threads / patch / 36709

patcht6006-*.sh: Fix truncation tests

Subject: [PATCH] t6006-*.sh: Fix truncation tests

## tl;dr

11 messages between May 20, 2014 and May 21, 2014. Diffs are folded; open one to read it.

replies: 10people: 3as markdown or json

Ramsay Jones· May 20, 2014, 13:54 UTC · lore

Signed-off-by: Ramsay Jones <ramsay@ramsay1.demon.co.uk>
---

Hi Alexey,

If you need to re-roll your 'as/pretty-truncate' branch, could you please squash the relevant parts of this patch into the corresponding patches of your patch series. (ie this is a patch against the head of the current pu branch ...).

Without this patch I get:

  $ ./t6006-rev-list-format.sh
  ok 1 - setup
  ok 2 - format percent
  ok 3 - format hash
  ok 4 - format tree
  ok 5 - format parents
  ok 6 - format author
  ok 7 - format committer
  ok 8 - format encoding
  ok 9 - format subject
  ./t6006-rev-list-format.sh: 152: ./t6006-rev-list-format.sh: Syntax error: "(" unexpected
  FATAL: Unexpected exit with code 2
  $

(if you have bash as /bin/sh you get different but related errors). The additional quoting suppresses the 'command redirection' errors, etc...

Thanks.

ATB Ramsay Jones

 t/t6006-rev-list-format.sh | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

Show changes to t/t6006-rev-list-format.sh +7 −7

diff --git a/t/t6006-rev-list-format.sh b/t/t6006-rev-list-format.sh
index 9bc089b..e1dec3e 100755
--- a/t/t6006-rev-list-format.sh
+++ b/t/t6006-rev-list-format.sh
@@ -149,7 +149,7 @@ commit $head1
 $added
 EOF
 
-test_format subject-truncated %<($truncate_count,trunc)%s <<EOF
+test_format subject-truncated %\<\($truncate_count,trunc\)%s <<EOF
 commit $head2
 changed (ge${changed_utf8_part}ndert)..
 commit $head1
@@ -256,7 +256,7 @@ commit $head1
 $added_iso88591
 EOF
 
-test_format complex-subject-trunc %<($truncate_count,trunc)%s <<EOF
+test_format complex-subject-trunc %\<\($truncate_count,trunc\)%s <<EOF
 commit $head3
 Test printing of c..
 commit $head2
@@ -265,7 +265,7 @@ commit $head1
 added (hinzugef${added_utf8_part_iso88591}gt..
 EOF
 
-test_format complex-subject-mtrunc %<($truncate_count,mtrunc)%s <<EOF
+test_format complex-subject-mtrunc %\<\($truncate_count,mtrunc\)%s <<EOF
 commit $head3
 Test prin..ex bodies
 commit $head2
@@ -274,7 +274,7 @@ commit $head1
 added (hi..f${added_utf8_part_iso88591}gt) foo
 EOF
 
-test_format complex-subject-ltrunc %<($truncate_count,ltrunc)%s <<EOF
+test_format complex-subject-ltrunc %\<\($truncate_count,ltrunc\)%s <<EOF
 commit $head3
 .. of complex bodies
 commit $head2
@@ -311,7 +311,7 @@ commit $head1
 $added
 EOF
 
-test_format complex-subject-commitencoding-unset-trunc %<($truncate_count,trunc)%s <<EOF
+test_format complex-subject-commitencoding-unset-trunc %\<\($truncate_count,trunc\)%s <<EOF
 commit $head3
 Test printing of c..
 commit $head2
@@ -320,7 +320,7 @@ commit $head1
 added (hinzugef${added_utf8_part}gt..
 EOF
 
-test_format complex-subject-commitencoding-unset-mtrunc %<($truncate_count,mtrunc)%s <<EOF
+test_format complex-subject-commitencoding-unset-mtrunc %\<\($truncate_count,mtrunc\)%s <<EOF
 commit $head3
 Test prin..ex bodies
 commit $head2
@@ -329,7 +329,7 @@ commit $head1
 added (hi..f${added_utf8_part}gt) foo
 EOF
 
-test_format complex-subject-commitencoding-unset-ltrunc %<($truncate_count,ltrunc)%s <<EOF
+test_format complex-subject-commitencoding-unset-ltrunc %\<\($truncate_count,ltrunc\)%s <<EOF
 commit $head3
 .. of complex bodies
 commit $head2

-- 
1.9.0

Alexey Shumkin· May 20, 2014, 14:19 UTC · re: Ramsay Jones · lore

Re: [PATCH] t6006-*.sh: Fix truncation tests

On Tue, May 20, 2014 at 02:54:20PM +0100, Ramsay Jones wrote:

Show 26 quoted lines

> 
> Signed-off-by: Ramsay Jones <ramsay@ramsay1.demon.co.uk>
> ---
> 
> Hi Alexey,
> 
> If you need to re-roll your 'as/pretty-truncate' branch, could
> you please squash the relevant parts of this patch into the
> corresponding patches of your patch series. (ie this is a patch
> against the head of the current pu branch ...).
> 
> Without this patch I get:
> 
>   $ ./t6006-rev-list-format.sh
>   ok 1 - setup
>   ok 2 - format percent
>   ok 3 - format hash
>   ok 4 - format tree
>   ok 5 - format parents
>   ok 6 - format author
>   ok 7 - format committer
>   ok 8 - format encoding
>   ok 9 - format subject
>   ./t6006-rev-list-format.sh: 152: ./t6006-rev-list-format.sh: Syntax error: "(" unexpected
>   FATAL: Unexpected exit with code 2
>   $

Ooops, my fault.

> 
> (if you have bash as /bin/sh you get different but related errors).
> The additional quoting suppresses the 'command redirection' errors, etc...

It's strange but I do have Bash as /bin/sh and unfortunately I have no this error

AFAIU, Junio already applied my patches (existance of a branch as/pretty-truncate tells us that). So, we can only send other patches that fix errors brought with former patches. You can send, too.

Show 78 quoted lines

> 
> Thanks.
> 
> ATB
> Ramsay Jones
> 
>  t/t6006-rev-list-format.sh | 14 +++++++-------
>  1 file changed, 7 insertions(+), 7 deletions(-)
> 
> diff --git a/t/t6006-rev-list-format.sh b/t/t6006-rev-list-format.sh
> index 9bc089b..e1dec3e 100755
> --- a/t/t6006-rev-list-format.sh
> +++ b/t/t6006-rev-list-format.sh
> @@ -149,7 +149,7 @@ commit $head1
>  $added
>  EOF
>  
> -test_format subject-truncated %<($truncate_count,trunc)%s <<EOF
> +test_format subject-truncated %\<\($truncate_count,trunc\)%s <<EOF
>  commit $head2
>  changed (ge${changed_utf8_part}ndert)..
>  commit $head1
> @@ -256,7 +256,7 @@ commit $head1
>  $added_iso88591
>  EOF
>  
> -test_format complex-subject-trunc %<($truncate_count,trunc)%s <<EOF
> +test_format complex-subject-trunc %\<\($truncate_count,trunc\)%s <<EOF
>  commit $head3
>  Test printing of c..
>  commit $head2
> @@ -265,7 +265,7 @@ commit $head1
>  added (hinzugef${added_utf8_part_iso88591}gt..
>  EOF
>  
> -test_format complex-subject-mtrunc %<($truncate_count,mtrunc)%s <<EOF
> +test_format complex-subject-mtrunc %\<\($truncate_count,mtrunc\)%s <<EOF
>  commit $head3
>  Test prin..ex bodies
>  commit $head2
> @@ -274,7 +274,7 @@ commit $head1
>  added (hi..f${added_utf8_part_iso88591}gt) foo
>  EOF
>  
> -test_format complex-subject-ltrunc %<($truncate_count,ltrunc)%s <<EOF
> +test_format complex-subject-ltrunc %\<\($truncate_count,ltrunc\)%s <<EOF
>  commit $head3
>  .. of complex bodies
>  commit $head2
> @@ -311,7 +311,7 @@ commit $head1
>  $added
>  EOF
>  
> -test_format complex-subject-commitencoding-unset-trunc %<($truncate_count,trunc)%s <<EOF
> +test_format complex-subject-commitencoding-unset-trunc %\<\($truncate_count,trunc\)%s <<EOF
>  commit $head3
>  Test printing of c..
>  commit $head2
> @@ -320,7 +320,7 @@ commit $head1
>  added (hinzugef${added_utf8_part}gt..
>  EOF
>  
> -test_format complex-subject-commitencoding-unset-mtrunc %<($truncate_count,mtrunc)%s <<EOF
> +test_format complex-subject-commitencoding-unset-mtrunc %\<\($truncate_count,mtrunc\)%s <<EOF
>  commit $head3
>  Test prin..ex bodies
>  commit $head2
> @@ -329,7 +329,7 @@ commit $head1
>  added (hi..f${added_utf8_part}gt) foo
>  EOF
>  
> -test_format complex-subject-commitencoding-unset-ltrunc %<($truncate_count,ltrunc)%s <<EOF
> +test_format complex-subject-commitencoding-unset-ltrunc %\<\($truncate_count,ltrunc\)%s <<EOF
>  commit $head3
>  .. of complex bodies
>  commit $head2
> -- 
> 1.9.0

