Affects Version/s: 1.5.7
Fix Version/s: None
Environment:it's more apparent on Windows
It's discussed in http://www.nabble.com/forum/ViewPost.jtp?post=21058369&framed=y
I had a UTF-8 template with non-latin content (Russian), Windows's default encoding is cp-1252 and template engine uses it on a certain stage, so non-latin characters can be broken and the resulting HTML ends up with garbage rather than Russian letters. See attachment for such a template.
Here is Guillaume's analysis:
After a quick examination, although the reader properly reads the
content of the template, we then generate some Groovy script
representing that template, and that template is transformed into an
inputstream of bytes using the engine's shell's encoding to be fed to
groovy classloader at some point.
The problem is when we get the bytes of the script, we use the shell's
encoding, and if none was provided (by means of CompilerConfiguration)
then the default system encoding is being used. So your UTF-8
characters are read as ISO something instead, hence the ugly
characters you see.