+++ This bug was initially created as a clone of Bug #619283 +++

Description of problem:
0952 is not rendering correctly. there are 2 different rendering for this character when typed alone.
1) when typed first character on any line the its invisible
2) when typed as non-first character on any line its not showing dotted circle.

Version-Release number of selected component (if applicable):
pango-1.28.0-1.fc13.i686

How reproducible:
always

Steps to Reproduce:
1.start gedit
2. try to type U0952 character
3. observe its rendering which looks invisible when typed as first character or is without dotted circle when typed as non-first character.
  
Actual results:
like other unicode characters 0953 or 0962 or 0942, is invisible as first character and not rendering with dotted circle

Expected results:
it should render with dotted circle and visible when typed as first character on any line in gedit

Additional info:

--- Additional comment from pnemade on 2010-07-29 03:45:51 EDT ---

Created attachment 435208 [details]
When typed 0952 as first character

--- Additional comment from pnemade on 2010-07-29 03:48:49 EDT ---

Created attachment 435209 [details]
When typed 0952 as non-first character

--- Additional comment from pnemade on 2010-07-29 03:49:37 EDT ---

Created attachment 435210 [details]
unicode chart 0952 rendering

--- Additional comment from psatpute on 2010-08-04 02:06:27 EDT ---

i tried to debug this last time
whenever we pass U+0952 to pango it is not getting processed by indic module (processing with basic module), thats is the basic problem
i am trying to debug it more to see why it is not getting processed by indic module

--- Additional comment from psatpute on 2010-08-12 13:04:31 EDT ---

Created attachment 438488 [details]
patch for fixing problem of U+0951 and U+0952

main problem is in glib2 function g_unichar_get_script()

this function returning "G_UNICODE_SCRIPT_INHERITED" for U+0951 and U=0952

that's why pango is processing these characters with Basicfc engine

g_unichar_get_script() should return G_UNICODE_SCRIPT_DEVANAGARI as it is doing returning for other Devanagari script characters

while more investigation i understood that this information is defined in gscripttable.h which is generated from Unicode Character Database Script.txt file

I am writing to Unicode for fixing it,

but it will take some time, attaching patch for fixing it in glib2

--- Additional comment from psatpute on 2010-08-16 03:02:31 EDT ---

i had discussion on same topic in Unicode mailing list

and the answer i got is,

if any unicode character used in multiple script, it will have G_UNICODE_SCRIPT_INHERITED property, it means characters will inherit script property from  previous character. so this is write things as per glib2

moving again bug to Pango, as other icu and qt handling it properly

http://www.unicode.org/mail-arch/unicode-ml/y2010-m08/0148.html

--- Additional comment from psatpute on 2010-08-17 02:07:32 EDT ---

upstream bug https://bugzilla.gnome.org/show_bug.cgi?id=627027

--- Additional comment from psatpute on 2010-09-14 04:58:54 EDT ---

This is very low priority bug presently, its is not affecting any thing.

This character not in used in any major Indian language. Even in real text this always comes with some characters preceding it, in that case it is rendering properly.

It is rendering improper only when it comes alone.