-- 
Alexey Shumkin
E-mail: Alex.Crezoff@gmail.com
ICQ: 118001447
Jabber (GoogleTalk): Alex.Crezoff@gmail.com
Skype: crezoff

Alexey Shumkin· May 20, 2014, 14:34 UTC · re: Alexey Shumkin · lore

Re: [PATCH] t6006-*.sh: Fix truncation tests

On Tue, May 20, 2014 at 06:19:36PM +0400, Alexey Shumkin wrote:

Show 57 quoted lines

> On Tue, May 20, 2014 at 02:54:20PM +0100, Ramsay Jones wrote:
> > 
> > Signed-off-by: Ramsay Jones <ramsay@ramsay1.demon.co.uk>
> > ---
> > 
> > Hi Alexey,
> > 
> > If you need to re-roll your 'as/pretty-truncate' branch, could
> > you please squash the relevant parts of this patch into the
> > corresponding patches of your patch series. (ie this is a patch
> > against the head of the current pu branch ...).
> > 
> > Without this patch I get:
> > 
> >   $ ./t6006-rev-list-format.sh
> >   ok 1 - setup
> >   ok 2 - format percent
> >   ok 3 - format hash
> >   ok 4 - format tree
> >   ok 5 - format parents
> >   ok 6 - format author
> >   ok 7 - format committer
> >   ok 8 - format encoding
> >   ok 9 - format subject
> >   ./t6006-rev-list-format.sh: 152: ./t6006-rev-list-format.sh: Syntax error: "(" unexpected
> >   FATAL: Unexpected exit with code 2
> >   $ 
> Ooops, my fault.
> > 
> > (if you have bash as /bin/sh you get different but related errors).
> > The additional quoting suppresses the 'command redirection' errors, etc...
> It's strange but I do have Bash as /bin/sh and unfortunately I have no
> this error
> 
> AFAIU, Junio already applied my patches (existance of a branch
> as/pretty-truncate tells us that). So, we can only send other patches that
> fix errors brought with former patches.
> You can send, too.
> > 
> > Thanks.
> > 
> > ATB
> > Ramsay Jones
> > 
> >  t/t6006-rev-list-format.sh | 14 +++++++-------
> >  1 file changed, 7 insertions(+), 7 deletions(-)
> > 
> > diff --git a/t/t6006-rev-list-format.sh b/t/t6006-rev-list-format.sh
> > index 9bc089b..e1dec3e 100755
> > --- a/t/t6006-rev-list-format.sh
> > +++ b/t/t6006-rev-list-format.sh
> > @@ -149,7 +149,7 @@ commit $head1
> >  $added
> >  EOF
> >  
> > -test_format subject-truncated %<($truncate_count,trunc)%s <<EOF
> > +test_format subject-truncated %\<\($truncate_count,trunc\)%s <<EOF

BTW, I would quoted that values rather than escaped

Ramsay Jones· May 20, 2014, 15:01 UTC · re: Alexey Shumkin · lore

Re: [PATCH] t6006-*.sh: Fix truncation tests

On 20/05/14 15:19, Alexey Shumkin wrote:

