Re: bidi in unipad

From: Asmus Freytag (
Date: Wed Feb 12 2003 - 01:34:00 EST

  • Next message: Andy White: "RE: Never say never"

    At 10:57 PM 2/11/03 +0100, Chris Jacobs wrote:

    >----- Original Message -----
    >From: "Doug Ewell" <>
    >To: "Unicode Mailing List" <>
    >Cc: "Chris Jacobs" <>; <>
    >Sent: Tuesday, February 11, 2003 8:11 AM
    >Subject: Re: bidi in unipad
    >[ ... ]
    > > Asking a text editor to display the ASCII sequence "\u05c0" as RTL, but
    > > to display the ASCII sequences "\x05c0" and "/u05c0" and "\uO5cO" (with
    > > capital letters O) and "\u04c0" as LTR, just seems bizarre and would
    > > surely be more confusing than the current state of affairs.
    >I did not mean to ask to display the ASCII sequence "\u05e0" as "05e0u\"
    >I meant something else.

    ... and I think Doug understood quite well what you meant.

    He's just pointing out that by your asking to treat \u05E0 as if it was the
    actual character U+05E0, and therefore affecting the text stream around it,
    as if it were, you would create a confusing situation where inserting a
    similar looking, but not identical sequence would result in vastly
    different text layout. For example, some people use the shorthand \x05E0 in
    C wide-character strings. To be consistent, this would have to be
    recognized by unipad as well.

    For a general purpose editor, such behavior seems out of place. But in a
    programming environment, where information about the (programming)
    'language' of a plain text file is kept and is used for coloring etc., it
    might indeed make sense to show strings correctly laid out, whether or not
    a character is inserted directly, or via an escape.

    But surely, that should be an option.


    PS: My preference would be to be able to toggle between viewing the escape
    and viewing the character as it would appear in the compiled program,
    without having to change what's actually in the source file. I'm not sure
    whether honoring the bidi layout by itself would buy me very much in that
    case. In fact, it might make more sense to turn off bidi altogether when I
    am focusing on the bit-level details of a string.

    This archive was generated by hypermail 2.1.5 : Wed Feb 12 2003 - 02:25:00 EST