It's not possible to cryptcheck a cache


#1

it’d be nice if I could cryptcheck a cache. even if all the command does is ignore the cache system entirely.

So right now I’m running a lot of commands
rclone copy -v --fast-list --drive-chunk-size 128M “localdirectory\localfile” “namecachecrypt:localdirectory”

then I run
rclone cryptcheck -v --fast-list --drive-chunk-size 128M “localdirectory\localfile” “namecrypt:localdirectory”

why? because running cryptcheck on a cache simply does not work at all ever.
now if it’s impossible to use a cache to run a cryptcheck, okay, fine, but i’d still like the command to know that. having to edit the name of my remote in all my command line commands is quite annoying.
there’s really no reason it has to work the way it currently does, since a cryptcheck makes 0 remote changes it doesn’t threaten the freshness of a cache at all.

Also while I’m at it. my cryptcheck usually uses --one-way in these situations, because I just want to check my most recent choices. I’d like --one-way as a flag to work for the copy command. It wouldn’t do anything, but why does it have to give me an error and refuse to run the command? The copy command is naturally always --one-way as such I’d like it to be the case that if you try to run the copy command with the --one-way flag the --one-way flag just has no effect. That means I can run rclone copy --one-way “source” “target” then run rclone cryptcheck --one-way “source” “target” and the commands are more similar and easier to type.

Personally I can’t see any harm in this, but, well, maybe I’m being shortsighted.


#2

It would be relatively easy for the crypcheck to unwrap the cache backend to look for a crypt backend, so check the underlying backend, rather than the cached version of it. Would that be useful?

I’m less keen on this change as the --one-way flag is a flag local to the cryptcheck command. To put it as a global flag would contribute to the cambrian explosion of flags in rclone which I’m trying to keep in check!


#3

Yes, that sounds like exactly what I’m asking for.

And yeah, I guess making the flag global when it literally doesn’t do anything in all those other commands would be bad.


#4

If you please make a new issue on github about the cryptcheck then we can take a look - thanks.