misc-mirrors/article_scraper
1
0
Fork 0
mirror of https://gitlab.com/news-flash/article_scraper.git synced 2025-07-07 08:05:31 +02:00
Code Releases Activity

no need for head

Browse source
This commit is contained in:
Jan Lukas Gernert 2023-08-10 02:06:52 +02:00
parent 8c7cdacd26
commit 6116ba38ae
1 changed files with 0 additions and 16 deletions
Download patch file Download diff file Expand all files Collapse all files

16
article_scraper/src/full_text_parser/mod.rs
View file

@ -122,8 +122,6 @@ impl FullTextParser {
Node::new("article", None, &document).map_err(|()| FullTextParserError::Xml)?;
document.set_root_element(&root);
Self::generate_head(&mut root, &document)?;
for page_html in pages {
self.parse_page(&mut article, &page_html, &mut root, config, global_config)?;
}
@ -1091,20 +1089,6 @@ impl FullTextParser {
}
}
fn generate_head(root: &mut Node, document: &Document) -> Result<(), FullTextParserError> {
if let Ok(mut head_node) = Node::new("head", None, document) {
if let Ok(()) = root.add_prev_sibling(&mut head_node) {
if let Ok(mut meta) = head_node.new_child(None, "meta") {
if meta.set_property("charset", "utf-8").is_ok() {
return Ok(());
}
}
}
}
Err(FullTextParserError::Xml)
}
fn prevent_self_closing_tags(context: &Context) -> Result<(), FullTextParserError> {
// search document for empty tags and add a empty text node as child
// this prevents libxml from self closing non void elements such as iframe