Oran Agra

not sure if it's as ugly as MIGRATE, or maybe uglier? in migrate we add an empty string as key, here you suggest we choose a flavor by checking the...

well in theory it could also be `z[m][rev]rand[withscore]`, but i'd rather not add so many commands, and instead i'd rather the `rev` and `withscore` be arguments. so maybe just the...

Looking at this again, trying to decide if it justifies special commands (we can add hundreds of them if we open that door), i wanna ask again why / if...

> And I remember that when adding GETEX/GETDEL commands I asked the same question, but you said new commands are better(https://github.com/redis/redis/pull/8327#issuecomment-762801080), I don't know why you have different opinions on...

@itamarhaber @yoav-steinberg i don't know if we can / should chuck SUBSTR (could break someone's app without a real good reason). i agree it's odd that it's in the redis...

i'd argue that even if we chuck it, redis.io should still document it, since it is being used as the documentation for old versions too.

you can't get that info from old redis versions.. it's simply not provided. what you can do is compile with a new redismodule.h and use RMAPI_FUNC_SUPPORTED to check if that...

I think this might be looking at redis passwords the wrong way (not the first time btw, since they're misnamed). the auth password in redis is more of a shared...

disclaimer: i didn't look at the code yet, i hope i'm not making any false assumptions. i don't recall saying i have an idea (maybe i did and i forgot?),...

@zcaudate yes, this is being worked on (see #8180), but it may only become a reality in redis 7.0.