Home | About | Sematext search-lucene.com search-hadoop.com
 Search Hadoop and all its subprojects:

Switch to Threaded View
Zookeeper >> mail # user >> ZooKeeper C++ Client

Copy link to this message
Re: ZooKeeper C++ Client
More wondering if you had seen it. Whichever way you go I don't think
we'd want both, so longer term it would be good to consolidate down to
a single implementation.


On Mon, Jun 4, 2012 at 4:11 PM, Michi Mutsuzaki <[EMAIL PROTECTED]> wrote:
> Hi Pat,
> Yes, but since my goal is to port the c client code to c++ with boost
> so that it's more portable, I felt it's easier to just start off with
> the c client. I borrowed some ideas (like logging) from zkfuse, tough.
> Do you think it's better if I somehow consolidate my work with zkfuze client?
> Thanks!
> --Michi
> On Mon, Jun 4, 2012 at 3:20 PM, Patrick Hunt <[EMAIL PROTECTED]> wrote:
>> Michi you looked at the C++ client wrapper that's included in zkfuse
>> under contrib?
>> Patrick
>> On Sun, Jun 3, 2012 at 1:31 AM, Michi Mutsuzaki <[EMAIL PROTECTED]> wrote:
>>> Hi,
>>> I've been working on a C++ binding for ZooKeeper. Right now it's a
>>> wrapper around the
>>> C client, but the intention is to port the code to C++ over time. More
>>> specifically:
>>> * Make it more portable. Replace pthread with boost::thread and socket
>>> with boost::ip::tcp.
>>> * Use a better logging system (either google-glog or log4cxx).
>>> * Use C++ version of jute generated code.
>>> * Use unordered_map instead of hashtable.
>>> It's still work in progress, but I'd like to get feedback from the
>>> community, especially regarding
>>> the API. Please let me know if you have any suggestions.
>>> github page: https://github.com/m1ch1/zookeeper
>>> doxygen: http://m1ch1.github.com/zkcpp/3.5.0/namespaceorg_1_1apache_1_1zookeeper.html
>>> Thanks!
>>> --Michi