My source files are in unicode text files, and Natural Docs cannot "understand" them.
If they are in UTF-16, this is a known issue with Perl. If they are in UTF-8 they should work, so put an example file into a zip file (so nothing gets changed by SourceForge) and post it here.
Yes, they are UTF-16 (1char = 2 bytes)
This is the default (and non-changeable) format of QTP for source files (vbs)
So a workaround is to convert UTF-16 -> UTF-8, then run NaturalDocs on the temp files.
Any better suggestion you have maybe?
New version of LineReader.pm
Use the attached file to replace LineReader.pm in 1.5 ([ND Dir]\Modules\NaturalDocs\LineReader.pm) and see if it works. You'll have to run Natural Docs with -r once to apply the changes everywhere. Unless there's issues, this will appear in 1.51.
I overwrote the prev LineReader.pm with your new one, re-generated the doc, and it works without any problem ! Thank you! Now I dont need to use the unicode->ascii converter anymore! Thanks again.