I have discovered FreeMind today, it’s a great tool, thanks to the developers and translators (the French doc is top notch!)
I have written a converter from Leo files to FreeMind; if someone has interest please ping me.
The only glitch is character encoding: apparently FreeMind supports UTF-8 encoded xml only when the file has a BOM and my program is not writing one, so you have to add it manually with Notepad++ for example.
I wonder why .mm files have no xml declaration on top, that would solve this encoding issue.
FreeMind natively stores unicode characters as XML character entities rather than as UTF-8 encoded characters. If you make your converter store unicode characters as XML character entities, there should be no problem.
Having a support for XML declaration at the top of the XML mind map file would be nice to have, but no one has so far implemented the support, and implementing the support seems far from critical and important.
I wonder which versions of FreeMind show the problem that you describe. There has been a regression bug introduced into FreeMind 0.9.0 RC4; reading of unicode files worked better in 0.9.0 RC3. Could it be that your problem is absent in 0.9.0 RC3?
See also the following bug:
RC4 regression: incorrect viewing of UTF-8 map
Ok, I understand now, I didn’t realize FreeMind used its own XML parser.
Because I don’t have a lot of .leo files to convert, I’ll just leave it that way.
In case someone needs the Leo to FreeMind converter, I’ve pushed the code there.