italian charset problem

Background:

We have an Italian Site setup on dynamo 5.1.1 with patch2 . The backend is Oracle 8.1.6 and the character set on the backend is UTF-8.

Issue:

When we are trying to extract the users registered through the italian site out of the database into a flat file on unix box, the italian character set that are
coming is funny i.e. garbage! (this we are able to notice by opening this generated flat file in the browser and viewing it in UTF-8 encoding)

Observation:

Through the dynamo administration (browser) interface , having the browser encoding set to UTF-8 , if we are querying the database, the records being displayed on the screen are correct.

Example:

This is the "italian city" that appears when we try to view the flat file (with extracted records) on the browser in UTF-8: Cantù

When we run the query (from the browser admin screen having the encoding to UTF-8) to pull out the city we get it as Cantù  (This is the correct city name)

We are using UTL_FILE package for extraction work. Some settings in init.ora OR database configuration OR registry OR environment variables need to be set?

Would really appreciate if you could help out in understanding why this is occuring.

Giving 100 points but more for good answers and points for expert for more than one good answer are not ruled out.

--- k_murli_krishna
LVL 17
k_murli_krishnaAsked:
Who is Participating?
 
mnicorasCommented:
Hi,

UTL_FILE has some problems when it tries to use specific national characters. Almost the same problem has been post it by somebody for chinese characters :-)
Anyhow I suggest the same thing now.
First of all, if you use the UTL_FILE try to set in your procedure which creates the flat file :

alter session set nls_language = 'ITALIAN'

and

alter session set nls_territory = 'ITALY'

this can be done by EXECUTE IMMEDIATE '....'  or using DBMS_UTILITY.EXEC_DDL_STATEMENT ('....') where  '....' = ddl command.

also on UNIX box should have the Italian fonts available to seen them from there.

any questions? I'm waiting...

kind regards,
Marius Nicoras
0
 
333Commented:
should be your text file UTF8 encoded? or some italian encoding?
also, check NLS_LANG environment variable on machine, where extraction runs. NLS_LANG's characterset portion should be set to either UTF8 or italian (depends on what you want). if you are extracting from win*, then NLS_LANG is in registry.
0
 
k_murli_krishnaAuthor Commented:
333 and marius nicoras, thanx: other offline suggestion was to set some name value pairs in returned html pages to browser working on it and will get back soon

--- k_murli_krishna
0
Question has a verified solution.

Are you are experiencing a similar issue? Get a personalized answer when you ask a related question.

Have a better answer? Share it in a comment.

All Courses

From novice to tech pro — start learning today.