[CXF-2937] character encoding in cxf-codegen is wrong

• Type: Bug
• Status: Closed
• Priority: Major
• Resolution: Fixed
• Affects Version/s: 2.2.9
• Fix Version/s: 2.5.4, 2.6.1
• Component/s: Tooling

Environment

windows, sun java 6

• Assignee: Daniel Kulp
• Reporter: Martin Goldhahn
• Created: 11/Aug/10 13:33
• Updated: 05/Jun/12 13:31
• Resolved: 03/May/12 16:51

Description

We develop on linux and Windows concurrently. Therefor we choose to use UTF-8 as source encoding. We also set the property project.build.sourceEncoding in maven to utf-8.

Further we have a utf-8 encoded WSDL with some comments in Norwegian (containing letters like åæø). On Windows the generated code contains these characters as ISO-8859-1 even though I set -Dfile.encoding=utf-8 on the maven command line.

A workaround is to set the property project.build.sourceEncoding in a maven profile and activate it by OS.

Issue Links

• relates to: CXF-2450 WSDL2java uses system encoding when generating java classes