postfix_log_parser/components/

discard.rs

//! # Discard 邮件丢弃组件解析器
//!
//! Discard 是 Postfix 的邮件丢弃代理，负责：
//! - 静默丢弃不需要的邮件（如垃圾邮件）
//! - 假装投递但实际丢弃邮件内容
//! - 提供策略性邮件丢弃功能
//! - 记录丢弃操作的详细统计信息
//!
//! ## 特点
//!
//! - 总是报告 `relay=none` 或 `relay=nonediscard`
//! - 状态始终为 `sent`，但实际邮件被丢弃
//! - DSN 通常为 `2.0.0` 表示成功处理
//! - 延迟时间通常很短，因为无实际网络投递
//!
//! ## 支持的事件类型
//!
//! - **邮件丢弃**: 邮件被丢弃的详细记录，包含延迟分析
//! - **配置事件**: 服务启动、传输映射、丢弃规则配置
//!
//! ## 示例日志格式
//!
//! ```text
//! # 邮件丢弃事件
//! queue_id: to=<user@example.com>, relay=none, delay=0.1, delays=0.1/0/0/0, dsn=2.0.0, status=sent (discarded)
//!
//! # 配置事件
//! starting discard service
//! transport mapping updated
//! ```

use regex::Regex;

use crate::error::ParseError;
use crate::events::discard::{DelayBreakdown, DiscardConfigType, DiscardEvent};
use crate::events::{base::BaseEvent, ComponentEvent};
use crate::utils::common_fields::CommonFieldsParser;

use super::ComponentParser;

/// DISCARD组件解析器
/// 
/// 基于Postfix DISCARD邮件丢弃代理的真实日志格式开发
/// DISCARD代理的特点：
/// - 假装投递但实际丢弃邮件
/// - 总是报告relay=nonediscard
/// - 状态总是为sent，但DSN为2.0.0表示成功丢弃
/// - 不进行实际网络投递，延迟时间通常很短
pub struct DiscardParser {
    /// 邮件丢弃事件解析 - 主要模式（95%+频率）
    /// 格式: "queue_id: to=&lt;recipient&gt;, relay=none, delay=X, delays=a/b/c/d, dsn=X.X.X, status=sent (reason)"
    message_discard_regex: Regex,
    
    /// 服务启动/配置事件解析
    /// 格式: 各种配置相关消息
    config_regex: Regex,
}

impl DiscardParser {
    pub fn new() -> Self {
        Self {
            // 主要的邮件丢弃事件模式
            message_discard_regex: Regex::new(
                r"^([A-F0-9]+):\s+to=<([^>]+)>,\s+relay=([^,]+),\s+delay=([0-9.]+),\s+delays=([0-9./]+),\s+dsn=([0-9.]+),\s+status=(\w+)\s+\(([^)]+)\)$"
            ).expect("DISCARD消息丢弃正则表达式编译失败"),
            
            // 配置和启动事件模式
            config_regex: Regex::new(
                r"^(starting|stopping|warning|configuration|transport).*$"
            ).expect("DISCARD配置正则表达式编译失败"),
        }
    }

    /// 解析DISCARD日志行
    pub fn parse_line(&self, line: &str, base_event: BaseEvent) -> Option<DiscardEvent> {
        // 尝试解析邮件丢弃事件（主要模式）
        if let Some(captures) = self.message_discard_regex.captures(line) {
            return self.parse_message_discard(captures, base_event);
        }
        
        // 尝试解析配置事件
        if let Some(captures) = self.config_regex.captures(line) {
            return self.parse_config_event(captures, base_event);
        }
        
        None
    }

    /// 解析邮件丢弃事件（使用公共字段解析器）
    fn parse_message_discard(
        &self,
        captures: regex::Captures,
        base_event: BaseEvent,
    ) -> Option<DiscardEvent> {
        let full_message = base_event.raw_message.as_str();
        let queue_id = captures.get(1)?.as_str().to_string();
        
