Options to access files in the community edition

[adihc]

As of now DBFS option is disabled in the Databricks community edition. 

What are the other ways to use file in the Databricks notebooks for learning? When I go to catalog it show default option only with the AWS S3. 

Is it the only option to access the files in the community edition now? For my learning I want to upload few types of file. Ex: CSV, JSON, Parquet etc... 

[gchandra]

The workaround will be to Mount the S3 bucket and use that as your source.

access_key = ""

secret_key = ""

encoded_secret_key = secret_key.replace("/", "%2F")

aws_bucket_name = "yourawsbucketname/"

mount_name = "youraliasmountname"

# #dbutils.fs.unmountf"/mnt/{mount_name}")

dbutils.fs.mount(f"s3a://{access_key}:{encoded_secret_key}@{aws_bucket_name}", f"/mnt/{mount_name}")

After mounting
goto next cell %fs ls /mnt/youraliasmountname

[Riporsa3011]

Any info on when the DBFS local file option will be back? It was the only option to keep studying without a paywall for me. Aprecciate your help.

[gchandra]

Signup for AWS Free https://aws.amazon.com/free/

You can use S3 buckets with IAM policies and use them with Databricks.

This is a good skill to have apart from using Databricks.

[Riporsa3011]

Unfortunately, the AWS Free requires Billing (cr***t card) information to proceed with the account, which I can't provide, since I don't have one. The DBFS option was working just fine a few days ago, do you anticipate this option coming back in the future? I'm sure it will help lots of other students.

[adihc]

Thanks for the answer @gchen. Apart from S3, there is no other option available in the community edition now?

[gchandra]

Try this technique

https://medium.com/@gchandra/databricks-how-to-load-data-from-google-drive-github-c98d6b34d1b5?sk=67...

[Riporsa3011]

This worked Ganesh, thanks a lot. For anyone having problems with uploading files, I recommend the Github pipe. Upload your files to github, set the repository as public, then following Ganesh's code should allow you to create dataframes with your own data in Databricks. Thanks again!

[gchandra]

Don't forget to accept the solution so others will benefit.

[gchandra · Accepted solution]

It's fixed. You can continue to use Upload.