gigi-cli 1.0.0

use anyhow::{Context, Result};
use async_trait::async_trait;
use reqwest::Client;
use serde::{Deserialize, Serialize};

use super::provider::ModelProvider;
use super::types::*;

// =============================================================================
// Groq Provider — Uses OpenAI-compatible API with Groq's endpoint
// =============================================================================

pub struct GroqProvider {
    client: Client,
    api_key: String,
    model: String,
    base_url: String,
}

impl GroqProvider {
    pub fn new(api_key: String, model: Option<String>, base_url: Option<String>) -> Self {
        Self {
            client: Client::new(),
            api_key,
            model: model.unwrap_or_else(|| "llama-3.3-70b-versatile".to_string()),
            base_url: base_url.unwrap_or_else(|| "https://api.groq.com/openai".to_string()),
        }
    }
}

#[async_trait]
impl ModelProvider for GroqProvider {
    async fn complete(&self, request: CompletionRequest) -> Result<CompletionResponse> {
        let url = format!("{}/v1/chat/completions", self.base_url);

        let messages = build_openai_messages(&request);
        let tools = build_openai_tools(&request);

        let mut body = serde_json::json!({
            "model": self.model,
            "messages": messages,
            "max_tokens": request.max_tokens,
        });

        if !tools.is_empty() {
            body["tools"] = serde_json::json!(tools);
        }

        let response = self
            .client
            .post(&url)
            .header("Authorization", format!("Bearer {}", self.api_key))
            .header("Content-Type", "application/json")
            .json(&body)
            .send()
            .await
            .context("Failed to send request to Groq API")?;

        if !response.status().is_success() {
            let status = response.status();
            let error_text = response.text().await.unwrap_or_default();
            anyhow::bail!("Groq API error ({}): {}", status, error_text);
        }

        let api_response: OpenAIResponse = response
            .json()
            .await
            .context("Failed to parse Groq API response")?;

        parse_openai_response(api_response)
    }

    fn name(&self) -> &str {
        "groq"
    }

    fn model_id(&self) -> &str {
        &self.model
    }

    fn supports_tools(&self) -> bool {
        true
    }
}

// =============================================================================
// Google AI Studio (Gemini) Provider
// =============================================================================

pub struct GoogleProvider {
    client: Client,
    api_key: String,
    model: String,
    base_url: String,
}

impl GoogleProvider {
    pub fn new(api_key: String, model: Option<String>, base_url: Option<String>) -> Self {
        Self {
            client: Client::new(),
            api_key,
            model: model.unwrap_or_else(|| "gemini-2.5-flash".to_string()),
            base_url: base_url
                .unwrap_or_else(|| "https://generativelanguage.googleapis.com".to_string()),
        }
    }
}

#[async_trait]
impl ModelProvider for GoogleProvider {
    async fn complete(&self, request: CompletionRequest) -> Result<CompletionResponse> {
        // Google Gemini uses its own API format
        let url = format!(
            "{}/v1beta/models/{}:generateContent?key={}",
            self.base_url, self.model, self.api_key
        );

        let mut parts: Vec<serde_json::Value> = Vec::new();

        // Add system instruction
        let system_instruction = if !request.system.is_empty() {
            Some(serde_json::json!({
                "parts": [{ "text": request.system }]
            }))
        } else {
            None
        };

        // Build contents array from messages
        let contents: Vec<serde_json::Value> = request
            .messages
            .iter()
            .filter_map(|msg| {
                let role = match msg.role {
                    Role::User => "user",
                    Role::Assistant => "model",
                    Role::System => return None, // Handled via systemInstruction
                };

                let parts: Vec<serde_json::Value> = msg
                    .content
                    .iter()
                    .filter_map(|block| match block {
                        ContentBlock::Text { text } => {
                            Some(serde_json::json!({ "text": text }))
                        }
                        ContentBlock::ToolUse { id, name, input } => {
                            Some(serde_json::json!({
                                "functionCall": {
                                    "name": name,
                                    "args": input
                                }
                            }))
                        }
                        ContentBlock::ToolResult {
                            tool_use_id: _,
                            content,
                            ..
                        } => Some(serde_json::json!({
                            "functionResponse": {
                                "name": "tool",
                                "response": { "result": content }
                            }
                        })),
                    })
                    .collect();

