065 * ok 4: commit complementing change from git
066
067 * expecting failure:

068            git-svn init file:///users/mspang/src/git/git/t/trash/svnrepo &&
069            git-svn fetch &&
070            git reset --hard refs/remotes/git-svn &&
071            svn co file:///users/mspang/src/git/git/t/trash/svnrepo t.svn &&
072            cd t.svn &&
073            echo fourth line from svn >> file &&
074            svn commit -m 'fourth line from svn' &&
075            cd .. &&
076            rm -rf t.svn &&
077            echo 'fourth line from git' >> file &&
078            git commit -a -m 'fourth line from git' &&
079            git-svn dcommit

080
081 Fetching git-svn
082 r0 = 103011a5e58371b0034e4655e82bbda686e3e60b
083 A file
084 r1 = 2555b1052c176cd6de86b775797b9e7924c1ea04
085 M file
086 r2 = 3e520e5a40bda6e97c37e1ae37dbb0583608e7e1
087 M file

088    r3 = e47289f12d30749c4b25762d402cc1762afbdd01
089    HEAD is now at e47289f... third line from git
090    A    t.svn/file
091    Checked out revision 3.

[missing]   Sending        file
[missing]   Transmitting file data .
[missing]   Committed revision 4.

092 Created commit b5ccb76480136befc4d702e1c8764897162cbc72
093 1 files changed, 1 insertions(+), 0 deletions(-)
094 diff-tree b5ccb76480136befc4d702e1c8764897162cbc72~1 b5ccb76480136befc4d702e1c8764897162cbc72
095 M file
096 Committed 4
097 M file

098    r4 = acaf5f49b02286596bd8f27139bf374082c7afaf
099    No changes between current HEAD and refs/remotes/git-svn
100    Resetting to the latest refs/remotes/git-svn
101    * FAIL 5: dcommit fails to commit because of conflict

[fails due to last missing commit]

102

103                    git-svn init file:///users/mspang/src/git/git/t/trash/svnrepo &&
104                    git-svn fetch &&
105                    git reset --hard refs/remotes/git-svn &&
106                    svn co file:///users/mspang/src/git/git/t/trash/svnrepo t.svn &&
107                   cd t.svn &&
108                    echo fourth line from svn >> file &&
109                    svn commit -m 'fourth line from svn' &&
110                    cd .. &&
111                    rm -rf t.svn &&
112                    echo 'fourth line from git' >> file &&
113                    git commit -a -m 'fourth line from git' &&
114                    git-svn dcommit
115
116
117    * expecting success:
118            git reset --hard refs/remotes/git-svn &&
119            echo 'index merge' > file2 &&
120            git update-index --add file2 &&
121            git commit -a -m 'index merge' &&
122            echo 'more changes' >> file2 &&
123            git update-index file2 &&
124            git commit -a -m 'more changes' &&
125            git-svn dcommit

126
127 HEAD is now at acaf5f4... fourth line from git
128 Created commit 4f898d08f8011241134e983cc30d98785e926f13
129 1 files changed, 1 insertions(+), 0 deletions(-)
130 create mode 100644 file2
131 Created commit 59cf4e5baa0fd3d444d3a084d5e8a87328ddfcae
132 1 files changed, 1 insertions(+), 0 deletions(-)
133 diff-tree 4f898d08f8011241134e983cc30d98785e926f13~1 4f898d08f8011241134e983cc30d98785e926f13
134 A file2
135 Committed 5
136 diff-tree 59cf4e5baa0fd3d444d3a084d5e8a87328ddfcae~1 59cf4e5baa0fd3d444d3a084d5e8a87328ddfcae
137 M file2
138 Committed 6
139 A file2
140 r5 = caed42b00cded960103b70a51b4c81194f94c960
141 M file2

142    r6 = 8e8a98024de9b2469e94815fea87492728ef1fc4
143    No changes between current HEAD and refs/remotes/git-svn
144    Resetting to the latest refs/remotes/git-svn
145    *   ok 6: dcommit does the svn equivalent of an index merge

146
147 * failed 2 among 6 test(s)

Eric Wong

2016-06-15 22:42 UTC #1 lore raw

Quoted 5 lines · Michael Spang

Some of the git-svn tests can fail on fast machines due to a race in
Subversion: if a file is modified in the same second it was checked out (or in for that matter), Subversion will not consider it modified. Apparently there is also a "racy Subversion" problem parallel to the "racy-git" problem. The machine is an Athlon 64 X2 3800+.

