Vladimir Iakovlev

Software Developer

[Vladimir Iakovlev @nvbn]

Does it allow to call a dirty function from a pure function?

Nope.

But it's super ugly, and unusable. It will be much nicer to have it as a sort of type annotation.

[Vladimir Iakovlev @nvbn]

A bit of a brain dump around functions colors thread 👻

Usually when talking about functions colors everyone mean async and regular (including ones with blocking io) functions.

An obvious example are Python functions defined with `async def` and `def`.

[Vladimir Iakovlev @nvbn]

If there are functions colored in pure color (ugh), there should be some dirty functions, right? And those dirty functions should be able to call pure functions inside.

It's doable with the same approach, and still ugly.

[Vladimir Iakovlev @nvbn]

After fancy cloud, terraform, k8s and things around, it feels liberating to just configure everything manually on a box and do deployment with fabric 💩

[Vladimir Iakovlev @nvbn]

@nikitonsky Lol, I live on that street 🙂

[Vladimir Iakovlev @nvbn]

AMS ✈️ TPE ✈️ HAN, 12 + 3.5 hours of flights approaching 🥶

[Vladimir Iakovlev @nvbn]

Deutsche Bahn/ice sucks, train to Düsseldorf decided to stop in Mönchengladbach and I needed to transfer to a slower train. The train back got cancelled 😐

[Vladimir Iakovlev @nvbn]

Seems like visit to Russia is going to be a little longer thanks to positive test 😐

[Vladimir Iakovlev @nvbn]

Today's advent of code problem was fun, and probably totally not optimal, and overkill to do with a parser combinator (nom for rust) https://t.co/PeADPjvnHg

[Vladimir Iakovlev @nvbn]

It feels like intellij idea is getting worse every year, now the whole window randomly freezes when I’m typing in rust on macos 😐

[Vladimir Iakovlev @nvbn]

Resident Evil 4 in vr is quite scary in some parts

[Vladimir Iakovlev @nvbn]

Computers are too smart, pojo generator from json schema created a class ‘Criterium’ for entries of a field ‘criteria’.

[Vladimir Iakovlev @nvbn]

@vladimarius It was only possible to make it work by installing tensorflow and keras from conda, pandas and jupyter from pip, matplotlib from git with manually downloaded qhull, and reinstalling numpy from conda after that. Eventually it works smoothly and kind of fast.

[Vladimir Iakovlev @nvbn]

Installing tensorflow, pandas, matplotlib and numpy on m1 laptop apparently is a real adventure