Show 33 quoted lines

> On Tue, May 20, 2014 at 02:54:20PM +0100, Ramsay Jones wrote:
>>
>> Signed-off-by: Ramsay Jones <ramsay@ramsay1.demon.co.uk>
>> ---
>>
>> Hi Alexey,
>>
>> If you need to re-roll your 'as/pretty-truncate' branch, could
>> you please squash the relevant parts of this patch into the
>> corresponding patches of your patch series. (ie this is a patch
>> against the head of the current pu branch ...).
>>
>> Without this patch I get:
>>
>>   $ ./t6006-rev-list-format.sh
>>   ok 1 - setup
>>   ok 2 - format percent
>>   ok 3 - format hash
>>   ok 4 - format tree
>>   ok 5 - format parents
>>   ok 6 - format author
>>   ok 7 - format committer
>>   ok 8 - format encoding
>>   ok 9 - format subject
>>   ./t6006-rev-list-format.sh: 152: ./t6006-rev-list-format.sh: Syntax error: "(" unexpected
>>   FATAL: Unexpected exit with code 2
>>   $ 
> Ooops, my fault.
>>
>> (if you have bash as /bin/sh you get different but related errors).
>> The additional quoting suppresses the 'command redirection' errors, etc...
> It's strange but I do have Bash as /bin/sh and unfortunately I have no
> this error

Hmm, I see this:

  $ bash t6006-rev-list-format.sh -i
  ok 1 - setup
  ok 2 - format percent
  ok 3 - format hash
  ok 4 - format tree
  ok 5 - format parents
  ok 6 - format author
  ok 7 - format committer
  ok 8 - format encoding
  ok 9 - format subject
  t6006-rev-list-format.sh: line 152: 20,trunc: command not found
  not ok 10 - format subject-truncated
  #	
  #			git rev-list --pretty=format:'%/dev/fd/63%s' master >output.subject-truncated &&
  #			test_cmp expect.subject-truncated output.subject-truncated
  #		
  $

(Since I'm on Linux Mint, my /bin/sh is dash, which produces the output in my previous mail).

ATB, Ramsay Jones

Alexey Shumkin· May 20, 2014, 16:02 UTC · re: Ramsay Jones · lore

Re: [PATCH] t6006-*.sh: Fix truncation tests

On Tue, May 20, 2014 at 04:01:22PM +0100, Ramsay Jones wrote:

Show 38 quoted lines

> On 20/05/14 15:19, Alexey Shumkin wrote:
> > On Tue, May 20, 2014 at 02:54:20PM +0100, Ramsay Jones wrote:
> >>
> >> Signed-off-by: Ramsay Jones <ramsay@ramsay1.demon.co.uk>
> >> ---
> >>
> >> Hi Alexey,
> >>
> >> If you need to re-roll your 'as/pretty-truncate' branch, could
> >> you please squash the relevant parts of this patch into the
> >> corresponding patches of your patch series. (ie this is a patch
> >> against the head of the current pu branch ...).
> >>
> >> Without this patch I get:
> >>
> >>   $ ./t6006-rev-list-format.sh
> >>   ok 1 - setup
> >>   ok 2 - format percent
> >>   ok 3 - format hash
> >>   ok 4 - format tree
> >>   ok 5 - format parents
> >>   ok 6 - format author
> >>   ok 7 - format committer
> >>   ok 8 - format encoding
> >>   ok 9 - format subject
> >>   ./t6006-rev-list-format.sh: 152: ./t6006-rev-list-format.sh: Syntax error: "(" unexpected
> >>   FATAL: Unexpected exit with code 2
> >>   $ 
> > Ooops, my fault.
> >>
> >> (if you have bash as /bin/sh you get different but related errors).
> >> The additional quoting suppresses the 'command redirection' errors, etc...
> > It's strange but I do have Bash as /bin/sh and unfortunately I have no
> > this error
> 
> Hmm, I see this:
> 
>   $ bash t6006-rev-list-format.sh -i

AFAIK, this is not "running Bash as /bin/sh" :)
Maybe I'm wrong but Bash as /bin/sh is:
  $ ls -l /bin/sh
  lrwxrwxrwx 1 root root 4 May 12 18:35 /bin/sh -> bash

But it does not matter here much, you've noticed my foolish mistake :)

Show 24 quoted lines

>   ok 1 - setup
>   ok 2 - format percent
>   ok 3 - format hash
>   ok 4 - format tree
>   ok 5 - format parents
>   ok 6 - format author
>   ok 7 - format committer
>   ok 8 - format encoding
>   ok 9 - format subject
>   t6006-rev-list-format.sh: line 152: 20,trunc: command not found
>   not ok 10 - format subject-truncated
>   #	
>   #			git rev-list --pretty=format:'%/dev/fd/63%s' master >output.subject-truncated &&
>   #			test_cmp expect.subject-truncated output.subject-truncated
>   #		
>   $ 
> 
> (Since I'm on Linux Mint, my /bin/sh is dash, which produces the output in my previous mail).
> 
> ATB,
> Ramsay Jones
> 
> 
>

