cube-js/cube.js

📊 Cube — The Semantic Layer for Building Data Applications

Website • Getting Started • Docs • Examples • Blog • Slack • Twitter

Cube is the semantic layer for building data applications. It helps data engineers and application developers access data from modern data stores, organize it into consistent definitions, and deliver it to every application.

Learn more about connecting Cube to data sources and analytics & visualization tools.

Cube was designed to work with all SQL-enabled data sources, including cloud data warehouses like Snowflake or Google BigQuery, query engines like Presto or Amazon Athena, and application databases like Postgres. Cube has a built-in relational caching engine to provide sub-second latency and high concurrency for API requests.

For more details, see the introduction page in our documentation.

Why Cube?

If you are building a data application—such as a business intelligence tool or a customer-facing analytics feature—you’ll probably face the following problems:

1. SQL code organization. Sooner or later, modeling even a dozen metrics with a dozen dimensions using pure SQL queries becomes a maintenance nightmare, which leads to building a modeling framework.
2. Performance. Most of the time and effort in modern analytics software development is spent providing adequate time to insight. In a world where every company’s data is big data, writing just SQL queries to get insight isn’t enough anymore.
3. Access Control. It is important to secure and govern access to data for all downstream data consuming applications.

Cube has the necessary infrastructure and features to implement efficient data modeling, access control, and performance optimizations so that every application—like embedded analytics, dashboarding and reporting tools, data notebooks, and other tools—can access consistent data via REST, SQL, and GraphQL APIs.

Getting Started 🚀

Cube Cloud

Cube Cloud is the fastest way to get started with Cube. It provides managed infrastructure as well as an instant and free access for development projects and proofs of concept.

For a step-by-step guide on Cube Cloud, see the docs.

Docker

Alternatively, you can get started with Cube locally or self-host it with Docker.

Once Docker is installed, in a new folder for your project, run the following command:

docker run -p 4000:4000 \
  -p 15432:15432 \
  -v ${PWD}:/cube/conf \
  -e CUBEJS_DEV_MODE=true \
  cubejs/cube

Then, open http://localhost:4000 in your browser to continue setup.

For a step-by-step guide on Docker, see the docs.

Resources

• Documentation
• Getting Started
• Examples & Tutorials
• Architecture

Contributing

There are many ways you can contribute to Cube! Here are a few possibilities:

• Star this repo and follow us on Twitter.
• Add Cube to your stack on Stackshare.
• Upvote issues with 👍 reaction so we know what's the demand for particular issue to prioritize it within road map.
• Create issues every time you feel something is missing or goes wrong.
• Ask questions on Stack Overflow with cube.js tag if others can have these questions as well.
• Provide pull requests for all open issues and especially for those with help wanted and good first issue labels.

All sort of contributions are welcome and extremely helpful 🙌 Please refer to the contribution guide for more information.

License

Cube Client is MIT licensed.

Cube Backend is Apache 2.0 licensed.

cube.js's Issues

Disable authentication check

As we are implementing our own authentication middleware, it would be great to disable the authentication check in @cubejs-backend/api-gateway.

I could imagine something like this:

import * as CubejsServerCore from "@cubejs-backend/server-core";

const config = {
  ...
  security: false
};
const core = CubejsServerCore.create(config);

What do you think?

Move logs from cubejs-server-core to cubejs-server

Hi all,

As we are using Cubejs in an embedded environment with our own Express server, the following logs are not correct anymore and a bit misleading:

• console.log(`🔒 Your temporary cube.js token: ${cubejsToken}`); (Link)
• console.log(`🦅 Dev environment available at ${localUrl}`); (Link)

Suggestion: These logs should not reside in cubejs-server-core, but in cubejs-server.

ResultSet series returns a series with no data

series method of ResultSet could return a series, which has no data at all. Plotting it would result into wrong legend as an on image below

There is no data for ∅ series.

