`hns` — Human Numeric Sort v0.1.0 (⏫︎2022-09-20)

© 2022 Fredrick R. Brennan and hns Authors
- Apache 2.0 licensed, see LICENSE.
man page

Packages

Commercial

Has this ever happened to you?

bash $ seq 1 30 | awk '{printf "data_%s.txt\n", $1}' | sort -h > important_filenames.txt $ sort -h < important_filenames.txt

data_10.txt data_11.txt data_12.txt data_13.txt data_14.txt data_15.txt data_16.txt data_17.txt data_18.txt data_19.txt data_1.txt data_20.txt data_21.txt data_22.txt data_23.txt data_24.txt data_25.txt data_26.txt data_27.txt data_28.txt data_29.txt data_2.txt data_30.txt data_3.txt data_4.txt data_5.txt data_6.txt data_7.txt data_8.txt data_9.txt

Oh no! You forgot that the -h flag of the GNU coreutils sort package doesn't actually do what it claims to, and can't be fixed for various historical reasons (stay with me now don't fall asleep)!

-h, --human-numeric-sort compare human readable numbers (e.g., 2K 1G)

If only there was a better way!

Hi, FREDDY MAYS here with another FUR-tastic invention. All your numbers, sorted just for you!

bash $ mv important_filenames.txt tests/data/README_example.txt $ hns < tests/data/README_example.shuf.txt data_1.txt data_2.txt data_3.txt data_4.txt data_5.txt data_6.txt data_7.txt data_8.txt data_9.txt data_10.txt data_11.txt data_12.txt data_13.txt data_14.txt data_15.txt data_16.txt data_17.txt data_18.txt data_19.txt data_20.txt data_21.txt data_22.txt data_23.txt data_24.txt data_25.txt data_26.txt data_27.txt data_28.txt data_29.txt data_30.txt

Wow!

But if you git pull in the next Unix epoch, you'll also get my super duper negative number-understanding version!

bash $ seq -10 10 | awk '{printf "data_%s.txt\n", $1}' | sort -h > tests/data/README_example2.shuf.txt

Before your numbers were sad and drab…

data_0.txt data_-10.txt data_10.txt data_-1.txt data_1.txt data_-2.txt data_2.txt data_-3.txt data_3.txt data_-4.txt data_4.txt data_-5.txt data_5.txt data_-6.txt data_6.txt data_-7.txt data_7.txt data_-8.txt data_8.txt data_-9.txt data_9.txt

But now they can be radically sequential! (Woah!)

bash $ hns < tests/data/README_example2.shuf.txt data_-10.txt data_-9.txt data_-8.txt data_-7.txt data_-6.txt data_-5.txt data_-4.txt data_-3.txt data_-2.txt data_-1.txt data_0.txt data_1.txt data_2.txt data_3.txt data_4.txt data_5.txt data_6.txt data_7.txt data_8.txt data_9.txt data_10.txt

So simple to use! No command line options! Just standard in and standard out, one size fits all! (Really, Freddy?) (Yes!)

And it's not only for short files, oh no no no! It's written in Rust so you know it can handle even the largest data overflows, such as an entire randomly sorted Class A network!

bash $ time RUST_LOG=INFO target/release/hns < /tmp/0.0.0.0／8.shuf.txt > /dev/null ini [2022-09-20T14:58:00Z INFO hns] Reading done; got 16777216 lines in 348513µs [2022-09-20T14:58:28Z INFO hns] Sorting done; sorted 16777216 lines in 27122184µs [2022-09-20T14:58:32Z INFO hns] Writing done; wrote in 4370570µs real 0m31.855s user 0m30.306s sys 0m1.548s

Sixteen million lines with four comparison points each sorted in under thirty seconds! And that's a Freddy Mays guarantee.

Benchmarking data

See humnumsort-test-data repository.

You may want to pull it as:

bash $ git pull https://github.com/ctrlcctrlv/humnumsort-test-data/ tests/data/expensive

TODO

Zero-padding via another binary?

License

Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. You may obtain a copy of the License at

http://www.apache.org/licenses/LICENSE-2.0