Ramsay Jones· May 20, 2014, 16:43 UTC · re: Alexey Shumkin · lore

Re: [PATCH] t6006-*.sh: Fix truncation tests

On 20/05/14 17:02, Alexey Shumkin wrote:

Show 43 quoted lines

> On Tue, May 20, 2014 at 04:01:22PM +0100, Ramsay Jones wrote:
>> On 20/05/14 15:19, Alexey Shumkin wrote:
>>> On Tue, May 20, 2014 at 02:54:20PM +0100, Ramsay Jones wrote:
>>>>
>>>> Signed-off-by: Ramsay Jones <ramsay@ramsay1.demon.co.uk>
>>>> ---
>>>>
>>>> Hi Alexey,
>>>>
>>>> If you need to re-roll your 'as/pretty-truncate' branch, could
>>>> you please squash the relevant parts of this patch into the
>>>> corresponding patches of your patch series. (ie this is a patch
>>>> against the head of the current pu branch ...).
>>>>
>>>> Without this patch I get:
>>>>
>>>>   $ ./t6006-rev-list-format.sh
>>>>   ok 1 - setup
>>>>   ok 2 - format percent
>>>>   ok 3 - format hash
>>>>   ok 4 - format tree
>>>>   ok 5 - format parents
>>>>   ok 6 - format author
>>>>   ok 7 - format committer
>>>>   ok 8 - format encoding
>>>>   ok 9 - format subject
>>>>   ./t6006-rev-list-format.sh: 152: ./t6006-rev-list-format.sh: Syntax error: "(" unexpected
>>>>   FATAL: Unexpected exit with code 2
>>>>   $ 
>>> Ooops, my fault.
>>>>
>>>> (if you have bash as /bin/sh you get different but related errors).
>>>> The additional quoting suppresses the 'command redirection' errors, etc...
>>> It's strange but I do have Bash as /bin/sh and unfortunately I have no
>>> this error
>>
>> Hmm, I see this:
>>
>>   $ bash t6006-rev-list-format.sh -i
> AFAIK, this is not "running Bash as /bin/sh" :)
> Maybe I'm wrong but Bash as /bin/sh is:
>   $ ls -l /bin/sh
>   lrwxrwxrwx 1 root root 4 May 12 18:35 /bin/sh -> bash

Ah yes, I keep forgetting that bash behaves differently when invoked as 'sh'. (i.e. it enters 'posix mode' in this case).

Indeed, this is (roughly) equivalent to:

  $ bash --posix t6006-rev-list-format.sh
  ok 1 - setup
  ok 2 - format percent
  ok 3 - format hash
  ...
  ok 52 - oneline with empty message
  ok 53 - single-character name is parsed correctly
  # passed all 53 test(s)
  1..53
  $

... which works.

Thanks again.

ATB Ramsay Jones

Junio C Hamano· May 20, 2014, 17:10 UTC · re: Alexey Shumkin · lore

Re: [PATCH] t6006-*.sh: Fix truncation tests

Alexey Shumkin <alex.crezoff@gmail.com> writes:

> AFAIU, Junio already applied my patches (existance of a branch
> as/pretty-truncate tells us that). So, we can only send other patches that
> fix errors brought with former patches.

No, NO, NOOOOO....

The existence of a branch merely means that I saw the patches, and that I thought that the series was not completely useless. In other words, it indicates that I wanted to make sure that I won't forget about the topic, and it was worth my time to create the branch and apply there for that purpose.

Please do not read anything more than that. Presense of a topic branch by itself does not say that I _read_ the patches or I thought they did not need reroll.

When such a branch is merged to 'next', that means I read the patches myself, or I saw somebody whose judgement I and others in the community trust read them and gave a positive response or an Ack, and that I decided that the topic is in a good enough shape to be worked on further with incremental updates.

You are talking about the latter state, but as/pretty-truncate is in the former state.

