[WICKET-6586] Broken JavaScript due to fix charsetName in JavaScriptPackageResource

• Type: Bug
• Status: Resolved
• Priority: Major
• Resolution: Fixed
• Affects Version/s: 7.10.0
• Fix Version/s: 7.11.0, 8.2.0, 9.0.0-M1
• Component/s: wicket
• Assignee: Martin Tzvetanov Grigorov
• Reporter: Andreas Müller
• Created: 13/Sep/18 08:42
• Updated: 16/Oct/18 07:58
• Resolved: 16/Oct/18 07:28

Description

When running Wicket in DEPLOYMENT Mode, DefaultJavaScriptCompressor will be set as JavaScript Compressor by default to strip Comments and Whitespaces from JavaScript files.

In JavaScriptPackageResource the processResponse method sets the charsetName fix to "UTF-8", when compressing JS files:

@Override
protected byte[] processResponse(final Attributes attributes, byte[] bytes)
{
 final byte[] processedResponse = super.processResponse(attributes, bytes);

 IJavaScriptCompressor compressor = getCompressor();

 if (compressor != null && getCompress())
 {
  try
  {
   String charsetName = "UTF-8";
...

If those JavaScript files are not in UTF-8, the Compressor can break those files.

E.g. when using ISO-8859-15 and having umlauts in the file, the umlauts will be replaced by the REPLACEMENT CHARACTER "\uefbfbd".

There should be a way to specify the charsetName in JavaScriptPackageResource.

E.g. by calling something like getResourceSettings().setJavaScriptCharset()

Issue Links

• links to: A Pull Request
• links to: GitHub Pull Request #295