| FazBrowse GitHub Viewer | Trending | | Home |
| Tools: [Download Repo ZIP] [Original HTTPS Page] |
|
E2E test should get updated to cover write features to cosmos db |
Sorry, something went wrong.
|
Can we add test cases to show this in working? |
Sorry, something went wrong.
|
The test failures are irrelevant hence merging the code. |
Sorry, something went wrong.
* WIP * Remove ununsed import * Rename collection to container * Typo * Typo * Use Sink as offline store * Add CI test case * CI test fix * SQL and CosmosDb doc * CosmosDb test * Resolve comments * Resolve comments * Environment variables in upper case * Get env from secrets * Databricks env
| Back | FazBrowse Home | New Git URL |
This PR is the client part corresponding to PR #475 , it contains Python code to generate config/jobparameters to enable the user to use CosmosDb as Data Source and Online/Offline storage.
It adds "CosmosDbSource" and "CosmosDbSink" in the client lib, the former is to use CosmosDb as data source, the latter is to use CosmosDb as online storage.
There are still some open questions about using CosmosDb as offline storage, the main one is, CosmosDb requires every row/doc to have a unique key, but in the feature joining job, we still don't know what should be used as a meaningful key.
For this reason, using CosmosDb as an offline store is not opened in the public interface, even the function is usable internally.