Testing and Vetting Web Data Connectors
Web Data Connectors (WDCs) contain JavaScript that typically connects to data on another site. Because of this, you should test and vet web data connectors before users use them as data sources for a workbook, and before you use them with Tableau Server.
This topic includes some suggestions for testing and vetting web data connectors.
Examine the source
The code in a web data connector is in JavaScript, so you can open the file (and any external files that the connector uses) and examine the source code.
Many connectors reference external JavaScript libraries, such as the jQuery library or API libraries for third parties. Validate
that the URL for external libraries points to a trusted location for the library. For example, if the connector
references the jQuery library, make sure that the library is on a site that is considered standard and safe. If it is practical
for you to change the source code of the connector, use HTTPS protocol (https://) to reference external
libraries (if the source site supports HTTPS) to help verify the site's authenticity.
To the extent possible, make sure you understand what the code is doing. In particular, try to understand how the code is constructing requests to external sites, and what information is being sent in the request.
Note: Experienced JavaScript programmers often compress (minify) their code to reduce the size of the code for download. Dense blocks of code that use cryptic function and variable names are not uncommon. While this can make it more difficult to examine the code, it is not a sign that the code was written to be deliberately difficult to understand.
Test the web data connector in an isolated environment
If possible, test the web data connector in an environment that is isolated from your production environment and from user computers. For example, add a web data connector to a safe list on a test computer or virtual machine that's running a version of Tableau Server that is not used for production.
Monitor the traffic created by the web data connector
When you test a web data connector, use a tool like Fiddler(Link opens in a new window), Charles HTTP proxy(Link opens in a new window), or Wireshark(Link opens in a new window) to examine the requests and responses that the connector makes. Make sure that you understand what sites the connector makes requests to and what content the connector is requesting. Similarly, examine the responses and their content to be sure that the connector is not reading data or code that is not directly related to the connector's purpose.
Test the performance and resource usage of the web data connector
When you test a web data connector, use tools to monitor its CPU and memory usage. Remember that the web data connector will run on Tableau Server, which is an environment in which many processes are already running. You want to make sure that when the connector fetches data, the connector does not have an undue impact on server performance.
Check whether the connector writes to disk. If it does, check how much disk space it occupies, and examine the output to make sure you understand what it's writing and why.