ref-filter: initialize empty name or email fields
PATCH 1/1 ref-filter: initialize empty name or email fields
Formatting $(taggername) on headerless tags such as v0.99 in Git causes a SIGABRT with error "munmap_chunk(): invalid pointer", because of an oversight in commit f0062d3b74 (ref-filter: free item->value and item->value->s, 2018-10-19).
---
If I understand correctly, such tags cannot be produced normally anymore. Therefore I'm unsure how to make tests, and if that is even warranted.
ref-filter.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)
ref-filter.c +3 -3
@@ -1028,7 +1028,7 @@ static const char *copy_name(const char *buf)if (!strncmp (cp," <" ,2 ))return xmemdupz (buf, cp - buf); } -return "" ; +return xstrdup ("" ); }static const char *copy_email (const char *buf)
@@ -1036,10 +1036,10 @@ static const char *copy_email(const char *buf)const char *email =strchr (buf,'<' );const char *eoemail;if (!email) -return "" ; +return xstrdup ("" ); eoemail =strchr (email,'>' );if (!eoemail) -return "" ; +return xstrdup ("" );return xmemdupz (email, eoemail +1 - email); }
Signature · 1 line
2.23.0
Quoted 9 lines · Mischa POSLAWSKY
Formatting $(taggername) on headerless tags such as v0.99 in Git causes a SIGABRT with error "
munmap_chunk(): invalid pointer", because of an oversight in commitf0062d3b74(ref-filter: freeitem->valueanditem->value->s, 2018-10-19).Signed-off-by Mischa POSLAWSKY <[email protected]>---
If I understand correctly, such tags cannot be produced normally anymore. Therefore I'm unsure how to make tests, and if that is even warranted.
Thanks for spotting.
I am not sure if the approach taken by this patch is the right one, though. I didn't follow the call/dataflow thoroughly, but if we replace unfree-able "" with NULL in these places, wouldn't fill_missing_values() take care of them?
Quoted 28 lines
ref-filter.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)ref-filter.c +3
-3@@ -1028,7 +1028,7 @@ static const char *copy_name(const char *buf)if (!strncmp (cp," <" ,2 ))return xmemdupz (buf, cp - buf); } -return "" ; +return xstrdup ("" ); }static const char *copy_email (const char *buf)@@ -1036,10 +1036,10 @@ static const char *copy_email(const char *buf)const char *email =strchr (buf,'<' );const char *eoemail;if (!email) -return "" ; +return xstrdup ("" ); eoemail =strchr (email,'>' );if (!eoemail) -return "" ; +return xstrdup ("" );return xmemdupz (email, eoemail +1 - email); }
Quoted 12 lines · Junio C Hamano
Quoted 9 lines · Mischa POSLAWSKY
Formatting $(taggername) on headerless tags such as v0.99 in Git causes a SIGABRT with error "
munmap_chunk(): invalid pointer", because of an oversight in commitf0062d3b74(ref-filter: freeitem->valueanditem->value->s, 2018-10-19).Signed-off-by Mischa POSLAWSKY <[email protected]>---
If I understand correctly, such tags cannot be produced normally anymore. Therefore I'm unsure how to make tests, and if that is even warranted.
Quoted 6 lines
Thanks for spotting.
I am not sure if the approach taken by this patch is the right one, though. I didn't follow the call/dataflow thoroughly, but if we replace unfree-able "" with NULL in these places, wouldn't
fill_missing_values()take care of them?
I think replacing these "" with NULL would be safe, but there are many places that return xstrdup("") from inside the callees of populate_value(), so the patch presented here would be more consistent with the current practice, I think.
So let's take the patch as is, at least for now. Thanks.
