Learn how easy it is to sync an existing GitHub or Google Code repo to a SourceForge project! See Demo


#860 RC4 regression: incorrect viewing of UTF-8 map

Roman Podolyan

Windows version without JRE.
The RC3 version opens map, coded in UTF-8 OK, and RC-4 does not (does not treat it as UTF-8).
.mm file and screenshots of correct and incorrect look attached


  • Roman Podolyan
    Roman Podolyan

    .mm file in case

  • Roman Podolyan
    Roman Podolyan

    Correct opening of .mm content.

  • Roman Podolyan
    Roman Podolyan

    Incorrect opening

  • I confirm the problem.

    I have downloaded the attached mind map. The attached mind map opens fine in 0.9.0 RC3 while it shows weird characters in RC4.

    Environment: Windows Vista, Java 1.6.0

  • Fixed in CVS.

    • status: open --> closed
  • Tests have shown that your bug is solved with the version 0.9.0
    of the FreeMind software.
    If you do not agree and can still reproduce the bug with this
    version of the software, please reopen this bug or create a new one.

    Thanks, Your FreeMind team

  • Bug repeat in RC6. Tested with attached mindmap also. Edited map saved incorrect: letter ė was ė (in notepad++)
    Version RC3 opens corect map.

  • I confirm that the attached mind map is shown in weird characters in 0.9.0 RC6 and 0.9.0 RC4, inlike in 0.9.0 RC3; Environment: Windows Vista, Java 1.6.0. --Dan

  • Hi all,

    this can be fixed from outside: freemind uses the standard java methods to read files. There is a default encoding, that can set by -Dfile.encoding=UTF8 in the java call. With this, I got the correct file under Windows, too.

    HTH, Chris

    • status: closed --> closed-works-for-me
  • Hi Chris,

    if I understand correctly, this presupposes that the user modifies FreeMind.bat or FreeMind.sh--the launching scripts. But this workaround of adding -Dfile.encoding=UTF8 to the Java call of FreeMind does not work with FreeMind.exe, which the user would have to recompile.

    FreeMind should better set the file encoding of UTF8 directly in FreeMind code rather than relying on an external workaround that only technically-minded users can be expected to be able to implement.

    Loading the attached file worked fine in 0.9.0 RC3. Why can't it work fine again in 0.9.0 RC8? The default encoding can be set in the Java code when opening the mind map file, doesn't it?