I don't think any of my machines are even half as fast. The git-svn tests take a long time on my dev machine, so we have entirely different issues.

Quoted 9 lines

For example, test #3 of t9106-git-svn-commit-diff-clobber.sh will fail if Subversion happens to fail to make any commit in test #2 of the same file. Test #2 will silently pass if no commit was made, as it is not an error to commit nothing. The commit in #3 is supposed to conflict with the one from #2, but obviously won't if that commit wasn't made. So in this case test #3's commit succeeds when failure is expected, and the test fails. The [annotated] output of a test run where this happens is attached. A couple of other tests have the same problem.

Quoted 7 lines

This may be a known issue, but my search of the archives was fruitless and it doesn't appear to be documented anywhere. It's obviously one that would need to ultimately be fixed in Subversion, although a workaround in the test suite might help those whose builds depend on a passing test suite. It's problematic for me to have the git test suite fail because the git package for Debian runs the test suite while building, and will abort the build if there are failures.

Not known to me. This is the first time I've heard of this issue; but I'm not at all surprised that this is happening, though.

Quoted 6 lines

Until this race is fixed in Subversion I guess I'm stuck either skipping the git-svn tests or inserting sleep 1 in a few places to work around it. A patch that works around this problem in all of the tests that failed for me is attached in case its useful to someone. Even faster machines may reveal more instances of the problem. I did not attempt to "fix" any tests that will still pass if commits are lost.

This is disconcerting. Given that hardware is still getting faster, I suspect there will be many more problems with the svn tests in the future. I have no plans for upgrading hardware in the near future; so I won't be hitting these problems myself.

I'm alright with adding the sleep 1 in several more places where this can be an issue. If it gets bad enough for people with slower computers, I'll probably just create a function that sleeps only if a variable is not set (TOO_SLOW_TO_RACE=1 :)

I've been considering rewriting the tests to use the Perl SVN::
libraries exclusively; but that runs the risk of introducing new bugs.

Quoted 6 lines

From 97e90fcd7cf600726ec468016eb37d1e1de38dde Mon Sep 17 00:00:00 2001
From: Michael Spang <[email protected]>
Date: Sun, 11 Feb 2007 20:56:22 -0500
Subject: [PATCH] Work around Subversion race in git-svn tests.
Signed-off-by Michael Spang <[email protected]>

It would've been good to have the above email explanation above in the commit message below so we know why the patch was needed (when it gets applied).

Signature · 1 line

Eric Wong

Michael Spang

2016-06-15 22:42 UTC #2 lore raw

Quoted 5 lines · Eric Wong

This is disconcerting. Given that hardware is still getting faster, I suspect there will be many more problems with the svn tests in the future. I have no plans for upgrading hardware in the near future; so I won't be hitting these problems myself.

Quoted 8 lines

I'm alright with adding the sleep 1 in several more places where this can be an issue. If it gets bad enough for people with slower computers, I'll probably just create a function that sleeps only if a variable is not set (TOO_SLOW_TO_RACE=1 :)
I've been considering rewriting the tests to use the Perl SVN::
libraries exclusively; but that runs the risk of introducing new bugs.

I've prepared a somewhat cleaner patch that solves the problem without using sleep. It increases the modification time by one second immediately prior to each commit. I added calls everywhere I think the problem could potentially occur.

I've had no failures with the patch (without it, they are quite frequent).

Cheers,
Michael Spang

Quoted 1 line

From f3a86250469df6607b71da235ac69ccb82d59fd9 Mon Sep 17 00:00:00 2001

From: Michael Spang <[email protected]>
Date: Mon, 12 Feb 2007 19:33:37 -0500
Subject: [PATCH] Work around Subversion race in git-svn tests.

