Strange issue creating a BCP xml format file with double dagger '‡' (alt + 0135) as column terminator with bcp.exe

Question

Hi,
I'm having issues generating a BCP XML format file using a fairly unusual column terminator, a double dagger symbol
‡ (alt + 0135) which I need to support.
I'm experiencing this problem with bcp.exe for SQL2008 R2 and SQL2012.
If I run the following command line:
bcp MyDB.TMP.Test_Extract format nul -c -x -f "C:\BCP\format_file_dagger_test.xml" -T -S localhost\SQL2012 -t‡
I end up with a XML format file like so:
<?xml version="1.0"?>
<BCPFORMAT xmlns="http://schemas.microsoft.com/sqlserver/2004/bulkload/format" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance">
 <RECORD>
  <FIELD ID="1" xsi:type="CharTerm" TERMINATOR="ç" MAX_LENGTH="255" COLLATION="SQL_Latin1_General_CP1_CI_AS"/>
  <FIELD ID="2" xsi:type="CharTerm" TERMINATOR="ç" MAX_LENGTH="50" 
.. and so on.
You will notice that the TERMINATOR="ç" (Minuscule c-cedilla) is output instead of TERMINATOR="‡". The
ç character, is strangely enough is alt + 135 (and not alt + 0135) so this might more than a coincidence! I know you can specify the codepage but this switch applies the data being imported or extracted and not for the format file
itself (I tried it anyway). 
In order to use the XML file to bulk import I manually did a text substitution of 'ç' character for '‡' and then BCP imports '‡' data fine. 
This character swap doesn't occur if I generate a non XML format file (the '‡' character is output in the format file correctly) however, this file produces other import errors, which I don't encounter if I use a standard delimiter like a comma. So I have stuck
with the working XML format file which I prefer.
Does anyone know why this is happening? I'm planning to automate the generation the of the XML format file and would like to avoid the additional step of text substitution if possible.
Thank you.

Answer 1

Hi Ham09,
According to your description , we do a test and find that the character of the terminator is changed due to the code page of Operation System. When you choose the different time zone in Data and Time bar, and do the same bcp test, you will find it will
export the different TERMINATOR in your XML format file. For example, you can import the character "ç" (alt + 135) in (UTC-12:00)International Date Line West time zone and (UTC+09:00)Osaka, Sapporo, Tokyo time zone, and check if the terminators are different.
By default, the field terminator is the tab character (represented as \t). To represent a paragraph mark, use \r\n.
For more information, there is detail about code page(Windows), you can review the following article.
http://msdn.microsoft.com/en-us/library/windows/desktop/dd317752(v=vs.85).aspx
Regards,
Sofiya Li
Sofiya Li
TechNet Community Support