[21:54:17] <lzia>	 ewulczyn: is your problem with Hue resolved?
[21:54:26] <lzia>	 I'm getting some errors in hive that may be related
[21:54:29] <ewulczyn>	 Yes!
[21:54:33] <leila>	 humm
[21:54:45] <leila>	 do you know what the problem was?
[21:55:06] <ewulczyn>	 somthing to do with the fact that I have 2 different shell names
[21:55:15] <leila>	 I see. thanks!
[21:55:34] <ottomata>	 lzia, what are your hive problems?
[21:55:44] <ottomata>	 ewulczyn: fyi, I puppetized that hive.support.concurrency t hing
[21:55:51] <ottomata>	 so you shouldn't have to set it manually now
[21:55:55] <ottomata>	 thanks for testing that fo rme
[21:56:22] <ewulczyn>	 np
[21:57:42] <leila>	 ottomata: I dropped the table we made yesterday and recreated it and now I don't see the error I was seeing earlier.
[21:58:01] <ottomata>	 ok
[21:58:24] <leila>	 however, I have another question for you: can you do select count(id) from leila.page_trees where project='fa'; ?
[21:58:32] <leila>	 ottomata, ^
[22:02:10] <ottomata>	 ja, NPE hm
[22:05:32] <ottomata>	 leila, i'm not sure what is up there
[22:05:53] <leila>	 yeah, let me look into it a bit more.
[22:06:14] <leila>	 I don't know if the data is read into the correctly, ottomata.
[22:06:42] <ottomata>	 is it possible the language stuff is messing up the json parser?  dunno.
[22:06:46] <ottomata>	 seems like that's not it
[22:06:51] <ottomata>	 have you tried it with a different partition?
[22:06:53] <ottomata>	 maybe en
[22:06:55] <ottomata>	 just to see?
[22:26:05] <leila>	 ottomata, in a meeting, will check it as soon as I'm out.
[22:30:15] <leila>	 same story with 'en' ottomata.
[22:33:14] <ottomata>	 that is more interesting!
[22:34:31] <ottomata>	 looking
[22:35:14] <ottomata>	 hmm
[22:40:48] <leila>	 ottomata: I think the tree structure is not transferred properly to the table.
[22:41:20] <ottomata>	 certainly not, you also don't have a children field
[22:41:29] <ottomata>	 but, i am trying that now, and it didnt' help
[22:41:51] <ottomata>	 leila, hive is not good with arbitrary schemas
[22:42:17] <ottomata>	 i think that's the problem.
[22:42:20] <ottomata>	 hmm
[22:42:29] <ottomata>	 unless json serde is really fance, but i don't think it is
[22:47:40] <ottomata>	 hm
[22:53:51] <ottomata>	 ok leila i got something
[22:54:12] <ottomata>	 rather than trying to a hive schema onto a variable json one
[22:54:14] <ottomata>	 hive has this:
[22:54:17] <ottomata>	 https://cwiki.apache.org/confluence/display/Hive/LanguageManual+UDF#LanguageManualUDF-get_json_object
[22:54:30] <ottomata>	 which allows you to sorta use JSON path like addressing of json objects in a hive query
[22:54:31] <ottomata>	 so
[22:55:33] <ottomata>	 http://www.codeshare.io/aITwc
[22:55:38] <ottomata>	 SELECT get_json_object(json_data, '$.id') FROM page_trees_ottoB where project='fa' limit 10;
[23:11:11] <leila>	 ottomata: this is working. thanks!
[23:25:32] <ottomata>	 yes! you are welcome.