threads / patch / 33494

patchtest-bzr: portable shell and utf-8 strings for Mac OS

Subject: [PATCH] test-bzr: portable shell and utf-8 strings for Mac OS

## tl;dr

3 messages between Apr 12, 2013 and Apr 12, 2013. Diffs are folded; open one to read it.

replies: 2people: 3as markdown or json

Torsten Bögershausen· Apr 12, 2013, 21:18 UTC · lore
Make the shell script more portable:
- Split export X=Y into 2 lines
- Use printf instead of echo -e
Use UTF-8 code points which are not decomposed by the filesystem:
 Code points like "á" will be decomposed by Mac OS X.
 bzr is unable to find the file "á" on disk.
 Use code points from unicode which can not be decomposed.
 In other words, the precompsed form use the same bytes as decomposed.
Signed-off-by: Torsten Bögershausen <tboegi@web.de>
---
 contrib/remote-helpers/test-bzr.sh | 38 +++++++++++++++++++-------------------
 1 file changed, 19 insertions(+), 19 deletions(-)
Show changes to contrib/remote-helpers/test-bzr.sh +19 −19
diff --git a/contrib/remote-helpers/test-bzr.sh b/contrib/remote-helpers/test-bzr.sh
index e800c97..34666e1 100755
--- a/contrib/remote-helpers/test-bzr.sh
+++ b/contrib/remote-helpers/test-bzr.sh
@@ -169,31 +169,30 @@ test_expect_success 'fetch utf-8 filenames' '
   mkdir -p tmp && cd tmp &&
   test_when_finished "cd .. && rm -rf tmp && LC_ALL=C" &&
 
-  export LC_ALL=en_US.UTF-8
-
+  LC_ALL=en_US.UTF-8
+  export LC_ALL
   (
   bzr init bzrrepo &&
   cd bzrrepo &&
 
-  echo test >> "áéíóú" &&
-  bzr add "áéíóú" &&
-  echo test >> "îø∫∆" &&
-  bzr add "îø∫∆" &&
-  bzr commit -m utf-8 &&
-  echo test >> "áéíóú" &&
-  bzr commit -m utf-8 &&
-  bzr rm "îø∫∆" &&
-  bzr mv "áéíóú" "åß∂" &&
-  bzr commit -m utf-8
+  echo test >> "ærø" &&
+  bzr add "ærø" &&
+  echo test >> "ø~?" &&
+  bzr add "ø~?" &&
+  bzr commit -m add-utf-8 &&
+  echo test >> "ærø" &&
+  bzr commit -m test-utf-8 &&
+  bzr rm "ø~?" &&
+  bzr mv "ærø" "ø~?" &&
+  bzr commit -m bzr-mv-utf-8
   ) &&
 
   (
   git clone "bzr::$PWD/bzrrepo" gitrepo &&
   cd gitrepo &&
-  git ls-files > ../actual
+  git -c core.quotepath=false ls-files > ../actual
   ) &&
-
-  echo "\"\\303\\245\\303\\237\\342\\210\\202\"" > expected &&
+  echo "ø~?" > expected &&
   test_cmp expected actual
 '
 
@@ -201,7 +200,8 @@ test_expect_success 'push utf-8 filenames' '
   mkdir -p tmp && cd tmp &&
   test_when_finished "cd .. && rm -rf tmp && LC_ALL=C" &&
 
-  export LC_ALL=en_US.UTF-8
+  LC_ALL=en_US.UTF-8
+  export LC_ALL
 
   (
   bzr init bzrrepo &&
@@ -216,15 +216,15 @@ test_expect_success 'push utf-8 filenames' '
   git clone "bzr::$PWD/bzrrepo" gitrepo &&
   cd gitrepo &&
 
-  echo test >> "áéíóú" &&
-  git add "áéíóú" &&
+  echo test >> "ærø" &&
+  git add "ærø" &&
   git commit -m utf-8 &&
 
   git push
   ) &&
 
   (cd bzrrepo && bzr ls > ../actual) &&
-  echo -e "content\náéíóú" > expected &&
+  printf "content\nærø\n" > expected &&
   test_cmp expected actual
 '
 
-- 
1.8.2.341.g543621f
Felipe Contreras· Apr 12, 2013, 21:27 UTC · re: Torsten Bögershausen · lore

Re: [PATCH] test-bzr: portable shell and utf-8 strings for Mac OS

On Fri, Apr 12, 2013 at 4:18 PM, Torsten Bögershausen <tboegi@web.de> wrote:
Show 9 quoted lines
> Make the shell script more portable:
> - Split export X=Y into 2 lines
> - Use printf instead of echo -e
>
> Use UTF-8 code points which are not decomposed by the filesystem:
>  Code points like "á" will be decomposed by Mac OS X.
>  bzr is unable to find the file "á" on disk.
>  Use code points from unicode which can not be decomposed.
>  In other words, the precompsed form use the same bytes as decomposed.
Looks good to me.
-- 
Felipe Contreras
Junio C Hamano· Apr 12, 2013, 22:40 UTC · re: Felipe Contreras · lore

Re: [PATCH] test-bzr: portable shell and utf-8 strings for Mac OS

Felipe Contreras <felipe.contreras@gmail.com> writes:
Show 12 quoted lines
> On Fri, Apr 12, 2013 at 4:18 PM, Torsten Bögershausen <tboegi@web.de> wrote:
>> Make the shell script more portable:
>> - Split export X=Y into 2 lines
>> - Use printf instead of echo -e
>>
>> Use UTF-8 code points which are not decomposed by the filesystem:
>>  Code points like "á" will be decomposed by Mac OS X.
>>  bzr is unable to find the file "á" on disk.
>>  Use code points from unicode which can not be decomposed.
>>  In other words, the precompsed form use the same bytes as decomposed.
>
> Looks good to me.
Thanks, both. Applied.

← back to recent threads