Re: Private Use Area

From: Peter_Constable@sil.org
Date: Wed Apr 30 2003 - 10:12:45 EDT

  • Next message: John H. Jenkins: "Re: TTC fonts"

    Eric:

    > >I believe these have just been
    > >codepoints used for internal purposes by GDI
    > >
    > And Netscape
    >
    <http://lxr.mozilla.org/seamonkey/source/intl/ctl/src/thaiShaper/thai-x.c>
    > and ICU
    > <http://oss.software.ibm.com/cvs/icu/icu/source/layout/ThaiShaping.cpp>,
    > and probably FreeType/Gnome...

    I looked at the Mozilla code. To the extent that I understand C (which,
    admittedly is only to a limited degree), I see 8-bit values used for
    presentation forms. For instance, in

    502 else if (is_char_type (cluster[0], SpltTailCons) &&
    503 is_char_type (cluster[1], BelowVowel|BelowDiac)) {
    504 glyph_lists[0] = PANGO_MOZ_MAKE_GLYPH (tailcutcons (ucs2tis
    (cluster[0]), shaping_table));
    505 glyph_lists[1] = PANGO_MOZ_MAKE_GLYPH (ucs2tis (cluster[1]));

    the Unicode codepoints are converted to TIS before being passed to the
    tailcutcons and PANGO_MOZ_MAKE_GLYPH macros, and there's a lookup into
    ThaiShapeTable, which is working in an 8-bit codespace (even if 16- or
    32-bit datatypes are being passed around, the values are all 0xA0..0xFF).
    So, I don't see reference to PUA codepoints. But then, I also don't know
    how Pango is getting from these character codes to glyph IDs, and perhaps
    somewhere along the way it's assuming U+F7xx code points in a Unicode cmap
    table.

    - Peter

    ---------------------------------------------------------------------------
    Peter Constable

    Non-Roman Script Initiative, SIL International
    7500 W. Camp Wisdom Rd., Dallas, TX 75236, USA
    Tel: +1 972 708 7485

    This archive was generated by hypermail 2.1.5 : Wed Apr 30 2003 - 11:10:41 EDT