# Connections

We advise to use Python to send data using our API. To help you get started, we have some examples for common use cases.

# Watch files

When data is available by frequently exported file dumps, a file watcher can be used to trigger sending data to the Touchbase API for every file creation or modification. This concept is detailed in touchbase-watcher. The idea is as follows

  1. Define which directory to watch
  2. Define a directory in which jobs are stored. A job is a Python script.
  3. Execute a certain job once a certain file has changed.

More details are available in the documentation of the repository.

# SQL

Connecting to SQL using a Python script can be tricky due to required local drivers. This can be solved by using Docker. Proof-of-concepts for connecting to MSSQL and Oracle with Python and Docker are detailed in these two repositories: