From: Rick Cameron (Rick.Cameron@businessobjects.com)
Date: Fri Dec 09 2005 - 11:31:56 CST
Can you explain your situation some more?
How is the data encoded in the file - as UTF-16 or UTF-32?
How do you want to store the data in memory - as UTF-16 or UTF-32?
If the answer to both questions is UTF-16, I don't think there is a real
problem here. wifstream will read in UTF-16 code points and store them
in (16-bit) wchar_t, without interpreting them. If there are surrogate
pairs in the text, it is up to you to interpret them correctly after
they've been read in.
You may want to look at ICU (http://icu.sourceforge.net/). It provides a
lot of help in dealing with Unicode, and is much more up-to-date than
the Microsoft runtime libraries.
[mailto:firstname.lastname@example.org] On Behalf Of Sajal Maity
Sent: Friday, 9 December 2005 2:42
Subject: Unicode file reading Problem??
I am getting problem while i am reading a unicode file using
wistream/wifstream in VC6.VC's wchar_t doesn't support unicode character
which is more than 2 bytes i.e. in surrogate area.My work is to read any
type of UTF text file. I got my result using CFile in MFC. But there i
used WideCharToMultiByte & the reverse one. But still i have some
confusion whether those function converts properly for every UTF?
Is there any other way to read unicode character?
Whether UNICODE CONSORTIUM itself has any readymade functions?
This archive was generated by hypermail 2.1.5 : Fri Dec 09 2005 - 11:35:25 CST