I have chosen the rest.li framework for creating the API. I will also have
a look at the frameworks you all have mentioned and keep you all posted.
Thanks again
Vaibhav
On 15-Apr-2015 2:40 am, "Josh Elser" <josh.el...@gmail.com> wrote:

> yes/no, IMO.
>
> Enabling SPNEGO would be a nice direct integration point with Accumulo 1.7
> (now that we have the Kerberos client auth groundwork in place), but I
> would imagine that reusing HTTP basic authentication (in conjunction with
> HTTPS) is probably a more standard way to go about this.
>
> Your choice of Java library (resteasy/jersey/etc) probably have their own
> way to automatically do this for you with a little bit of wiring up.
>
> Sean Busbey wrote:
>
>> For the auth part, using SPNEGO will put you in line with  most other
>> components.
>>
>> --
>> Sean
>>
>> On Apr 14, 2015 9:37 AM, "Kurt Christensen" <hoo...@hoodel.com
>> <mailto:hoo...@hoodel.com>> wrote:
>>
>>
>>     Accumulo seems to invite the following structure for URIs.
>>
>>     https://host/table/row
>>     https://host/table/row/__columnFamily
>>     <https://host/table/row/columnFamily>
>>     https://host/table/row/__columnFamily/column
>>     <https://host/table/row/columnFamily/column>
>>
>>     On the other hand, because the various keys (row, columnFamily,
>>     column) can be arbitrarily long, URIs built this way can be
>>     problematic. Some or all of the key data may need to be URLencoded
>>     as well.
>>
>>     For CRUD, the following seems to be common practice:
>>          Create: Method="PUT" [or "POST"]
>>          Read: Method="GET"
>>          Update: Method="POST"
>>     Delete: Method="DELETE"
>>
>>     It seems to me that the whole user
>>     identification/authentication/__authority determination is the hard
>>     part for the general audience. You can't really count on having a
>>     PKI/CAS infrastructure, which would solve that issue. Stateful
>>     authentication could be done via log-in and made part of the
>>     client's signature (perhaps a cookie).
>>
>>     I'm just starting to go through "REST API Design Rulebook" by Mark
>>     Massé (O'Reilly 2012). You might want to check it out. Here are the
>>     first few "rules" from Chapter 2 (note the difference between the
>>     use of MUST versus SHOULD). Rules continue throughout the book.
>>     Chapter 3: "Interaction Design with HTTP" covers CRUD operations.
>>
>>     Rule: Forward slash separator (/) must be used to indicate
>>     hierarchical relationship
>>
>>     Rule: A trailing forward slash (/) should not be included in URIs
>>
>>     Rule: Hyphens (-) should be used to improve the readability of URIs
>>
>>     Rule: Underscores (_) should not be used in URIs
>>            [underscore is masked by underline]
>>
>>     Rule: Lowercase letters should be preferred in URI paths
>>
>>     Rule: File extensions should not be included in URIs
>>            [I don't like this one. The author says that negotiation
>>     should be according to an Accept request header.]
>>
>>     Rule: Consistent subdomain names should be used for your APIs
>>            (http://api.soccer.restapi.org__)
>>
>>     Rule: Consistent subdomain names should be used for your client
>>     developer portal
>>            (http://developer.soccer.__restapi.org
>>     <http://developer.soccer.restapi.org>)
>>
>>     I hope that helps.
>>
>>     Kurt
>>
>>     On 4/13/15 3:33 PM, vaibhav thapliyal wrote:
>>
>>
>>         Dear all,
>>         I am trying to build a Rest-api for accumulo. I would like to
>>         know the minimum features that should be there in the rest api?
>>
>>         Also are there any plans to make a rest api for accumulo by the
>>         development team at accumulo itself?
>>         Thanks and regards,
>>         Vaibhav Thapliyal
>>
>>
>>     --
>>
>>     Kurt Christensen
>>     P.O. Box 811
>>     Westminster, MD 21158-0811
>>
>>     ------------------------------__----------------------------
>> --__------------
>>     "Perfection is not attainable, but if we chase perfection we can
>>     catch excellence." -Vince Lombardi
>>
>>

Reply via email to