Update order format

Query object could contain an order property, but its format needs to be changed. Once it is changed it should be added to the Query Format documentation page.

Giving me weird error

I have created 2 Schemas/Tables Orders and Users. In index.js file I am trying to render User's table attributes. I don't have this 'status' attribute in any of my sql table and schema. Why I am getting this even though,I have no where mentioned 'status' and I am trying to render Users schema.

This is my index.js File

 `const CubejsServer = require('@cubejs-backend/server');
  import cubejs from '@cubejs-client/core';
  import Chart from 'chart.js';
  import chartjsConfig from './toChartjsData';

   const cubejsApi = cubejs('eyJ0eXAiOiJKV1QiLCJhbGciOiJIUzI1NiJ9.eyJpIjozODU5NH0.5wEbQo- 
   VG2DEjR2nBpRpoJeIcE_oJqnrm78yUo9lasw'); 

   const server = new CubejsServer();


  const resultSet = await cubejsApi.load({
      measures: ['Users.paying'],
      timeDimensions: [{
        dimension: 'Users.createdAt',
       dateRange: ['2018-01-01', '2018-07-31'],
       granularity: 'month'
    }]
 })
 const context = document.getElementById("myChart");
 new Chart(context, chartjsConfig(resultSet));


 server.listen().then(({ port }) => {
 console.log(`🚀 Cube.js server is listening on ${port}`);
 });`

mongobi nested field issue while filter query

Describe the bug
mongobi nested field issue while filter query

To Reproduce
Filter query on bases of nested field .

Expected behavior
work perfectly if we wrap nested field by backtick

issue with query generator for Mongobi for nested fields in document

error while executing query with nested fields.

Error while querying: {"queryKey":["SELECT details.name flow.detailsname, count(*) flow.count FROM flow AS flow GROUP BY 1 ORDER BY 2 DESC LIMIT 10000",[],[]],"error":"Error: Unknown column 'details.name' in 'field list'

For schema file below

cube(Flow, { sql: SELECT * FROM 1224.flow`,

joins: {

},

measures: {
count: {
type: count,
drillMembers: [createdbyuser, details.name, projectid, createddate, modifieddate]
}
},

dimensions: {
details.owner: {
sql: details.owner,
type: string
},

createdbyuser: {
  sql: `createdByUser`,
  type: `string`
},

details.description: {
  sql: `details.description`,
  type: `string`
},

details.name: {
  sql: `details.name`,
  type: `string`
},
`

For mongobi query, the nested fields needs to be inside `` I tested this with mysql workbench connecting to mongo over same BI Connectors.
I think, may be query generator needs to handle this.

working version of same query from mysql test
SELECT details.name flow.detailsname, count(*) flow.count FROM flow AS flow GROUP BY 1 ORDER BY 2 DESC LIMIT 10000

Validate for empty query object

Now, if you pass empty query it will give a weird error. Query needs to be validated.

segments missing in request validation. issue

Hi, it seems that segments missing in request validation.

Can you add segments validation to cubejs-api-gateway.

Now its not working when request contains segments

Downloading reactjs dependencies for cubejs

cubejs create hello-world -d postgres

I am using postgres for database. After starting localhost on my vm and accessing it via my host machine then some dependencies got downloaded for reactjs app(Also creating a reactjs app inside a dashboard-app folder) later I wanted to add dependencies

$ npm i --save @cubejs-client/core
$ npm i --save @cubejs-client/react

Don't know where to add in hello-world or dashboard-app?

