As NCTIR looked towards upgrading their FME Server from the 2017 to 2018 version, it wasn’t only the versions that where changing, but the entire platform. While understanding that a number of major components had been installed, it was hard to know what python libraries where being used in FME workspaces and what libraries were needed in new environment. Besides knowing the Python libraries that were in use, it also gave the NCTIR GIS team the opportunity to look at what python had been implemented in the past and what things could be changed in future FME process.
The key point of the process was understand what Python libraries would be needed in the new FME Server. Abley built a process to work with the FME Server Rest API to pull all the workspace details and then read each transformer within the workspaces to pull out any Python processes. These processes were then automatically scanned for imports of libraries.
The outcome is a process that can read all the FME Server workspaces and pull all Python libraries out of both Python-based transformers and start and stop Python scripts. It has allowed for the NCTIR GIS team to implement a thorough upgrade process, as well as highlighting processes that may require changes. As this is useful knowledge on an ongoing basis, we have built in into our Audit Dashboard capability.