It should not be called directly: externall callers should be calling flush()
instead.
To be sure it doesn't happen again, make seal_active_memtable private.
Signed-off-by: Glauber Costa <glommer@cloudius-systems.com>
When we create a column family, we can pass as an extra parameter, the
commitlog - or lack thereof. Because the commitlog is optional to begin with -
it won't exist if we don't call init_commitlog, we can have this to be empty
meaning no commit log.
The creation of a column family should be always done through
add_column_family. And if that is the case, we have the database's commitlog
right there and can get the pointer through the db. Only tests are not creating
the column family this way, and for them, it is fine.
We want to do that, because some column family operations will use the commit log.
Right now, they are forcing us to add parameters to APIs that would be much cleaner
without it. So while separation is good, this level of coupling is a net win as it
allows us to clean up some visible APIs.
Signed-off-by: Glauber Costa <glommer@cloudius-systems.com>
"Fixes for ORDER BY clauses" from Pawel.
"The patches fix several issues in CQL3 frontend related to ORDER BY
clauses. Also, column component indexes are now handled properly and it
is possible to create tables with more than one column in clustering key."
The logger class constructor registers itself with the logger registry,
in order to enable dynamically setting log levels. However, since
thread_local variables may be (and are) initialized at the time of first
use, when the program starts up no loggers are registered.
Fix by making loggers global, not thread_local. This requires that the
registry use locking to prevent registration happening on different threads
from corrupting the registry.
Note that technically global variables can also be initialized at the
point of first use, and there is no portable way for classes to self-register.
However this is the best we can do.
We don't really handle expiration of entrie rows yet, due to issue #17.
Fixing a bug in 'insert' made this test start to fail because the
expired row started to appear in the results. Until #17 is fixed,
let's use 'update' to create the row, so that the expectations are
met.
"(Partial?) implementation of BatchLogManager.
Requires the token function/restriction series.
Functional as in that it can create batchlog mutations, and do replay
of data in this system table.
Since range queries does not yet work, it only handles a very small
table contents.
It is not used yet either, but will eventually be needed for batch statements
etc."
The messaging service is initialized very early, before we have the
proxy or query processor initialized. It is mostly fine, except for
the fact that the messaging service also finishes the initialization
of the storage service. That part will issue queries agains the system
tables, (as soon as we support them), and need to happen later.
Signed-off-by: Glauber Costa <glommer@cloudius-systems.com>
What "internal" really means here, is queries that operate on the system tables.
Up until now, there wasn't really an issue because we weren't really checking
on this flag, and the internal calls were all unimplemented.
Now that we are about to hook up the internal calls, leaving this state marked
as internal will lead us to fail when we try to issue statements - like create
table - that are not, and will not be implemented in their internal
counterparts.
Signed-off-by: Glauber Costa <glommer@cloudius-systems.com>
"We don't want multiple shards to respond with the same data. Higher level code
assumes that shard data is non-overlapping. It's cheaper to drop duplicates as
soon as possible. Memtable reader for example will never have overlapping
data, so cache hitting queries will never need to pay for this. Compaction
process may also rely on this."
For simplicity in expiration time computations 0 is used as current
timestamp, make sure that sstable write code is aware of that and
doesn't consider cells as already expired.
Signed-off-by: Paweł Dziepak <pdziepak@cloudius-systems.com>
Currently if rpc handler returns smart pointer rpc will try to serialize
the pointer object as opposite to an object the ptr is pointing to.
This patch fixes it by serializing real object instead of a pointer.
