finalytics 0.9.0

use anyhow::Result;
use cached::proc_macro::cached;
use chrono::{NaiveDate, NaiveDateTime};
use once_cell::sync::Lazy;
use polars::prelude::*;
use reqwest::header::{HeaderMap, HeaderValue, USER_AGENT};
use reqwest::{Client, StatusCode};
use select::document::Document;
use select::predicate::Name;
use std::error::Error;
use std::time::Duration;
use tokio::task::spawn_blocking;
use vader_sentiment::SentimentIntensityAnalyzer;

pub static REQUEST_CLIENT: Lazy<Client> = Lazy::new(|| {
    let mut headers = HeaderMap::new();
    headers.insert(USER_AGENT, HeaderValue::from_static("Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/58.0.3029.110 Safari/537.36"));
    headers.insert(
        "Accept",
        HeaderValue::from_static("application/json, text/plain, */*"),
    );

    Client::builder()
        .default_headers(headers)
        .cookie_store(true)
        .build()
        .unwrap()
});

pub async fn fetch_news(
    token: &str,
    start_date: NaiveDate,
    end_date: NaiveDate,
    compute_sentiment: bool,
) -> Result<DataFrame, Box<dyn Error + Send + Sync>> {
    let url = format!(
        "https://news.google.com/rss/search?q=allintext:{}+after:{}+before:{}",
        token,
        start_date.format("%Y-%m-%d"),
        end_date.format("%Y-%m-%d")
    );
    let body = fetch_html(url).await?;
    let df = spawn_blocking(move || extract_news_details(body, compute_sentiment)).await?;
    Ok(df)
}

#[cached(
    result = true,
    time = 3600 // Cache Google News Results for 1 Hour
)]
async fn fetch_html(url: String) -> Result<String, Box<dyn Error + Send + Sync>> {
    let response = REQUEST_CLIENT.get(&url).send().await?;

    if response.status() != StatusCode::OK {
        let status = response.status();
        let body = response.text().await?;
        return Err(format!(
            "Google News request failed with status {status} for URL {url}: {body}"
        )
        .into());
    }

    let body = response.text().await?;
    Ok(body)
}

fn extract_news_details(body: String, compute_sentiment: bool) -> DataFrame {
    let document = match Document::from_read(body.as_bytes()) {
        Ok(doc) => doc,
        Err(e) => {
            eprintln!("Failed to parse news RSS body: {e}");
            return DataFrame::default();
        }
    };

    // Collect data into vectors
    let mut titles = Vec::new();
    let mut sources = Vec::new();
    let mut links = Vec::new();
    let mut pub_dates = Vec::new();
    let mut sentiment_scores = Vec::new();

    for item in document.find(Name("item")) {
        let title = item.children().next().map(|n| n.text()).unwrap_or_default();
        let source = item.last_child().map(|n| n.text()).unwrap_or_default();
        let link = item.children().nth(2).map(|n| n.text()).unwrap_or_default();
        let pub_date = item.children().nth(4).map(|n| n.text()).unwrap_or_default();
        if title.is_empty() || link.is_empty() || pub_date.is_empty() {
            continue;
        }
        let pub_date = match NaiveDateTime::parse_from_str(&pub_date, "%a, %d %b %Y %H:%M:%S GMT") {
            Ok(dt) => dt,
            Err(_) => {
                eprintln!("Skipping news item with unparseable date: {pub_date}");
                continue;
            }
        };
        titles.push(title.clone());
        links.push(format!(
            r#"<a href="{}">{}</a>"#,
            link,
            title.replace(format!("- {source}").as_str(), "")
        ));
        sources.push(source);
        pub_dates.push(pub_date);
        if compute_sentiment {
            let analyzer = SentimentIntensityAnalyzer::new();
            let sentiment = analyzer.polarity_scores(&title);
            sentiment_scores.push(sentiment["compound"]);
        }
    }

    let df = match DataFrame::new(vec![
        Column::new("Published Date".into(), pub_dates),
        Column::new("Source".into(), sources),
        Column::new("Title".into(), titles),
        Column::new("Link".into(), links),
    ]) {
        Ok(df) => df,
        Err(e) => {
            eprintln!("Failed to build news DataFrame: {e}");
            return DataFrame::default();
        }
    };

    if compute_sentiment {
        let mut new_df = df.clone();
        match new_df.with_column(Series::new("Sentiment Score".into(), sentiment_scores)) {
            Ok(_) => new_df,
            Err(e) => {
                eprintln!("Error Computing Sentiment Scores: {e}");
                df
            }
        }
    } else {
        df
    }
}