Building NSIS as a pure Unicode application
I am wondering: Can NSIS be built as a pure Unicode application for Windows instead of as an ANSI application? (Of course, I'm not trying to target Windows 9x/Me in this case.)
Why do I need to do this? I am endeavoring to produce a data-centric application, like an encyclopedia, where I wish for the user interface to appear in the same language as the data, regardless of the language setting of the host system. (And I don't want to require the user to change their system settings in order to run my program.) For example, say that the data is Japanese and the user's Windows XP system is configured to default to English. I still want the user interface for the program to appear in readable Japanese. Everything works great for my program since it is built for Unicode. However, the NSIS installer is still an ANSI application. So when the installer is run, the internationalized text that appears in the title bar and the rest of the user interface is unreadable if the system language is English instead of Japanese. (Again, I don't want to require the user to change their system settings in order to install the program. Please also assume that they have the required fonts and language packs installed on their system for Japanese or whatever other language I need to support.)
Please let me know if this is possible or if there is anyone else who is doing or attempting this. If it is not possible now, what do you think it would take on someone's part to modify the NSIS source code to accomplish it?
Thank you for help and efforts with NSIS.
Regards,
Vargas