move markdown handling to this crate

This commit is contained in:
Gabriel 2025-07-15 09:59:38 -04:00
parent d6fa7898ab
commit dbb3dd5d27
4 changed files with 135 additions and 10 deletions

1
.gitignore vendored
View file

@ -1 +1,2 @@
/target /target
.vscode

2
Cargo.lock generated
View file

@ -1193,7 +1193,7 @@ dependencies = [
[[package]] [[package]]
name = "rss_content" name = "rss_content"
version = "0.1.2" version = "0.1.3"
dependencies = [ dependencies = [
"ego-tree", "ego-tree",
"reqwest", "reqwest",

View file

@ -1,6 +1,6 @@
[package] [package]
name = "rss_content" name = "rss_content"
version = "0.1.2" version = "0.1.3"
edition = "2024" edition = "2024"
[dependencies] [dependencies]

View file

@ -5,15 +5,115 @@ use scraper::{ElementRef, Html, Node};
paragraphs with fancy formatting are turned into markdown, same with paragraphs with fancy formatting are turned into markdown, same with
*/ */
//Supported content
#[derive(Debug,Clone)]
pub enum Content {
Markdown(String),
Image(String),
Audio(String),
Video(String)
}
//double recursion? This seems dumb.
fn markdownify(item: &Item) -> String{
match markdown_content(item) {
Content::Markdown(s) => {
s.to_owned()
}
_ => {"".to_owned()}
}
}
fn markdown_content(item: &Item) -> Content {
let mut markdown = String::new();
match item {
Item::Title(n,t) => {
markdown = markdown + &"#".repeat(*n);
let _ = t.iter().map(|i|{
markdown = "".to_owned() + &markdown + &markdownify(i);
});
},
Item::BoldedText(b) => {
let _ = b.iter().map(|i|{
markdown = "**".to_owned() + &markdown + &markdownify(i) + "**";
});
},
Item::EmphasisText(e) => {
let _ = e.iter().map(|i|{
markdown = "*".to_owned() + &markdown + &markdownify(i) + "*";
});
}
Item::Text(s) => {
markdown = markdown + s;
},
Item::Link(href, children) => {
markdown = markdown + &markdownify(item);
}
Item::Paragraph(p) => {
let _ = p.iter().map(|i|{
markdown = "".to_owned() + &markdown + &markdownify(i);
});
}
Item::UnorderedList(u) => {
let _ = u.iter().map(|i|{
markdown = "".to_owned() + &markdown + &markdownify(i);
});
}
Item::OrderedList(o) => {
let _ = o.iter().map(|i|{
markdown = "".to_owned() + &markdown + &markdownify(i);
});
}
_ => {}
}
Content::Markdown(markdown)
}
fn media_content(item: &Item) -> Content{
Content::Markdown("Media not supported yet".to_owned())
}
pub fn process_content(content: &str) -> Vec<Content> {
let items = itemize_content(content);
let mut result: Vec<Content> = Vec::new();
let _ = items.iter().map(|i| {
match i {
Item::Paragraph(children) => {
result.push(markdown_content(i));
},
Item::UnorderedList(children) => {
result.push(markdown_content(i));
}
Item::OrderedList(children) => {
result.push(markdown_content(i));
}
Item::Image(src) => {
result.push(Content::Image(src.to_owned()));
}
Item::Video(children) => {
result.push(media_content(i));
}
Item::Audio(children) => {
result.push(media_content(i));
}
_ => {}
}
});
[Content::Markdown("Ayy lmao".to_owned())].to_vec()
}
#[derive(Debug)] #[derive(Debug)]
pub enum Item { enum Item {
Ignore, Ignore,
Title(usize,Vec<Item>),
Text(String), Text(String),
//text, links, formatting are all markdown //text, links, formatting are all markdown
//arguably, for better control it will be best to turn markdown into its own set of items //arguably, for better control it will be best to turn markdown into its own set of items
Image(String), Image(String),
Gif(String), //can't detect gif from image, has to be handled on front-end
Svg(String),// wont' support for a while I think. Svg(String),// wont' support for a while I think.
Video(Vec<Item>), Video(Vec<Item>),
Audio(Vec<Item>), Audio(Vec<Item>),
@ -25,6 +125,7 @@ pub enum Item {
ListItem(Vec<Item>), ListItem(Vec<Item>),
Paragraph(Vec<Item>),//gotta replace this with specific items, needlessly flexible Paragraph(Vec<Item>),//gotta replace this with specific items, needlessly flexible
Link(String,Vec<Item>), Link(String,Vec<Item>),
Table(Vec<Item>)
} }
@ -35,7 +136,7 @@ pub fn itemize_content(content: &str) -> Vec<Item> {
}).collect() }).collect()
} }
pub fn get_children(el: &ElementRef) -> Vec<Item>{ fn get_children(el: &ElementRef) -> Vec<Item>{
el.children().map(|c|{parse_items(c)}).collect() el.children().map(|c|{parse_items(c)}).collect()
} }
@ -49,6 +150,14 @@ fn parse_items(n: ego_tree::NodeRef<'_,Node>) -> Item{
let tag_name = el.value().name(); let tag_name = el.value().name();
let mut item: Item; let mut item: Item;
match tag_name { match tag_name {
"h1" => {return Item::Title(1, get_children(&el))},
"h2" => {return Item::Title(2, get_children(&el))},
"h3" => {return Item::Title(3, get_children(&el))},
"h4" => {return Item::Title(4, get_children(&el))},
"h5" => {return Item::Title(5, get_children(&el))},
"h6" => {return Item::Title(6, get_children(&el))},
"strong" => {return Item::BoldedText(get_children(&el))},
"em" => {return Item::EmphasisText(get_children(&el))},
"br" => {return Item::Text("\n".to_owned())}, "br" => {return Item::Text("\n".to_owned())},
"hr" => {return Item::Text("---".to_owned())} "hr" => {return Item::Text("---".to_owned())}
"p" => { "p" => {
@ -124,15 +233,17 @@ mod tests {
*/ */
#[cfg(test)] #[cfg(test)]
mod tests { mod tests {
use core::panic;
use rss::Channel;
mod example_data; mod example_data;
use crate::{itemize_content, process_content, tests::example_data::FEEDS};
use crate::{itemize_content, tests::example_data::FEEDS};
fn get_feed(u: &str) -> rss::Channel {
rss::Channel::read_from(u.as_bytes()).unwrap()
}
#[test] #[test]
fn real_feeds(){ fn itemize_feeds(){
let _ = FEEDS.map(|u|{ let _ = FEEDS.map(|u|{
let feed = rss::Channel::read_from(u.as_bytes()).unwrap(); let feed = get_feed(u);
let results: Vec<_> = feed.items.into_iter().map(|item| { let results: Vec<_> = feed.items.into_iter().map(|item| {
itemize_content(&item.content.unwrap()); itemize_content(&item.content.unwrap());
}).collect(); }).collect();
@ -140,4 +251,17 @@ mod tests {
println!("Evaluated feed\nScanned {} items without errors",results.len()) println!("Evaluated feed\nScanned {} items without errors",results.len())
}); });
} }
#[test]
fn markdownify_feeds(){
let _ = FEEDS.map(|u|{
let feed = get_feed(u);
let results: Vec<_> = feed.items.into_iter().map(|item|{
process_content(&item.content.unwrap());
}).collect();
println!("Processed {} items without errors",results.len())
});
}
} }