misc-mirrors/article_scraper
1
0
Fork 0
mirror of https://gitlab.com/news-flash/article_scraper.git synced 2025-07-10 09:25:29 +02:00
Code Releases Activity

refactor & more testing

Browse source
This commit is contained in:
Jan Lukas Gernert 2023-02-25 00:42:26 +01:00
parent 7ae98904d4
commit e3246af28b
14 changed files with 1969 additions and 101 deletions
Download patch file Download diff file Expand all files Collapse all files

2
src/constants.rs
View file

@ -4,6 +4,8 @@ use once_cell::sync::Lazy;
use regex::Regex;
pub const DEFAULT_CHAR_THRESHOLD: usize = 500;
pub static IS_IMAGE: Lazy<Regex> =
Lazy::new(|| Regex::new(r#"/\.(jpg|jpeg|png|webp)/i"#).expect("IS_IMAGE regex"));
pub static SIBLING_CONTENT: Lazy<Regex> =
Lazy::new(|| Regex::new(r#"/\.( |$)/"#).expect("SIBLING_CONTENT regex"));
pub static BYLINE: Lazy<Regex> = Lazy::new(|| {