Images are now grouped by archives prior to processing, to avoid
downloading archives several times.
Error handling was added for several different scenarios and may seem to be a bit all over the place:
no longer existing or accessible archives (will lead to a (FAIL) notice after the archive in the program output)
updated archives where image names are no longer matching (will result in an Error: […] output including the image name that could not be found)
aborted or failed downloads, and I considered two sub-scenarios, one of which I was not able to reproduce: a) wget returns an error code (problem with the FTP server, download will be re-tried), b) wget returns no error code but the downloaded archive is not complete, which will lead to EOF errors during extraction – I'm not sure if or how often this happens, but I included error handling for this case anyway.
If downloading a package fails for X (-r parameter) times, the program execution will be halted because it is assumed that something is wrong with the FTP server (user was locked out, temporary downtime etc.). The other errors described above will only produce a log event because they are considered to be permanent.
Images are now grouped by archives prior to processing, to avoid downloading archives several times.
Error handling was added for several different scenarios and may seem to be a bit all over the place:
(FAIL)notice after the archive in the program output)Error: […]output including the image name that could not be found)If downloading a package fails for X (
-rparameter) times, the program execution will be halted because it is assumed that something is wrong with the FTP server (user was locked out, temporary downtime etc.). The other errors described above will only produce a log event because they are considered to be permanent.