Xtract Universal 6.0.0

Breaking changes

• Web API
  • /log?req_type=all has been removed
    • To fetch the full logs for an extraction the prefered workflow is:
      • Get the timestamp for the extraction from the xu.exe output or the header of the HTTP response.
      • Request the run log for the extraction with log?req_type=extraction&name=extractionname&timestamp=extractionTimestamp
      • If that does not hold the desired information:
        • There now is a ‘WebLog’ column when listing the extractions runs with log?req_type=extraction&name=extractionname
        • Use this timestamp to request the corresponding server log for the extraction with log?req_type=server&timestamp=serverTimestamp
  • /log?req_type=server now yields the worker instead of listener log
  • /metadata?name=extractionName has been removed
    • Use json endpoint instead
  • /parameters?name=extractionName has been removed
    • Use json endpoint instead
  • /clearResultCache has been removed
  • /CurrentVersion does now serve a json payload
    • it also does only serve one version, since the kernel dependency was dropped

We aim to provide a more consistent and useful Web API as part of upcoming releases.

Deprecations

• The ‘clearBuffer’ parameter for extraction requests was renamed to ‘ignoreCache’ and does just that instead of purging the whole cache
  • ‘clearBuffer’ will be removed in a future release

Other

• Web server, result cache and the engine which runs the extractions (runtime) were rewritten.
• When running extractions the web server now responds with an HTTP error code and an error message instead of a 200 without any data or information if the SAP system yielded an error or preparing the destination failed
  • an error while writing the data to the destination will still only show up in the logs
• Web server authentication: HTTPS + Designer Read access is now the same as HTTPS + Anonymous if Designer access is not restricted
  • Before one had to supply (any) Basic credentials but they were not checked.