add build scripts about druidplugin

[abhisant]

Sorry for the late response. I am currently traveling out of the country for a month, (on vacation with intermittent internet connectivity). I will try to find some time out of my vacation to add the build scripts.

[hmermerkaya]

I need build scripts too, I could make some contributions to the project.
Based on our needs, I added the following features to Kenji Noguchi work.

1. Shifting time forward so that one can compare for an instance the data of the last week with that of this week in the same time period.
2. Implementing predefined post aggregations and their relevant aggregations read from json file sitting in dist folder with the same name as that of druid datasource.
3. Implementing post aggregation in which either of the aggregations could exist in different datasources. Aggregations from different datasource can be invoked by object-property notation. For example in a datasource labeled as "A" and an aggreation named "agg1", it is 'A.agg1'.

Please see my github repo https://github.com/hmermerkaya/druidplugin

[abhisant]

I am back today late afternoon. I will start working on this repo tomorrow onwards one bullet at a time. On Mon, Feb 6, 2017 at 12:59 PM Hamit Mermerkaya <[email protected]> wrote: I need build scripts too, I could make some contributions to the project. Based on our needs, I added the following features to Kenji Noguchi work. 1. Shifting time forward so that one can compare for an instance the data of the last week with that of this week in the same time period. 2. Implementing predefined post aggregations and their relevant aggregations read from json file sitting in dist folder with the same name as that of druid datasource. 3. Implementing post aggregation in which either of the aggregations could exist in different datasources. Aggregations from different datasource can be invoked by object-property notation. For example in a datasource labeled as "A" and an aggreation named "agg1", it is 'A.agg1'. Sample usage is in the DruidPanel screenshot. Please see my github repo https://github.com/hmermerkaya/druidplugin — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#4 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AF098B4_J3i0OVZgv2YhiQmw8legJx3Pks5rZ4mVgaJpZM4LZ8f3> .

[knoguchi]

This plugin made its way to the official plugin repo. It means this repo is responsible for the backward compatibility since the general users started using it. Though it still lacks Druid's critical features such as PostAggregator, granularity=all, timezone handling, the filter expressions that support AND OR, etc etc.

It was too much work for me to keep my fork mergeable state to the upstream while maintaining the backward compatibility, adding new features, and working around Druid's weird design all at the same time. Also my fork introduced math expression parsers so it can support calculations beyond the simple binary operation e.g. +-*/ . That's the time I decided to take my repo off the Github. I've never meant to create a hostile fork.

If we build v2.0 I would contribute back where we don't have to maintain the backward compatibility. Going forward we better come up with roadmap to support the Druid's features thoroughly, or design it to be extensible to the least. Maybe we can start it for Grafana v4.0. What do you think?

By the way @hmermerkaya, the SQL query is not working. You may want to remove it. Back then I was planning to add the PlyQL nodejs module to the plugin. It turned out to be difficult to integrate.

[abhisant]

Added build scripts.