        // 使用公共字段解析器提取字段
        let recipient = CommonFieldsParser::extract_to_email(full_message)
            .map(|email| email.address)
            .unwrap_or_else(|| captures.get(2).map_or_else(String::new, |m| m.as_str().to_string()));
        
        let relay_info = CommonFieldsParser::extract_relay_info(full_message);
        let relay = relay_info.as_ref()
            .map(|r| r.hostname.clone())
            .unwrap_or_else(|| captures.get(3).map_or_else(String::new, |m| m.as_str().to_string()));
        
        let delay_info = CommonFieldsParser::extract_delay_info(full_message);
        let delay = delay_info.as_ref()
            .map(|d| d.total)
            .unwrap_or_else(|| {
                captures.get(4)
                    .and_then(|m| m.as_str().parse().ok())
                    .unwrap_or(0.0)
            });
        
        // 使用公共解析器的延迟分解，回退到自定义解析
        let delays = delay_info.as_ref()
            .and_then(|d| d.breakdown.as_ref())
            .and_then(|breakdown| {
                DelayBreakdown::from_delays_string(&format!("{}/{}/{}/{}", 
                    breakdown[0], breakdown[1], breakdown[2], breakdown[3]))
            })
            .or_else(|| {
                captures.get(5)
                    .and_then(|m| DelayBreakdown::from_delays_string(m.as_str()))
            })?;
        
        let status_info = CommonFieldsParser::extract_status_info(full_message);
        let dsn = status_info.as_ref()
            .and_then(|s| s.dsn.clone())
            .unwrap_or_else(|| captures.get(6).map_or_else(String::new, |m| m.as_str().to_string()));
        
        let status = status_info.as_ref()
            .map(|s| s.status.clone())
            .unwrap_or_else(|| captures.get(7).map_or_else(String::new, |m| m.as_str().to_string()));
        
        let discard_reason = status_info.as_ref()
            .and_then(|s| s.description.clone())
            .unwrap_or_else(|| captures.get(8).map_or_else(String::new, |m| m.as_str().to_string()));
        
        Some(DiscardEvent::MessageDiscard {
            base: base_event,
            queue_id,
            recipient,
            relay,
            delay,
            delays,
            dsn,
            status,
            discard_reason,
        })
    }
    
    /// 解析配置事件
    /// 处理DISCARD服务的配置、启动、停止等事件
    fn parse_config_event(
        &self,
        captures: regex::Captures,
        base_event: BaseEvent,
    ) -> Option<DiscardEvent> {
        let message = captures.get(0)?.as_str();
        
        // 根据消息内容确定配置类型
        // 注意：优先级很重要！更具体的匹配应该放在前面
        let config_type = if message.contains("starting") || message.contains("stopping") {
            DiscardConfigType::ServiceStartup
        } else if message.contains("transport") {
            DiscardConfigType::TransportMapping
        } else if message.contains("discard") || message.contains("rule") {
            DiscardConfigType::DiscardRules
        } else {
            DiscardConfigType::Other
        };
        
        Some(DiscardEvent::Configuration {
            base: base_event,
            config_type,
            details: message.to_string(),
        })
    }
}

impl ComponentParser for DiscardParser {
    fn parse(&self, message: &str) -> Result<ComponentEvent, ParseError> {
        // 创建一个临时的BaseEvent，用于解析
        // 在实际使用中，这些字段会被MasterParser正确填充
        let base_event = BaseEvent {
            timestamp: chrono::Utc::now(),
            hostname: "temp".to_string(),
            component: "discard".to_string(),
            process_id: 0,
            log_level: crate::events::base::PostfixLogLevel::Info,
            raw_message: message.to_string(),
        };

        if let Some(discard_event) = self.parse_line(message, base_event) {
            Ok(ComponentEvent::Discard(discard_event))
        } else {
            Err(ParseError::ComponentParseError {
                component: "discard".to_string(),
                reason: "无法识别的discard日志格式".to_string(),
            })
        }
    }
    
    fn component_name(&self) -> &'static str {
        "discard"
    }

    fn can_parse(&self, message: &str) -> bool {
        // 检查是否包含discard特征
        self.message_discard_regex.is_match(message) || self.config_regex.is_match(message)
    }
}