Alexey Shumkin· May 21, 2014, 04:11 UTC · re: Junio C Hamano · lore

Re: [PATCH] t6006-*.sh: Fix truncation tests

On Tue, May 20, 2014 at 10:10:46AM -0700, Junio C Hamano wrote:

Show 13 quoted lines

> Alexey Shumkin <alex.crezoff@gmail.com> writes:
> 
> > AFAIU, Junio already applied my patches (existance of a branch
> > as/pretty-truncate tells us that). So, we can only send other patches that
> > fix errors brought with former patches.
> 
> No, NO, NOOOOO....
> 
> The existence of a branch merely means that I saw the patches, and
> that I thought that the series was not completely useless.  In other
> words, it indicates that I wanted to make sure that I won't forget
> about the topic, and it was worth my time to create the branch and
> apply there for that purpose.

Oh, I got it! I'll fix my typo (ISO8895-1 vs ISO8859-1) and quotes issue then. And will reroll patches.

Thanks!

Show 13 quoted lines

> 
> Please do not read anything more than that.  Presense of a topic
> branch by itself does not say that I _read_ the patches or I thought
> they did not need reroll.
> 
> When such a branch is merged to 'next', that means I read the
> patches myself, or I saw somebody whose judgement I and others in
> the community trust read them and gave a positive response or an
> Ack, and that I decided that the topic is in a good enough shape to
> be worked on further with incremental updates.
> 
> You are talking about the latter state, but as/pretty-truncate is in
> the former state.

Alexey Shumkin· May 20, 2014, 14:48 UTC · re: Ramsay Jones · lore

[PATCH] t6006 (rev-list-format): quote format strings to avoid error on some shells

Added in 0a144b3 (t4205, t6006: add failing tests for the case when i18n.logOutputEncoding is set, 2014-05-19) tests give no error (somehow) with Bash as /bin/sh but fail for some other shells.

Quote format strings to avoid errors.

Signed-off-by: Alexey Shumkin <Alex.Crezoff@gmail.com>
Suggested-by: Ramsay Jones <ramsay@ramsay1.demon.co.uk>
---
 t/t6006-rev-list-format.sh | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

Show changes to t/t6006-rev-list-format.sh +7 −7

diff --git a/t/t6006-rev-list-format.sh b/t/t6006-rev-list-format.sh
index c6e9a73..19434ad 100755
--- a/t/t6006-rev-list-format.sh
+++ b/t/t6006-rev-list-format.sh
@@ -149,7 +149,7 @@ commit $head1
 $added
 EOF
 
-test_format subject-truncated %<($truncate_count,trunc)%s <<EOF
+test_format subject-truncated "%<($truncate_count,trunc)%s" <<EOF
 commit $head2
 changed (ge${changed_utf8_part}ndert)..
 commit $head1
@@ -259,7 +259,7 @@ commit $head1
 $added_iso88591
 EOF
 
-test_format complex-subject-trunc %<($truncate_count,trunc)%s <<EOF
+test_format complex-subject-trunc "%<($truncate_count,trunc)%s" <<EOF
 commit $head3
 Test printing of c..
 commit $head2
@@ -268,7 +268,7 @@ commit $head1
 added (hinzugef${added_utf8_part_iso88591}gt..
 EOF
 
-test_format complex-subject-mtrunc %<($truncate_count,mtrunc)%s <<EOF
+test_format complex-subject-mtrunc "%<($truncate_count,mtrunc)%s" <<EOF
 commit $head3
 Test prin..ex bodies
 commit $head2
@@ -277,7 +277,7 @@ commit $head1
 added (hi..f${added_utf8_part_iso88591}gt) foo
 EOF
 
-test_format complex-subject-ltrunc %<($truncate_count,ltrunc)%s <<EOF
+test_format complex-subject-ltrunc "%<($truncate_count,ltrunc)%s" <<EOF
 commit $head3
 .. of complex bodies
 commit $head2
@@ -314,7 +314,7 @@ commit $head1
 $added
 EOF
 
-test_format complex-subject-commitencoding-unset-trunc %<($truncate_count,trunc)%s <<EOF
+test_format complex-subject-commitencoding-unset-trunc "%<($truncate_count,trunc)%s" <<EOF
 commit $head3
 Test printing of c..
 commit $head2
@@ -323,7 +323,7 @@ commit $head1
 added (hinzugef${added_utf8_part}gt..
 EOF
 
-test_format complex-subject-commitencoding-unset-mtrunc %<($truncate_count,mtrunc)%s <<EOF
+test_format complex-subject-commitencoding-unset-mtrunc "%<($truncate_count,mtrunc)%s" <<EOF
 commit $head3
 Test prin..ex bodies
 commit $head2
@@ -332,7 +332,7 @@ commit $head1
 added (hi..f${added_utf8_part}gt) foo
 EOF
 
