Implement autocomplete for ACE editor

Settings tab is crashing in the context of variable editor.

https://grafana.com/docs/grafana/latest/developers/plugins/sign-a-plugin/

Rework the column type detect method. Eventually query Druid for columns data types.

while adding the plugin getting following exception:

Unexpected token '<' Evaluating https://quanta.goshd.com/public/plugins/grafadruid-druid-datasource/module.js Loading plugins/grafadruid-druid-datasource/module

Implement bloom filter: https://druid.apache.org/docs/latest/development/extensions-core/bloom-filter.html

The query inspector doesn't display response from Druid as JSON instead just a binary blob:

It is really helpful for users to be able to see JSON response in order to figure out what/if anything is wrong with their queries.

grafadruid/go-druid@8186874 brings structured intervals spec instead of the shorthand "intervals": ["start/end"], it allows for copy/paste converted sql queries to native queries from druid web console for instance. it's so a better structure.
need to update the intervals component accordingly before updating go-druid dependency.

Most plugins let you skip TLS verification, though not recommended, this might be helpful in some scenarios. Changes needed in frontend, backend and go-druid

Instead of having to write the table name it would be great to retrieve the list form the query SELECT * FROM INFORMATION_SCHEMA.TABLES WHERE TABLE_SCHEMA = 'druid' and render it in a AsyncMultiSelect component (https://developers.grafana.com/ui/latest/index.html?path=/docs/forms-select--multi-select-async)

At the moment there are several copy / paste for each component, like https://github.com/grafadruid/druid-grafana/blob/master/src/builder/datasource/Inline.tsx#L14 this resetBuilder method, it would be better to extract into a separate component (maybe using some HoC https://capgemini.github.io/frontend/react-HoCs/)

Add options to add a delay after the last keypress before submitting the query, including an infinite delay so that queries will only be submitted by click on the run query button.

What ?

Add an option to configure the response rows limit.

Why ?

I got a use case where the grafana panel is a map, and too much points on this map can make the laptop crash.
I can set LIMIT in the sql request but then we'll won't be notice the response is limited, imo, having a message that explain limit is a good ux feature.

How ?

It could be a numeric field in "Settings" -> "Response options" container just after "Hide empty columns".
Default to 0 means no limit.
If the limit is exceeded, then it returns the limited response + an error at grafana response level.

In the Druid datasource configuration you can specify default query settings.

These are respected but the set values don't show in the query settings pane when you are running queries.

Steps to reproduce:

1. Set a query setting in the datasource config. For example I set the format to log and the hide empty columns to on.
2. Go to the explore tab and perform a query.
3. See that the settings are used. In my example, I get back log formatted results with empty columns hidden.
4. Pop out the query settings. See that they don't show the defaults set at the datasource level.

Implement probably some method that returns a container (make([]whattypeever, 0)) and its related appender func based on column type)

Add an option to filter out empty columns. Should require adding an option to control whether this is turned on and modifications to prepareResponse to filter out the empty columns.

What happened:

When I was trying to create an alter and test rules, there seemed to be a bug. The request and response is as below.

