I have a database say 'X' whose charset is - WE8ISO8859P1
After SQL*Loader inserts data into X from a windoze box, when queried from X some characters are displayed as reverse question mark.
0x27 - standard ASCII apostrophe (')
0x91 - left single quotation mark (‘)
0x92 - right single quotation mark (’)
0xb4 - acute accent (´)
Windoze box has charset - WE8MSWIN1252
1. What is the possible issue? Is it the incompatibity of character sets between them?
2. While fetching can I use CONVERT function to fix and display this properly?
3. Just trying to understand - is the problem while inserting the data or just while displaying?
Please give some pointers in this regard.
( on Oracle 220.127.116.11.0 )
What is the encoding or the characters that you have inserted? Are they characters which belong to WE8ISO8859P1 encoding?
What is the NLS_LANG defined in your windows environment?
What is the characterset set in the controlfile of the SQL*Loader if you have defined one. If you haven't defined one in the controlfile then the NLS_LANG then sqlldr uses the characterset defined by the NLS_LANG.
++ Note that the CHARACTERSET (or, if not defined, for sqlldr the NLS_LANG) needs to reflect the characterset of the FILE you try to load, not the database NLS_CHARACTERSET.
Can you have SQL Developer (a unicode tool) installed on your DB server and check if you still observe reverse question mark?