With the runner spawning the handler process, debugging test failures is currently a bit involved. Since handlers take input from stdin, they can very easily be manually tested on command line with echo <request> | <handler-bin>.
I propose adding -v (verbose on failure) and -vv (always verbose) args (potentially also a --verbosity={0,1,2}), where:
-v will print the full request and response for each failed test
-vv will print the full request and response for each test
This way, the failed tests can easily be replicated manually with a simple copy-paste.
With the runner spawning the handler process, debugging test failures is currently a bit involved. Since handlers take input from
stdin, they can very easily be manually tested on command line withecho <request> | <handler-bin>.I propose adding
-v(verbose on failure) and-vv(always verbose) args (potentially also a--verbosity={0,1,2}), where:-vwill print the full request and response for each failed test-vvwill print the full request and response for each testThis way, the failed tests can easily be replicated manually with a simple copy-paste.