Hi,
Looking at the ImportJavaSourceMavenCollector indicates that this is an issue on our side.
I've created a ticket to fix it asap :
http://jira.codehaus.org/browse/SONAR-948.
However providing a sample small helloworld maven project (1 pom, 1 unit test and 1 class) with Shift_JIS encoded sources would help.
Fix the recording into db with correct charset will be easy but maybe not rendering this into the webapp.
Regards
2009/6/30 Nikolay Elenkov
<nick@...>
Freddy Mallet wrote:
> As your source code is
> not UTF8 encoded, I guess you should specify the Shift_JIS character
> encoding when you create the Mysql schema.
> Could you give a try with the following command :
>
> CREATE DATABASE sonar CHARACTER SET sjis COLLATE sjis_japanese_ci;
>
Thanks. I tried it but I still get the same error?
I am right to assume that what gets inserted into snapshot_sources#data is
java source code? I haven't look at the source (no time right now...), but if
the file encoding is specified properly (in pom.xml or with a system property),
once it is read successfully, one should be able to save it again in UTF-8.
BTW, is there a way to specify trace logging for ImportJavaSourceMavenCollector,
so that I can at least track down what file the string that fails is coming
from?