Quoted 28 lines
Quoted 28 lines
ref-filter.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)ref-filter.c +3
-3@@ -1028,7 +1028,7 @@ static const char *copy_name(const char *buf)if (!strncmp (cp," <" ,2 ))return xmemdupz (buf, cp - buf); } -return "" ; +return xstrdup ("" ); }static const char *copy_email (const char *buf)@@ -1036,10 +1036,10 @@ static const char *copy_email(const char *buf)const char *email =strchr (buf,'<' );const char *eoemail;if (!email) -return "" ; +return xstrdup ("" ); eoemail =strchr (email,'>' );if (!eoemail) -return "" ; +return xstrdup ("" );return xmemdupz (email, eoemail +1 - email); }
Quoted 21 lines · Junio
Quoted 12 lines · Junio C Hamano
Quoted 9 lines · Mischa POSLAWSKY
Formatting $(taggername) on headerless tags such as v0.99 in Git causes a SIGABRT with error "
munmap_chunk(): invalid pointer", because of an oversight in commitf0062d3b74(ref-filter: freeitem->valueanditem->value->s, 2018-10-19).Signed-off-by Mischa POSLAWSKY <[email protected]>---
If I understand correctly, such tags cannot be produced normally anymore. Therefore I'm unsure how to make tests, and if that is even warranted.Quoted 6 lines
Thanks for spotting.
I am not sure if the approach taken by this patch is the right one, though. I didn't follow the call/dataflow thoroughly, but if we replace unfree-able "" with NULL in these places, wouldn't
fill_missing_values()take care of them?
Quoted 4 lines
I think replacing these "" with NULL would be safe, but there are many places that return
xstrdup("")from inside the callees ofpopulate_value(), so the patch presented here would be more consistent with the current practice, I think.
Indeed, I just copied the existing style. Returning NULL seems to work, but not something I'm confident to clean up here.
Quoted 1 line
So let's take the patch as is, at least for now. Thanks.
Thank you!
Quoted 28 lines
Quoted 28 lines
Quoted 28 lines
ref-filter.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)ref-filter.c +3
-3@@ -1028,7 +1028,7 @@ static const char *copy_name(const char *buf)if (!strncmp (cp," <" ,2 ))return xmemdupz (buf, cp - buf); } -return "" ; +return xstrdup ("" ); }static const char *copy_email (const char *buf)@@ -1036,10 +1036,10 @@ static const char *copy_email(const char *buf)const char *email =strchr (buf,'<' );const char *eoemail;if (!email) -return "" ; +return xstrdup ("" ); eoemail =strchr (email,'>' );if (!eoemail) -return "" ; +return xstrdup ("" );return xmemdupz (email, eoemail +1 - email); }
Quoted 6 lines · Junio C Hamano
Quoted 2 lines · Mischa POSLAWSKY
If I understand correctly, such tags cannot be produced normally anymore. Therefore I'm unsure how to make tests, and if that is even warranted.
Thanks for spotting.
A quick trial to recreate a tag object seems to succeed:
$ git cat-file tag v0.99 |
Quoted 2 lines
sed -e '/-----BEGIN/,$d' |
git hash-object --stdin -w -t tag
667d141b478eee5e53d2ee05acd61bb1f640249a
$ git cat-file tag 667d141b47
object a3eb250f996bf5e12376ec88622c4ccaabf20ea8
type commit
tag v0.99
Test-release for wider distribution.
I'll make the first public RPM's etc, thus the tag.So we should be able to do something along the above line. Here is my quick-n-dirty one.
t/t6300-for-each-ref.sh | 12 ++++++++++++ 1 file changed, 12 insertions(+)
t/t6300-for-each-ref.sh +12 -0
@@ -869,4 +869,16 @@ test_expect_success 'for-each-ref --ignore-case ignores case' ' test_cmp expect actual ' +test_expect_success 'show a taggerless tag' ' + test_commit tagged && + git tag -a -m "a normal tag" to-be-shown-0 HEAD && + another=$(git cat-file tag to-be-shown-0 | + sed -e "/^tagger /d" \ + -e "/^tag to-be-shown/s/0/1/" \ + -e "s/a normal tag/a broken tag/" | + git hash-object --stdin -w -t tag) && + git tag to-be-shown-1 $another && + git for-each-ref --format="%(refname:short) %(taggername)" refs/tags/to-be-shown\* +' + test_done
[PATCH 2/1] t6300: format missing tagger
Quoted 42 lines · Junio
Quoted 6 lines · Junio C Hamano
Quoted 2 lines · Mischa POSLAWSKY
If I understand correctly, such tags cannot be produced normally anymore. Therefore I'm unsure how to make tests, and if that is even warranted.