  "firing": true,
  "state": "pending",
  "conditionEvals": " = true",
  "timeMs": "179.965ms",
  "error": "tsdb.HandleRequest() response error &{POST http://proxy.druid.ke.com/druid/v2 giving up after 6 attempt(s) POST http://proxy.druid.ke.com/druid/v2 giving up after 6 attempt(s) A <nil> [] [] 0xc001913e30}",
  "logs": [
    {
      "message": "Condition[0]: Query",
      "data": {
        "from": 1622531328676,
        "queries": [
          {
            "refId": "A",
            "model": {
              "builder": {
                "aggregations": [
                  {
                    "name": "all_cnt",
                    "type": "count"
                  }
                ],
                "dataSource": {
                  "name": "commerce_adx_digv_empty_log",
                  "type": "table"
                },
                "filter": {
                  "dimension": "ip",
                  "type": "selector",
                  "value": "10.26.12.246"
                },
                "granularity": "minute",
                "intervals": {
                  "intervals": [
                    "${__from:date:iso}/${__to:date:iso}"
                  ],
                  "type": "intervals"
                },
                "postAggregations": [],
                "queryType": "timeseries",
                "virtualColumns": []
              },
              "expr": "{\"builder\":{\"queryType\":\"timeseries\",\"dataSource\":{\"name\":\"commerce_adx_digv_empty_log\",\"type\":\"table\"},\"intervals\":{\"type\":\"intervals\",\"intervals\":[\"${__from:date:iso}/${__to:date:iso}\"]},\"granularity\":\"minute\",\"filter\":{\"dimension\":\"ip\",\"type\":\"selector\",\"value\":\"10.26.12.246\"},\"aggregations\":[{\"name\":\"all_cnt\",\"type\":\"count\"}],\"postAggregations\":[],\"virtualColumns\":[]},\"settings\":{}}",
              "hide": true,
              "refId": "A",
              "settings": {}
            },
            "datasource": {
              "id": 18,
              "name": "Druid-server-alert"
            },
            "maxDataPoints": 0,
            "intervalMs": 0
          }
        ],
        "to": 1622531628676
      }
    }
  ]
}

As you can see, the parameter "intervals" seems not work properly. At lease, I think, It should not be something like "${__from:date:iso}/${__to:date:iso}".

However, When I specified a certain time range like the pic below, the problem was solved.

What you expected to happen:
The alter works successfully without specifying the intervals.

For an instance, when I use official grafana-prometheus. The request of an alter is like this:

The interval is "" (nothing here, only "from/to" specify the time range )

How to reproduce it:
Build an timeseries query without specifying the intervals. Then, create an alter and test rule. The error message will come out.

Environment:

• Grafana version: v7.3.5
• Data source type: druid
• OS Grafana is installed on: Linux
• User OS & Browser: Chrome

Probably need to extract per-query preprocessor and postprocessor into a per-query file.
Load those "plugins" (ak. QueryProcessor ?) into a register and then do something like plugins[q.Type()].preprocess(q) and plugins[q.Type()].postprocess(r) ?

This is in Druid

This is in Grafana

Add an option to return results in a format that supports Grafana's log viewer. This seems to require modifying prepareRespose to return results with dataframes shaped to work with logs. See https://grafana.com/docs/grafana/latest/developers/plugins/build-a-logs-data-source-plugin/ for details on what the response should look like.

In case of references inceptions the trash (delete) button become hidden. Need to adjust the css.

would be great to have a kind of hook mechanism in order to be able to have modules that does alter the query from the builder before passing it to the backend.
see https://github.com/grafadruid/druid-grafana/blob/master/src/QueryEditor.tsx#L34 for context

Provide a way to use a multi-values variables within a rune query.
Ideally making it possible to merge with static options, like:

So ["${countries:druidmultiple}", "France"] would expand into something like ["USA", "Italy", "France"]

Instead of having to write the lookup table name it would be great to retrieve the list form the query SELECT * FROM INFORMATION_SCHEMA.TABLES WHERE TABLE_SCHEMA = 'lookup' and render it in a AsyncSelect component (https://developers.grafana.com/ui/latest/index.html?path=/docs/forms-select--basic-select-async)

Hello,

Can we use druid metadata information like Datasource name, metric name and dimension name as drop down for forming the query instead of keying in all these information through UI. There are many chances user may have many numer of datasources/metrics or dimension which needs to be selected while generating dashboard.

Thanks,
Akhilesh

At the moment the "query builder component selector" components are mostly copy/paste (cf https://github.com/grafadruid/druid-grafana/blob/master/src/builder/query/Query.tsx and https://github.com/grafadruid/druid-grafana/blob/master/src/builder/datasource/DataSource.tsx for instance), it would be better to use an abstracted component "QueryBuilderComponentSelector" instead.

Join data source restricts data source types for right and left data sources (cf https://druid.apache.org/docs/latest/querying/datasource.html#joins-in-native-queries).
At the moment every data source type is rendered in the select list nevermind the restrictions, it would be better to list only allowed ones instead.

