Cassandra client
Jonas Bonér
pooling etc. Bases on Viktor's code.
But it is a pain to maintain. Now I found Hector. Seems to impl all we
have plus reconnect, loadbalancing etc.
http://prettyprint.me/2010/03/03/load-balancing-and-improved-failover-in-hector/
Should we drop our own and use Hector?
Add a Scala API around.
Anyone feel it would be fun to do?
--
Jonas Bonér
http://jayway.com
http://akkasource.org
Viktor Klang
As you know Akka has it's own Cassandra client adding scala API, pooling etc. Bases on Viktor's code.
But it is a pain to maintain.
Sorry to hear that :-( I admit to have dropped the ball on Cassidy, I simply have no time to spare for that project :(
Now I found Hector. Seems to impl all we have plus reconnect, loadbalancing etc.
http://prettyprint.me/2010/03/03/load-balancing-and-improved-failover-in-hector/
Should we drop our own and use Hector?
Add a Scala API around.
Anyone feel it would be fun to do?
--
Jonas Bonér
http://jayway.com
http://akkasource.org
--
You received this message because you are subscribed to the Google Groups "Akka User List" group.
To post to this group, send email to akka...@googlegroups.com.
To unsubscribe from this group, send email to akka-user+...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/akka-user?hl=en.
--
Viktor Klang
| "A complex system that works is invariably
| found to have evolved from a simple system
| that worked." - John Gall
Akka - the Actor Kernel: Akkasource.org
Twttr: twitter.com/viktorklang
Jonas Bonér
Viktor Klang
It is not painful due to your code but that Cassandra is changing API all the time and I don't have time to do it. Better for me to outsource the thing, eg use another lib. It also has some features we currently not have bur which are nice.
+1 on all you said.
I'm totally cool with this.
Thanks for bringing it up :-)
Tupshin Harper
integrating it into Cassandra.
-Tupshin
Ike
resulting code is not very Scala-ish. I wrote a little bit about it
here: http://theikester.wordpress.com/2010/04/07/access-cassandra-from-scala-hector/.
I also considered wrapping it in Scala but, given Cassandra's and
Hector's fast dev pace, that would introduce yet another moving target
to maintain. Then there's the question about what "the right" API
should be Scala. Viktor had:
// session => session | (key, new ColumnPath("Standard1", null,
column1))
which borrows from the Thrift API. Scalandra does:
// cassandra.ColumnFamily("Standard1")("key")("column1")
which looks pretty clean, but I couldn't get it to work with the
latest version of Cassandra. And then there's the new kid on the block
Cascal which does:
// session.get("Test" \ "Standard1" \ "key" \ "column1")
Any thoughts from the more Scala savvy among us? I'm not a fan of the
back-slash character as separator but not sure that argument
parenthesis are better. Should we also consider "->" and ">"?
On Mar 29, 8:28 am, Jonas Bonér <jo...@jonasboner.com> wrote:
> As you know Akka has it's own Cassandra client adding scala API,
> pooling etc. Bases on Viktor's code.
>
> But it is a pain to maintain. Now I found Hector. Seems to impl all we
> have plus reconnect, loadbalancing etc.
>
> http://prettyprint.me/2010/03/03/load-balancing-and-improved-failover...
Jonas Bonér
> Getting Hector to work in Scala is very straightforward but the
> resulting code is not very Scala-ish. I wrote a little bit about it
> here: http://theikester.wordpress.com/2010/04/07/access-cassandra-from-scala-hector/.
>
> I also considered wrapping it in Scala but, given Cassandra's and
> Hector's fast dev pace, that would introduce yet another moving target
> to maintain. Then there's the question about what "the right" API
Then it is probably better to wait. What we have is working ok.
> should be Scala. Viktor had:
>
> // session => session | (key, new ColumnPath("Standard1", null,
> column1))
>
> which borrows from the Thrift API. Scalandra does:
>
> // cassandra.ColumnFamily("Standard1")("key")("column1")
>
> which looks pretty clean, but I couldn't get it to work with the
> latest version of Cassandra. And then there's the new kid on the block
> Cascal which does:
>
> // session.get("Test" \ "Standard1" \ "key" \ "column1")
>
> Any thoughts from the more Scala savvy among us? I'm not a fan of the
> back-slash character as separator but not sure that argument
I actually like the backslash, make it very easy to understand.
Navigate like a filesystem.
> parenthesis are better. Should we also consider "->" and ">"?
>
> On Mar 29, 8:28 am, Jonas Bonér <jo...@jonasboner.com> wrote:
>> As you know Akka has it's own Cassandra client adding scala API,
>> pooling etc. Bases on Viktor's code.
>>
>> But it is a pain to maintain. Now I found Hector. Seems to impl all we
>> have plus reconnect, loadbalancing etc.
>>
>> http://prettyprint.me/2010/03/03/load-balancing-and-improved-failover...
>>
>> Should we drop our own and use Hector?
>> Add a Scala API around.
>> Anyone feel it would be fun to do?
>>
>> --
>> Jonas Bonérhttp://jayway.comhttp://akkasource.org
>
> --
> You received this message because you are subscribed to the Google Groups "Akka User List" group.
> To post to this group, send email to akka...@googlegroups.com.
> To unsubscribe from this group, send email to akka-user+...@googlegroups.com.
> For more options, visit this group at http://groups.google.com/group/akka-user?hl=en.
>
>
--
Jonas Bonér
work: http://jayway.com
code: http://akkasource.org
blog: http://jonasboner.com
twitter: @jboner
Ike
On Apr 9, 3:15 am, Jonas Bonér <jbo...@gmail.com> wrote:
> On 7 April 2010 17:33, Ike <ikes...@gmail.com> wrote:
>
> > Getting Hector to work in Scala is very straightforward but the
> > resulting code is not very Scala-ish. I wrote a little bit about it
> > here:http://theikester.wordpress.com/2010/04/07/access-cassandra-from-scal....
>
> > I also considered wrapping it in Scala but, given Cassandra's and
> > Hector's fast dev pace, that would introduce yet another moving target
> > to maintain. Then there's the question about what "the right" API
>
> Then it is probably better to wait. What we have is working ok.
>
>
>
>
>
> > should be Scala. Viktor had:
>
> > // session => session | (key, new ColumnPath("Standard1", null,
> > column1))
>
> > which borrows from the Thrift API. Scalandra does:
>
> > // cassandra.ColumnFamily("Standard1")("key")("column1")
>
> > which looks pretty clean, but I couldn't get it to work with the
> > latest version of Cassandra. And then there's the new kid on the block
> > Cascal which does:
>
> > // session.get("Test" \ "Standard1" \ "key" \ "column1")
>
> > Any thoughts from the more Scala savvy among us? I'm not a fan of the
> > back-slash character as separator but not sure that argument
>
> I actually like the backslash, make it very easy to understand.
> Navigate like a filesystem.
>
Yes, like an MS-DOS filesystem :) But kidding aside I agree. I like
the hierarchy implication, even if it's only 2 or 3 levels deep. I
think it is a good metaphor for the dimensions in Cassandra.
>
>
>
> > parenthesis are better. Should we also consider "->" and ">"?
>
> > On Mar 29, 8:28 am, Jonas Bonér <jo...@jonasboner.com> wrote:
> >> As you know Akka has it's own Cassandra client adding scala API,
> >> pooling etc. Bases on Viktor's code.
>
> >> But it is a pain to maintain. Now I found Hector. Seems to impl all we
> >> have plus reconnect, loadbalancing etc.
>
> >>http://prettyprint.me/2010/03/03/load-balancing-and-improved-failover...
>
> >> Should we drop our own and use Hector?
> >> Add a Scala API around.
> >> Anyone feel it would be fun to do?
>
> >> --
> >> Jonas Bonérhttp://jayway.comhttp://akkasource.org
>
> > --
> > You received this message because you are subscribed to the Google Groups "Akka User List" group.
> > To post to this group, send email to akka...@googlegroups.com.
> > To unsubscribe from this group, send email to akka-user+...@googlegroups.com.
> > For more options, visit this group athttp://groups.google.com/group/akka-user?hl=en.
Jonas Bonér
>
>
> On Apr 9, 3:15 am, Jonas Bonér <jbo...@gmail.com> wrote:
>> On 7 April 2010 17:33, Ike <ikes...@gmail.com> wrote:
>>
>> > Getting Hector to work in Scala is very straightforward but the
>> > resulting code is not very Scala-ish. I wrote a little bit about it
>> > here:http://theikester.wordpress.com/2010/04/07/access-cassandra-from-scal....
>>
>> > I also considered wrapping it in Scala but, given Cassandra's and
>> > Hector's fast dev pace, that would introduce yet another moving target
>> > to maintain. Then there's the question about what "the right" API
>>
>> Then it is probably better to wait. What we have is working ok.
>>
>>
>>
>>
>>
>> > should be Scala. Viktor had:
>>
>> > // session => session | (key, new ColumnPath("Standard1", null,
>> > column1))
>>
>> > which borrows from the Thrift API. Scalandra does:
>>
>> > // cassandra.ColumnFamily("Standard1")("key")("column1")
>>
>> > which looks pretty clean, but I couldn't get it to work with the
>> > latest version of Cassandra. And then there's the new kid on the block
>> > Cascal which does:
>>
>> > // session.get("Test" \ "Standard1" \ "key" \ "column1")
>>
>> > Any thoughts from the more Scala savvy among us? I'm not a fan of the
>> > back-slash character as separator but not sure that argument
>>
>> I actually like the backslash, make it very easy to understand.