and also after running it in both places I am getting

npm ERR! Unexpected end of JSON input while parsing near '...r\n-----END PGP SIGNA'

npm ERR! A complete log of this run can be found in:
npm ERR! /home/navs/.npm/_logs/2019-04-17T10_26_24_301Z-debug.log
navs@kafka:~/CubeJS/hello-world$ npm i --save @cubejs-client/core
npm ERR! Unexpected end of JSON input while parsing near '...r\n-----END PGP SIGNA'

npm ERR! A complete log of this run can be found in:
npm ERR! /home/navs/.npm/_logs/2019-04-17T10_28_53_779Z-debug.log
navs@kafka:/CubeJS/hello-world$ cd dashboard-app/
navs@kafka:/CubeJS/hello-world/dashboard-app$ npm i --save @cubejs-client/core
npm ERR! Unexpected end of JSON input while parsing near '...r\n-----END PGP SIGNA'

npm ERR! A complete log of this run can be found in:
npm ERR! /home/navs/.npm/_logs/2019-04-17T10_29_21_857Z-debug.log
navs@kafka:/CubeJS/hello-world/dashboard-app$ cd ..
navs@kafka:/CubeJS/hello-world$ npm i --save @cubejs-client/core
npm ERR! Unexpected end of JSON input while parsing near '...r\n-----END PGP SIGNA'

npm ERR! A complete log of this run can be found in:
npm ERR! /home/navs/.npm/_logs/2019-04-17T10_30_37_847Z-debug.log

Allow hot reload for cubes. Feature request

Now server starts and loads all cubes from schema folder ( or another folder )
when I change my schema I need to restart start server.

hot reload feature:

• when the server cant find some dimension or measure or segment, it can reload cube again and check is it present in the cube. do just 1 reload and return validation result when cant find required property.

profit:
it`ll allow changing schema in runtime.

Integer column not getting mapped in the schema file for BQ

Describe the bug
I am currently using BigQuery as DB. I have column 'age' set as 'integer' in the table.
But the schema file in cube.js, does not show this column in dimension or in measures.

Expected behavior
All the columns that are present in the table, should be mapped in the dimension.

Version:
0.5.2

Mongodb BI SSL configuration broken

Describe the bug
The current implementation of cubejs-mongobi-driver has broken SSL connection support.
It passes ssl config to underlying node-mysql2 driver like so:
https://github.com/statsbotco/cube.js/blob/3202508c98f7c2f342e5936b7f219d78876a2117/packages/cubejs-mongobi-driver/driver/MongoBIDriver.js#L9-L17

https://github.com/statsbotco/cube.js/blob/3202508c98f7c2f342e5936b7f219d78876a2117/packages/cubejs-mongobi-driver/driver/MongoBIDriver.js#L26

But underlying mysql driver used to connect to mongo bi (mongosqld) accepts ssl config as an object:
https://github.com/sidorares/node-mysql2/blob/b38c3a45c887a38180896c3a6f256296018511dd/lib/connection.js#L299

const secureContext = Tls.createSecureContext({
      ca: this.config.ssl.ca,
      cert: this.config.ssl.cert,
      ciphers: this.config.ssl.ciphers,
      key: this.config.ssl.key,
      passphrase: this.config.ssl.passphrase
    });

To Reproduce
Try configuring SSL

Expected behavior
SSL configuration should work.

Version:
[e.g. 0.4.5]

Additional context
Opened a PR.

Add ESLint, any other auto-formatting config

Per our Slack conversation, this ticket is a placeholder for the work needed to move any existing ESLint, Prettier, or any other dot-files / config related to linting and formatting into this public Github repo.

Let me know if I can be of any help on this - happy to!

MUTEX for requests

Stop sending old requests if query has been changed

BigQuery?

Hi,
first of all this project looks extremely interesting!

I was studying it and looking for BigQuery support. It is not clear to me whether using big query as db is supported or not, as the BQ is listed as supported in the home, but in the docs it is not mentioned in the -d option flag. Moreover, the CLI help says that valid options for that flags are

-d, --db-type <db-type> Preconfigure for selected database. Options: postgres, mysql, athena

Add TypeScript support

Hi all,

As quickly discussed with @paveltiunov in Slack, it would be awesome to support TypeScript for Cube.js. This can be achieved by adding a declaration file and bundling it with the npm package. Here you can find more information.

