bugFree UCS Outline Fonts - Bugs: bug #50706, Oriya candrabindu overlaps...

 
 

bug #50706: Oriya candrabindu overlaps ascenders

Submitter:  David Corbett <dscorbett>
Submitted:  Sun 02 Apr 2017 05:48:43 PM UTC
   
 
Category:  individual character(s) Severity:  3 - Normal
Item Group:  letter positioning poor Status:  Need info
Privacy:  Public Assigned to:  Stevan_White
Open/Closed:  Open Release:  * development
* Mandatory Fields

Add a New Comment Rich Markup
   

Jump to the original submission

Sun 23 Apr 2017 10:32:09 AM UTC, comment #11: 

Note: nice resources at http://www.odia.org/
Note: I wrote to Mr. Bayaryn who made the Dhatupatha file, but have not yet heard back.

Steve White <Stevan_White>
Group administrator
Sat 22 Apr 2017 07:19:36 AM UTC, comment #10: 

Hi,

(I know this is a side issue but, having seen an application of a feature, I want to support it.)

I went ahead and replaced those characters meant to be vedic marks, and then added anchors to the Oriya range to support them.  See attachments.

(file #40473, file #40474)

Steve White <Stevan_White>
Group administrator
Thu 20 Apr 2017 11:38:41 PM UTC, comment #9: 

The text uses U+030D COMBINING VERTICAL LINE ABOVE but I think it is meant to be U+0951 DEVANAGARI STRESS SIGN UDATTA. Similarly, it uses U+0331 COMBINING MACRON BELOW for U+0952 DEVANAGARI STRESS SIGN ANUDATTA.

David Corbett <dscorbett>
Thu 20 Apr 2017 10:26:21 PM UTC, comment #8: 

This is already a very nice text.

You mention ଗିଁ -- and I agree that could be better rendered. 
If you could find a picture of the book it's taken from, that would settle that question.

But there are still the very high letters.  Does the candrabindu ever interact with them.

While you're at it, this text uses U+030d.  I had never seen that, didn't expect it, and it isn't rendered well.  What does it do, how  should it be positioned?

Steve White <Stevan_White>
Group administrator
Thu 20 Apr 2017 08:14:05 PM UTC, comment #7: 

Oh, it’s definitely not wrongly encoded; I just can’t find an image of it. A plain text source is http://www.odia.org/uneditedSanskrit/major_works/dhatupatha_index_svara_odia.txt. It contains reph + candrabindu (e.g. ⟨ର୍ଣ୍ଣଁ⟩) and high vowel sign + candrabindu (e.g. ⟨ଗିଁ⟩).

David Corbett <dscorbett>
Thu 20 Apr 2017 07:22:47 PM UTC, comment #6: 

Hi David,

I will certainly do something if we can find a rule, or even an example of such a sequence.  Without that however, I have to suspect that any such string is just wrongly coded -- if it's wrongly coded, I think it's wrong to try to make it look better.

I'll do some more searches tonight for combinations of consonant + high vowel + candrabindu.

Steve White <Stevan_White>
Group administrator
Thu 20 Apr 2017 05:44:11 PM UTC, comment #5: 

Reph doesn’t make sense on an independent vowel letter, but it does make sense on a consonant letter with a vowel sign; for example, ⟨ର୍ଭୌ⟩ “rbhau”.

I think the candrabindu goes above the letter, perhaps hanging a bit off the right edge. For example, nearly every page of https://archive.org/details/TranslationOfTheMeaningsOfTheNobleQuranInTheOriyapdf has a candrabindu, and each is directly above its base letter. http://www.odialanguage.com/help/lekhaodia/images/OR08.JPG shows the candrabindu offset to the right. http://languagelog.ldc.upenn.edu/myl/ldc/IndianScriptsUnicode.html even says it “is sometimes treated as a spacing mark”. Unfortunately I haven’t found a instance of a candrabindu combined with another high diacritic (reph or vowel sign). I suggest putting the candrabindu directly above the base letter, unless there is another high diacritic, in which case I suggest putting the candrabindu above the base letter and to the right of the other diacritic. I don’t have any evidence, but it is what other Indic scripts do, so it might be right, and it is at least less wrong than the current behavior.