-test_format complex-subject-commitencoding-unset-ltrunc %<($truncate_count,ltrunc)%s <<EOF
+test_format complex-subject-commitencoding-unset-ltrunc "%<($truncate_count,ltrunc)%s" <<EOF
 commit $head3
 .. of complex bodies
 commit $head2

-- 
1.9.2-17

Ramsay Jones· May 20, 2014, 15:03 UTC · re: Alexey Shumkin · lore

Re: [PATCH] t6006 (rev-list-format): quote format strings to avoid error on some shells

On 20/05/14 15:48, Alexey Shumkin wrote:

Show 80 quoted lines

> Added in 0a144b3 (t4205, t6006: add failing tests for the case when
> i18n.logOutputEncoding is set, 2014-05-19) tests give no error
> (somehow) with Bash as /bin/sh but fail for some other shells.
> 
> Quote format strings to avoid errors.
> 
> Signed-off-by: Alexey Shumkin <Alex.Crezoff@gmail.com>
> Suggested-by: Ramsay Jones <ramsay@ramsay1.demon.co.uk>
> ---
>  t/t6006-rev-list-format.sh | 14 +++++++-------
>  1 file changed, 7 insertions(+), 7 deletions(-)
> 
> diff --git a/t/t6006-rev-list-format.sh b/t/t6006-rev-list-format.sh
> index c6e9a73..19434ad 100755
> --- a/t/t6006-rev-list-format.sh
> +++ b/t/t6006-rev-list-format.sh
> @@ -149,7 +149,7 @@ commit $head1
>  $added
>  EOF
>  
> -test_format subject-truncated %<($truncate_count,trunc)%s <<EOF
> +test_format subject-truncated "%<($truncate_count,trunc)%s" <<EOF
>  commit $head2
>  changed (ge${changed_utf8_part}ndert)..
>  commit $head1
> @@ -259,7 +259,7 @@ commit $head1
>  $added_iso88591
>  EOF
>  
> -test_format complex-subject-trunc %<($truncate_count,trunc)%s <<EOF
> +test_format complex-subject-trunc "%<($truncate_count,trunc)%s" <<EOF
>  commit $head3
>  Test printing of c..
>  commit $head2
> @@ -268,7 +268,7 @@ commit $head1
>  added (hinzugef${added_utf8_part_iso88591}gt..
>  EOF
>  
> -test_format complex-subject-mtrunc %<($truncate_count,mtrunc)%s <<EOF
> +test_format complex-subject-mtrunc "%<($truncate_count,mtrunc)%s" <<EOF
>  commit $head3
>  Test prin..ex bodies
>  commit $head2
> @@ -277,7 +277,7 @@ commit $head1
>  added (hi..f${added_utf8_part_iso88591}gt) foo
>  EOF
>  
> -test_format complex-subject-ltrunc %<($truncate_count,ltrunc)%s <<EOF
> +test_format complex-subject-ltrunc "%<($truncate_count,ltrunc)%s" <<EOF
>  commit $head3
>  .. of complex bodies
>  commit $head2
> @@ -314,7 +314,7 @@ commit $head1
>  $added
>  EOF
>  
> -test_format complex-subject-commitencoding-unset-trunc %<($truncate_count,trunc)%s <<EOF
> +test_format complex-subject-commitencoding-unset-trunc "%<($truncate_count,trunc)%s" <<EOF
>  commit $head3
>  Test printing of c..
>  commit $head2
> @@ -323,7 +323,7 @@ commit $head1
>  added (hinzugef${added_utf8_part}gt..
>  EOF
>  
> -test_format complex-subject-commitencoding-unset-mtrunc %<($truncate_count,mtrunc)%s <<EOF
> +test_format complex-subject-commitencoding-unset-mtrunc "%<($truncate_count,mtrunc)%s" <<EOF
>  commit $head3
>  Test prin..ex bodies
>  commit $head2
> @@ -332,7 +332,7 @@ commit $head1
>  added (hi..f${added_utf8_part}gt) foo
>  EOF
>  
> -test_format complex-subject-commitencoding-unset-ltrunc %<($truncate_count,ltrunc)%s <<EOF
> +test_format complex-subject-commitencoding-unset-ltrunc "%<($truncate_count,ltrunc)%s" <<EOF
>  commit $head3
>  .. of complex bodies
>  commit $head2
>

Yep, actually using string quotes works just as well! :-P

Thanks!