Thanks a lot and keep up the great work!

CLI doesn't scaffold correct .env file based on -d option

Describe the bug
CLI doesn't scaffold correct .env file based on -d option

To Reproduce
Steps to reproduce the behavior:

1. Create new Cube.js app with cubejs create hello-world -d bigquery
2. Inspect .env
3. It has wrong credential placeholders for BigQuery.

Expected behavior
Correct environment variables placeholders.

Update normalizePivotConfig method

• Do not fail if x or y is not passed.
• Check to display all data, which has been requested.
• fillMissingDates should work even if x or y isn't passed.

Add dependency "@cubejs-backend/server-core" installs "@cubejs-backend"

When I run yarn add @cubejs-backend/server-core I get a @cubejs-backend in my node_modules folder.

I would expect the following modules instead:

• node_modules/@cubejs-backend/server-core
• node_modules/@cubejs-backend/api-gateway (dependency of server-core)
• node_modules/@cubejs-backend/query-orchestrator (dependency of server-core)
• node_modules/@cubejs-backend/schema-compiler (dependency of server-core)

Add feature to dynamically define measures based on SQL commands in query

Hi!

It would be nice if we could pass custom measures based on SQL commands with a query. This could look for example like this:

{
  "measures": ["statistics.count"],
  "additionalMeasures": {
    "id": "weekday",
    "sql": "DAYNAME(statistics.createdat)"
  },
  "dimensions": ["statistics.weekday"],
  "filters": [
    {
      "dimension": "statistics.key",
      "operator": "equals",
      "values": ["timeInRoom"]
    }
  ]
}

At the moment we can achieve the same by adding the sql based measures in the schema on the server. However, it would be easier and more flexible if the client could pass these custom measures along with the query.

Pivot, series name generation. Do not concat with measure name if it is only one measure

issue with schema generator for mongobi

1. Using mysql driver for MongoDB BI

For MongoDB BI generate schema file contains DB name as prefix in query which causes an issue.
For Example:
Schema file for TableOne generate SQL = > sql: SELECT * FROM DBANE_PREFIX.TableOne,
If DBANE_PREFIX is removed it works fine.

2. Using monogbi driver for MongoDB BI

Schema file still contains PREFIX however it seems server can not transform the schema. (May be mongobi handling needs to be done for rendering on UI ).
In this case error as below

Error: Error: Compile errors: TypeError: Cannot read property 'aliasName' of null at CubeToMetaTransformer.measureConfig (/home/kru/antphant/git/cube.js/p1/node_modules/@cubejs-backend/schema-compiler/compiler/CubeToMetaTransformer.js:105:31) at config.measures.R.compose.R.map (/home/kru/antphant/git/cube.js/p1/node_modules/@cubejs-backend/schema-compiler/compiler/CubeToMetaTransformer.js:32:40) at _map (/home/kru/antphant/git/cube.js/p1/node_modules/@cubejs-backend/schema-compiler/node_modules/ramda/src/internal/_map.js:6:19) at map (/home/kru/antphant/git/cube.js/p1/node_modules/@cubejs-backend/schema-compiler/node_modules/ramda/src/map.js:57:14) at /home/kru/antphant/git/cube.js/p1/node_modules/@cubejs-backend/schema-compiler/node_modules/ramda/src/internal/_dispatchable.js:39:15 at /home/kru/antphant/git/cube.js/p1/node_modules/@cubejs-backend/schema-compiler/node_modules/ramda/src/internal/_curry2.js:20:46 at f1 (/home/kru/antphant/git/cube.js/p1/node_modules/@cubejs-backend/schema-compiler/node_modules/ramda/src/internal/_curry1.js:17:17) at /home/kru/antphant/git/cube.js/p1/node_modules/@cubejs-backend/schema-compiler/node_modules/ramda/src/internal/_pipe.js:3:14 at /home/kru/antphant/git/cube.js/p1/node_modules/@cubejs-backend/schema-compiler/node_modules/ramda/src/internal/_arity.js:5:45 at CubeToMetaTransformer.transform (/home/kru/antphant/git/cube.js/p1/node_modules/@cubejs-backend/schema-compiler/compiler/CubeToMetaTransformer.js:35:10)

Introduce Query class in ResultSet

We need to introduce Query class, which instance should returned by query getter in ResultSet. Query class should provide a set of convenient methods to access information about the query, such as hasDimensions.

