bugFree UCS Outline Fonts - Bugs: bug #37138, Devanagari: some substitutions...

 
 

bug #37138: Devanagari: some substitutions don't work in Windows

Submitted by:  Steve White <Stevan_White>
Submitted on:  Sat 18 Aug 2012 09:44:01 AM UTC  
 
Category: character rangeSeverity: 4 - Important
Item Group: character substitution issueStatus: Proceeding
Privacy: PublicAssigned to: Steve White <Stevan_White>
Open/Closed: OpenRelease: 2012-05-03

Add a New Comment (Rich MarkupRich Markup):
   

You are not logged in

Please log in, so followups can be emailed to you.

 

Sat 22 Sep 2012 11:55:30 AM UTC, comment #3:

Regarding Sanskrit substitutions: I have no idea how to turn these features on in Windows, at least in web browsers. The 'lang="sa"' attribute seems to have no effect. In Linux it works.

Steve White <Stevan_White>
Project AdministratorIn charge of this item.
Wed 19 Sep 2012 11:09:49 PM UTC, comment #2:

There are two standards for OpenType Devanagari font shaping, named 'deva' and 'dev2', the latter replacing the former some years ago. Windows 7 and Vista use 'dev2'. I am told Linux font rendering will be going in this direction too.
They differ in that 'dev2' has more Indic font features, and somehow, evidently, there is a difference in the order the tables are applied, or just what they mean, because simply making all the 'deva' tables also executed under 'dev2' doesn't work for all the conjuncts.
So I disabled 'dev2' in some of the existing tables, made 'rkrf' and 'cjct' tables, added some dev2-specific 'blwf' tables, populated them as best as I could understand from the MS documentation.
It seems the new standard should be just easier to work with (except we're stuck also maintaining the new one too for the forseeable future).
The Rakaar forms 'rkrf' specifically handle conjuncts with ra for consonants that have a half-form (basically, the ones with a vertical descending stem), while the Conjunct 'cjct' table is meant for consonants with no half-form (ones with no vertical descending stem). The former happens before the half-form substitutions are applied, the latter after. This makes that distinction easier to sort out.
Since the bar-shaped and caron-shaped rakaars typical of this distinction are applied in a different way in 'dev2', a different 'blwf' table is in order to change the rakaar shape.
Problem: haven't figured out how to display Sanskrit in a web page on Windows yet. So far 'lang="sa"' seems to be ignored.

Steve White <Stevan_White>
Project AdministratorIn charge of this item.
Sat 15 Sep 2012 05:35:34 PM UTC, comment #1:

This may also affect some handheld e-book readers.

Steve White <Stevan_White>
Project AdministratorIn charge of this item.
Sat 18 Aug 2012 09:44:01 AM UTC, original submission:

In a short run of Hindi text viewed in Explorer or Firefox on Windows 7, viramas are still appearing.

It is strange... some replacements are clearly working, others not.

Happens in both FreeSerif and FreeSans.

The same text is displayed perfectly by Pango, Firefox, and XeLaTeX on Linux.

Steve White <Stevan_White>
Project AdministratorIn charge of this item.

 

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

Attach File(s):
   
   
Comment:
   

Attached Files

 

Depends on the following items: None found

Items that depend on this one: None found

 

Carbon-Copy List
  • -unavailable- added by Stevan_White (Submitted the item)
  •  

    Do you think this task is very important?
    If so, you can click here to add your encouragement to it.
    This task has 0 encouragements so far.

    Only logged-in users can vote.

     

    Please enter the title of George Orwell's famous dystopian book (it's a date):

     

     

    Follow 2 latest changes.

    Date Changed By Updated Field Previous Value => Replaced By
    Wed 19 Sep 2012 11:09:49 PM UTCStevan_WhiteStatusConfirmed=>Proceeding
    Sat 18 Aug 2012 09:44:01 AM UTCStevan_WhiteAttached File-=>Added Hindi-UDHR1.html, #26380

    Back to the top


    Powered by Savane 3.1-cleanup