                Some(serde_json::json!({
                    "role": role,
                    "parts": parts
                }))
            })
            .collect();

        let mut body = serde_json::json!({
            "contents": contents,
            "generationConfig": {
                "maxOutputTokens": request.max_tokens,
            }
        });

        if let Some(si) = system_instruction {
            body["systemInstruction"] = si;
        }

        // Add tool declarations if any
        if !request.tools.is_empty() {
            let function_declarations: Vec<serde_json::Value> = request
                .tools
                .iter()
                .map(|t| {
                    serde_json::json!({
                        "name": t.name,
                        "description": t.description,
                        "parameters": t.input_schema
                    })
                })
                .collect();

            body["tools"] = serde_json::json!([{
                "functionDeclarations": function_declarations
            }]);
        }

        let response = self
            .client
            .post(&url)
            .header("Content-Type", "application/json")
            .json(&body)
            .send()
            .await
            .context("Failed to send request to Google AI Studio")?;

        if !response.status().is_success() {
            let status = response.status();
            let error_text = response.text().await.unwrap_or_default();
            anyhow::bail!("Google AI Studio error ({}): {}", status, error_text);
        }

        let api_response: serde_json::Value = response
            .json()
            .await
            .context("Failed to parse Google AI Studio response")?;

        // Parse Gemini response format
        let candidates = api_response["candidates"]
            .as_array()
            .context("No candidates in Google response")?;

        let first_candidate = candidates.first().context("Empty candidates array")?;
        let resp_parts = first_candidate["content"]["parts"]
            .as_array()
            .context("No parts in candidate")?;

        let mut content_blocks: Vec<ContentBlock> = Vec::new();
        let mut has_tool_use = false;

        for part in resp_parts {
            if let Some(text) = part["text"].as_str() {
                content_blocks.push(ContentBlock::Text {
                    text: text.to_string(),
                });
            }
            if let Some(fc) = part.get("functionCall") {
                has_tool_use = true;
                let name = fc["name"].as_str().unwrap_or("unknown").to_string();
                let args = fc.get("args").cloned().unwrap_or(serde_json::json!({}));
                content_blocks.push(ContentBlock::ToolUse {
                    id: format!("toolu_{}", uuid::Uuid::new_v4()),
                    name,
                    input: args,
                });
            }
        }

        let stop_reason = if has_tool_use {
            StopReason::ToolUse
        } else {
            match first_candidate["finishReason"].as_str() {
                Some("MAX_TOKENS") => StopReason::MaxTokens,
                _ => StopReason::EndTurn,
            }
        };

        Ok(CompletionResponse {
            content: content_blocks,
            stop_reason,
            usage: None, // Gemini usage parsing is different; skip for now
        })
    }

    fn name(&self) -> &str {
        "google"
    }

    fn model_id(&self) -> &str {
        &self.model
    }

    fn supports_tools(&self) -> bool {
        true
    }
}

// =============================================================================
// Shared OpenAI-compatible helpers (used by Groq and local providers)
// =============================================================================

/// Build OpenAI-format messages from our internal types.
pub fn build_openai_messages(request: &CompletionRequest) -> Vec<serde_json::Value> {
    let mut messages = Vec::new();

    // Add system message
    if !request.system.is_empty() {
        messages.push(serde_json::json!({
            "role": "system",
            "content": request.system,
        }));
    }

    for msg in &request.messages {
        let role = match msg.role {
            Role::User => "user",
            Role::Assistant => "assistant",
            Role::System => "system",
        };

        // Check if this message has tool calls (assistant) or tool results (user)
        let has_tool_use = msg.content.iter().any(|b| matches!(b, ContentBlock::ToolUse { .. }));
        let tool_results: Vec<_> = msg
            .content
            .iter()
            .filter_map(|b| match b {
                ContentBlock::ToolResult { tool_use_id, content, is_error } => {
                    Some((tool_use_id, content, is_error))
                }
                _ => None,
            })
            .collect();

        if has_tool_use {
            // Assistant message with tool calls
            let text_content: String = msg
                .content
                .iter()
                .filter_map(|b| match b {
                    ContentBlock::Text { text } => Some(text.as_str()),
                    _ => None,
                })
                .collect::<Vec<_>>()
                .join("");

