Bug 207275

Summary: [si_LK] All Repaya form of the conjuncts in sinhala with 200D is appearing wrongly - Priority - C
Product: [Fedora] Fedora Reporter: Satyabrata Maitra <smaitra>
Component: fonts-sinhalaAssignee: Rahul Bhalerao <b.rahul.pm>
Status: CLOSED RAWHIDE QA Contact:
Severity: low Docs Contact:
Priority: medium    
Version: 7CC: aalam, behdad, eng-i18n-bugs, lizhang, llch, snavin, yshao
Target Milestone: ---Keywords: i18n
Target Release: ---   
Hardware: All   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2007-11-16 06:06:07 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Bug Depends On:    
Bug Blocks: 150223, 208066, 211958    
Attachments:
Description Flags
Patch to fix all rephaya problems
none
Tested Images for verification
none
Tested Image2 for verification none

Description Satyabrata Maitra 2006-09-20 11:46:30 UTC
Description of problem: this problem is quite confusing, so describing it
throughly. Repaya form of the conjuncts can be prosuced in two ways. 1st is with
200D (Zero width Non-Joiner) and 2nd is without 200D. The shape of any
combination for both the cases should be different.
Now for Both, with and without 200D, the shape appearing, is the same.
Now the 1st problem is the shape appearing now for both, should be for the
combination which is without 200D only.
And 2nd problem is there is not correct shape available with the combination
which is with 200D.

For Example :

Combination with 200D : 0D9A + 0DBB + 0DCA + 200D + 0D9A : Actually for this
type of combination, there is no proper shape available now. This is to be
implemented still.

Combination without 200D : 0D9A + 0DBB + 0DCA + 0D9A : The shape now showing, is
the proper shape for this type of combination which is without 200D.

Version-Release number of selected component (if applicable):
fonts-sinhala-0.2-4
pango-1.14.4-1.cf6
scim-sinhala-0.2.0-1.fc6

How reproducible:
Always

Steps to Reproduce:
1. Open gedit
2. Activate SCIM with CTRL+SPACE.
3. Select RAWCODE from the scim-table in Others
4. Type the unicode sequentially as given in the 1st comment but without 'U' and 
   '+'.
5. Observe the result.
  
Actual results:
Shapes are showing same for both the set of repaya form

Expected results:
Shape should different for each set of Repaya form.

Additional info:
The list of Combinations with and without 200D is given in the first comment as
its a big list!!

Comment 1 Satyabrata Maitra 2006-09-20 11:57:00 UTC
Unicode Combinations with 200D :

