Skip to content

WM_CHAR does not function properly when _UNICODE is defined #283

@GitMensch

Description

@GitMensch

My app that uses PDCursesMod is compiled with the Use Unicode Character Set option with MSVC++, which will ensure all Win32 API calls use the wide-character variants, all of which use UTF16 encoded characters and strings regardless of the user's selected locale.

That means that when _UNICODE is defined the character sent to the WM_CHAR message is UTF16 encoded -- and PDCursesMod adds it to key queue directly via add_key_to_queue((int) wParam); My app defines PDC_FORCE_UTF8, so I was assuming that internally this character would be converted to a sequence of UTF8 bytes to be received by getch() calls.

However, this was not the case, and parsing some non-ascii characters was problematic.

Originally posted by @clangen in #260 where @Bill-Gray presented a different (still unfinished) approach.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions