Question marks instead of unicode characters

[GoogleCodeExporter]

What steps will reproduce the problem?

    conn = pyodbc.connect('Driver=FreeTDS;TDS_Version=8.0;Server={};Port={};UID={};PWD={}')
    cur = conn.cursor()
    rows = cur.execute("select * from data order by row_id").fetchall()

What is the expected output? What do you see instead?

Column data contains unicode characters (cyrillic and turkish). System default 
encoding and locale set to "utf-8". Server (MSSQL2008) shows collation 
«cyrillic_general_ci_as» (if that should actually do something with it).

What version of the product are you using? On what operating system?

Server: Win2k8 server + MSSQL 2k8

Client:
FreeBSD 8.2
freetds-0.64_6,1
libiodbc-3.52.7
pyodbc==2.1.8

Please provide any additional information below.

tsql console session shows text correctly.

Original issue reported on code.google.com by aenor.realm on 25 Apr 2011 at 7:06

[GoogleCodeExporter]

Perhaps I should write a replacement for FreeTDS...  There are really only 2 
encoding supported in the ODBC specification: ANSI and UCS2, which is the 
2-byte Unicode used by Windows NT, 2-byte Python builds, etc.

Can you test setting FreeTDS to use UCS-2?  If it works, I'll update the FAQ 
appropriately.  (I realize it suggests UTF-8, but apparently that only works 
for some databases.  I will reword it.)  I don't have FreeTDS setup at the 
moment.

Original comment by mkleehammer on 20 May 2011 at 7:30

• Changed state: Investigating

[GoogleCodeExporter]

I've tried to put UCS2, UCS-2 and UTF-8 into client charset settings - no 
effect.

Original comment by aenor.realm on 20 May 2011 at 8:15

[GoogleCodeExporter]

Fixed with pyodbc 3.0.6 and FreeTDS 0.91.

There were lots of Unicode fixes in pyodbc 3.0.x and FreeTDS 0.91 both.

Original comment by mkleehammer on 27 Sep 2012 at 10:22

• Changed state: Complete