I am developing unicode to general text program in VSTO (VB.NET).  Iam 
working on Gurmukhi/Punkabi unicode i.e. Raavi. There are many chars in it 
which are formed ny combining two or more diferant chars like ਕ + ਿ = ਕਿ or ਕ 
+ ਿ + ਂ = ਕਿਂ. Earlier i ws not able to search any single instance of ਿ or ਂ 
but later i tried use wildcard option along with [ਿ] . This char is sarched 
perfectly but while replacing it with general char "i" replaces whole "ਕਿ" or 
 "ਕਿਂ" with single char "i". Can anyone help How can i search single chars 
and replace with some other chars to normalize Unicode to General font 
convertion ? ? ?

"Chand" <Chand@discussions.microsoft.com> wrote:
> I am developing unicode to general text program in VSTO (VB.NET).
> Iam working on Gurmukhi/Punkabi unicode i.e. Raavi.
> There are many chars in it which are formed ny combining two or more
> diferant chars like ਕ + ਿ = ਕਿ or ਕ + ਿ + ਂ = ਕਿਂ.
> Earlier i ws not able to search any single instance of ਿ or ਂ
> but later i tried use wildcard option along with [ਿ] .

That's a problem I know from diacritics in other languages. The diacritic 
alone, or the letter it's combined with alone, isn't found.
The work-around I use is the same as yours.
I've filed a bug report years ago, but didn't hear back if it's going to be 
fixed.

> This char is sarched perfectly but while replacing it with general
> char "i" replaces whole "ਕਿ" or "ਕਿਂ" with single char "i".
> Can anyone help How can i search single chars and replace
> with some other chars to normalize Unicode to General font
> convertion ? ? ?

I'm not sure from your description what you want to replace with what.
Since you're doing a wildcard replacement, you can re-use anything matched 
if you put parentheses around it in "Find what", and then use the 
appropriate placeholder in "Replace with" (\1 for the first expression in 
parentheses, \2 for the second...).

I have no idea about Gurmukhi/Punkabi unicode. In other scripts using 
ligatures and diacritics like say Arabic, the ligatures form automatically 
if a well-designed font is used. The glyphs for ligatures may be only in 
Unicode for compatibility reasons -- because old fonts don't do the 
ligatures, or because old files used the ligatures since fonts back when 
didn't do them automatically.
So maybe ask in a group with knowledgeable people (say 
microsoft.public.word.international.features) if the replacements you are 
trying to make are sensible, or if instead you can use a font that handles 
the ligatures automatically?

Regards,
Klaus 

I'm unsure if this addresses the problem, but the code that I wrote to 
replace both ANSI and Unicode character strings with ligatures required:
  1.  Specify match case (prevents character variants from matching)
  2.  Exclude small caps
  3.  Enable format matching (to detect caps and bold/italic properly)

Cheers

"Klaus Linke" wrote:

> "Chand" <Chand@discussions.microsoft.com> wrote:
> > I am developing unicode to general text program in VSTO (VB.NET).
> > Iam working on Gurmukhi/Punkabi unicode i.e. Raavi.
> > There are many chars in it which are formed ny combining two or more
> > diferant chars like ਕ + ਿ = ਕਿ or ਕ + ਿ + ਂ = ਕਿਂ.
> > Earlier i ws not able to search any single instance of ਿ or ਂ
> > but later i tried use wildcard option along with [ਿ] .
> 
> That's a problem I know from diacritics in other languages. The diacritic 
> alone, or the letter it's combined with alone, isn't found.
> The work-around I use is the same as yours.
> I've filed a bug report years ago, but didn't hear back if it's going to be 
> fixed.
> 
> > This char is sarched perfectly but while replacing it with general
> > char "i" replaces whole "ਕਿ" or "ਕਿਂ" with single char "i".
> > Can anyone help How can i search single chars and replace
> > with some other chars to normalize Unicode to General font
> > convertion ? ? ?
> 
> I'm not sure from your description what you want to replace with what.
> Since you're doing a wildcard replacement, you can re-use anything matched 
> if you put parentheses around it in "Find what", and then use the 
> appropriate placeholder in "Replace with" (\1 for the first expression in 
> parentheses, \2 for the second...).
> 
> I have no idea about Gurmukhi/Punkabi unicode. In other scripts using 
> ligatures and diacritics like say Arabic, the ligatures form automatically 
> if a well-designed font is used. The glyphs for ligatures may be only in 
> Unicode for compatibility reasons -- because old fonts don't do the 
> ligatures, or because old files used the ligatures since fonts back when 
> didn't do them automatically.
> So maybe ask in a group with knowledgeable people (say 
> microsoft.public.word.international.features) if the replacements you are 
> trying to make are sensible, or if instead you can use a font that handles 
> the ligatures automatically?
> 
> Regards,
> Klaus 
> 
> .
>