JEditorPane should improve its text, especially HTML text, support by
providing more flexible character encoding support. There are two
different requirements:

 1) HTML "charset" detection

    HTML (4.0) supports "charset" so that it can specify the character
    encoding in which the document is coded. There is RFE 4436969 for this
    request which was closed as the generic HTML 4.0 RFE 4308782. However,
    "charset" support is crucial in some countries where different
    character encodings are often used, like Japan. Currently, you can
    import text encoded in the platform default encoding into
    JEditorPane. For example, you can't edit Shift-JIS (SJIS) encoded HTML
    in EUC-based JIS (EUC-JIS) environment on Solaris. The request here
    is to support "charset" separately from (or before) HTML 4.0 support.

 2) Add methods to be able to specify a character converter

    JEditorPane lacks APIs for specifying a character converter to be able
    to read/write text in a different character encoding from the default
    one. It's necessary to be able to specify converters separately for
    input and output.

    Sometimes character encoding of HTML text is unknown. In this case, an
    "auto detect" converter should determine its encoding and perform
    necessary conversion. But output requires a different converter. See
    RFEs 4320428 and 4426483.