I'm trying to build the architecture for a Power BI custom data connector, that will consume a GraphQL API. I'm currently thinking about 2 approaches;
Creating a wrapper REST layer, to avoid having to deal with GraphQl in the custom data connector
Consuming the GraphQl API directly in the custom data connector.
Is option 2 possible at all, i.e. can you make GraphQl requests with the M language?
Related
I am trying to use REST API to connect data into Tableau Desktop. Which option do I select when choosing the datasource when opening Tableau Desktop? I have tested the API in Postman, and it successfully connects.
The REST API is not a data source that can be accessed from Tableau Desktop. To use the REST API you have to export the data from an API call to csv or other flat file. Then you can connect Tableau Desktop to that file. You can automate this using Python. You can also export directly from Postman. At the top of your results after you run your GET, download the file from here
I do strongly recommend you do this all in Python rather than Postman though so you can consume the json file much more easily with pandas. I would also recommend looking at tableauserverclient within Python rather than using the API directly. tableauserverclient is more limited than the API but will likely have the functionality you need.
Tableau cannot connect to REST API directly.
One way is to use Web Data Connector, BUT you have to build a web page for each REST API and paste the web page link in Web Data Connector.
Alternatively, you can use ETL tools that support extracting data from API and loading it to Tableau. (I personally recommend Acho. Its API connector is very useful)
I am new to GraphQl. I would like to implement the Persistent query support for my API. I want to know whether we can achieve it using GraphQL Java. We are not using apolo or any other third party libraries.
Would like to know , how can i implement this support in GraphQL Java .
Lets say client send me a query , i would like to persist it and needs to return a hash for it
Then clients can call my API using this Persisted query hash.
Thanks in Advanace
I have azure web api app with number of web apis of type post/get to get data. these apis are having basic authentication (bearer token)
I want to use 2-3 APIs as a data source in power bi, so that I can use these APIs JSON response as a table.
In Power BI Desttop > Get Data > Web > add API URL with sample GET API method name, will work but then it is asking for authentication.
If I tried to put API Key, then it is saying: A web API key can only be specified when a web API key name is provided.
How to connect web api using authentication in power bi as a data source.
PowerBI web seems to currently have little or no support for connecting to data via webapi, or web queries generally.
PowerBI Desktop allows you to get data from the web. I have been able to access data from webapis managed by Azure API Management in two different ways:
Use Advanced rather than Basic in the From Web dialog box. Put the webapi query in the url, but add a header with name Ocp-Apim-Subscription-Key and put in as value the value of the api key.
Alternatively, instead of putting the api key in the header, with APIs managed by Azure API management, it seems you can add them to as a parameter to the apiquery, subscription-key=PutYourApiKey here.
However, I have found that although you can publish your data source or report to an online workspace, PowerBI online will give an error ("credentials not recognized" or similar) when you try and refresh the data! So as mentioned above, PowerBI online seems currently (Nov 2020) useless for connecting to WebApi data.
I'm looking for advice choosing a back-end architecture for a web app. In the app, users upload tabular data from multiple files. Their data is then processed, aggregated and visualized. Data is private and each user has their own dashboard.
I believe Cube.js is an excellent choice for the dashboard, but I am wondering what back-end web framework I should integrate it with. I have experience of Django, but would use Express if it had significant advantages.
Thanks for any advice!
Cube.js is designed to run as a microservice. It means there's no requirement for you to use Node.js or Express as a stack for the rest of your backend implementation. However usually you'd need to have separate backend for storing data about users, reports and dashboards if those are dynamic. Cube.js server handles only analytic queries.
Cube.js dashboard templates are designed to work with GraphQL. So any backend web framework where there's a solid support of GraphQL will be a good choice. Most notable ones in Node.js world are apollo-server, Prisma, Hasura. Seems like Python has it's own champion as well: https://github.com/graphql-python/graphene.
I am trying to implement a webservice in Azure which will get the data through an APP via RESTful interface and the same data will be provided to Spark for analysis. Do I have a connector available for this. I do not want to use Kaffka or Cassendra or influxDB or any other extra component.