Some of the git-svn tests can fail on fast machines due to a race in
Subversion: if a file is modified in the same second it was checked out (or in for that matter), Subversion will not consider it modified. This works around the problem by increasing the timestamp by one second before each commit.

Signed-off-by Michael Spang <[email protected]>

---
 t/lib-git-svn.sh                       |    3 +++
 t/t9101-git-svn-props.sh               |    7 +++++--
 t/t9103-git-svn-graft-branches.sh      |    1 +
 t/t9104-git-svn-follow-parent.sh       |    2 ++
 t/t9106-git-svn-commit-diff-clobber.sh |    2 ++
 5 files changed, 13 insertions(+), 2 deletions(-)

t/lib-git-svn.sh +3 -0

index bb1d7b8..71e8010 100644
@@ -45,3 +45,6 @@ fi
 svnrepo="file://$svnrepo"
 
 
+poke() {
+	touch -r "$1" -d +1sec "$1"
+}

t/t9101-git-svn-props.sh +5 -2

index a2c4dc3..e8133d8 100755
@@ -56,11 +56,14 @@ test_expect_success 'checkout working copy from svn' "svn co $svnrepo test_wc"
 test_expect_success 'setup some commits to svn' \
 	'cd test_wc &&
 		echo Greetings >> kw.c &&
+		poke kw.c &&
 		svn commit -m "Not yet an Id" &&
 		echo Hello world >> kw.c &&
+		poke kw.c &&
 		svn commit -m "Modified file, but still not yet an Id" &&
 		svn propset svn:keywords Id kw.c &&
-		svn commit -m "Propset Id"
+		poke kw.c &&
+		svn commit -m "Propset Id" &&
 	cd ..'
 
 test_expect_success 'initialize git-svn' "git-svn init $svnrepo"

@@ -83,7 +86,7 @@ test_expect_success "propset CR on crlf files" \
 		svn propset svn:eol-style CR empty &&
 		svn propset svn:eol-style CR crlf &&
 		svn propset svn:eol-style CR ne_crlf &&
-		svn commit -m "propset CR on crlf files"
+		svn commit -m "propset CR on crlf files" &&
 	 cd ..'
 
 test_expect_success 'fetch and pull latest from svn and checkout a new wc' \

t/t9103-git-svn-graft-branches.sh +1 -0

index 4e55778..183ae3b 100755
@@ -16,6 +16,7 @@ test_expect_success 'initialize repo' "
 	svn co $svnrepo wc &&
 	cd wc &&
 	echo feedme >> branches/a/readme &&
+	poke branches/a/readme &&
 	svn commit -m hungry &&
 	cd trunk &&
 	svn merge -r3:4 $svnrepo/branches/a &&

t/t9104-git-svn-follow-parent.sh +2 -0

index 8d2e2fe..405b555 100755
@@ -16,11 +16,13 @@ test_expect_success 'initialize repo' "
 	svn co $svnrepo wc &&
 	cd wc &&
 	echo world >> trunk/readme &&
+	poke trunk/readme &&
 	svn commit -m 'another commit' &&
 	svn up &&
 	svn mv -m 'rename to thunk' trunk thunk &&
 	svn up &&
 	echo goodbye >> thunk/readme &&
+	poke thunk/readme &&
 	svn commit -m 'bye now' &&
 	cd ..
 	"

t/t9106-git-svn-commit-diff-clobber.sh +2 -0

index 59b6425..6f132f2 100755
@@ -18,6 +18,7 @@ test_expect_success 'commit change from svn side' "
 	svn co $svnrepo t.svn &&
 	cd t.svn &&
 	echo second line from svn >> file &&
+	poke file &&
 	svn commit -m 'second line from svn' &&
 	cd .. &&
 	rm -rf t.svn

@@ -45,6 +46,7 @@ test_expect_failure 'dcommit fails to commit because of conflict' "
 	svn co $svnrepo t.svn &&
 	cd t.svn &&
 	echo fourth line from svn >> file &&
+	poke file &&
 	svn commit -m 'fourth line from svn' &&
 	cd .. &&
 	rm -rf t.svn &&

Signature · 1 line

1.5.0.rc4.26.gcc46a

Eric Wong

Acked-by

2016-06-15 22:42 UTC #3 lore raw

Quoted 6 lines · Michael Spang

I've prepared a somewhat cleaner patch that solves the problem without using sleep. It increases the modification time by one second immediately prior to each commit. I added calls everywhere I think the problem could potentially occur.
I've had no failures with the patch (without it, they are quite frequent).

