Mastodawn

Nathan Dykes 1d ago

New entry in the Int. Encyclopedia of Language and Linguistics with James Tauber:

Regular Expressions for Corpus Queries.

We introduce #regex with actual #corpuslinguistics examples instead of making everything look like \]\[

Happy to send a copy to anyone interested.
https://doi.org/10.1016/B978-0-323-95504-1.01539-8

Michael T Babcock 2d ago

One of the most powerful commands on your #linux system is 'find' ... and one of the most annoying commands on your system is also find.

Do you need to use -and between parameters?

Did you specify the right #regex type before -regex search? And "./"?

Can you remember the before and after date syntax?

If you want to do something to a certain subset of files, find is probably the right way. And you'll spend fifteen minutes fixing the command-line before getting it right.

#rant #sysadmin #tips

Klaus Frank 3d ago

It has been 0 days since I fucked up regex.

To be fair I wasn't aware that the filter sequence is also detecting regex. (Ok I probably should have looked into the full documentation instead of just the examples on a 3rd party page, but still)

Here it is btw: https://rclone.org/filtering/#delete-excluded-delete-files-on-dest-excluded-from-sync

#rclone #regex

Rclone Filtering

Rclone filtering, includes and excludes

Rclone

𝕂𝚞𝚋𝚒𝚔ℙ𝚒𝚡𝚎𝚕 3d ago

finding all regex matches has always been O(n²). even in the engines built to prevent it

every regex engine that promises linear time breaks that promise the moment you ask for all matches. the problem has been there since the 70s, hiding in plain sight.
— by ian erik varatalu

🤔 https://iev.ee/blog/the-quadratic-problem-nobody-fixed/

#regex #problem #coding #70s #blogpost #find #blog #liner #lineartime #time

finding all regex matches has always been O(n²). even in the engines built to prevent it | ian erik varatalu

every regex engine that promises linear time breaks that promise the moment you ask for all matches. the problem has been there since the 70s, hiding in the iteration loop.

ian erik varatalu

doyonseb 3d ago

finding all regex matches has always been O(n²). even in the engines built to prevent it
https://iev.ee/blog/the-quadratic-problem-nobody-fixed/
#regex #rust

finding all regex matches has always been O(n²). even in the engines built to prevent it | ian erik varatalu

every regex engine that promises linear time breaks that promise the moment you ask for all matches. the problem has been there since the 70s, hiding in the iteration loop.

ian erik varatalu

N-gated Hacker News 3d ago

Ah, the #classic 🤦 "we discovered something obvious and decided to blog about it" trope. Apparently, #regex engines have been lying to us for decades, promising #speed but delivering #molasses 🐢. Who knew a computer science relic from the '70s would be today's tech bloggers' favorite bandwagon? 🎉
https://iev.ee/blog/the-quadratic-problem-nobody-fixed/ #techblogging #computerScience #tropes #HackerNews #ngated

finding all regex matches has always been O(n²). even in the engines built to prevent it | ian erik varatalu

every regex engine that promises linear time breaks that promise the moment you ask for all matches. the problem has been there since the 70s, hiding in the iteration loop.

ian erik varatalu

GripNews 6d ago

🌗 正規表示式衝擊波：高效重構與優化指南
➤ 從繁雜到極簡，重新定義你的正規表示式效能
✤ https://mdp.github.io/regex-blaster/
本文探討了「Regex Blaster」這一工具的開發理念。該工具旨在解決開發者在處理複雜正規表示式時常遇到的效能瓶頸與可讀性低落問題。作者透過引入自動化重構機制，將難以維護的冗長表達式拆解為模組化邏輯，並結合效能評測套件，確保優化後的表達式不僅運行速度更快，且具備極佳的程式碼維護性。這對於依賴數據處理與文字分析的工程師而言，是一項不可或缺的效能利器。
+ 終於有工具能自動整理這些像亂碼一樣的 Regex 了，這對維護舊程式碼簡直是救星。
+ 概念很棒，但希望能看到更多關於「回溯（Backtracking）」問題的處理機制，畢竟那纔是效能殺手。
#程式開發 #技術優化 #Regex