On Fri, Jul 8, 2011 at 2:48 PM, Grant Ingersoll <[email protected]> wrote:
> But are you keeping member variables or do you put everything in the > context? Anything that you want to remember needs to put in the context. PIG makes sure that the constructor is called with the same arguments on front-end and backend. In addition, for loaders and storage, setContext API is invoked with the same same context on frontend and backend. Anything else you need to put in the context. If something is derived from constructor arguments, you don't need to put into context for e.g. not sure if I understood the question correctly, but PIG does not transfer your object, so what you store in the member variables does not matter. Raghu. On Jul 8, 2011, at 3:21 PM, Raghu Angadi wrote: > > > yes. that is exactly how HBaseStorage uses context. > > > > On Fri, Jul 8, 2011 at 10:19 AM, Jeremy Hanna < > [email protected]>wrote: > > > >> In CassandraStorage, we had been using some load/store URL specific > >> information (keyspace, column family names) to make the > >> UDFContext.properties key unique, but with what Grant said was in the > docs, > >> we just wrote a patch to instead use the udf context signatures for > those > >> keys when setting and getting those property values. Is that the way to > go > >> then? I'm setting those as member variables and then using them later. > >> > >> @Override > >> public void setUDFContextSignature(String signature) > >> { > >> this.loadSignature = signature; > >> } > >> > >> /* StoreFunc methods */ > >> public void setStoreFuncUDFContextSignature(String signature) > >> { > >> this.storeSignature = signature; > >> } > >> > >> > >> On Jul 8, 2011, at 7:24 AM, Grant Ingersoll wrote: > >> > >>> What is the guidance here on using member variables when implementing > >> UDFs and passing properties? That is, what are the semantics for using > them > >> to store properties for a UDF instance? The docs talk a lot about > making > >> sure that no side effects happen from multiple calls to a UDF instance, > but > >> it is not clear whether that means it's doing things like changing the > >> Location for a given instance of a UDF or just calling it multiple > times. > >> PigStorage suggests not (since it keeps a member var location), but the > >> UDFContext docs suggests that one keep all state in the UDFContext under > an > >> appropriate signature. > >>> > >>> See also https://issues.apache.org/jira/browse/CASSANDRA-2869 for > >> another case where this has reared it's head in an improper > implementation. > >>> > >>> -Grant > >>> > >>> On Jul 7, 2011, at 3:24 AM, Jeremy Hanna wrote: > >>> > >>>> > >>>> On Jul 6, 2011, at 11:10 PM, Raghu Angadi wrote: > >>>> > >>>>> On Wed, Jul 6, 2011 at 7:20 PM, Jeremy Hanna < > >> [email protected]>wrote: > >>>>> > >>>>>> > >>>>>> On Jul 6, 2011, at 12:47 PM, Dmitriy Ryaboy wrote: > >>>>>> > >>>>>>> I think this is the same problem we were having earlier: > >>>>>>> http://hadoop.markmail.org/thread/kgxhdgw6zdmadch4 > >>>>>>> > >>>>>>> One workaround is to use defines to explicitly create different > >>>>>>> instances of your UDF, and use them separately.. it's ugly but it > >>>>>>> works. > >>>>>> > >>>>>> Thanks Dmitriy. > >>>>>> > >>>>>> I tried doing something like: > >>>>>> define ToCassandraBag1 org.pygmalion.udf.ToCassandraBag(); > >>>>>> define ToCassandraBag2 org.pygmalion.udf.ToCassandraBag(); > >>>>>> > >>>>> > >>>>> This still does not work since you can't distinguish the two. The way > I > >> was > >>>>> thinking of doing this is to let user pass in some unique sting as a > >>>>> substitute for context: > >>>>> > >>>>> define ToCassandraBag1 ToCassandraBag('1'); > >>>>> define ToCassandraBag2 