ATB, Ramsay Jones

Alexey Shumkin· May 21, 2014, 12:59 UTC · re: Alexey Shumkin · lore

Re: [PATCH] t6006 (rev-list-format): quote format strings to avoid error on some shells

This patch is redundant then. It will be squashed into next patch series.

On Tue, May 20, 2014 at 06:48:43PM +0400, Alexey Shumkin wrote:

Show 79 quoted lines

> Added in 0a144b3 (t4205, t6006: add failing tests for the case when
> i18n.logOutputEncoding is set, 2014-05-19) tests give no error
> (somehow) with Bash as /bin/sh but fail for some other shells.
> 
> Quote format strings to avoid errors.
> 
> Signed-off-by: Alexey Shumkin <Alex.Crezoff@gmail.com>
> Suggested-by: Ramsay Jones <ramsay@ramsay1.demon.co.uk>
> ---
>  t/t6006-rev-list-format.sh | 14 +++++++-------
>  1 file changed, 7 insertions(+), 7 deletions(-)
> 
> diff --git a/t/t6006-rev-list-format.sh b/t/t6006-rev-list-format.sh
> index c6e9a73..19434ad 100755
> --- a/t/t6006-rev-list-format.sh
> +++ b/t/t6006-rev-list-format.sh
> @@ -149,7 +149,7 @@ commit $head1
>  $added
>  EOF
>  
> -test_format subject-truncated %<($truncate_count,trunc)%s <<EOF
> +test_format subject-truncated "%<($truncate_count,trunc)%s" <<EOF
>  commit $head2
>  changed (ge${changed_utf8_part}ndert)..
>  commit $head1
> @@ -259,7 +259,7 @@ commit $head1
>  $added_iso88591
>  EOF
>  
> -test_format complex-subject-trunc %<($truncate_count,trunc)%s <<EOF
> +test_format complex-subject-trunc "%<($truncate_count,trunc)%s" <<EOF
>  commit $head3
>  Test printing of c..
>  commit $head2
> @@ -268,7 +268,7 @@ commit $head1
>  added (hinzugef${added_utf8_part_iso88591}gt..
>  EOF
>  
> -test_format complex-subject-mtrunc %<($truncate_count,mtrunc)%s <<EOF
> +test_format complex-subject-mtrunc "%<($truncate_count,mtrunc)%s" <<EOF
>  commit $head3
>  Test prin..ex bodies
>  commit $head2
> @@ -277,7 +277,7 @@ commit $head1
>  added (hi..f${added_utf8_part_iso88591}gt) foo
>  EOF
>  
> -test_format complex-subject-ltrunc %<($truncate_count,ltrunc)%s <<EOF
> +test_format complex-subject-ltrunc "%<($truncate_count,ltrunc)%s" <<EOF
>  commit $head3
>  .. of complex bodies
>  commit $head2
> @@ -314,7 +314,7 @@ commit $head1
>  $added
>  EOF
>  
> -test_format complex-subject-commitencoding-unset-trunc %<($truncate_count,trunc)%s <<EOF
> +test_format complex-subject-commitencoding-unset-trunc "%<($truncate_count,trunc)%s" <<EOF
>  commit $head3
>  Test printing of c..
>  commit $head2
> @@ -323,7 +323,7 @@ commit $head1
>  added (hinzugef${added_utf8_part}gt..
>  EOF
>  
> -test_format complex-subject-commitencoding-unset-mtrunc %<($truncate_count,mtrunc)%s <<EOF
> +test_format complex-subject-commitencoding-unset-mtrunc "%<($truncate_count,mtrunc)%s" <<EOF
>  commit $head3
>  Test prin..ex bodies
>  commit $head2
> @@ -332,7 +332,7 @@ commit $head1
>  added (hi..f${added_utf8_part}gt) foo
>  EOF
>  
> -test_format complex-subject-commitencoding-unset-ltrunc %<($truncate_count,ltrunc)%s <<EOF
> +test_format complex-subject-commitencoding-unset-ltrunc "%<($truncate_count,ltrunc)%s" <<EOF
>  commit $head3
>  .. of complex bodies
>  commit $head2