ResourceRequest timeout on db connection problems

There should be test connection phase before any db operation.

Check if date in dateRange is invalid or dateRange itself is invalid

See dateRange in the docs for Query Format: https://cube.dev/docs/query-format#time-dimensions-format

Make API path namespace configurable

In the official blog post Node Express Analytics Dashboard with Cube.js it's mentioned that
Cube.js is

mounted into the /cubejs-api/v1/ path namespace. But you can change it and a lot of other things by passing the configuration object to the CubejsServerCore.create() method.

In the API Gateway the path seems to be hardcoded (see L196, L235 and L248). Nevertheless, I like the idea to have it configurable.

Fix dependency warning in schema compiler (joi)

I get the following warning for @cubejs-backend/server-core: @cubejs-backend/schema-compiler > [email protected]: This version is no longer maintained. Please upgrade to the latest version.

Make `.query` method a getter for for ResultSet

Unify how time filters work with different date and time formats

Create an issue and pull request templates

GitHub now supports issue templates via a markdown file called issue_template.md in the root or .github/ folder of your repo!

There is also added support for pull request templates via a markdown file called pull_request_template.md in the root or .github/ folder of your repo!

• Example: https://github.com/OSBI/saiku/blob/development/.github/ISSUE_TEMPLATE.md
• About: https://github.com/blog/2111-issue-and-pull-request-templates

ReferenceError: self is not defined - Issue with whatwg-fetch when using the core lib with Node.js

The core library is not compatible with Node.js because of the whatwg-fetch module. I'd suggest moving to an isomorphic alternative that works with both browsers and server.

Cube.js Error while connecting to postgres db

Hi,

I m trying to connect to postgres db (which comes with Tableau installation). I am able to connect to this postgres db from tableau whereas when I tried to establish connectivity from cubejs I got below error message
Error: connect ECONNREFUSED XXX.XX.XXX.XX:5432 at TCPConnectWrap.afterConnect [as oncomplete] (net.js:1113:14)

Please note the tableau server which has postgres db is SSL enabled. I m not sure whether that is the problem.

Please help fix this issue. I m facing same issue while connecting to mongoDB via mongoDB BI Connector

Thanks
Murugan

ResultSet.chartPivot keys issue

It looks like ResultSet.chartPivot uses weird combination of keys and titles as keys Object returning by chartPivot.

// What is returned by resultSet.seriesNames()
[
  {title: "Reports, Events Any Event - Total", key: "Reports:Events.anyEvent"},
  {title: "Dashboard, Events Any Event - Total", key: "Dashboard:Events.anyEvent"},
  {title: "∅, Events Any Event - Total", key: "∅:Events.anyEvent"}
]

// What is returned by resultSet.chartPivot()
[
{category: "2019-02-05T00:00:00.000", x: "2019-02-05T00:00:00.000", ∅, Events.anyEvent: 0, Reports, Events.anyEvent: 0, Dashboard, Events.anyEvent: 0, …},
{category: "2019-02-06T00:00:00.000", x: "2019-02-06T00:00:00.000", ∅, Events.anyEvent: 0, Reports, Events.anyEvent: 0, Dashboard, Events.anyEvent: 0, …},
{category: "2019-02-07T00:00:00.000", x: "2019-02-07T00:00:00.000", ∅, Events.anyEvent: 0, Reports, Events.anyEvent: 0, Dashboard, Events.anyEvent: 0, …},
{category: "2019-02-08T00:00:00.000", x: "2019-02-08T00:00:00.000", ∅, Events.anyEvent: 0, Reports, Events.anyEvent: 0, Dashboard, Events.anyEvent: 0, …}
...
]

The key in the chartPivot returning object is Reports, Events.anyEvent which is neither key nor title from seriesNames. I'd suggest using keys.

