Looks fantastic Lewis. I hope to find some time in the next few weeks to look at this.
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ Chris Mattmann, Ph.D. Chief Architect Instrument Software and Science Data Systems Section (398) NASA Jet Propulsion Laboratory Pasadena, CA 91109 USA Office: 168-519, Mailstop: 168-527 Email: [email protected] WWW: http://sunset.usc.edu/~mattmann/ ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ Adjunct Associate Professor, Computer Science Department University of Southern California, Los Angeles, CA 90089 USA ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ -----Original Message----- From: Lewis John Mcgibbney <[email protected]> Reply-To: "[email protected]" <[email protected]> Date: Saturday, November 8, 2014 at 9:35 PM To: "<[email protected]>" <[email protected]> Subject: Re: Gora REST API >Hi Folks, >OK, I've finished a first pass at implementation of all of the CRUD and >Schema operations associated with the DataStore API [0]. >The draft REST API spec and documentation can be found here [1]. >The functionality I have neglected to implement so far includes the >following list. Please keep reading after this list as there is more >content on my own criticisms of the current API as defined in the >documentation. > >void close() >Close the DataStore. > >Result<K,T> execute(Query<K,T> query) >Executes the given query and returns the results. > >void flush() >Forces the write caches to be flushed. > >BeanFactory<K,T> getBeanFactory() >Returns the BeanFactory used by the DataStore > >Class<K> getKeyClass() >Returns the class of the keys > >List<PartitionQuery<K,T>> getPartitions(Query<K,T> query) >Partitions the given query and returns a list of PartitionQuerys, which >will execute on local data. > >Class<T> getPersistentClass() >Returns the class of the persistent objects > >void initialize(Class<K> keyClass, Class<T> persistentClass, >Properties properties) >Initializes this DataStore. > >K newKey() >Returns a new instance of the key object. > >T newPersistent() >Returns a new instance of the managed persistent object. > >Query<K,T> newQuery() >Constructs and returns a new Query. > >void setBeanFactory(BeanFactory<K,T> beanFactory) >Sets the BeanFactory to use by the DataStore. > >void setKeyClass(Class<K> keyClass) >Sets the class of the keys > >void setPersistentClass(Class<T> persistentClass) >Sets the class of the persistent objects > >If we wish to implement any of the above then i am happy to do so, but my >understanding is that these are all taken care of programmatically within >the server side code execution. The exceptions to this comment are >possibly >close and flush. > >I have some concerns here, these have mainly to do with the current design >of the DataStore API and the limitations it poses for us when implementing >a clean and intuitive REST API, namely > > - Once I've created a schema within the DataStore how do I delete one >or > more schema(s) when the method call to void deleteSchema() [2] does not > permit me to define which schema I wish to delete? > - The same goes for getting one or more schema(s) information for one > particular schema(s) > - The same exists for determining if one or more particular schema(s) > exist > - The same exists for truncating data which exists under one or more > schema(s) > >The above questions are my main pain points right now as I feel that >extending the DataStore API in order to address such requirements is quite >a major step seeing as DataStore is THE main Gora API! >I have recently been tracking a very promising project [6] which is >described as "...a RESTful web service for storing and serving mappings >between schema identifiers and schema definitions." This may help us out >for storing schema information which may be potentially leveraged to >reduce >direct impact on the Gora Datastore API but this is still slightly unclear >to me right now. > >Anyways, I hope that this is a decent start towards us implementing the >REST API for Gora. > > >BTW, if anyone wishes to comment on the apiary documentation I can grant >you perms so please send me your details if you wish to be added. >Thanks >Lewis > >[0] >https://builds.apache.org/view/All/job/gora-trunk/javadoc/index.html?org/a >pache/gora/store/DataStore.html >[1] http://docs.apachegoraapi.apiary.io/ >[2] >https://github.com/apache/gora/blob/master/gora-core/src/main/java/org/apa >che/gora/store/DataStore.java#L98 >[3] >https://github.com/apache/gora/blob/master/gora-core/src/main/java/org/apa >che/gora/store/DataStore.java#L83 >[4] >https://github.com/apache/gora/blob/master/gora-core/src/main/java/org/apa >che/gora/store/DataStore.java#L110 >[5] >https://github.com/apache/gora/blob/master/gora-core/src/main/java/org/apa >che/gora/store/DataStore.java#L104 >[6] https://github.com/schema-repo/schema-repo > >On Thu, Nov 6, 2014 at 9:12 AM, Lewis John Mcgibbney < >[email protected]> wrote: > >> Hi Folks, >> I've started documenting how I think we could implement a REST API in >>Gora. >> This is in early stages but when i good a good crack at it we can move >>it >> on quickly. >> The documentation can be found here >> http://docs.apachegoraapi.apiary.io/ >> I would REALLY like some help on this if anyone is obliged to do so. >> The aim here is that this will be the GOTO for documentation, we can >>then >> implement the actual API in Apache CXF [0] (which we already pull to >> satisfy Hadoop dependencies) and implement it via CXF's JAXRS >> implementation[1]. >> Lewis >> >> [0] http://cxf.apache.org >> [1] http://cxf.apache.org/docs/jax-rs.html >> >> -- >> *Lewis* >> > > > >-- >*Lewis*
