The database settings are OK for French characters. If want us to find the cause of your problem, you need to describe the generation of the Web report, especially the used technologies/tools in detail. You should also precisely define what you mean by "are not showing". Are they missing entirely from the screen?
Thank you for your help.
I created a web report using Visual Studio 2010. The report is pulling the information from the Oracle database and has been deployed on a web site.
In the web browser (Internet Explorer), the French characters on the report are just missing from fields defined as NVARCHAR2 and showing as a spaces for fields defined as VARCHAR2.
I checked the browser and it displays correctly French characters on different sites.
1. Using the latest Oracle Data Access components for VB.Net ver. 220.127.116.11.20 release 5
2. Development on Windows Server 2008
Could you check the encoding of the page that is displayed and also click on View Source to see what the HTML page contains in place of the French letters? Let's concentrate on VARCHAR2 and forget NVARCHAR2.
Encoding is Unicode (UTF-8) but I also tried with 'Western European': same problem, no special characters are displayed.
Unfortunately, I cannot check the incorrect words via 'View Source' in Internet Explorer. The reason is that I browse the Oracle database through Microsoft Report Viewer that is designed by Microsoft in a way that doesn't allow you to view the database content.
When I open 'View Source' in IE, I can only see strings of random characters, it is probably some encryption...
Note: I have the same problem when I open the Oracle tables in Ms Access through ODBC. Instead of French characters I can only see question marks.
However, the value in the database does not seem to make sense to me either. Could you tell me what is the expected string that corresponds to the DUMP result and also, what you exactly see on the screen where this string should be displayed.
I just realized there is something wrong with the CSV file I upload to the database table. I get this file from an AS/400 database and if I open it in Notepad, I can see small squares instead of French characters. I verified the ASCII code for such a square that corresponds to an E with accent and I got: 144. This is correct. So I assumed the CSV file is correct.
However, I was wrong. If I copy and paste an E with accent from a web page directly into the Oracle table (using TOAD), it shows correctly on the report!
Why the ASCII code behind the French characters in Notepad is right but the characters are showing as small squares, it' unclear...
As I mentioned, we get the CSV file from a DB2 database using an IBM application (iSeries) so I'll have to investigate what characters sets are supported, etc.
Thank you for your help.
OK it is now resolved! It was the character set ID in the ISeries apllication that had to be set to a special value, 819.
Actually, if it is supported, you should use 1252 instead of 819. 819 corresponds to WE8ISO8859P1, 1252 corresponds to WE8MSWIN1252, your database character set. WE8ISO8859P1 is a subset of WE8MSWIN1252 and does not support some characters, such as '€' (Euro sign), or 'œ' (oe ligature). This may or may not be an issue, depending on what data you get from AS/400. Code 144 for 'É' is valid, indeed, however not in WE8MSWIN1252 but in US8PC437 or WE8PC850.