Import cubejs from '@cubejs-client/core'; error

Hi all,

I'm trying the VanillaJS Javascript integration (inside of an Angular component built with Angular CLI):

import cubejs from '@cubejs-client/core';
import Chart from 'chart.js';
import chartjsConfig from './toChartjsData';

@Component({
  selector: 'home',
  templateUrl: './home.html'
})
export class HomeComponent implements OnInit {

  public ngOnInit() {

    const cubejsApi = cubejs('dd44575826bc7113dbbbd06e7f490a5030f42c630da052913499956b6c20a993a9c72655f0c1d3921bad8b1047b411ed8f426d6342d526e082176aadc41a3e49');

    const resultSet = await cubejsApi.load({
       measures: ['Stories.count'],
       timeDimensions: [{
         dimension: 'Stories.time',
         dateRange: ['2015-01-01', '2015-12-31'],
         granularity: 'month'
       }]
    });

    const context = document.getElementById('myChart');
    new Chart(context, chartjsConfig(resultSet));
  }
}

But at cubejs(...) I get the following error: Uncaught Error: Uncaught (in promise): TypeError: core_2.default is not a function

When I change the import to import * as cubejs from '@cubejs-client/core'; it's fine.

Support for multiple databases & Database name specification at runtime

Is your feature request related to a problem? Please describe.
We provide analytics to healthcare professionals. Due to the sensitive nature of our clients' data, we have opted to separate each of our clients' data into separate database schemas. Unfortunately, even though all of our schemas have the same structure, Cube.js's single-database model prevents us from using Cube.js as an analytics engine on top of our data warehouses.

Describe the solution you'd like
Add the ability to specify multiple data sources in the configuration of the Cube.js service, and add the ability to specify which database schema to use at query time.

Suggested implementation

1. Due to the flat structure of .env files, it would be hard to specify multiple databases without custom names. Instead, it is proposed to use yaml environment files with the following structure:

 api_secret: secret
 dbs: 
   - alias: <YOUR_DB_ALIAS_NAME>
      host: <YOUR_DB_HOST_HERE>
      name: <YOUR_DB_NAME_HERE>
      user: <YOUR_DB_USER_HERE>
      password: <YOUR_DB_PASS_HERE>
      type: postgres

Here alias serves for easier identification for the step 3.

2. If host, name, type & user are the same for any two dbs, group these dbs into a singular connection. Otherwise deploy one connection per individual database.

3. At query time, allow the use of the alias specified in step 1 to specify the database to interact with.

{
  database: '<YOUR_DB_ALIAS_HERE>'
  measures: ['Stories.count'],
  dimensions: ['Stories.category'],
  filters: [{
    dimension: 'Stories.isDraft',
    operator: 'equals',
    values: ['No']
  }],
  timeDimensions: [{
    dimension: 'Stories.time',
    dateRange: ['2015-01-01', '2015-12-31'],
    granularity: 'month'
  }],
  limit: 100
}

which would translate to picking the right connection + adding a USE database statement:

USE '<YOUR_DB_ALIAS_HERE>';

-- SQL equivalent query

4. (Optional): The next natural evolution of this idea is to allow to define multiple Cube.js schema and assign them to different database aliases.

