[ZEPPELIN-381] sqlContext doesn't work with tables created by external jars (e.g. com.databricks.spark.csv); spark <1.5.1

• Type: Bug
• Status: Resolved
• Priority: Major
• Resolution: Fixed
• Affects Version/s: 0.6.0
• Fix Version/s: 0.6.0
• Component/s: Core

Environment

this bug is true for Spark 1.4; I tried it with newest 1.5.1 and it worked fine

• Assignee: Unassigned
• Reporter: Tomas Hudik
• Created: 31/Oct/15 20:57
• Updated: 03/Feb/16 04:00
• Resolved: 01/Feb/16 02:09

Description

this works fine:

%dep
z.load("com.databricks:spark-csv_2.10:1.2.0")

val df = sqlContext.load("com.databricks.spark.csv", Map("path" -> "bank-full.csv","header"->"true","delimiter"->";"))
df.registerTempTable("bank")

however, when I try to do select in SQL interpreter I got error:

%sql
select age, count(1) from hmm where age < 30 group by age order by age

java.lang.ClassNotFoundException: com.databricks.spark.csv.CsvRelation$$anonfun$tokenRdd$1$$anonfun$1 at java.net.URLClassLoader.findClass(URLClassLoader.java:381) at java.lang.ClassLoader.loadClass(ClassLoader.java:424) at sun.misc.Launcher$AppClassLoader.loadClass(Launcher.java:331) at java.lang.ClassLoader.loadClass(ClassLoader.java:357)

Question: if temp table is registered why Sql command still requires it?

Issue Links

• duplicates: ZEPPELIN-648 jar dependencies loaded in %spark but not %sql in spark 1.6/zeppelin 0.5.6
• is superceded by: ZEPPELIN-630 Introduce new way of dependency loading to intepreter