Hi moon, In the end I managed to refactor my code so I didn’t need the InterpreterContext or the paragraph ID at all. As my use case for it has sort of disappeared I won’t file an issue for it now. Other people might need to have access to the ZeppelinContext or InterpreterContext for their use cases though, so don’t let me stop you raising an issue of you want… Thanks again for your help, Lucas.
From: moon soo Lee [mailto:m...@apache.org] Sent: 13 October 2015 16:27 To: users@zeppelin.incubator.apache.org Subject: Re: How to find out if a user's Scala commands are executing in the same Zeppelin paragraph? We can think some mechanism to get instance of InterpreterContext from any user library. For example, store instance of InterpreterContext to some staticfield everytime before run interpret(). If you think it would be useful, please don't hesitate to file an issue. Best, moon On Tue, Oct 13, 2015 at 3:50 PM Partridge, Lucas (GE Aviation) <lucas.partri...@ge.com<mailto:lucas.partri...@ge.com>> wrote: Hi moon, Thanks very much for the workaround. Rather than go through all my methods to add an extra parameter for the InterpreterContext I decided to pass it in once to a new setup() method which then keeps a reference to it for use by the other methods. It’s not as user-friendly as before because now the user has to call setup(z.getInterpreterContext()) before they call any of the other methods but hopefully it will work… Hmm, on testing it looks like I will need to pass in InterpreterContext once *per paragraph* because the InterpreterContext instance appears to change between paragraphs too… Oh well! Thanks for your help, Lucas. From: moon soo Lee [mailto:m...@apache.org<mailto:m...@apache.org>] Sent: 13 October 2015 12:31 To: users@zeppelin.incubator.apache.org<mailto:users@zeppelin.incubator.apache.org> Subject: Re: How to find out if a user's Scala commands are executing in the same Zeppelin paragraph? Hi, "org.apache.zeppelin" % "zeppelin-interpreter" % "0.5.0-incubating' Will provide class for InterpreterContext. However ZeppelinContext is inside of "org.apache.zeppelin" % "spark" and it is not published to public maven repository. So, you can add "org.apache.zeppelin" % "zeppelin-interpreter" % "0.5.0-incubating" as dependency of your scala code. And everytime you call your scala code in the paragraph, you can pass InterpreterContext to your class/function. Is this approach okay for your case? Best, moon On Tue, Oct 13, 2015 at 1:05 PM Partridge, Lucas (GE Aviation) <lucas.partri...@ge.com<mailto:lucas.partri...@ge.com>> wrote: Hi moon, Sorry – I should’ve asked this before. How can I make use of “z.getInterpreterContext().getParagraphId()” in my own Scala code which I’m building outside of Zeppelin, please? I’ve added this line to the scala file where I make the calls: import org.apache.zeppelin.spark.ZeppelinContext and I’ve added this line to my libraryDependencies in my sbt build file: "org.apache.zeppelin" % "zeppelin" % "0.5.0-incubating" but when I try to compile I get this error first: [error] …. object zeppelin is not a member of package org.apache [error] import org.apache.zeppelin.spark.ZeppelinContext [error] ^ When I look in my .ivy2 cache I see there are no jars under org.apache.zeppelin/zeppelin. Also, there are no jars on maven central at http://repo1.maven.org/maven2/org/apache/zeppelin/zeppelin/0.5.0-incubating/<https://urldefense.proofpoint.com/v2/url?u=http-3A__repo1.maven.org_maven2_org_apache_zeppelin_zeppelin_0.5.0-2Dincubating_&d=BQMFaQ&c=IV_clAzoPDE253xZdHuilRgztyh_RiV3wUrLrDQYWSI&r=c1CCNND4PG-Q_V2AJWDWrugZAXQ8Y3EE_f_mAHcpXcs&m=kANvf78ADS7-6Flu-9_2opb242HYlQzuWjzb-umWlK8&s=uusoVdVbG2ds4AKU5qO0jiG2haP4FxOIfho98YsS0lA&e=> either! Is this what would you expect, or do the binaries for that release need publishing to maven central please? Or should I build zeppelin-0.5.0-incubating locally instead? Many thanks, Lucas. From: Partridge, Lucas (GE Aviation) Sent: 13 October 2015 09:56 To: users@zeppelin.incubator.apache.org<mailto:users@zeppelin.incubator.apache.org> Subject: RE: How to find out if a user's Scala commands are executing in the same Zeppelin paragraph? Thanks moon; that’s exactly what I wanted☺. I confirm it works for me in standalone zeppelin-0.5.0-incubating. I get strings back with date/time stamps and a unique ID that differs between paragraphs. Thanks again, Lucas. From: moon soo Lee [mailto:m...@apache.org] Sent: 12 October 2015 17:17 To: users@zeppelin.incubator.apache.org<mailto:users@zeppelin.incubator.apache.org> Subject: Re: How to find out if a user's Scala commands are executing in the same Zeppelin paragraph? Hi Lucas, You can get current paragraph Id by calling %spark z.getInterpreterContext().getParagraphId() I'm not quite sure i understand your question correctly, but hope this helps. Thanks, moon On Mon, Oct 12, 2015 at 5:29 PM Partridge, Lucas (GE Aviation) <lucas.partri...@ge.com<mailto:lucas.partri...@ge.com>> wrote: Does anyone please know how to find out what paragraph your Scala code is running in at runtime? I’m using the standalone distribution of Zeppelin, zeppelin-0.5.0-incubating.tgz. I want to write some Scala code that does one thing if it’s being executed in the same paragraph as another call to my code, but another if it’s executing in a different paragraph to the other call. (Specifically I want to create a new figure or chart if it’s running in a new paragraph, or add to an existing figure if it’s running in the same paragraph.) Many thanks, Lucas.