// single database setup
cube(`Users`, {
  sql: `SELECT * FROM users`
});

// multiple database setup
cube(`<YOUR_DB_ALIAS_HERE>`, `Users`, {
  sql: `SELECT * FROM users`
});

Describe alternatives you've considered
Create one Cube.js serverless deployment per database schema (which means managing 10s-100s of deployments) + HIPAA compliance requires us to have our PHI data within a VPC thereby greatly reducing the performance of lambdas due to the extended warming up period.

I know this may seem like a niche use case at first, but I'm sure that the valid use-cases will grow along as you get to encounter more situations. This feature essentially federates all of these different database APIs into a single server, a bit a la GraphQL.

Let me know what you think :)

Have a great day ☀️

Cheers,

Philippe

Timeseries databases support

It would be cool if this project could add support for timeseries databases like graphite or influxDB

Change dimension in the filter object in query to member

Name dimension is misleading, since filter could be applied to both measures and dimensions. Support dimension key for the back compatibility.

QueryRenderer: fix prop types warnings

Self reference guard in schema compiler

Handle RangeError: Maximum call stack size exceeded for cases like

   foo: {
      sql: `${foo}`,
      type: `string`
    },

ResultSet pivot methods documentation

How to define `cube()` in Typescript

Hello I am trying to define a cube schema in a typescript based project but I cannot just define cube() files in and also cannot define a .js files.

Any documentation or support for TypeScript?

Thanks!

Add option "year" for granularity in timeDimensions

Hi cube.js dev team,

It would be great if the granularity in timeDimensions of a query could also be set to "year". We need that to show a multidimensional bar chart grouped by years.

Thank you!

Correctly handle queries to rollups without date ranges

Describe the bug
Rollups pre-aggregations require the dateRange to be passed. Currently, if it is not passed, the backend returns 500 internal error, but it should be 400 error with a clear message of the problem.

To Reproduce
Steps to reproduce the behavior:

1. Create a rollup pre-aggregations like this

  preAggregations: {
   main: {
     type: `rollup`,
     measureReferences: [amount],
     timeDimensionReference: createdAt,
     dimensionReferences: [Users.city],
     granularity: `day`,
     partitionGranularity: `day`
   }
  }

2. Send a query without date range

{"query":"{\"measures\":[\"Orders.amount\"],\"timeDimensions\":[{\"dimension\":\"Orders.createdAt\",\"granularity\":\"day\"}],\"dimensions\":[\"Users.city\"],\"filters\":[]}"

Expected behavior
The server should handle this case and return a User Error with a clear message.

Version:
0.5

Microsoft SQL Server Database Support

It can help connect to Azure SQL Database, Azure MS-SQL Server, etc.

Athena: SYNTAX_ERROR: IN value and list items must be the same type

This error happens when you set a filter for a numeric (or other non-string) dimension or measure with multiple values. It looks like athena requires to cast all values in the IN statement to the type of dimension/measure.

Cassandra Driver: Create cubejs Hello world project with Cassandra

Describe the bug
Permission denied to build using node-gyp

What I did affter successfully installing cli: cubejs create hello-world -d cassandra

**- Installing DB driver dependencies

> [email protected] install /home/navs/CubeJS/hello-world/node_modules/java
> node-gyp rebuild

sh: 1: node-gyp: Permission denied
npm ERR! file sh
npm ERR! code ELIFECYCLE
npm ERR! errno ENOENT
npm ERR! syscall spawn
npm ERR! [email protected] install: `node-gyp rebuild`
npm ERR! spawn ENOENT
npm ERR!
npm ERR! Failed at the [email protected] install script.
npm ERR! This is probably not a problem with npm. There is likely additional logging output above.

npm ERR! A complete log of this run can be found in:
npm ERR!     /root/.npm/_logs/2019-04-16T12_10_36_327Z-debug.log

Cube.js Error ---------------------------------------

Error: npm install --save @cubejs-backend/jdbc-driver node-java-maven failed with exit code 1
    at ChildProcess.child.on.code (/root/.nvm/versions/node/v10.13.0/lib/node_modules/cubejs-cli/cubejsCli.js:31:16)
    at ChildProcess.emit (events.js:182:13)
    at maybeClose (internal/child_process.js:962:16)
    at Process.ChildProcess._handle.onexit (internal/child_process.js:251:5)**

