What kind of mouse is it? Upon log-in my USB keyboard doesn't work every time. Then I have to switch the workspace with mouse and then it works. Weirdest bug ever.

It's the trackpad built into the ThinkPad X1

I'll try switching workspaces with the mouse if it happens again 🙂 Thanks

(I use Manjaro+i3wm and this keyboard switch-off happens only with this keyboard)

(no (truly) external keyboard or mouse here..) Manjaro here too (and KDE plasma)

Actually it's exceptionally weird what it does. The keyboard works and Super key is actually working, but all other keys don't work 😄 But then I switch to a different i3wm tag (workspace) and it starts to work. Weeird. And I have to use mouse for this. So Super+Mouse click or Super+mouse wheel to switch workspace. And I use my i3bar in auto-hide mode, so the Super key is needed to show it

Hmmm. Computers!

I recall having some weird input issues when I enabled the power-saving features for my USB peripherals and hubs with powertop.

Does anyone know why would Thunderbird's calendar fail when importing an update to an event? It didn't fail on my last laptop but upon changing I can't make it work anymore.

I love KDE Plasma - the look and feel of the desktop environment, the features, the plasmoids and plug-ins - but I also hate how annoyingly buggy it is, specially when I disconnect and reconnect peripherals and when it comes back from sleep mode... I am almost replacing it, but unfortunately other envs do not come close...

On a t460 and fedora gnome I had to use the nub after the laptop woke up in order to use the trackpad buttons. Not sure if that would have any effect in your case. (The issue went away when I switched to sway)

that's interesting. what causes it?

Using next, conj and no transient https://github.com/clojure/clojure/blob/clojure-1.10.1/src/clj/clojure/core.clj#L1540

You could also provide an arity without a map which does a transducer like fusion of ops

But composing a bunch of TABA functions is hard 😄

TABA?

There And Back Again

Took me a minute to figure out the shape of the problem and an "accumulation" path

Which is the return path and not the call path like we're usually used to

I should add it to clj-fast

not going into clj-fast, will release a new small lib tomorrow

Not worth it or not suitable?

Very suitable, but this is going to be a tiny library focused only on select-keys and maybe rename-keys. The idea is to provide something smaller with limited scope for people who don't want to pull in all of clj-fast

I've really done this so many times it didn't take long to write, in the long run it would probably be better to provide a dedicated module

an http://ask.clojure.org issue is welcome

probably higher priority for select-keys than rename-keys

There's already an ask and a jira for select-keys, and I assume the core team's approach will be more conservative than mine. If you like my implementation you're hereby granted permission to take it as is or adapt it to your taste for core 🙂

https://ask.clojure.org/index.php/1913/use-transients-with-select-keys-if-possible @U050ECB92

Here you go https://clojurians.slack.com/archives/C06MAR553/p1671528873325009

It kinda depends on what your goal is. would you be accepting or rejecting whole commits, or would you also be applying commits with edits?

accepting commits that touch clojure files

without any edits outside of the commits?

correct

how many commits?

125 or so

I assume there would be conflicts if you just accepted all commits or is there some other factor where you know that they're all compatible?

want to exclude any frontend js changes

just want all backend changes

Are the files for backend and frontend mutually exclusive? I can't tell if you're trying to just find the commits that are backend changes or you have a list of commits that you want to apply, but you just want a way to review them.

hope so. part of the reason i want the output to be another patch file is to inspect it. But can proceed on the assumption that this is the case

The way I would do it is just a cut branch with all the commits applied and then you can do something like:

git diff main...new-main-with-cherry-picks --stat

i know that i do not want all of the commits. lots of FE changes i cannot take

so that just brings me back to how to filter out commits that hit the FE or filter to commits that hit the BE

If the files between backend and frontend are mutually exclusive, then

git diff main...new-main-with-cherry-picks --stat

I’m not following. I don’t have a list of commits from the branch i want to cherry pick because FE and BE commits are intermingled

(hopefully no commit touches both, and will proceed as if that is the case for now)

Yea, there's a bunch of different ways to go about it depending on what you're filtering by and your code base. You could use a programmatic approach using something like jgit. It's maybe similar amount of time to do something more manual like:

git log -n 1 -- path/to/backend

git log -n 1 abcdef^ -- path/to/backend

I’m looking at jgit. And it can parse a patch file but it doesn’t have a notion of separate commits.

patch=> (with-open [is (io/input-stream "patch.patch")]
          (let [p (Patch.)]
            (.parse p is)
            (count (.getFiles p))))
862

Don't you have a patch for each commit? I believe git format-patch will do that for you. You can also just walk the commit history with jgit and compare shas.

i am on a draft PR and I’m downloading the patch file which can contain multiple commits

eg here is a patch file from ring with two commits in it: https://patch-diff.githubusercontent.com/raw/ring-clojure/ring/pull/236.patch

If the draft PR is from a branch, then you should be able to do:

git format-patch main..draft-branch

aha. and then i can parse each of those in turn asking for their files

@U7RJTCH6J splitting the single patch into multiple with git format-match main..branch made this quite simple. Thank you very much

awesome!

Oh, nice! @U050CBXUZ