Unicode INI file, read on NT 4 crashes app

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts
  • jlm13
    Forum Member
    • Aug 2003
    • 2

    Unicode INI file, read on NT 4 crashes app

    Hi all,

    I have a UNICODE text file containing strings for every message, image string, and error. The reason it's unicode is I plan on having multiple languages contained within the same file, including Korean (hence my unicode requirement).

    The file is structured as an INI file, and I use the INIFile object to read specific entries out as I go. This works fine under Win2k and up -- however, when I try to run under NT4 SP6A, the autorun app inevitably crashes.

    How do I know it's the unicode format causing the problem? When I change the file to ANSI, everything works fine.

    Please HELP!

    Thanks,
    John
  • McGargle
    Forum Member
    • Jun 2003
    • 8

    #2
    Re: Unicode INI file, read on NT 4 crashes app

    Are you using any unusual fonts? Not all fonts support all Unicode characters but would support most ANSI characters, so that might explain why ANSI works. I believe -- but could be wrong -- that under WinNT only the Arial, Times, and Courier New fonts support the whole Unicode character set. You also might need to use OpenType fonts instead of TrueType fonts. You would definitely need fonts that support each writing system (Kanji, etc.)

    Another possibility is that you're using supplementary characters (see http://www.i18nguy.com/surrogates.html, for example). If so, you might need to install UniScribe (see http://msdn.microsoft.com/library/de...crib_44f9.asp, for example).

    Comment

    • jlm13
      Forum Member
      • Aug 2003
      • 2

      #3
      Re: Unicode INI file, read on NT 4 crashes app

      First, thanks for your reply!

      I don't think that the fonts are the issue -- I use Arial everywhere, and what I'm seeing is not a display issue but an out-and-out wholesale collapse of the software. I'm betting the improved Unicode support found in Win2k and up might have something to do with why it works there and not in WinNT (for example, I did a bit more testing, and found that Korean text shows up correctly on a Korean W2K OS, which makes me happy!).

      So for now, depending on which OS I'm loading on, I use ANSI if it's NT and Unicode if it's 2K and up (fortunately, we'll be phasing out support for NT by next December).

      Thanks again!

      Comment

      Working...
      X