Delta Share Recipient

[lynchdhl]

Is it possible to catalog a delta share as a recipient in Databricks Free?

[edonaire]

not in Databricks Free Edition... 😞

To catalog a Delta Share (Databricks-to-Databricks), you need a Unity Catalog-enabled workspace with permissions to use providers and create catalogs from shares.

Those capabilities are not fully available in the Free tier, which has limited governance and sharing features.

In Free Edition, you can still consume shared data using the open Delta Sharing protocol (token-based), but you won’t be able to mount it as a catalog in your workspace.

So it’s more “external access” than native catalog integration.

[lynchdhl]

Thanks for the response. Any docs you can point me to on how I can connect to an external delta share? I have received a credentials file with a bearer token, 

[edonaire]

If you already have the credentials file with a bearer token, you're pretty much there! 😛

On Free Edition you can't register the share in Unity Catalog, but you can still access the data using the Delta Sharing protocol directly. Two ways to do it:

1. Python (easiest for exploration)

Install the client:

pip install delta-sharing

Then point it at your credentials file (.json or .share):

python

import delta_sharing
profile_file = "/path/to/your/credentials.file"

# List available tables
client = delta_sharing.SharingClient(profile_file)
print(client.list_all_tables())

# Load a table into pandas
df = delta_sharing.load_as_pandas(
    profile_file + "#share.schema.table"
)
df.head()

2. Spark (if your environment supports it)

python

spark.read.format("deltaSharing") \    .load("/path/to/credentials.file#share.schema.table") \    .show()

A few Free Edition limitations worth knowing:

• No Unity Catalog integration — you can't register the share as a catalog
• No persistent metadata layer for the shared data
• Access is session-based, straight from the credentials file

So the workflow is more "connect and read" than "govern and catalog."

---

Full docs covering both Python and Spark usage: https://docs.delta.io/latest/delta-sharing.html

If you let me know which environment you're on (local Python, Databricks Free, etc.), I can tailor the example a bit more.

[Ashwin_DSA · Accepted solution]

Hi @lynchdhl,

Yes. You can use Delta Sharing as a recipient on Databricks Free Edition, and you can catalog a share as long as your Free workspace is Unity Catalog-enabled. Functionally, this works the same as on paid workspaces. Delta Sharing is generally supported, with no difference between the trial and Free tiers.

Because you’ve received a credentials file with a bearer token, you’re using the open sharing flow. The high‑level steps in your Free workspace are:

1. Open Catalog Explorer → gear icon → Delta Sharing (or Share → Delta Sharing).
2. On Shared with me, click Import data, give the provider a name (no spaces), and upload the credential file you received.
3. On the Shares tab for that provider, click Create catalog on the share you want to expose; this creates a UC catalog backed by the shared data.
4. Grant permissions and query it like any other UC catalog, e.g. SELECT * FROM your_share_catalog.schema.table;

Unfortunately, when I tried to test this in my free edition, it failed for a different reason. Having checked internally, it looks like it could be a backend issue. So, I would recommend you try the above with your login once, but if, for some reason, you don’t have UC or cannot use Catalog Explorer, you can still read from the share in a notebook using the delta-sharing connector and the same credential file (Python / Scala examples in the docs).

Here are some useful links. 

Access data shared with you using Delta Sharing (for recipients)  HERE

Step‑by‑step for credential file + bearer token (open sharing): HERE

These instructions apply to Free Edition the same way they do to standard workspaces, as long as the workspace has Unity Catalog enabled.

If this answer resolves your question, could you mark it as “Accept as Solution”? That helps other users quickly find the correct fix.