David Corbett <dscorbett>
Thu 20 Apr 2017 05:29:36 PM UTC, comment #4: 

The om behavioor in fig 12-23 was already supported.
But it turned out there were already two ligature glyphs for Oriya om in the font, so I made both accessible.

ଓଁ with ZWJ ଓ‍ଁ with ZWNJ ଓ‌ଁ

Steve White <Stevan_White>
Group administrator
Thu 20 Apr 2017 05:07:31 PM UTC, comment #3: 

According to the Unicode Standard ch. 12, the section on Oriya, under

"The sacred syllable om is formed by U+0B13 oriya letter o and U+0B01 oriya sign candrabindu."

Rules for ZWJ and ZWNJ are given.

But that's the only case listed.  And the o wasn't ever problematic anyway, because it doesn't have an ascender.

Steve White <Stevan_White>
Group administrator
Thu 20 Apr 2017 04:25:38 PM UTC, comment #2: 

Hi David,

I don't think the reph makes sense on a vowel.  As I understand it, the phonetic effect of the reph is to put an r sound before a consonant with no intervening vowel sound, like rk or rs.  I find no occurrences of any of the combinations < ର୍ଐ ର୍ଔ ର୍କୈ ର୍କୌ ର୍କୖ ର୍କୗ >

The candrabindu isn't very clear to me.  It seems to be generally rather rare in normal text.  Web searches turned up instances only of the combination ଐଁ, for instance
https://or.m.wikipedia.org/wiki/%E0%AC%B8%E0%AC%B0%E0%AC%B8%E0%AD%8D%E0%AC%B5%E0%AC%A4%E0%AD%80

So the question is: where does the candrabindu go in this case?  It might go over the letter or to the right.  I just don't know

Steve White <Stevan_White>
Group administrator
Tue 18 Apr 2017 06:13:57 PM UTC, comment #1: 

The reph (U+0B30 ORIYA LETTER RA, U+0B4D ORIYA SIGN VIRAMA) overlaps those vowel signs too.

David Corbett <dscorbett>
Sun 02 Apr 2017 05:48:43 PM UTC, original submission:  

In FreeSerif, U+0B01 ORIYA SIGN CANDRABINDU overlaps the ascenders of:

  • U+0B10 ORIYA LETTER AI
  • U+0B14 ORIYA LETTER AU
  • U+0B48 ORIYA VOWEL SIGN AI
  • U+0B4C ORIYA VOWEL SIGN AU
  • U+0B56 ORIYA AI LENGTH MARK
  • U+0B57 ORIYA AU LENGTH MARK
David Corbett <dscorbett>

 

(Note: upload size limit is set to 16384 kB, after insertion of the required escape characters.)

Attach Files:
   
   
Comment:
   

Attached Files
file #40473:  dhatupatha-fix.py added by Stevan_White (369B - text/x-python)
file #40474:  Dhatupatha - Vedic Oriya.pdf added by Stevan_White (519KiB - application/pdf)

 

Depends on the following items: None found

Items that depend on this one: None found

 

Carbon-Copy List
  • -email is unavailable- added by Stevan_White (Posted a comment)
  • -email is unavailable- added by dscorbett (Submitted the item)
  •  

    There are 0 votes so far. Votes easily highlight which items people would like to see resolved in priority, independently of the priority of the item set by tracker managers.

    Only logged-in users can vote.

     

    Follow 4 latest changes.

    Date Changed by Updated Field Previous Value => Replaced by
    2017-04-22 Stevan_White Attached File- Added dhatupatha-fix.py, #40473
        Attached File- Added Dhatupatha - Vedic Oriya.pdf, #40474
    2017-04-20 Stevan_White StatusNone Need info
        Assigned toNone Stevan_White

    Back to the top

    Powered by Savane 3.13-f8d8.
    Corresponding source code