Return to Question

added 262 characters in body

Source Link

edited Feb 10, 2010 at 15:45

2.8k
4
25
28

I want to write a large dataset from a SQL Server 2005 database using BCP (bulk copy program).

Ideally I'd like to do the following:

bcp MyDatabase..MyTable OUT STDOUT -c -t, |gzip -c c:\temp\dataset.csv.gz

However BCP just writes this to a literal file named STDOUT.

I also tried:

bcp MyDatabase..MyTable OUT CON: -c -t, |gzip -c c:\temp\dataset.csv.gz

but this returns and error of Error = [Microsoft][SQL Native Client]Unable to open BCP host data-file.

So, can I create a named pipe anywhere? I've seen hints on the web of somehow starting gzip with one end tied to a filename like \\.\named_pipe - but how is this done?

Update: Note that Microsoft themselves acknowedge that they really don't care about efficiency with SQL Server: http://connect.microsoft.com/SQLServer/feedback/details/337702/support-bcp-out-in-and-bulk-insert-compressed-or-named-pipe-support-at-least

I want to write a large dataset from a SQL Server 2005 database using BCP (bulk copy program).

Ideally I'd like to do the following:

bcp MyDatabase..MyTable OUT STDOUT -c -t, |gzip -c c:\temp\dataset.csv.gz

However BCP just writes this to a literal file named STDOUT.

I also tried:

bcp MyDatabase..MyTable OUT CON: -c -t, |gzip -c c:\temp\dataset.csv.gz

but this returns and error of Error = [Microsoft][SQL Native Client]Unable to open BCP host data-file.

So, can I create a named pipe anywhere? I've seen hints on the web of somehow starting gzip with one end tied to a filename like \\.\named_pipe - but how is this done?

I want to write a large dataset from a SQL Server 2005 database using BCP (bulk copy program).

Ideally I'd like to do the following:

bcp MyDatabase..MyTable OUT STDOUT -c -t, |gzip -c c:\temp\dataset.csv.gz

However BCP just writes this to a literal file named STDOUT.

I also tried:

bcp MyDatabase..MyTable OUT CON: -c -t, |gzip -c c:\temp\dataset.csv.gz

but this returns and error of Error = [Microsoft][SQL Native Client]Unable to open BCP host data-file.

So, can I create a named pipe anywhere? I've seen hints on the web of somehow starting gzip with one end tied to a filename like \\.\named_pipe - but how is this done?

Source Link

asked Feb 3, 2010 at 10:14

PP.

2.8k
4
25
28

Creating a named pipe on Windows

I want to write a large dataset from a SQL Server 2005 database using BCP (bulk copy program).

Ideally I'd like to do the following:

bcp MyDatabase..MyTable OUT STDOUT -c -t, |gzip -c c:\temp\dataset.csv.gz

However BCP just writes this to a literal file named STDOUT.

I also tried:

bcp MyDatabase..MyTable OUT CON: -c -t, |gzip -c c:\temp\dataset.csv.gz

but this returns and error of Error = [Microsoft][SQL Native Client]Unable to open BCP host data-file.

So, can I create a named pipe anywhere? I've seen hints on the web of somehow starting gzip with one end tied to a filename like \\.\named_pipe - but how is this done?