lcsm29 / project-euler

My solutions for Project Euler problems in Python, C, C++, C#, F#, Go, Haskell, Java, JavaScript, PHP, Ruby, SQL

Solutions for Project Euler

These are my solutions for Project Euler problems. I've listed many languages in the About section on the right. However, I can barely code in Python and absolutely nothing in any other language at this moment, since I've started coding about 2 months ago. Hence, that's more like a wishlist than an actual description, at least for now.

All code in this repo is runnable, and gives you a comparison of the execution times of each implementation (overall execution times of each implementation for the predetermined number of iterations, as well as min/avg/max of single run). It is tested whether the solution provides a correct answer for the given variable, but may produce an error on different variable. I'm planning to add more testing.

Main files on each folder (e.g. main.py) will run every function implemented on every problems I've solved so far. By default, each problem takes about a second to finish the entire iterations on my computer, so it'll take about a couple of minutes to run if I ever write the solutions for every problems. Right now, it takes 10 seconds.

Current Progress

Languages	Progress
Python	30 / 747
C	0 / 747
C++	0 / 747
C#	0 / 747
F#	0 / 747
Go	0 / 747
Haskell	0 / 747
Java	0 / 747
JavaScript	0 / 747
PHP	0 / 747
Ruby	0 / 747
SQL	0 / 747

Scoreboard

This section hasn't been fully implemented yet. This is my attempt to compare the crunching speed of each language, but there are many limitations. The first and foremost limitation would be: each implementation of each language may use slightly different algorithms (or substantially different one in some cases), due to various reasons, including the differences in features, data types, or styles of each language, my own limited capability, et cetera. Also, each language may have a varying degree of optimization. As such, this is more like the comparison between apple and orange, rather than apple to apple comparison.

Average execution time (in nanoseconds, unless specified) (lower is better)

This table shows how much nanoseconds it took for the fastest algo to complete one iteration on average. I haven't implemented the auto-update feature yet (except for Python), but once implemented, running main files on each folder (e.g. main.py) will trigger the automatic update of this table.

Problems	py	pypy	c	cpp	cs	fs	go	hs	java	js	php	rb	sql
1	581	25										982
2	2,160	131
3	314,980	5,755
4	521,115	46,393
5	50,807	17,003
6	220	35
7	9,436μs	1,311μs
8	1,406μs	165,874
9	19,685μs	61,736
10	163,380μs	33,435μs
11	801,136	103,612
12	155,919μs	30,274μs
13	1,935	2,570
14	777,981μs	407,915μs
15	1,702	404
16	23,152	13,352
17	1,472μs	584,939
18	38,705	21,386
19	145,267	5,966
20	12,553	7,740
21	92,111μs	117,632μs
22	3,092μs	2,909μs
23	1,111ms	604,644μs
24	13,230μs	27,619μs
25	22,750μs	88,062μs
26	14,905μs	5,679μs
27	40,082μs	4,081μs
28	239,247	7,370
29	4,068μs	8,549μs
30	716,723μs	67,669μs

• The table above is auto-generated from the following environments;
  • py: Python 3.9.5.final.0 (64 bit), macOS 11.4 (arm64), Apple M1
  • pypy: PyPy 7.3.5 with GCC Apple LLVM 12.0.5 (clang-1205.0.22.9), macOS 11.4 (x86_64), Apple M1
  • rb: Ruby 3.0.1p64, Windows 10 Professional (AMD64), Intel(R) Core(TM) i7-9700K CPU @ 3.60GHz

Number of Iterations (higher is better)

This table shows how much iterations it computes for about a second.

Problems	py	pypy	c	cpp	cs	fs	go	hs	java	js	php	rb	sql
1	65,000	600,000										4,000
2	150,000	1,000,000
3	1,500	26,000
4	420	2,200
5	20,000	60,000
6	44,000	1,500,000
7	110	780
8	700	7,000
9	50	17,000
10	6	32
11	1,250	9,500
12	7	31
13	25,000	10,000
14	2	3
15	560,000	2,300,000
16	43,000	80,000
17	680	1,650
18	26,000	27,000
19	3,200	31,000
20	34,000	65,000
21	11	10
22	320	350
23	1	2
24	77	35
25	45	11
26	12	6
27	25	240
28	4,500	130,000
29	250	125
30	2	14

• Each main script not just runs the fastest algo, but runs every implementations. Thus, doing poorly on some algo could hamper the result greatly, even if the language is pretty fast on most implementations.