Clojurians Log v2

Does anyone use a SLOC counter for Clojure that ignores (comment ) forms? tokei for example considers only ;; to be a comment

@dimitar.ouzounoff How about empty lines?

it counts them separately

nice. I think rewrite-clj could be suited to implement a more specialized sloc counter for Clojure, taking into account comment and #_

See #rewrite-clj

@dimitar.ouzounoff Just wondering: for Clojure, would you be interested in the number of lines, or the number of forms? I feel the latter should probably give better information. Depending on what you want to use it for of course, but with a lisp I guess I care less about lines.

I’m not sure to be honest, especially if one uses a lot of threading macros.. unless they are unwound

Same with test coverage: form coverage is interesting, not so much line coverage.

and keeping SLOC as a measure allows for comparison with other languages

Yeah that’s what I meant with “depending on what you want to use it for”, but I guess I’m not so much interested in this metric at allo so I’m the wrong person to talk to about that 😉

In some teams we've run linters against a targeted LOC count. Some friends target <150, I tend to be fine with <350 it's a good thing for keeping namespaces focused/maintainable and also yielding dep trees that are faster to reload (with require :reload or tools.namespace)

YMMV though and all that :)

For a minute I thought you meant 350kloc code-bases

also good to lint for <100kloc codebases 😂

My last job our main app was ~5.5 million lines of code 😞

But not CLojure, C# .NET

I have a macbook pro with a 250gb harddrive and 16gb of memory. I use docker, multipass, and virtualbox extensively. I'm a bit surprised that this hasn't been sufficient. I can get the job done, but I find myself freeing up disk space and killing services far too often. My next laptop will be 500gb and 32gb

don't take away my 1kLOC core.cljs 😾

Are there separate terms to describe these two phases of "parsing" data from an outside source: - transforming json into "richer" data, ie date strings into Instants or UUID strings to objects - then transforming those maps into domain objects used by my application

would the first one be conforming?

second one I would consider calling "reifying"

words like raw and intermediate come to mind. naming things IL or intermediate language and realizing the boundary of it can be quite helpful

good one's thanks! "Richify" kept popping into my head for the first one

How about ‘type enriching’ and ‘conforming’?

Yeah "enrich" is probably better 🙂

I think the first phase is usually called coercing

e.g. https://github.com/exoscale/coax also calls it this

cool

hydration comes to mind for the latter step

@ctamayo We have a channel for this: #news-and-articles

thanks, I'll move it there

If you haven't noticed this already. Share across your team. https://www.linkedin.com/posts/karthikeyan-annamalai-0aa3003a_free-april-2021-activity-6783417946850123776-grMh

Should you continue to use a library from an author who adhere to conspiracy theories? Or broadly speaking what is the place of politics in software engineering?

Depends on whether the library is any good or not

Maybe it is the same question as why we keep using WhatsApp or Instagram while knowing the damage from FB did on some minorities...

Its kinda interesting because unlike supporting commercial organizations, you're not really helping someone by using their library, you can generally keep the fact you're using it fairly secret to even avoid giving them a negligible amount of credibility. Still I could see it feeling wrong if their views are particularly awful

Also to the extent that you bog them down with bug reports that they work on, you have some ability to shift their time from being used on being awful to being useful to the public

If its a single person, I don't really care about their political affiliation or their views. If the code is good, it's good. But I try not to use anything by big companies like Facebook, WhatsApp, Twitter, Instagram etc as I think they are really damaging society. But if a guy writes a good useful library, I don't care if he's a fruitloop

If I don't financially support a business, it is because of that business's actions, not the views of its employees. I don't have a problem with people having opinions I disagree with

Yeah, I mean, I have a mate who doesn't really believe in dinosaurs, still love the dude

meaning what, exactly? that they never existed?

Yeah

That all the fossils are fake

Fascinating. I thought I’d heard everything before, but nope.

How much of your 16gb of system memory is allocated to the linux vm that is the docker host?

@tvaughan there is a setting which IIRC defaults to ~25%

if you're doing most of your work in docker you might get away with 16gb with that setting turned up...

I think the relationship I have with a library maintainer is different than the relationship I have with a business.

I am much more likely to communicate directly with the maintainer of a library. personally I enjoy the community aspect of participating in open source - e.g. discussing features, reporting bugs, contributing code and docs, answering questions

Unfortunately I have to work with a proprietary image that requires ~8gb (it's poorly architected). At other times I also have to run a Windows VM which also requires a substantial amount of memory. I understand this may not be typical, but when running a "normal" set of applications like mail, web browser, emacs, and other smaller containers, there's not enough room left for even just one of these abnormal situations

yeah, 16gb used to feel like enough but it does not anymore, my workstation has 32gb and that feels like enough... for now

with a corporation like WhatsApp or Instagram, I'm calculating how much harm the company does and whether I want to tacitly support that. It's very indirect; I am unlikely to be harmed, it's more about trying to act in concert with many others who share my beliefs in order to limit the corps impact and my own culpability.

with a library maintainer, I'm less worried about widespread impact (unless they are a VERY popular lib) and more worried about if they're going to tell me to get bent if I report a bug, or if they are arrested for storming a capitol and I have to take over maintenance of a library 😄

"for now" exactly 😢