Lee moon soo created ZEPPELIN-1698:
--------------------------------------
Summary: Restrict of interpreter setting on GUI for each
interpreter.
Key: ZEPPELIN-1698
URL: https://issues.apache.org/jira/browse/ZEPPELIN-1698
Project: Zeppelin
Issue Type: Improvement
Reporter: Lee moon soo
Zeppelin interpreter setting page has options.
* Interpreter mode : {{Globally|Per Note|Per User}} in {{Scoped|Isolated}}
process
* User Impersonate
But not all interpreter necessarily require those options.
For example, markdown interpreter will need only {{Globally}}, JDBC interpreter
might need {{Globally|Per Note}} while Spark interpreter require all options
including "User Impersonate".
However, unless user fully read documentation or source code, it's very
difficult to know which 1) combination of options to use, 2) in which
circumstance.
So it'll be great each interpreter setting displays restricted options
supported by the interpreter, with description about the combination.
Information about the restriction can be provided by interpreter-setting.json
of each interpreter. Here's example
{code}
{
"options" : {
"globally" : {
"description" : "all notebook shares a single spark context"
},
"per user" : {
"description": "each user has it's own REPL session",
"mode" : {
"scoped" : { description: "Spark context is being shared by all
users"},
"isolated" : {
"description": "Each user will have it's own SparkContext",
"impersonate" : "Process will run as a logged in user"
}
},
},
"per note" : {
"description": "each note has it's own REPL session",
"mode" : {
"scoped" : { description: "Spark context is being shared by all
notes"},
"isolated" : {
"description": "Each note will have it's own SparkContext",
"impersonate" : "Process will run as a logged in user"
}
},
}
}
}
{code}
another example for md, which restrict to 'globally'
{code}
{
"options" : {
"globally" : {
"description" : "all notebook shares a single spark context"
}
}
}
{code}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