Implement ACE editor similar (or even same) than on web console: https://github.com/apache/druid/blob/master/web-console/src/views/query-view/query-input/query-input.tsx

I'm attempting to connect to a Druid cluster that uses both basic auth and CA cert. I see that there's a way to pass the username and pass, but no way to pass a cert. How can I enable connectivity between the Druid cluster and Grafana?

CF #65 (review)

Currently, included dashboard that is based on example druid data does not work when one does ./mage env:start. It is a very common question that noobs ask (see discussion here). It will be nice if we can import the dashboard at startup and also ingest the example data when druid container comes up.

Support virtual columns: https://druid.apache.org/docs/latest/querying/virtual-columns.html

Hi,

Is there a way currently we can define variables for creating multiple aggregations and dimension. Based on the variables selected need to build multiple aggregation and dimensions. Tried using "${Variable:druid:json}" but it doesnt add multiple aggregations nor dimensions. This is for query builder. Please let me know of there are any work around for this.

Regards,
Akhilesh

grafana version: V7.0.3
plugin version: V1.0.3(V1.0.2、V1.0.1、V1.0.0)
when i unzip this plugin in data/plugin , and restart grafana
the log show msg="Failed to load plugin" logger=plugins error="plugin \"grafadruid-druid-datasource\" is unsigned"
show i add allow_loading_unsigned_plugins=grafadruid-druid-datasource in configure
but when i use this plugin ,there are some error happen

maybe it's a frontEnd Error
please help me or give me some advise

sample PR
grafana/grafana-plugin-repository#183

Trying to play around with this as we've been longing for Druid support in Grafana for what feels like ages. Thanks for building this great plugin and putting it out there.

While I was able to put together some basic metrics displays (e.g. simple queries for singular metrics) I was struggling with and failed to build some more complex graphs e.g. referencing global variables and such. I assume they would be supported after studying the readme.

Of course having a proper tutorial would make sense in the long run. For the moment even seeing some screenshots of possible graph configurations/query types/advanced usage examples like variables and formulas along with the results would help kickstart everyone and provide a starting point from which to build upon. I assume the people using your plugin at the moment are probably advanced Grafana users to say the least so this can go a long way.

Instead of having to write the table name it would be great to retrieve the list form the query SELECT * FROM INFORMATION_SCHEMA.TABLES WHERE TABLE_SCHEMA = 'druid' and render it in a AsyncSelect component (https://developers.grafana.com/ui/latest/index.html?path=/docs/forms-select--basic-select-async)

The demo dashboard link appears to brooken.

https://github.com/grafadruid/druid-grafana/blob/master/dashboard.json

Grafana 8 gonna provides a brand new variables API that will need to be implemented: grafana/grafana#29674

It was a big help at Poshmark to have the auto complete. Made the querying so much more easier.

It was added in the following PRs
grafana-druid-plugin/druidplugin#48
grafana-druid-plugin/druidplugin#28
grafana-druid-plugin/druidplugin#30
grafana-druid-plugin/druidplugin#36

I did not understand the use case of having interval option as part of the UI.

I always thought that interval should come from grafana and we should use that

Cascade extraction function use a new kind of component that allows to add multiple sub-components: https://github.com/grafadruid/druid-grafana/blob/master/src/builder/extractionfn/Cascade.tsx
It probably deserve some abstraction (like a QueryBuilderComponents component) in order to be easilly re-used.

Have to implement the extensions based aggregations

Hello, we used to use the first druid plugin for Grafana (https://github.com/societe-generale/druidplugin ) and as you can see when we do a group by (or topn) query , Druid give us 2 metrics : one for each series

When we try do to the same with your plugin, the asnwer is only one metric instead of 2, when we used the graph panel. We don't understand why because when we do the same query with a table panel panel, we can see all the data. For example :

Best regards.

when i use postAggregations, I select the type is "arithmetic", but the post data is "cardinality"

Implement experimental FilterTuning

At the moment no field is required on client side, no validation at all.
Would be great to, at least, add required validation.