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

#[cfg(test)]
mod tests {
    use super::*;
    use crate::events::base::BaseEvent;
    use chrono::{DateTime, Utc};
    
    fn create_test_base_event() -> BaseEvent {
        BaseEvent {
            timestamp: DateTime::parse_from_rfc3339("2024-04-07T10:51:05+00:00")
                .unwrap()
                .with_timezone(&Utc),
            hostname: "m01".to_string(),
            component: "discard".to_string(),
            process_id: 85,
            log_level: crate::events::base::PostfixLogLevel::Info,
            raw_message: "test message".to_string(),
        }
    }
    
    #[test]
    fn test_parse_message_discard() {
        let parser = DiscardParser::new();
        let base_event = create_test_base_event();
        
        let message = "5A4DF1C801B0: to=<six@nextcloud.games>, relay=none, delay=0.05, delays=0.04/0/0/0, dsn=2.0.0, status=sent (nextcloud.games)";
        
        let result = parser.parse_line(message, base_event);
        assert!(result.is_some());
        
        if let Some(DiscardEvent::MessageDiscard { 
            queue_id, 
            recipient, 
            relay,
            delay,
            dsn,
            status,
            discard_reason,
            .. 
        }) = result {
            assert_eq!(queue_id, "5A4DF1C801B0");
            assert_eq!(recipient, "six@nextcloud.games");
            assert_eq!(relay, "none");
            assert_eq!(delay, 0.05);
            assert_eq!(dsn, "2.0.0");
            assert_eq!(status, "sent");
            assert_eq!(discard_reason, "nextcloud.games");
        } else {
            panic!("解析结果类型不正确");
        }
    }
    
    #[test]
    fn test_parse_various_delays() {
        let parser = DiscardParser::new();
        let base_event = create_test_base_event();
        
        // 测试不同的延迟格式
        let test_cases = vec![
            ("delays=0.04/0/0/0", 0.04),
            ("delays=0/0/0/0", 0.0),
            ("delays=0.01/0.02/0/0", 0.03),
        ];
        
        for (delays_part, expected_total) in test_cases {
            let message = format!("5A4DF1C801B0: to=<test@example.com>, relay=none, delay=0.05, {}, dsn=2.0.0, status=sent (example.com)", delays_part);
            
            let result = parser.parse_line(&message, base_event.clone());
            assert!(result.is_some());
            
            if let Some(DiscardEvent::MessageDiscard { delays, .. }) = result {
                assert!((delays.total_delay() - expected_total).abs() < 0.001);
            }
        }
    }
    
    #[test]
    fn test_parse_config_event() {
        let parser = DiscardParser::new();
        let base_event = create_test_base_event();
        
        let message = "starting mail discard service";
        
        let result = parser.parse_line(message, base_event);
        assert!(result.is_some());
        
        if let Some(DiscardEvent::Configuration { config_type, details, .. }) = result {
            assert!(matches!(config_type, DiscardConfigType::ServiceStartup));
            assert_eq!(details, "starting mail discard service");
        } else {
            panic!("解析结果类型不正确");
        }
    }
    
    #[test]
    fn test_delay_breakdown_parsing() {
        // 测试DelayBreakdown的解析功能
        let delay_breakdown = DelayBreakdown::from_delays_string("0.04/0/0/0").unwrap();
        assert_eq!(delay_breakdown.queue_wait, 0.04);
        assert_eq!(delay_breakdown.connection_setup, 0.0);
        assert_eq!(delay_breakdown.connection_time, 0.0);
        assert_eq!(delay_breakdown.transmission_time, 0.0);
        assert_eq!(delay_breakdown.total_delay(), 0.04);
        assert!(delay_breakdown.is_fast_discard());
        
        let delay_breakdown = DelayBreakdown::from_delays_string("0.01/0.02/0.03/0.04").unwrap();
        assert_eq!(delay_breakdown.total_delay(), 0.10);
        assert!(!delay_breakdown.is_fast_discard());
    }
}