Charles,
I'm not a big Thrift user but by reading the code I think you are right:
byte [][] columnsText = new byte[columns.size()][];
for (int i = 0; i < columns.size(); ++i) {
columnsText[i] = getText(columns.get(i));
}
Scanner scanner = table.getScanner(columnsText,
getText(startRow));
Current IDL is not as complete as it should be, feel free to open a new
issue in Jira <https://issues.apache.org/jira/browse/HBASE> so that someone
can post a patch.
Thx,
J-D
On Thu, Sep 18, 2008 at 12:31 PM, Charles Mason <[EMAIL PROTECTED]>wrote:
> Hi All,
>
> I have been working on interfacing HBase with C# using thrift. I am
> currently writing a lightweight ORM style wrapper to make interfacing
> with data in HBase easier. I am currently trying to implement a
> scanner and I am running in to a problem.
>
> The scannerOpen function seems to always require a column list, wear
> as with the REST interface it can be left empty in which case all a
> rows columns are returned. If I provide an empty column list via
> Thrift no columns are returned.
>
> Is there something I am missing or this just a missing future with the
> thrift interface? I am running the 0.2 release.
>
> Charlie M
>
