Re: [perl #58656]: unicore/mktables expects wrong syntax from CaseFolding.txtfor Turkish

[karl williamson <public@khwilliamson.com>]

I did some research on this.  The change to the syntax of the 
CaseFolding.txt file occurred between Unicode 3.1 and 3.2.  I could not 
find a discussion of the changes on the internet, and especially not in 
the Unicode's discussion about the changes.  This change was snuck in 
silently.

Anyway, it appears that the Unicode folks decided they were doing the 
wrong thing in 3.1, and changed it in 3.2.  Perl is still looking for 
the 3.1 syntax.

And the change introduces an incompatibility between Turkic and 
non-Turkic handling.  In other words, it now becomes locale dependent. 
The change was to remove the I entries in the file and replace them with 
modified versions marked instead by T.  One of the T entries now maps an 
ASCII capital I to a lower case dotless I, clearly the wrong thing for 
non-Turkic languages.  So we get to choose whether Perl supports Turkic, 
or whether it supports every other language which uses an uppercase I. 
It is clear that Turkic loses this one.

I see in the archives of p5p that people were looking for perl to 
support Turkic.  I didn't see a resolution to that.  But it would 
require special handling.

Anyway, I'll submit a patch to remove the references to the obsolete 
syntax.  The recommendation in the 3.1 file was to include its I 
entries.  The recommendation in the 3.2 and later versions of the file 
is to not include the T entries (because it screws up non-Turkic 
languages).

---

[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index][Thread Index][Top&Search][Original]

---