misc-mirrors/article_scraper
1
0
Fork 0
mirror of https://gitlab.com/news-flash/article_scraper.git synced 2025-07-07 16:15:32 +02:00
Code Releases Activity

fmt

Browse source
This commit is contained in:
Jan Lukas Gernert 2023-03-04 17:55:17 +01:00
parent e2b804d00a
commit 2528aa3e18
1 changed files with 7 additions and 5 deletions
Download patch file Download diff file Expand all files Collapse all files

12
src/full_text_parser/readability/mod.rs
View file

@ -131,10 +131,12 @@ impl Readability {
} else if !Util::is_whitespace(&child) { } else if !Util::is_whitespace(&child) {
let mut new_node = Node::new("p", None, &document) let mut new_node = Node::new("p", None, &document)
.map_err(|()| FullTextParserError::Readability)?; .map_err(|()| FullTextParserError::Readability)?;
let mut old_node = node_ref.replace_child_node(new_node.clone(), child).map_err(|error| { let mut old_node = node_ref
log::error!("{error}"); .replace_child_node(new_node.clone(), child)
FullTextParserError::Readability .map_err(|error| {
})?; log::error!("{error}");
FullTextParserError::Readability
})?;
new_node.add_child(&mut old_node).map_err(|error| { new_node.add_child(&mut old_node).map_err(|error| {
log::error!("{error}"); log::error!("{error}");
@ -345,7 +347,7 @@ impl Readability {
// non_significant_whitespace: false, // non_significant_whitespace: false,
// }); // });
// std::fs::write("doc.html", &html).unwrap(); // std::fs::write("doc.html", &html).unwrap();
// The scores shouldn't get too low. // The scores shouldn't get too low.
let score_threshold = last_score / 3.0; let score_threshold = last_score / 3.0;