Hi.
I tried:
select ( select expand( select date, COUNT(ip) as IpNo from Downloads
group by date order by date asc) ) where IpNo = 5
It still returns nothing.
On Tuesday, December 23, 2014 12:06:47 AM UTC+2, Lvc@ wrote:
Mihai,
You should use expand() in the inner query.
Lvc@
On 22
Is auto sharding working in 2.0 RC1?
I have 2 servers nodes with orientdb distributed mode running. Both
database are synced. I have tried to enable autosharding following the
manual
at
http://www.orientechnologies.com/docs/last/orientdb.wiki/Distributed-Sharding.html
and create my claases.
Thanks for the response.
So then, the only tcp-ip calls would be for the replication.
And if I want to write to more masters, I have to start the same java app
on each of the different master machines I want to be writing to,
with the same configuration file (hazelcast.xml) ?
On Monday,
Using PHPOrient to connect to OrientDB, I'm trying to create a Profile
record and then a User record that refers to it in the same transaction.
The problem is that the temporary record ID (RID) doesn't get updated for
the link. Code snippet:
$client = new PhpOrient('localhost', 2424);
Andrey, thank you!
I've seen recommendation to use not much memory (512MB) for heap before and
I know, that this was made to reduce GC pauses.
I just misunderstood some of recent discussions (can't find it), were I've
seen kind of calculations of heap size corresponding to cache.
Sorry and
Hi Aleksandar,
Luigi is working on it (using JavaCC). This is the branch:
https://github.com/orientechnologies/orientdb/tree/javacc_parser
Our plan are to provide new parser syntax checking and optimizer in 2.1
(we're at good point), then provide new engine for 3.0.
Lvc@
On 6 January 2015 at
Brilliant, thanks for the update Luca... I'll follow then Luigi's work.
Cheers,
Aleks
On Tuesday, January 6, 2015 1:01:57 PM UTC, Lvc@ wrote:
Hi Aleksandar,
Luigi is working on it (using JavaCC). This is the branch:
https://github.com/orientechnologies/orientdb/tree/javacc_parser
Our
Hey
I just did a little test with my own data, and it seems if I create a
string field on a vertex e.g. foo, and enter a value that looks like an
@RID (and lets say for arguments sake it's valid), then using a fetchplan
select against this vertex, then foo is populate with that record it was
Hi, I have a strange situation. I have two preexisting objects for which I
wish to associate one object to another i.e. via a Link relationship which
has been predefined in the schema.
1. find both pre-existing objects specifically find them by RID
2. set the second object at a specific
Hi guys.
I come from DDD and RDBMS background.
Usually when I create my domain model I use UUID as business key to
uniquely distinguish my domain entities. In RDBMS I create surrogate auto
incremented primary key and indexed business key with unique constraint. I
do it to decouple my domain
Thanks Luca ...
must say orientdb is just awesome !!!
the example is the People / Address where People can work/live at the same
Address; so the Address Id is unique for each address, while multiple
People hold the same Address Id.
What I want is for People to link to Address @rid (via Address
Thanks Keith, yes that approach works and I guess is a good work around;
I am migrating data from mysql so it's somewhat slower perhaps as it
requires more queries/loops but is doable :) ...
On Wednesday, January 7, 2015 4:38:18 AM UTC+11, Keith Freeman wrote:
This works for me:
import
I have a simple class geo with location property with string items like
united states
houston, texas, united states
...
select geo where location containstext united
Returns records as expected
However...
select geo where location containstext united
select geo where location containstext
Correction. Works without fulltext index. Fails with fulltext index.
On Tuesday, January 6, 2015 8:54:52 PM UTC-7, Erik Peterson wrote:
I have a simple class geo with location property with string items like
united states
houston, texas, united states
...
select geo where location
I think I may have figured it out. Seems like the best way is to create a
function that takes the list of 'names' and concatenates them with slashes.
On Monday, January 5, 2015 5:38:16 PM UTC-5, Kyle Hendricks wrote:
If I traverse a graph with vertices with 'name' properties, is there any
Andrey- Does this mean that in the new automatic mode it is never
necessary to call detach() or attach()?
On Thursday, November 20, 2014 6:13:22 AM UTC-7, Andrey Lomakin wrote:
Hi all.
Small announcement of changes in attachment/detachment policy.
Now we support 2 modes manual and
Hi,
Is it possible to index a whole EMBEDDEDMAP with Lucene and, if so, how to
query it?
Regards,
-Stefan
--
---
You received this message because you are subscribed to the Google Groups
OrientDB group.
To unsubscribe from this group and stop receiving emails from it, send an email
to
Hi Stefan
not yet see here
https://github.com/orientechnologies/orientdb-lucene/issues/25
Enrico
2015-01-06 16:53 GMT+01:00 ste...@activitystream.com:
Hi,
Is it possible to index a whole EMBEDDEDMAP with Lucene and, if so, how to
query it?
Regards,
-Stefan
--
---
You received this
Hi George,
link traversal is O(1) operation because the RID is like a physical pointer
to data. Using UUID would mean relying on an index to find the RID
corresponding to the UUID.
SBTree index you will give you O(logN) performance.
HashIndex will give you a near O(1) performance in the average
for example, there are two classes
Price with *amount* and *currency *properties (this class is abstract);
Price.amount is indexed NONUNIQUE
SaleItem with notImportant, price1 (embedded Price) price2 (embedded
Price), etc
querying SaleItem by embedded document price.amount is very slow
When
Hi Enrico,
I have sent you some memory profile data by email. Please check your gmail.
Thanks,
Jing
--
---
You received this message because you are subscribed to the Google Groups
OrientDB group.
To unsubscribe from this group and stop receiving emails from it, send an email
to
This works for me:
import com.orientechnologies.orient.core.db.document.ODatabaseDocumentTx;
import com.orientechnologies.orient.core.record.impl.ODocument;
public class O2DocumentHW
{
public static void main(String[] args)
{
long start = System.currentTimeMillis();
I've tried as many different permutations of the above as I can think of.
- I trued to change the order of the 'attach' and and command creation;
shuffling seems to have no effect (as expected).
- I also tried nested transactions, which do seem like they're supported in
some cases, but don't
Exactly, it's automatic.
Lvc@
On 6 January 2015 at 16:51, Keith Freeman 8fo...@gmail.com wrote:
Andrey- Does this mean that in the new automatic mode it is never
necessary to call detach() or attach()?
On Thursday, November 20, 2014 6:13:22 AM UTC-7, Andrey Lomakin wrote:
Hi all.
The CREATE LINK command want unique key, but the value
2831b7f5-28ba-457d-b871-21be494f509c is in more than 1 record in database,
so it can't create the LINK.
Lvc@
On 6 January 2015 at 18:38, Keith Freeman 8fo...@gmail.com wrote:
This works for me:
import
25 matches
Mail list logo