Thanks for spotting.
A quick trial to recreate a tag object seems to succeed:
$ git cat-file tag v0.99 |Quoted 2 lines
sed -e '/-----BEGIN/,$d' |
git hash-object --stdin -w -t tag667d141b478eee5e53d2ee05acd61bb1f640249a $ git cat-file tag 667d141b47 object a3eb250f996bf5e12376ec88622c4ccaabf20ea8 type commit tag v0.99 Test-release for wider distribution. I'll make the first public RPM's etc, thus the tag.So we should be able to do something along the above line. Here is my quick-n-dirty one.
t/t6300-for-each-ref.sh | 12 ++++++++++++ 1 file changed, 12 insertions(+)t/t6300-for-each-ref.sh +3
-0@@ -869,4 +869,16 @@ test_expect_success 'for-each-ref --ignore-case ignores case' ' test_cmp expect actual ' +test_expect_success 'show a taggerless tag' ' + test_commit tagged && + git tag -a -m "a normal tag" to-be-shown-0 HEAD &&
t/t6300-for-each-ref.sh test_expect_success 'for-each-ref --ignore-case ignores case' ' · line 884
Quoted 10 lines
t/t6300-for-each-ref.sh +9
-0+ another=$(git cat-file tag to-be-shown-0 | + sed -e "/^tagger /d" \ + -e "/^tag to-be-shown/s/0/1/" \ + -e "s/a normal tag/a broken tag/" | + git hash-object --stdin -w -t tag) && + git tag to-be-shown-1 $another && + git for-each-ref --format="%(refname:short) %(taggername)" refs/tags/to-be-shown\* +' + test_done
Alright, thanks for the pointer.
Here's a batch of tests on all pertaining atoms.
-- >8 --
Strip an annotated tag of its tagger header and verify it's ignored correctly in all cases, as fixed in commit e2a81276e8 (ref-filter:
initialize empty name or email fields, 2019-08-19).
--- t/t6300-for-each-ref.sh | 19 +++++++++++++++++++ 1 file changed, 19 insertions(+)
t/t6300-for-each-ref.sh +19 -0
@@ -526,6 +526,25 @@ test_expect_success 'Check ambiguous head and tag refs II (loose)' ' test_cmp expected actual ' +test_expect_success 'create tag without tagger' ' + git tag -a -m "Broken tag" taggerless && + git tag -f taggerless $(git cat-file tag taggerless | + sed -e "/^tagger /d" | + git hash-object --stdin -w -t tag) +' + +test_atom refs/tags/taggerless type 'commit' +test_atom refs/tags/taggerless tag 'taggerless' +test_atom refs/tags/taggerless tagger '' +test_atom refs/tags/taggerless taggername '' +test_atom refs/tags/taggerless taggeremail '' +test_atom refs/tags/taggerless taggerdate '' +test_atom refs/tags/taggerless committer '' +test_atom refs/tags/taggerless committername '' +test_atom refs/tags/taggerless committeremail '' +test_atom refs/tags/taggerless committerdate '' +test_atom refs/tags/taggerless subject 'Broken tag' + test_expect_success 'an unusual tag with an incomplete line' ' git tag -m "bogo" bogo &&
Signature · 1 line
2.23.0
Re: [PATCH 2/1] t6300: format missing tagger
Quoted 2 lines · Mischa POSLAWSKY
Alright, thanks for the pointer.
Here's a batch of tests on all pertaining atoms.