Version:
CubeJS version: 0.7.0

issue with Time Field for mongoBI

While using mysql driver for mongoBI

Time fields are mapped correctly in "time" field. But when selected any of them this error is thorwn

Error: Error: scalar function 'convert_tz' is not supported

mongobi issue with parsing schema file with nested fields (Scaffolding bug)

Consider MongoDB Document
{ "_id" : ObjectId("5c8a215c46e0fb0001055061"), "details" : { "name" : "asdfsadf", "status" : "asdfdsf", "severity" : "asdfsadf" }, "projectId" : "5c6eb78046e0fb0001ee0b32", . . . }

Schema file generate by cubejs contains measure names with "." for nested fields as below

cube(Flow, { sql: SELECT * FROM 1224.flow`,

joins: {

},

measures: {
count: {
type: count,
drillMembers: [createdbyuser, details.name, projectid, createddate, modifieddate]
}
},

dimensions: {
details.owner: {
sql: details.owner,
type: string
},

createdbyuser: {
  sql: `createdByUser`,
  type: `string`
},

**details.description**: {
  sql: `details.description`,
  type: `string`
},

**details.name**: {
  sql: `details.name`,
  type: `string`
},

**details.severity**: {
  sql: `details.severity`,
  type: `string`
},

.
.
.
`

This throws error while parsing

Recompiling schema: {"version":"default_schema_version_bc194c45a5c745c8a93e4b0515c9dcc4"}
Internal Server Error: {"error":"Error: Compile errors:\nSyntaxError: Unexpected token, expected , (16:11)\n at Parser.pp$5.raise (/home/kru/antphant/git/poc/node_modules/babylon/lib/index.js:4454:13)\n at Parser.pp.unexpected (/home/kru/antphant/git/poc/node_modules/babylon/lib/index.js:1761:8)\n at Parser.pp.expect (/home/kru/antphant/git/poc/node_modules/babylon/lib/index.js:1749:33)\n at Parser.pp$3.parseObj (/home/kru/antphant/git/poc/node_modules/babylon/lib/index.js:3978:12)\n at Parser.pp$3.parseExprAtom (/home/kru/antphant/git/poc/node_modules/babylon/lib/index.js:3719:19)\n at Parser.pp$3.parseExprSubscripts (/home/kru/antphant/git/poc/node_modules/babylon/lib/index.js:3494:19)\n at Parser.pp$3.parseMaybeUnary (/home/kru/antphant/git/poc/node_modules/babylon/lib/index.js:3474:19)\n at Parser.pp$3.parseExprOps (/home/kru/antphant/git/poc/node_modules/babylon/lib/index.js:3404:19)\n at Parser.pp$3.parseMaybeConditional (/home/kru/antphant/git/poc/node_modules/babylon/lib/index.js:3381:19)\n at Parser.pp$3.parseMaybeAssign (/home/kru/antphant/git/poc/node_modules/babylon/lib/index.js:3344:19)\n at ErrorReporter.throwIfAny (/home/kru/antphant/git/poc/node_modules/@cubejs-backend/schema-compiler/compiler/DataSchemaCompiler.js:42:13)\n at DataSchemaCompiler.throwIfAnyErrors (/home/kru/antphant/git/poc/node_modules/@cubejs-backend/schema-compiler/compiler/DataSchemaCompiler.js:149:23)\n at repository.dataSchemaFiles.then.then (/home/kru/antphant/git/poc/node_modules/@cubejs-backend/schema-compiler/compiler/DataSchemaCompiler.js:99:14)","authInfo":{"iat":1553003564,"exp":1553089964}}

Add a method to ResultSet to retrieve the format for query member