Currently we try fetching metadata for all supported builds from production TaskCluster.
We should mirror the requests made and commit them in so we can mock them with the built-in HTTP server implementation. We should also add stripped down data files (zip/tar.bz2/dmg) with 0-length contents to test extraction.
Currently we try fetching metadata for all supported builds from production TaskCluster.
We should mirror the requests made and commit them in so we can mock them with the built-in HTTP server implementation. We should also add stripped down data files (zip/tar.bz2/dmg) with 0-length contents to test extraction.