ZEPPELIN-217 : Adding documentation on adding new chart types and Data validation serivce

[Madhuka]

Documentation on

• How to add new chart libraries for zeppelin
• How add new data validation schema
• Tutorial with Map Visualization

[Madhuka]

Documents are under Development

Data Validation

Introducing New Chart Library

[felixcheung]

typo: Zeppelin

[Leemoonsoo]

Thanks for good documentation. LGTM!

[elbamos]

I don't mean to be a naysayer, and I agree that documentation on these topics would be very helpful. I also agree that this is a good starting point. I find both pages, though, very difficult to understand. Are these directions for implementing new interpreters? For users? For developing applications on top of Zeppelin?

On Aug 16, 2015, at 12:36 PM, Lee moon soo notifications@github.com wrote:

Thanks for good documentation. LGTM!

—
Reply to this email directly or view it on GitHub.

[corneadoug]

The Data Validation documentation is too complicated and not complete I think.

I think the description should be reworked to emphasized where it is used in Zeppelin and for what purpose.

There should also have a part explaining exactly what it is doing (How the validation is done).
Some link to example, and/or an use case explanation.

Also the part about beyond data type validation is pretty obscure, it should be made so its easier to understand and apply. If you want to say that you can extend the schema, then a simple example would be helpful too.

Seems like this branch also need to be rebased (merge conflicts)

[Madhuka]

@elbamos Thanks for the feedback. A validator is important to determine the data maps to the schema without any errors beforehand. Yes it is complicated but I will write more simpler manner so that everybody can understand. It's not an interpreter. It's useful for users as well as for developers. Users can use the validator before visualization to check if the data is in accordance with the data model. It will not break and the user can take the right decision. For zeppelin developers when they introduce a new chart library they can add a validator for it so that users can validate the data before they visualize it.

[Madhuka]

@corneadoug
Yes I will write more descriptive documentation and add few sample code for it.

[Madhuka]

Data Validation Document is Updated with below sub titles.

• Where the data validator is used in zeppelin?
• Why the data validator is used?
• Why the data validator is important?
• How Data Validation is done?
• Adding new Data Range Validation

[Madhuka]

Documentation updated with Tutorial with Map Visualization. Added screenshots for docs.

Tutorial with Map Visualization

[elbamos]

This looks really helpful! I withdraw my prior objection.

On Aug 17, 2015, at 4:37 AM, madhuka notifications@github.com wrote:

@elbamos Thanks for the feedback. A validator is important to determine the data maps to the schema without any errors beforehand. Yes it is complicated but I will write more simpler manner so that everybody can understand. It's not an interpreter. It's useful for users as well as for developers. Users can use the validator before visualization to check if the data is in accordance with the data model. It will not break and the user can take the right decision. For zeppelin developers when they introduce a new chart library they can add a validator for it so that users can validate the data before they visualize it.

—
Reply to this email directly or view it on GitHub.