I have two nested loops like this: loop1 do ... loop2 do ... break if cond ... end ... end Here, the break just gets me out of the inner loop2 but I want to get out of both. I have a cludge way of doing that, but does Crystal provide an elegant way to break out of both?

@asterite apparently so. I reran it using @beta-ziliani version and it took ~266s too, so it seems it’s not jumping out of the inner loop? I think @straight-shoota suggestion of doing it PHP’s way by giving the number of loops to back out of would be the cleanest visually and functionally. Oh wel…

There’s a part you missed from my last comment: your code isn’t raising, so there’s a problem with the logic of the algorithm.

I don’t know what you mean by that. There is no error in the algorithm that creates an exception to normal operation. It just needs to terminate those loops when prime > Math.isqrt(val) You shouldn’t conceptualize that as an exception, it’s normal operational flow. So I want to break out those l…

I see the error now, that I copied into my code: you weren’t raising the Break.new(:here). You should add the raise: catch(:here) do prms.each_with_index do |resgroup, k| res.each_with_index do |prm_r, i| ... raise Break.new(:here) if prime > Math.isqrt(val) # ^--- he…

[image] beta-ziliani: I see the error now, that I copied into my code: you weren’t raising the Break.new(:here) . You should add the raise : Adding the raise reduced the times to ~64s, about the same as the nested breaks. That implementation is overkill for what I want. I just want a break t…

I understand but I don’t think we’ll ever add something like that.

Goto’s in Nim. https://forum.nim-lang.org/t/8963

In that discussion they arrive at the same discussion we have here. Basically, there’s no such thing as goto in Nim, and according to a member there shouldn’t be: Nim’s optimizer does not understand control flow hiding in emit and is free to break your program. Instead of goto, use structured prog…

That throw/catch code is really nice and concise. I would love to see that in the standard library.

Uh. On my AMD 3950x I get: linus@dunk[prog] - (master *=): crystal run --release --mcpu znver2 sozpg.cr throw/catch 308.38k ( 3.24µs) (± 0.95%) 4.08kB/op 1.92× slower return 553.48k ( 1.81µs) (± 1.15%) 4.01kB/op 1.07× slower meth 591.63k ( 1.69µs) (± 0.92%) 3.96kB/op fa…

Elegant breaking from nested loops

Help & Support

jzakiya December 21, 2023, 6:58pm 38

Topic		Replies	Views
Kudos to Crystal! News	12	2314	December 29, 2021
Twin Primes Segmented Sieve of Zakiya (SSoZ) Explained News	16	864	June 29, 2022
On the Infinity of Twin Primes and other K-tuples	22	1340	November 27, 2019
There is a way to optimize this program? Help & Support	62	2207	January 18, 2026
Improve Crystal's profile on Rosetta Code Community	100	6562	February 28, 2024

Elegant breaking from nested loops

Related topics