>> Navigate like a filesystem.
>>
>
> Yes, like an MS-DOS filesystem :) But kidding aside I agree. I like
/ is equally good to me
> For more options, visit this group at http://groups.google.com/group/akka-user?hl=en.
Chris Shorrock
Ike
> Hey guys - I'm the guy who wrote cascal. Just a few notes on the syntax - I
> really struggled with what to use when defining a path. A few considerations
> with what lead me to using "\":
>
> * Originally I wanted to use "->", but I thought this may run into issue
> with already existing using of "->" to form a tuple, so I didn't want to
> overload that operator as it appeared to already be spoken for.
> * I then thought about using "/". My only issue with this was within cascal
> when you do "Keyspace" \ "ColumnFamily", you also do "Keyspace" \\ "SuperCF"
> - using "//" wouldn't of worked as it would of commented out everything :) -
> I could of used something else to create a super column family, however I
> like the consistancy of "\"
> * scala.xml.NodeSeq had already defined the use of "\" for traversal of
> nodes so I thought the analogy was somewhat similar - so it seemed like a
> logical choice.
>
> Anyhow. My 2 cents on the issue when I was thinking on this.
Thanks for your insights Chris. I haven't looked at your code yet but
I'm just curious: do you absolutely need a different delimiter for
supercolumns? Isn't there any other way to determine the kind of
Column Family we're dealing with? Like number of arguments, context,
etc. (i.e. if 1 token after the Key then StandardColumn , if 2 tokens
then SuperColumn)? Couldn't that be achieved with a simple overloaded
method signature or am I missing something?
For example:
session.get("Test" \ "Standard" \ "Key" \ "ColumnName") // standard
column family
session.get("Test" \ "Super" \ "Key" \ "SuperColumn" \
"ColumnName") // super column family
By the way, I think some of the stuff you're doing with the object
mapping is really practical. Perhaps we should continue this
discussion on the cascal list?
>
>
>
> On Fri, Apr 9, 2010 at 6:00 AM, Jonas Bonér <jo...@jonasboner.com> wrote:
> > akka-user+...@googlegroups.com<akka-user%2Bunsubscribe@googlegroups .com>
> > .
> > >> > For more options, visit this group athttp://
> > groups.google.com/group/akka-user?hl=en.
>
> > >> --
> > >> Jonas Bonér
>
> > >> work: http://jayway.com
> > >> code: http://akkasource.org
> > >> blog: http://jonasboner.com
> > >> twitter: @jboner
>
> > > --
> > > You received this message because you are subscribed to the Google Groups
> > "Akka User List" group.
> > > To post to this group, send email to akka...@googlegroups.com.
> > > To unsubscribe from this group, send email to
> > akka-user+...@googlegroups.com<akka-user%2Bunsubscribe@googlegroups .com>
> > .
> > > For more options, visit this group at
> >http://groups.google.com/group/akka-user?hl=en.
>
> > --
> > Jonas Bonér
>
> > work: http://jayway.com
> > code: http://akkasource.org
> > blog: http://jonasboner.com
> > twitter: @jboner
>
> > --
> > You received this message because you are subscribed to the Google Groups
> > "Akka User List" group.
> > To post to this group, send email to akka...@googlegroups.com.
> > To unsubscribe from this group, send email to
> > akka-user+...@googlegroups.com<akka-user%2Bunsubscribe@googlegroups .com>
Chris Shorrock
To unsubscribe from this group, send email to akka-user+...@googlegroups.com.
Viktor Klang
sorry for necroposting, but I found this and think it might be viable: http://github.com/cliffmoon/scromium
WDYT?
Viktor Klang
| "A complex system that works is invariably
| found to have evolved from a simple system
| that worked." - John Gall
Akka - the Actor Kernel: Akkasource.org
Twttr: twitter.com/viktorklang
Jonas Bonér
I like it. Do you like it better than yours ?
If someone feel urged to help out integrating it I would be most grateful.
On 22 Apr 2010 17:35, "Viktor Klang" <viktor...@gmail.com> wrote:
Hey guys,
sorry for necroposting, but I found this and think it might be viable: http://github.com/cliffmoon/scromium
WDYT?
On Sat, Apr 10, 2010 at 11:09 PM, Chris Shorrock <chris.s...@gmail.com> wrote:
>
> Thanks for ...
--
Viktor Klang
| "A complex system that works is invariably
| found to have evolved from a simple sys...
You received this message because you are subscribed to the Google Groups "Akka User List" group.
To...
Viktor Klang
I like it. Do you like it better than yours ?
Looks good. And also we don't need to maintain the code ourselves, which to me is a big plus.
I'll add it as a ticket in assembla.
| that worked." - John Gall
Akka - the Actor Kernel: Akkasource.org
Twttr: twitter.com/viktorklang