zine/src/web_fetchers.rs

use feed_rs::model::Entry;
use feed_rs::parser;

use reqwest::blocking::get;
use scraper::{Html, Selector};

use std::error::Error;

pub fn fetch_feed(url: &str) -> Result<Vec<Entry>, Box<dyn Error>> {
    let content = get(url)?.text()?;
    let feed = parser::parse(content.as_bytes())?;
    if feed.entries.is_empty() {
        log::warn!("Feed {url} returned no items!");
    } else {
        log::info!("Feed {} returned {} items", url, feed.entries.len());
    }
    Ok(feed.entries)
}

pub fn fetch_social_image(url: String) -> Result<String, Box<dyn std::error::Error>> {
    let html = reqwest::blocking::get(url)?.text()?;
    let document = Html::parse_document(&html);
    let selector = Selector::parse("meta[property=\"og:image\"]").unwrap();

    let image_url = document
        .select(&selector)
        .next()
        .and_then(|element| element.value().attr("content"));

    if let Some(url) = image_url {
        Ok(url.to_string())
    } else {
        Err("No social image found".into())
    }
}

pub fn is_valid_image_url(url: &str) -> Result<bool, Box<dyn std::error::Error>> {
    let client = reqwest::blocking::Client::new();
    let response = client.head(url).send()?;

    let status = response.status();
    let content_type = response.headers().get(reqwest::header::CONTENT_TYPE);

    Ok(status.is_success()
        && content_type.map_or(false, |ct| {
            ct.to_str().map_or(false, |s| s.starts_with("image/"))
        }))
}