-
Notifications
You must be signed in to change notification settings - Fork 6
Description
I often like for programs that show me their recommendation to also give me all the data for me to come to my own conclusion. It's conceivable you could have 50 servers effectively at the same load, within 1% of each other, but that the 50th place one load-wise had a ping value at half as much as others with a technically lesser load. So, I'd propose two changes.
I'd probably then use these changes to manually run speed tests on a couple possibilities.
Output in columns would be important here for quick skimming.
-v, --verbose Prints considered servers, their loads, and ping values if tested
This could always print the servers matching any given filters and their loads, in addition to only showing the one(s) it recommends, probably sorted by load. Without ping tests, depending on the filters given, if any, this could give an awful lot of output.
In combination with simultaneous or linear ping options, this could also show the ping value for each, limited to the number of servers to ping, probably sorted by ping.
I'd probably sort by worst to best, so no scrolling is needed if there's lots of results to see the best ones. (Again, I'd think sorted by load without a ping test, and by ping with a ping test.)
Also, if I'm correct that -a, --amount <AMOUNT> lets you adjust the number of servers for a ping test, accepting ALL for <AMOUNT> might be nice. (Although giving 10000 basically does this.)
Interestingly, -s and -s -a 100 give me the same server, but -s -a 10000 does give me a different one.
Could be worth having this be incompatible with -p, at least if given a value too high.