U+0D9A U+0DBB U+0DCA U+200D U+0D9A
U+0D9B U+0DBB U+0DCA U+200D U+0D9A
U+0D9C U+0DBB U+0DCA U+200D U+0D9A
U+0D9D U+0DBB U+0DCA U+200D U+0D9A
U+0D9E U+0DBB U+0DCA U+200D U+0D9A
U+0D9F U+0DBB U+0DCA U+200D U+0D9A
U+0DA0 U+0DBB U+0DCA U+200D U+0D9A
U+0DA1 U+0DBB U+0DCA U+200D U+0D9A
U+0DA2 U+0DBB U+0DCA U+200D U+0D9A
U+0DA3 U+0DBB U+0DCA U+200D U+0D9A
U+0DA4 U+0DBB U+0DCA U+200D U+0D9A
U+0DA5 U+0DBB U+0DCA U+200D U+0D9A
U+0DA6 U+0DBB U+0DCA U+200D U+0D9A
U+0DA7 U+0DBB U+0DCA U+200D U+0D9A
U+0DA8 U+0DBB U+0DCA U+200D U+0D9A
U+0DA9 U+0DBB U+0DCA U+200D U+0D9A
U+0DAA U+0DBB U+0DCA U+200D U+0D9A
U+0DAB U+0DBB U+0DCA U+200D U+0D9A
U+0DAC U+0DBB U+0DCA U+200D U+0D9A
U+0DAD U+0DBB U+0DCA U+200D U+0D9A
U+0DAE U+0DBB U+0DCA U+200D U+0D9A
U+0DAF U+0DBB U+0DCA U+200D U+0D9A
U+0DB0 U+0DBB U+0DCA U+200D U+0D9A
U+0DB1 U+0DBB U+0DCA U+200D U+0D9A
U+0DB3 U+0DBB U+0DCA U+200D U+0D9A
U+0DB4 U+0DBB U+0DCA U+200D U+0D9A
U+0DB5 U+0DBB U+0DCA U+200D U+0D9A
U+0DB6 U+0DBB U+0DCA U+200D U+0D9A
U+0DB7 U+0DBB U+0DCA U+200D U+0D9A
U+0DB8 U+0DBB U+0DCA U+200D U+0D9A
U+0DB9 U+0DBB U+0DCA U+200D U+0D9A
U+0DBA U+0DBB U+0DCA U+200D U+0D9A
U+0DBB U+0DBB U+0DCA U+200D U+0D9A
U+0DBD U+0DBB U+0DCA U+200D U+0D9A
U+0DC0 U+0DBB U+0DCA U+200D U+0D9A
U+0DC1 U+0DBB U+0DCA U+200D U+0D9A
U+0DC2 U+0DBB U+0DCA U+200D U+0D9A
U+0DC3 U+0DBB U+0DCA U+200D U+0D9A
U+0DC4 U+0DBB U+0DCA U+200D U+0D9A
U+0DC5 U+0DBB U+0DCA U+200D U+0D9A
U+0DC6 U+0DBB U+0DCA U+200D U+0D9A
U+0D9A U+0DBB U+0DCA U+200D U+0DBA U+0DCA U+200D U+0DBA

Unicode Combinations without 200D :

U+0D9A U+0DBB U+0DCA U+0D9A
U+0D9B U+0DBB U+0DCA U+0D9A
U+0D9C U+0DBB U+0DCA U+0D9A
U+0D9D U+0DBB U+0DCA U+0D9A
U+0D9E U+0DBB U+0DCA U+0D9A
U+0D9F U+0DBB U+0DCA U+0D9A
U+0DA0 U+0DBB U+0DCA U+0D9A
U+0DA1 U+0DBB U+0DCA U+0D9A
U+0DA2 U+0DBB U+0DCA U+0D9A
U+0DA3 U+0DBB U+0DCA U+0D9A
U+0DA4 U+0DBB U+0DCA U+0D9A
U+0DA5 U+0DBB U+0DCA U+0D9A
U+0DA6 U+0DBB U+0DCA U+0D9A
U+0DA7 U+0DBB U+0DCA U+0D9A
U+0DA8 U+0DBB U+0DCA U+0D9A
U+0DA9 U+0DBB U+0DCA U+0D9A
U+0DAA U+0DBB U+0DCA U+0D9A
U+0DAB U+0DBB U+0DCA U+0D9A
U+0DAC U+0DBB U+0DCA U+0D9A
U+0DAD U+0DBB U+0DCA U+0D9A
U+0DAE U+0DBB U+0DCA U+0D9A
U+0DAF U+0DBB U+0DCA U+0D9A
U+0DB0 U+0DBB U+0DCA U+0D9A
U+0DB1 U+0DBB U+0DCA U+0D9A
U+0DB3 U+0DBB U+0DCA U+0D9A
U+0DB4 U+0DBB U+0DCA U+0D9A
U+0DB5 U+0DBB U+0DCA U+0D9A
U+0DB6 U+0DBB U+0DCA U+0D9A
U+0DB7 U+0DBB U+0DCA U+0D9A
U+0DB8 U+0DBB U+0DCA U+0D9A
U+0DB9 U+0DBB U+0DCA U+0D9A
U+0DBA U+0DBB U+0DCA U+0D9A
U+0DBB U+0DBB U+0DCA U+0D9A
U+0DBD U+0DBB U+0DCA U+0D9A
U+0DC0 U+0DBB U+0DCA U+0D9A
U+0DC1 U+0DBB U+0DCA U+0D9A
U+0DC2 U+0DBB U+0DCA U+0D9A
U+0DC3 U+0DBB U+0DCA U+0D9A
U+0DC4 U+0DBB U+0DCA U+0D9A
U+0DC5 U+0DBB U+0DCA U+0D9A
U+0DC6 U+0DBB U+0DCA U+0D9A
U+0D9A U+0DBB U+0DCA U+0DBA U+0DCA U+0DBA