Good to see that you made it much more thorough than my q-n-d illustration patch ;-)
Quoted 5 lines
-- >8 --
Strip an annotated tag of its tagger header and verify it's ignored correctly in all cases, as fixed in commit
e2a81276e8(ref-filter:
initialize empty name or email fields, 2019-08-19).
I am inclined to squash this test part of the update into the said commit; you'd lose one commit count, but hopefully you do not mind?
My motivation for doing so is that it would allow us to lose the "as fixed in commit X" comment in a log message, which in turn would mean that the code-fix patch can later be rebased safely without having to remember that this one needs to be adjusted ("git rebase" does not do such a rewrite for us, and I personally do not think "git rebase" should do such a rewrite silently, as I cannot quantify the risk of false positives).
Quoted 36 lines
Signed-off-by Mischa POSLAWSKY <[email protected]>--- t/t6300-for-each-ref.sh | 19 +++++++++++++++++++ 1 file changed, 19 insertions(+)t/t6300-for-each-ref.sh +19
-0@@ -526,6 +526,25 @@ test_expect_success 'Check ambiguous head and tag refs II (loose)' ' test_cmp expected actual ' +test_expect_success 'create tag without tagger' ' + git tag -a -m "Broken tag" taggerless && + git tag -f taggerless $(git cat-file tag taggerless | + sed -e "/^tagger /d" | + git hash-object --stdin -w -t tag) +' + +test_atom refs/tags/taggerless type 'commit' +test_atom refs/tags/taggerless tag 'taggerless' +test_atom refs/tags/taggerless tagger '' +test_atom refs/tags/taggerless taggername '' +test_atom refs/tags/taggerless taggeremail '' +test_atom refs/tags/taggerless taggerdate '' +test_atom refs/tags/taggerless committer '' +test_atom refs/tags/taggerless committername '' +test_atom refs/tags/taggerless committeremail '' +test_atom refs/tags/taggerless committerdate '' +test_atom refs/tags/taggerless subject 'Broken tag' + test_expect_success 'an unusual tag with an incomplete line' ' git tag -m "bogo" bogo &&
Re: [PATCH 2/1] t6300: format missing tagger
Quoted 9 lines · Junio
Quoted 3 lines · Mischa POSLAWSKY
Strip an annotated tag of its tagger header and verify it's ignored correctly in all cases, as fixed in commit
e2a81276e8(ref-filter:
initialize empty name or email fields, 2019-08-19).I am inclined to squash this test part of the update into the said commit; you'd lose one commit count, but hopefully you do not mind?
Quoted 7 lines
My motivation for doing so is that it would allow us to lose the "as fixed in commit X" comment in a log message, which in turn would mean that the code-fix patch can later be rebased safely without having to remember that this one needs to be adjusted ("git rebase" does not do such a rewrite for us, and I personally do not think "git rebase" should do such a rewrite silently, as I cannot quantify the risk of false positives).
Of course. Might get one commit back if you pick it into maint :)
Re: [PATCH 2/1] t6300: format missing tagger
Quoted 18 lines · Mischa POSLAWSKY
Quoted 9 lines · Junio
Quoted 3 lines · Mischa POSLAWSKY
Strip an annotated tag of its tagger header and verify it's ignored correctly in all cases, as fixed in commit
e2a81276e8(ref-filter:
initialize empty name or email fields, 2019-08-19).I am inclined to squash this test part of the update into the said commit; you'd lose one commit count, but hopefully you do not mind?
Quoted 7 lines
My motivation for doing so is that it would allow us to lose the "as fixed in commit X" comment in a log message, which in turn would mean that the code-fix patch can later be rebased safely without having to remember that this one needs to be adjusted ("git rebase" does not do such a rewrite for us, and I personally do not think "git rebase" should do such a rewrite silently, as I cannot quantify the risk of false positives).
Quoted 1 line
Of course. Might get one commit back if you pick it into maint :)
Actually you won't; I generally do not cherry-pick, even though I merge down relevant fixes to older maintenance tracks.