CreativePro Forum

Return to Member Forum

• Author
  Posts
• • December 2, 2015 at 10:57 am #79844
    Nasim Ali

    Member

    So I’m trying to convert certain text encoded in legacy encoding to Unicode. For that I’ve made a script to search and replace the characters. The conversion is mostly fine except for certain replacements. I deal with Indic text.
    Suppose I’m converting The legacy text

    “@eþi” where

    @ = ?, e = ?, þ = ?+U200C , i = ?

    It should convert to ???? but instead shows something else. Any further ?? that I type gets immediately converted too. When I copy this text and paste anywhere else it appears okay. Even when I switch to a fresh inDesign document and type the same thing in Unicode it appears fine. Interestingly this happens on only 2 fonts, if I select other fonts the text shows up normal. Unfortunately these two fonts are my staple and author of one is retired and other is a free one.

    View post on imgur.com

    
    The image shows: Top- Normal text that appears when I type or copy in a fresh document, Middle- Jumbled up text when I run the script, Lower- the legacy text is selected

    I’m using inDesign CS6 on Windows 7 machine.

  • December 3, 2015 at 3:11 am #79863
    Chris Thompson

    Member

    It’s hard to see exactly what’s happening as the forum software has converted the (Oriya?) characters to ???? question marks.

    A couple of thoughts –
    1. Have you checked “missing glyph protection”?
    2. You said of those two fonts that you are using: “author of one is retired and other is a free one”. I would suspect something not totally correct in the font, which causes problems for InDesign, but is accepted by other programs, perhaps?

    U200C = ZWNJ – presumably you’re trying to *prevent* a ligature/conjunct there.
    There’s a mention of ZWNJ problems in Oriya on a google groups forum here: https://groups.google.com/forum/#!topic/oriya_it/k6ysvNZOLMg

    I have to sympathise – those legacy Indic scripts are a real nightmare to deal with.
    Good luck,
    Chris.

  • December 3, 2015 at 6:09 am #79866
    Nasim Ali

    Member

    Well! My mistake! It turns out in the top image where the font apparently works fine is because ‘World Ready Composer’ is not turned on. When I turn it on those two fonts display the same problems. Obviously the problem is with Open Type GSUB tables in the two fonts which have been poorly defined, so they work out fine in other programs like Word etc but start their shenanigans when inDesign enforces strict rules.

    Delving into GSUB rules and fixing the culprit will take too long. So I just copied the glyphs onto another font that works. Looks fine except that baselines and Ascent are different. Gotta fix that now :(

  • December 3, 2015 at 8:39 am #79877
    Chris Thompson

    Member

    Very good!

    I’ve had similar behaviour/shenanigans from a Nastaliq font for Urdu – it behaved differently in different programs. I gave up eventually.

    C

• Author
  Posts