            let tool_calls: Vec<serde_json::Value> = msg
                .content
                .iter()
                .filter_map(|b| match b {
                    ContentBlock::ToolUse { id, name, input } => Some(serde_json::json!({
                        "id": id,
                        "type": "function",
                        "function": {
                            "name": name,
                            "arguments": input.to_string(),
                        }
                    })),
                    _ => None,
                })
                .collect();

            let mut msg_json = serde_json::json!({
                "role": "assistant",
                "tool_calls": tool_calls,
            });
            if !text_content.is_empty() {
                msg_json["content"] = serde_json::json!(text_content);
            }
            messages.push(msg_json);
        } else if !tool_results.is_empty() {
            // Tool result messages
            for (tool_use_id, content, _is_error) in tool_results {
                messages.push(serde_json::json!({
                    "role": "tool",
                    "tool_call_id": tool_use_id,
                    "content": content,
                }));
            }
        } else {
            // Regular text message
            let text: String = msg
                .content
                .iter()
                .filter_map(|b| match b {
                    ContentBlock::Text { text } => Some(text.as_str()),
                    _ => None,
                })
                .collect::<Vec<_>>()
                .join("");

            messages.push(serde_json::json!({
                "role": role,
                "content": text,
            }));
        }
    }

    messages
}

/// Build OpenAI-format tool definitions.
pub fn build_openai_tools(request: &CompletionRequest) -> Vec<serde_json::Value> {
    request
        .tools
        .iter()
        .map(|t| {
            serde_json::json!({
                "type": "function",
                "function": {
                    "name": t.name,
                    "description": t.description,
                    "parameters": t.input_schema,
                }
            })
        })
        .collect()
}

/// Parse an OpenAI-compatible response into our internal types.
pub fn parse_openai_response(response: OpenAIResponse) -> Result<CompletionResponse> {
    let choice = response
        .choices
        .into_iter()
        .next()
        .context("No choices in API response")?;

    let mut content_blocks: Vec<ContentBlock> = Vec::new();
    let mut has_tool_calls = false;

    // Add text content if present
    if let Some(text) = &choice.message.content {
        if !text.is_empty() {
            content_blocks.push(ContentBlock::Text { text: text.clone() });
        }
    }

    // Add tool calls if present
    if let Some(tool_calls) = &choice.message.tool_calls {
        has_tool_calls = !tool_calls.is_empty();
        for tc in tool_calls {
            let input: serde_json::Value =
                serde_json::from_str(&tc.function.arguments).unwrap_or(serde_json::json!({}));
            content_blocks.push(ContentBlock::ToolUse {
                id: tc.id.clone(),
                name: tc.function.name.clone(),
                input,
            });
        }
    }

    let stop_reason = if has_tool_calls {
        StopReason::ToolUse
    } else {
        match choice.finish_reason.as_deref() {
            Some("length") => StopReason::MaxTokens,
            Some("stop") => StopReason::EndTurn,
            Some("tool_calls") => StopReason::ToolUse,
            _ => StopReason::EndTurn,
        }
    };

    let usage = response.usage.map(|u| Usage {
        input_tokens: u.prompt_tokens,
        output_tokens: u.completion_tokens,
    });

    Ok(CompletionResponse {
        content: content_blocks,
        stop_reason,
        usage,
    })
}

// =============================================================================
// OpenAI-compatible wire types (shared by Groq and local providers)
// =============================================================================

#[derive(Deserialize)]
pub struct OpenAIResponse {
    pub choices: Vec<OpenAIChoice>,
    pub usage: Option<OpenAIUsage>,
}

#[derive(Deserialize)]
pub struct OpenAIChoice {
    pub message: OpenAIMessage,
    pub finish_reason: Option<String>,
}

#[derive(Deserialize)]
pub struct OpenAIMessage {
    pub content: Option<String>,
    pub tool_calls: Option<Vec<OpenAIToolCall>>,
}

#[derive(Deserialize)]
pub struct OpenAIToolCall {
    pub id: String,
    pub function: OpenAIFunction,
}

#[derive(Deserialize)]
pub struct OpenAIFunction {
    pub name: String,
    pub arguments: String,
}

#[derive(Deserialize)]
pub struct OpenAIUsage {
    pub prompt_tokens: u32,
    pub completion_tokens: u32,
}