Thanks for the patch. Just one portability nit below, otherwise:

Acked-by Eric Wong <[email protected]>

t/lib-git-svn.sh fi · line 50

Quoted 3 lines

t/lib-git-svn.sh +3 -0

+poke() {
+	touch -r "$1" -d +1sec "$1"
+}

I'm not sure that -d is portable. However,

perl -e '@x = stat($ARGV[0]); utime($x[8], $x[9] + 1, $ARGV[0])' "$1"

should work on any platform git-svn runs on.

Signature · 1 line

Eric Wong

Michael Spang

Acked-by Eric Wong

2016-06-15 22:42 UTC #4 lore raw

Quoted 21 lines · Eric Wong

Quoted 7 lines · Michael Spang
I've prepared a somewhat cleaner patch that solves the problem without using sleep. It increases the modification time by one second immediately prior to each commit. I added calls everywhere I think the problem could potentially occur.
I've had no failures with the patch (without it, they are quite frequent).
Thanks for the patch. Just one portability nit below, otherwise:
Acked-by Eric Wong <[email protected]>
Review
Quoted 4 lines
(quoted diff) +3 -0
+poke() {
+	touch -r "$1" -d +1sec "$1"
+}
I'm not sure that -d is portable. However,

Quoted 5 lines

perl -e '@x = stat($ARGV[0]); utime($x[8], $x[9] + 1, $ARGV[0])' "$1"
should work on any platform git-svn runs on.

Doh. I knew that option to touch was too good to be true. Having it in Perl may also make it easier for you if you decide to go ahead with rewriting the tests to use only the Perl bindings. Assuming a workaround is still needed, anyway--I've never used Subversion with Perl. Thanks for the tips.

Michael Spang

Quoted 1 line

From e632a69547531eb3459a55513ebc2e020b19eb57 Mon Sep 17 00:00:00 2001

From: Michael Spang <[email protected]>
Date: Mon, 12 Feb 2007 19:33:37 -0500
Subject: [PATCH] Work around Subversion race in git-svn tests.

Some of the git-svn tests can fail on fast machines due to a race in
Subversion: if a file is modified in the same second it was checked out (or in for that matter), Subversion will not consider it modified. This works around the problem by increasing the timestamp by one second before each commit.

Signed-off-by Michael Spang <[email protected]>

Acked-by Eric Wong <[email protected]>

---
 t/lib-git-svn.sh                       |    3 +++
 t/t9101-git-svn-props.sh               |    7 +++++--
 t/t9103-git-svn-graft-branches.sh      |    1 +
 t/t9104-git-svn-follow-parent.sh       |    2 ++
 t/t9106-git-svn-commit-diff-clobber.sh |    2 ++
 5 files changed, 13 insertions(+), 2 deletions(-)

t/lib-git-svn.sh +3 -0

index bb1d7b8..67d08cf 100644
@@ -45,3 +45,6 @@ fi
 svnrepo="file://$svnrepo"
 
 
+poke() {
+	perl -e '@x = stat($ARGV[0]); utime($x[8], $x[9] + 1, $ARGV[0])' "$1"
+}

t/t9101-git-svn-props.sh +5 -2

index a2c4dc3..e8133d8 100755
@@ -56,11 +56,14 @@ test_expect_success 'checkout working copy from svn' "svn co $svnrepo test_wc"
 test_expect_success 'setup some commits to svn' \
 	'cd test_wc &&
 		echo Greetings >> kw.c &&
+		poke kw.c &&
 		svn commit -m "Not yet an Id" &&
 		echo Hello world >> kw.c &&
+		poke kw.c &&
 		svn commit -m "Modified file, but still not yet an Id" &&
 		svn propset svn:keywords Id kw.c &&
-		svn commit -m "Propset Id"
+		poke kw.c &&
+		svn commit -m "Propset Id" &&
 	cd ..'
 
