Fork me on GitHub
#emacs
<
2017-09-01
>
lmergen12:09:20

anyone knows whether it’s possible to customise cljr’s behaviour of cljr-clean-ns so that it never attempts to remove unused imports ? i’m running into the situation where it’s removing imports where i only use i.e. multimethod declarations, and no specific function calls

dpsutton12:09:27

cljr-libspec-whitelist

dpsutton12:09:36

i'm seeing that in the code, looks like it would be useful to you

ag19:09:48

any magit experts here? Can someone help me squashing commits in the middle of a log?

jeff.terrell19:09:33

Yep. I do that kind of thing frequently. Want to describe your problem, or pair on it?

ag19:09:03

when I press, r i while cursor being exactly how indicated it says: Proceed despite merge in rebase range

ag19:09:58

I think my problem is that it's not straight history, those commits are from a feature branch

jeff.terrell19:09:31

Yeah, sounds right. So can you proceed anyway or does it stop you?

ag19:09:08

I pressed c - continue

ag19:09:52

but not it shows bunch of commits that I don't expect, and at the end the log looks kinda screwed up (maybe)

jeff.terrell19:09:15

Does the magit process buffer (`$`) help illuminate what's going on?

ag19:09:51

this is what it's doing git … rebase -i 3a87a0bc which is correct

jeff.terrell19:09:27

In general, I recommend dividing and conquering here: create a branch from HEAD^ (i.e. 4b874df6). Then you just have to squash the top 3 commits on the branch. After that you can rebase or even cherry-pick your final commit (`Tian`-something) atop the squashed commit and push that new history up.

ag19:09:55

yah, probably that's what I need to do

ag19:09:07

alright thanks Jeff!

jeff.terrell19:09:28

Sure, good luck! 👍 If you still run into problems (e.g. b/c of the merge commit), let me know.

ag19:09:03

I think my problem is that in my team instead of choosing one strategy: either "rebase all the way" or "merge all the time" we mix it and create tangled mess

jeff.terrell19:09:24

Yeah, I could see that being problematic. simple_smile FWIW I put a lot of thought into the git best practices at my company: https://github.com/RoleModel/BestPractices/

ag20:09:05

Thanks. The problem with "best practices" is that hard to make people follow them without slapping on their hands when they don't

jeff.terrell20:09:57

True…people are the hardest problem in programming. 😉

ag20:09:46

and customers.

ag20:09:18

software development would be so great if there were no customers/clients/consumers

ag20:09:53

whiny, demanding, little sons of bitches