Unicode CSV
From: futbridge (futbridge_at_discussions.microsoft.com)
Date: 10/25/04
- Next message: Olivier Matrot: "Re: Ado & oracle ( again )"
- Previous message: Steve Hoberecht [Microsoft]: "RE: Query file system using ADO?"
- Messages sorted by: [ date ] [ thread ]
Date: Mon, 25 Oct 2004 10:53:04 -0700
I am trying to connect to a Unicode (small endian) CSV file using ADO. I have
succesfully connected to the file but the recordset ADO is giving me gives
the wrong data.
I have tried to modify the files schema.ini entry and set
CharacterSet=Unicode.
This worked fine until I have tried using different text on the CSV mixing
Japanese, Chinese, Korean, Arabic, Sanskrit, etc. The non-English texts were
replaced by question marks ('?'). My language setting in the Control Panel's
Regional Options was set to English, and so I set it to Japanese. I was now
able to see Japanese and some Chinese texts but not the Arabic, Sanskrit...
And then I changed the language setting to Arabic and now the Arabic texts
are returned but not the Japanese, Chinese texts.
I have compiled my application and already set UNICODE in the build setting
and using the wide-char string of the returned _bstr_t.
I have tried to debug the application anf found out that the question marks
are actual '?' characters and not just limitation of the fonts or anything.
Also, I have noticed that the column name returned was correct regardless of
what locale/language setting is set. I have tried mxing different charaters
as always get the correct field names.
Please send me information regarding this issue. Any feedbackwill be
greately appreciated.
- Next message: Olivier Matrot: "Re: Ado & oracle ( again )"
- Previous message: Steve Hoberecht [Microsoft]: "RE: Query file system using ADO?"
- Messages sorted by: [ date ] [ thread ]
Relevant Pages
|
|