Are you using thrift 0.5 as downloaded from there: http://incubator.apache.org/thrift/download/ ?
On Thu, Mar 17, 2011 at 10:20 PM, Michael Fortin <mi...@m410.us> wrote: > Thanks for the response, sorry if my initial question wasn't clear. > > When using thrift, I call > client.get_slice(keyBytes, columnParent, range, level) > > i get a list of ColumnOrSuperColumns back. When I iterate over them and and > call: > byte[] nameBytes = columnOrSuperColumn.getSuper_column().getName() > > I seem to be getting a byte array that contains, not only the name of the > super column, but all of the child columns as well. The output that start '? > get_slice' is the byte array converted to a string. Shouldn't nameBytes only > return the name of the superColumn? In my case 'super-col-0' in byte form? > > This is using cassandra 0.7.0 & 0.7.4. > > cheers, > M!ke > > On Mar 17, 2011, at 9:43 AM, Sylvain Lebresne wrote: > >> Are you sure you don't have a problem with handling ByteBuffers ? >> What do you mean by 'deserialized string' ? >> >> -- >> Sylvain >> >> On Thu, Mar 17, 2011 at 4:20 AM, Michael Fortin <mi...@m410.us> wrote: >>> Hi, >>> >>> I've been working on a scala based api for cassandra. I've built it >>> directly on top of thrift. I'm having a problem getting a slice of a >>> superColumn. When I get a columnOrSuperColumn back, and call >>> 'cos.super_column.name' and deserialize the bytes I'm not getting the >>> expected output. >>> >>> Here's whats in cassandra >>> ------------------- >>> RowKey: key >>> => (super_column=super-col-0, >>> (column=column, value=76616c756530, timestamp=1300330948240) >>> (column=column1, value=76616c756530, timestamp=1300330948244)) >>> …. >>> >>> and this is the deserialized string >>> >>> ? get_slice super-col-0 column >>> value0 >>> .?æ? column1 value0 >>> .?æ? super-col-1 column value1 >>> .?æ? column1 value1 >>> .?æ? super-col-2 column value2 >>> .?æ? column1 value2 >>> .?æ? super-col-3 column value3 >>> .?æ? column1 value3 >>> .?æ? >>> >>> I would expect >>> super-col-0 >>> >>> Any ideas on what I'm doing wrong? >>> >>> Thanks, >>> Mike > >
