Fluentcassandra Filter Problem

Question

i'm new in FluentCassandra and Cassandra.

I have a Problem to filter data from a range of value. In my opinion i can use a Generic List to filter the result, but i don't no how?!

In MSSQL i use this (SELECT * FROM TABLE WHERE Row1 like '%search%').

CassandraSuperColumnFamily<UTF8Type, UTF8Type> familyname= db.GetColumnFamily<UTF8Type, UTF8Type>("Messages");
var results= familyname.Get("key")
                .Take(5)
                .FirstOrDefault()
                .AsDynamic();

Maybe some one can help me?!

Thanks calimero

score 2 · Answer 1 · answered Apr 13 '11 at 20:56

Searching for a substring requires an inefficient sequential scan. Cassandra doesn't make that easy, because it's usually the wrong thing to do:

if you want full-text search, you should use Solandra
if you want to do analytical queries, you should use Pig or Hive on top of Hadoop, which will parallelize the work across the cluster

If you absolutely must do a non-parallel seq scan, you'll have to page through the rows manually and check for your substring in C# code.

Thank you jbellis. I will test the Solandra API. – Calimero Apr 14 '11 at 06:38 — Calimero, Apr 14 '11 at 06:38

score 1 · Answer 2 · answered Jun 22 '11 at 01:42

With what jbellis said, column name scanning is now supported in Cassandra 0.7 and greater. To scan the column names for a specific value you can do:

var results = familyname.Get(startKey: "key", keyCount: 30, family => family["last_name"] == "Smith")
    .Take(5)
    .FirstOrDefault()
    .AsDynamic()

The above code will start at the specified key and scan the next 30 keys for a column with the name "last_name" and the value of "Smith".

Fluentcassandra Filter Problem

2 Answers2