ToCassandraBag('2'); > >>>> > >>>> Ah yes. I had misunderstood. Thanks for the clarification. Now the > >> pig docs also make more sense in the Passing Configurations to UDFs > section: > >>>> > >> > http://pig.apache.org/docs/r0.8.1/udf.html#Passing+Configurations+to+UDFs > >>>> It says: > >>>> "The UDF can pass its constructor arguments, or some other identifying > >> strings. This allows each instantiation of the UDF to have a different > >> properties object thus avoiding name space collisions between > instantiations > >> of the UDF." > >>>> and the HBaseStorage example was helpful to see that in action. > >>>> > >>>> Thanks both to Raghu and Dmitriy. > >>>> > >>>>> > >>>>> inside the UDF, you would use this arg to make a 'contextString' (see > >>>>> HBaseStorage.java for example use) to store any state. > >>>>> > >>>>> ideally UDFs shouldn't have to do this.. They should have the same > >> context > >>>>> info that is available for loaders and storage. > >>>>> > >>>>> Raghu. > >>>>> > >>>>> > >>>>>> > >>>>>> at the top and then using each one only once. That still produces > the > >> same > >>>>>> error. I guess in this case we'll just have to require the field > >> names be > >>>>>> entered into the UDF and it won't introspect them. Ah well. Would > be > >> nice > >>>>>> to be able to use it but I don't really see another way around this > >> bug with > >>>>>> the shared UDF context. > >>>>>> > >>>>>>> > >>>>>>> D > >>>>>>> > >>>>>>> On Wed, Jul 6, 2011 at 9:42 AM, Jeremy Hanna < > >> [email protected]> > >>>>>> wrote: > >>>>>>>> We have a UDF that introspects the output schema and gets the > field > >>>>>> names there and use that in the exec method. > >>>>>>>> > >>>>>>>> The UDF is found here: > >>>>>> > >> > https://github.com/jeromatron/pygmalion/blob/master/udf/src/main/java/org/pygmalion/udf/ToCassandraBag.java > >>>>>>>> > >>>>>>>> A simple example is found here: > >>>>>> > >> > https://github.com/jeromatron/pygmalion/blob/master/scripts/from_to_cassandra_bag_example.pig > >>>>>>>> > >>>>>>>> It takes the relation's aliases and uses them in the output so > that > >> the > >>>>>> user doesn't have to specify them. However we've been noticing that > >> if you > >>>>>> have more than one ToCassandraBag call in a pig script, sometimes > they > >> are > >>>>>> run at the same time and the key is the same in the UDF context: > >>>>>> cassandra.input_field_schema. So we think there is an issue there > >> (array > >>>>>> out of bounds exceptions when running the script, but when running > in > >> grunt > >>>>>> one at a time, it doesn't do that). > >>>>>>>> > >>>>>>>> Is there a right way to do this parameter passing so that we don't > >> get > >>>>>> these errors when multiple calls exist? > >>>>>>>> > >>>>>>>> We thought of using the schema hash code as a suffix (e.g. > >>>>>> cassandra.input_field_schema.12344321) but we don't have access to > the > >>>>>> schema in the exec method. > >>>>>>>> > >>>>>>>> We thought of having the first parameter of the input tuple be a > >> unique > >>>>>> name that the script specifies, like 'filename.relationalias' as a > >>>>>> convention to make them unique to the file. However in the > >> outputSchema, we > >>>>>> don't have access to the input tuple, just the schema itself, so it > >> couldn't > >>>>>> get that value in there. > >>>>>>>> > >>>>>>>> Any ideas on how to make this so it doesn't stomp on each other > >> within > >>>>>> the pig script? Is there a best way to do that? > >>>>>>>> > >>>>>>>> Thanks! > >>>>>>>> > >>>>>>>> Jeremy > >>>>>> > >>>>>> > >>>> > >>> > >>> -------------------------- > >>> Grant Ingersoll > >>> > >>> > >>> > >> > >> > > -------------------------- > Grant Ingersoll > > > >