 test_expect_success 'initialize git-svn' "git-svn init $svnrepo"

@@ -83,7 +86,7 @@ test_expect_success "propset CR on crlf files" \
 		svn propset svn:eol-style CR empty &&
 		svn propset svn:eol-style CR crlf &&
 		svn propset svn:eol-style CR ne_crlf &&
-		svn commit -m "propset CR on crlf files"
+		svn commit -m "propset CR on crlf files" &&
 	 cd ..'
 
 test_expect_success 'fetch and pull latest from svn and checkout a new wc' \

t/t9103-git-svn-graft-branches.sh +1 -0

index 4e55778..183ae3b 100755
@@ -16,6 +16,7 @@ test_expect_success 'initialize repo' "
 	svn co $svnrepo wc &&
 	cd wc &&
 	echo feedme >> branches/a/readme &&
+	poke branches/a/readme &&
 	svn commit -m hungry &&
 	cd trunk &&
 	svn merge -r3:4 $svnrepo/branches/a &&

t/t9104-git-svn-follow-parent.sh +2 -0

index 8d2e2fe..405b555 100755
@@ -16,11 +16,13 @@ test_expect_success 'initialize repo' "
 	svn co $svnrepo wc &&
 	cd wc &&
 	echo world >> trunk/readme &&
+	poke trunk/readme &&
 	svn commit -m 'another commit' &&
 	svn up &&
 	svn mv -m 'rename to thunk' trunk thunk &&
 	svn up &&
 	echo goodbye >> thunk/readme &&
+	poke thunk/readme &&
 	svn commit -m 'bye now' &&
 	cd ..
 	"

t/t9106-git-svn-commit-diff-clobber.sh +2 -0

index 59b6425..6f132f2 100755
@@ -18,6 +18,7 @@ test_expect_success 'commit change from svn side' "
 	svn co $svnrepo t.svn &&
 	cd t.svn &&
 	echo second line from svn >> file &&
+	poke file &&
 	svn commit -m 'second line from svn' &&
 	cd .. &&
 	rm -rf t.svn

@@ -45,6 +46,7 @@ test_expect_failure 'dcommit fails to commit because of conflict' "
 	svn co $svnrepo t.svn &&
 	cd t.svn &&
 	echo fourth line from svn >> file &&
+	poke file &&
 	svn commit -m 'fourth line from svn' &&
 	cd .. &&
 	rm -rf t.svn &&

Signature · 1 line

1.5.0.rc4.26.gcc46a

Junio C Hamano

2016-06-15 22:42 UTC #5 lore raw

In reply to Eric Wong

Quoted 9 lines · Eric Wong

Thanks for the patch. Just one portability nit below, otherwise:
Acked-by Eric Wong <[email protected]>
t/lib-git-svn.sh fi · line 50
Quoted 3 lines
t/lib-git-svn.sh +3 -0
+poke() {
+	touch -r "$1" -d +1sec "$1"
+}
I'm not sure that -d is portable. However,

Quoted 3 lines

perl -e '@x = stat($ARGV[0]); utime($x[8], $x[9] + 1, $ARGV[0])' "$1"
should work on any platform git-svn runs on.

Sounds sensible, as git-svn wants perl-svn libraries already anyway.

Junio C Hamano

2016-06-15 22:42 UTC #6 lore raw

In reply to Michael Spang

Quoted 2 lines · Michael Spang

I've had no failures with the patch (without it, they are quite frequent).

I've seen the failures myself without the patch, and working it around is a good thing, however...

t/lib-git-svn.sh fi · line 50

Quoted 10 lines

Cheers,
Michael Spang
From f3a86250469df6607b71da235ac69ccb82d59fd9 Mon Sep 17 00:00:00 2001
From: Michael Spang <[email protected]>
Date: Mon, 12 Feb 2007 19:33:37 -0500
Subject: [PATCH] Work around Subversion race in git-svn tests.
t/lib-git-svn.sh +3 -0
+poke() {
+	touch -r "$1" -d +1sec "$1"
+}

I wonder how portable this might be.