Add a helper class for using the Databricks ODBC driver #615
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
hadley
reviewed
Nov 8, 2023
|
Updated to use a new S4 class instead of a helper function. |
atheriel
changed the title
bschwedler
reviewed
Nov 8, 2023
| user_agent <- paste0("r-odbc/", utils::packageVersion("odbc")) | ||
| if (nchar(Sys.getenv("SPARK_CONNECT_USER_AGENT")) != 0) { | ||
| # Respect the existing user-agent if present. Normally we'd append, but the | ||
| # Databricks ODBC driver does not yet support space-separated entries in |
There was a problem hiding this comment.
Well that is most unfortunate.
Comment on lines
+151
to
+166
| default_path <- "" | ||
| if (Sys.info()["sysname"] == "Linux") { | ||
| default_path <- "/opt/simba/spark/lib/64/libsparkodbc_sb64.so" | ||
| } else if (Sys.info()["sysname"] == "Darwin") { | ||
| default_path <- "/Library/simba/spark/lib/libsparkodbc_sbu.dylib" | ||
| } | ||
|
|
||
| if (file.exists(default_path)) default_path else "Simba Spark ODBC Driver" | ||
| } |
There was a problem hiding this comment.
Yay for sensible defaults.
For linux, would we want to default to /opt/rstudio-drivers/spark/bin/lib/libsparkodbc_sb64.so and fallback to the /opt/simba/spark location?
There was a problem hiding this comment.
Is that the expected install location for the Pro driver? That could make sense.
There was a problem hiding this comment.
Yes, that is the expected location of the driver. I don't quite understand bin/lib in the path, but it follows the convention of where the other drivers are installed.
hadley
approved these changes
Nov 10, 2023
|
I'm happy to merge once the build falures are resolved. |
The official Databricks ODBC driver supports a number of authentication
methods but these are difficult to use correctly and require users to
pass them manually rather than picking up on ambient credentials as
other Databricks SDKs do.
This commit adds a high-level helper class that implements (a subset of)
the Databricks client unified authentication model [0] in its
dbConnect() method and should make connection code that runs on
Databricks-aware environments work automatically without manual
configuration.
Compare the interface from this commit:
DBI::dbConnect(
odbc::databricks(),
http_path = "sql/protocolv1/o/4425955464597947/1026-023828-vn51jugj"
)
with filling all of the required parameters out manually:
DBI::dbConnect(
odbc::odbc(),
driver = "/opt/simba/spark/lib/64/libsparkodbc_sb64.so",
host = gsub("https://", "", Sys.getenv("DATABRICKS_HOST")),
HTTPPath = "sql/protocolv1/o/4425955464597947/1026-023828-vn51jugj"
ThriftTransport = 2,
UserAgentEntry = Sys.getenv("SPARK_CONNECT_USER_AGENT"),
port = 443,
Protocol = "https",
SSL = 1,
AuthMech = 11,
Auth_Flow = 0,
Auth_AccessToken = "<your OAuth token>"
)
[0]: https://docs.databricks.com/en/dev-tools/auth.html#databricks-client-unified-authentication
Signed-off-by: Aaron Jacobs <[email protected]>
|
CI issues are fixed and |
|
Thanks! |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
The official Databricks ODBC driver supports a number of authentication methods but these are difficult to use correctly and require users to pass them manually rather than picking up on ambient credentials as other Databricks SDKs do.
This commit adds a high-level helper class that implements (a subset of) the Databricks client unified authentication model in its
dbConnect()method and should make connection code that runs on Databricks-aware environments work automatically without manual configuration.Compare the interface from this commit:
with filling all of the required parameters out manually: