try using my UDF: https://github.com/gerritjvv/pigutils/tree/master/pigudfs/udfs
it loads json as a map, and all nested objects as maps. I've not released the jar yet so you'll need to compile a jar from source. here's an example of how to use it: l = load 'myfile.json.gz' using org.nts.pigutils.udfs.JSONLoader(); r = foreach l generate m#'user'#'age', m#'name'; On Thu, Aug 29, 2013 at 3:19 PM, jamal sasha <[email protected]> wrote: > Hi, > > I have json file in follwoing format: > { "_id" : "foo.com", "categories" : [], "h1" : { "bar==" : { "first" : > 1281916800, "last" : 1316995200 }, "foo==" : { "first" : 1281916800, "last" > : 1316995200 } }, "name2" : [ "foobarl.com", "foobar2.com" ], "rep" : > null } > So, how do i parse this json in pig.. > > also, the categories and rep can have some char in it..and might not be > always empty. > > Thanks > > > > This message may contain confidential and/or privileged information. If it > has > been sent to you in error, please reply to advise the sender of the error > and > then immediately delete this message.
