Teradata 13.10 is rejecting my insert with the error:
The string contains an untranslatable character. (6706)
Here is my INSERT statement:
Insert Into CCDW_T.UTest (ucol, i) SELECT 'ɂ',578;
This is my CREATE Table:
CREATE TABLE CCDW_T.utest
ucol CHAR(1) CHARACTER SET UNICODE
PRIMARY index PI_utest(ucol)
In the INSERT statement that *thing* between the quotes is Unicode code point 0242 (hex) = decimal 578
My client setting is UTF-8 (tried with both SQL assistant and BTEQ). My insert statement is saved as a file in UTF8 encoding.
I save the file to disk and verified with a hex editor that character is correctly encoded in the file as two bytes UTF 8: C9 82 - That is correct for the Unicode code point 0242
We've added support for U+0242 (LATIN SMALL LETTER GLOTTAL STOP) in Teradata 14.0. That release includes support for all Unicode Standard 6.0 BMP characters in data. Note that U+0242 was added in Unicode version 5.0.
Here is an example from 15.10:
sel U&'#0242' uescape '#';
*** Query completed. One row found. One column returned.
*** Total elapsed time was 1 second.
It seems I have a similar problem with a French character
LATIN SMALL LETTER O WITH CIRCUMFLEX
FILE_READER: TPT19003 Delimited Data Parsing error: Invalid multi-byte character in row 571, col 0
or with fastload **** 12:35:31 Error on piom GET ROW: 39, Text: Invalid UTF8 character
encountered !ERROR! Delimited Data Parsing error: Invalid
multi-byte character in row 571, col 1
**** 12:35:31 Error at record number 571
Could you confirm that it might be unsupported?
I am using this SET SESSION CHARSET 'UTF8'; with the fastload from the Windows 7 file saved in UTF8.
U+00F4 has been supported for a very long time.
Double-check your input file. In UTF-8, this character should be represented as the two-byte sequence C3 B4.
Joe & Dave,
I had the same problem with the character at U+0245 (inverted "V"), on Teradata 13.0. Is there a list (or range) of characters that were unsupported in version 13 and/or added with 14?
U+0245 is not supported in TD13.0, it was added in TD14.0.
There is a UNCDUNCD.txt file that documents the supported Unicode characters. It's on the documentation CD (index entry is "Unicode Server Character Set") or you can get it from www.info.teradata.com. Search for Product ID B035-1056 and you will find one for TD12/13 and one for TD14/15.
Thanks Fred. Exactly what I was looking for.
I considered posting a diff of those but there's nearly 3000 new characters(!).