On 06/14/2011 02:47 PM, Pavel Tisnovsky wrote:
> Hi all,
>
> I'd like to push new patch containing fix for a regression test
> openjdk/jdk/test/sun/net/www/protocol/file/EncodedMultiByteChar.java
> into IcedTea6 HEAD. This patch ensures, that the test deletes its
> temporary file when it finishes.
>
>
> Here's ChangeLog entry (reformatted by my mail client to 80 chars/line,
> real entry is formatted properly ;-):
>
> 2011-06-14 Pavel Tisnovsky<ptisnovs at redhat.com>
>
> * Makefile.am: added new patch
> * patches/jtreg-EncodedMultiByteChar.patch:
> Make sure that the regression test
> openjdk/jdk/test/sun/net/www/protocol/file/EncodedMultiByteChar.java
> deletes all its work files.
>
>
> Can anybody please review this patch?
>
> Thank you in advance,
> Pavel T.
>
>
> hg_diff.patch
>
>
> diff -r e71f6f90abb6 Makefile.am
> --- a/Makefile.am Tue Jun 14 13:00:21 2011 +0200
> +++ b/Makefile.am Tue Jun 14 13:56:38 2011 +0200
> @@ -353,7 +353,8 @@
> patches/openjdk/mutter.patch \
> patches/fonts-rhel-version.patch \
> patches/jtreg-ChangeDir.patch \
> - patches/jtreg-TempBuffer.patch
> + patches/jtreg-TempBuffer.patch \
> + patches/jtreg-EncodedMultiByteChar.patch
>
> if WITH_ALT_HSBUILD
> ICEDTEA_PATCHES += \
> diff -r e71f6f90abb6 patches/jtreg-EncodedMultiByteChar.patch
> --- /dev/null Thu Jan 01 00:00:00 1970 +0000
> +++ b/patches/jtreg-EncodedMultiByteChar.patch Tue Jun 14 13:56:38 2011 +0200
> @@ -0,0 +1,24 @@
> +--- openjdk.orig/jdk/test/sun/net/www/protocol/file/EncodedMultiByteChar.java 2011-02-28 17:07:06.000000000 +0100
> ++++ openjdk/jdk/test/sun/net/www/protocol/file/EncodedMultiByteChar.java 2011-06-14 13:21:36.000000000 +0200
> +@@ -61,6 +61,9 @@
> + } catch (IOException e) {
> + System.out.println("Failed to create test file ");
> + e.printStackTrace();
> ++ if (file != null) {
> ++ file.delete();
> ++ }
> + return;
> + }
> +
this makes to much sense! (in case that null file was not part of the test case itself!)
If not - this part ok.
> +@@ -81,5 +84,11 @@
> + } catch (IOException ioe) {
> + ioe.printStackTrace();
> + }
> ++ finally {
> ++ System.out.println("test OK");
Is this "test ok" really ok both when exception is thrown even when it is not thrown(when it is in finally bnlok)? Also maybe "jtreg-EncodedMultiByteChar.patch main method test ok" can be better then just "test ok".
I
> ++ if (file != null) {
> ++ file.delete();
> ++ }
> ++ }
> + }
> + }
Regards J.