I did this
A command line that uses globbing for the upload file, with a specified second URL that does a plain GET, quite mysteriously fetches the second URL the same number of times as it made PUT requests for the first URL. In this example three!
curl -T 'upload{1,2,3}' http://localhost/upload --output first http://localhost/download --output second-file
Note: doing upload globbing like this and storing the output in a file makes the same file get overwritten three times so only the third and final upload results are left after the invoke. I'm thinking we can perhaps enhance this by expanding the # [num] support that can extract parts from the URL glob to be able to extract something from the upload glob as well. But that's separate.
I expected the following
This command line should do three PUTs and one GET.
curl/libcurl version
curl git master
operating system
independent
I did this
A command line that uses globbing for the upload file, with a specified second URL that does a plain GET, quite mysteriously fetches the second URL the same number of times as it made PUT requests for the first URL. In this example three!
Note: doing upload globbing like this and storing the output in a file makes the same file get overwritten three times so only the third and final upload results are left after the invoke. I'm thinking we can perhaps enhance this by expanding the
# [num]support that can extract parts from the URL glob to be able to extract something from the upload glob as well. But that's separate.I expected the following
This command line should do three PUTs and one GET.
curl/libcurl version
curl git master
operating system
independent