Comment 2 Rahul Bhalerao 2006-09-26 23:01:45 UTC
1. The entire bug can be summerized as:

Combination of:
     consonant + u0DCA + u0D9A 
does not produce the repaya form.

2. The bug description says, 
      consonant + u0DCA + u0D9A
    should not produce any substitution and
     
      consonant + u0DCA + u200D + u0D9A should produce substitution.

This description is logically wrong since it demands use of
Zero-Width-Non-Joiner(u200D) for combining two consonants and a straight
combination to be desplayed without any substitution. Though its technically
possible, it can not be recommanded.

I will need to study this issue in detail and verify the implementations in the
font for other similar cases.


Comment 3 Rahul Bhalerao 2006-09-26 23:33:01 UTC
CORRRECTION:
In previous comment, the combinations are:

1. u0DBB + u0DCA + consonant (and not consonant + u0DCA + u0D9A )
2. u0DBB + u0DCA + u200D + consonant (and not consonant + u0DCA + u200d + u0D9A )


Comment 4 Rahul Bhalerao 2006-09-27 07:35:00 UTC
The character u0DBB gets a post-base *reph* form. In pango its class bit is
currently '_ct', it should be '_rv' similar to u0930 of devanagari.

Comment 5 Rahul Bhalerao 2006-09-27 07:35:46 UTC
The character u0DBB gets a above-base *reph* form. In pango its class bit is
currently '_ct', it should be '_rv' similar to u0930 of devanagari.

Comment 6 Rahul Bhalerao 2006-09-27 07:37:03 UTC
The Comment #5 is correction to Comment #4.

Comment 7 Rahul Bhalerao 2006-09-27 18:14:57 UTC
According to the standards accepted by govt. agencies in Srilanka, the rephaya
form should be displayed as a combination of 

0DBB + 0DCA + 200D + Consonant

Though it technically adds lot of redendancy work in font development, this
standard is accepted by most of the Sinhala fonts. Also its not very convincing
to use rephaya(reph) forms in this way while all other Indic scripts have a
general standard for it. Though this is what Sinhala community has been using
for developing i18n tools. Changing it all would be lot difficult.

I found the glyphs for most rephaya forms present in the font but they are
missing the ligature rules. Thus this bug can be solved in font, no modification
needed in Pango.


Comment 8 Rahul Bhalerao 2006-09-28 09:34:10 UTC
Created attachment 137294 [details]
Patch to fix all rephaya problems

Font was missing GSUB rules for the rephaya forms. I have added these rules.

Comment 9 Leon Ho 2006-09-28 11:04:45 UTC
Thanks Rahul. Built into fonts-sinhala-0.2.1-1

Comment 10 A S Alam 2007-05-02 12:23:12 UTC
Created attachment 153939 [details]
Tested Images for verification

Comment 11 A S Alam 2007-05-02 12:24:07 UTC
Created attachment 153941 [details]
Tested Image2 for verification

Comment 12 A S Alam 2007-05-02 12:37:53 UTC
snavin,
can you please check above combination whether those are correct or not?

Comment 13 Danishka Navin 2007-05-03 11:20:14 UTC
A S Alam,
I have checked above combination, those are correct.

Comment 14 Jens Petersen 2007-10-23 07:09:59 UTC
So this is fixed now?

Comment 15 A S Alam 2007-11-16 06:06:07 UTC
based on comment #13, it is marked fixed, if found further issue, please reopen
issue,
thanks