gigi-cli 1.0.0

use anyhow::Result;
use async_trait::async_trait;
use reqwest::header::USER_AGENT;
use serde::{Deserialize, Serialize};
use serde_json::Value;
use std::collections::BTreeSet;

use super::{Tool, ToolOutput};

#[derive(Debug, Clone, Serialize, Deserialize)]
pub struct WebSearchInput {
    pub query: String,
    pub allowed_domains: Option<Vec<String>>,
    pub blocked_domains: Option<Vec<String>>,
}

#[derive(Debug, Clone, Serialize, Deserialize)]
pub struct SearchResult {
    pub title: String,
    pub url: String,
    pub snippet: String,
}

#[derive(Debug, Clone, Serialize, Deserialize)]
pub struct WebSearchOutput {
    pub query: String,
    pub results: Vec<SearchResult>,
}

pub struct WebSearchTool {
    client: reqwest::Client,
}

impl WebSearchTool {
    pub fn new() -> Self {
        Self {
            client: reqwest::Client::new(),
        }
    }
}

impl Default for WebSearchTool {
    fn default() -> Self {
        Self::new()
    }
}

// Strips HTML tags from a string
fn strip_html_tags(html: &str) -> String {
    let mut result = String::new();
    let mut in_tag = false;
    for ch in html.chars() {
        if ch == '<' {
            in_tag = true;
        } else if ch == '>' {
            in_tag = false;
        } else if !in_tag {
            result.push(ch);
        }
    }
    // Decode HTML entities
    result
        .replace("&amp;", "&")
        .replace("&lt;", "<")
        .replace("&gt;", ">")
        .replace("&quot;", "\"")
        .replace("&#x27;", "'")
        .replace("&#39;", "'")
}

#[async_trait]
impl Tool for WebSearchTool {
    fn name(&self) -> &str {
        "WebSearch"
    }

    fn description(&self) -> &str {
        "Search the web for current information and return cited results."
    }

    fn parameters_schema(&self) -> Value {
        serde_json::json!({
            "type": "object",
            "properties": {
                "query": { "type": "string", "minLength": 2 },
                "allowed_domains": {
                    "type": "array",
                    "items": { "type": "string" }
                },
                "blocked_domains": {
                    "type": "array",
                    "items": { "type": "string" }
                }
            },
            "required": ["query"],
            "additionalProperties": false
        })
    }

    async fn execute(&self, input: Value) -> Result<ToolOutput> {
        let search_input: WebSearchInput = serde_json::from_value(input)?;

        let search_url = reqwest::Url::parse_with_params(
            "https://html.duckduckgo.com/html/",
            &[("q", &search_input.query)],
        )?;

        // Query DuckDuckGo using a desktop User-Agent to ensure HTML table rendering
        let res = self
            .client
            .get(search_url)
            .header(
                USER_AGENT,
                "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 \
                 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36",
            )
            .send()
            .await?;

        let html_content = res.text().await?;
        let mut results = Vec::new();

        let allowed_set: Option<BTreeSet<String>> = search_input
            .allowed_domains
            .map(|list| list.into_iter().map(|d| d.to_lowercase()).collect());
        let blocked_set: Option<BTreeSet<String>> = search_input
            .blocked_domains
            .map(|list| list.into_iter().map(|d| d.to_lowercase()).collect());

        // Parse organic DuckDuckGo HTML results using string token scanning
        let mut cursor = 0;
        while let Some(start_idx) = html_content[cursor..].find("<a class=\"result__a\"") {
            let absolute_start = cursor + start_idx;
            let current_slice = &html_content[absolute_start..];

            // Extract the URL
            let href_tag = "href=\"";
            if let Some(href_start) = current_slice.find(href_tag) {
                let url_start = href_start + href_tag.len();
                if let Some(url_end) = current_slice[url_start..].find('"') {
                    let mut url = current_slice[url_start..(url_start + url_end)].to_string();

                    // Clean redirect URLs if present
                    if url.contains("uddg=") {
                        if let Some(uddg_idx) = url.find("uddg=") {
                            let encoded_url = &url[uddg_idx + 5..];
                            let decoded_url = urlencoding::decode(encoded_url).unwrap_or_default().into_owned();
                            url = decoded_url.split('&').next().unwrap_or(&decoded_url).to_string();
                        }
                    }

                    // Extract the Title
                    let closing_bracket = ">";
                    if let Some(bracket_idx) = current_slice[url_start + url_end..].find(closing_bracket) {
                        let title_start = url_start + url_end + bracket_idx + closing_bracket.len();
                        if let Some(a_close_idx) = current_slice[title_start..].find("</a>") {
                            let raw_title = &current_slice[title_start..(title_start + a_close_idx)];
                            let title = strip_html_tags(raw_title).trim().to_string();

                            // Locate snippet inside the result snippet container
                            let mut snippet = String::new();
                            if let Some(snippet_container_idx) = current_slice.find("result__snippet") {
                                let snippet_slice = &current_slice[snippet_container_idx..];
                                if let Some(snippet_content_start) = snippet_slice.find('>') {
                                    let snippet_start = snippet_container_idx + snippet_content_start + 1;
                                    if let Some(snippet_content_end) = current_slice[snippet_start..].find('<') {
                                        let raw_snippet = &current_slice[snippet_start..(snippet_start + snippet_content_end)];
                                        snippet = strip_html_tags(raw_snippet).trim().to_string();
                                    }
                                }
                            }

                            // Verify domain constraints if present
                            let matches_allowed = allowed_set.as_ref().map_or(true, |set| {
                                set.iter().any(|domain| url.to_lowercase().contains(domain))
                            });
                            let matches_blocked = blocked_set.as_ref().map_or(false, |set| {
                                set.iter().any(|domain| url.to_lowercase().contains(domain))
                            });

                            if matches_allowed && !matches_blocked && !url.is_empty() && !title.is_empty() {
                                results.push(SearchResult { title, url, snippet });
                            }
                        }
                    }
                }
            }

            // Move cursor past this result__a match
            if let Some(closing_a) = current_slice.find("</a>") {
                cursor = absolute_start + closing_a + 4;
            } else {
                cursor = absolute_start + 20;
            }

            // Cap at 6 results to stay concise
            if results.len() >= 6 {
                break;
            }
        }

        let output = WebSearchOutput {
            query: search_input.query,
            results,
        };

        let serialized = serde_json::to_string_pretty(&output)?;
        Ok(ToolOutput::success(serialized))
    }
}