imapdown/email.go

package main

import (
	"archive/zip"
	"bytes"
	"fmt"
	"io"
	"mime"
	"mime/multipart"
	"net/mail"
	"os"
	"path/filepath"
	"strings"
	"time"
)

// EmailMessage represents a parsed email message
type EmailMessage struct {
	UID     uint32
	Date    time.Time
	Subject string
	Raw     []byte
	Parsed  *mail.Message
}

// ParseEmailMessage parses raw RFC822 email data
func ParseEmailMessage(raw []byte, uid uint32) (*EmailMessage, error) {
	msg, err := mail.ReadMessage(bytes.NewReader(raw))
	if err != nil {
		return nil, fmt.Errorf("failed to parse email: %w", err)
	}

	return &EmailMessage{
		UID:     uid,
		Date:    GetMessageDate(msg),
		Subject: GetMessageSubject(msg),
		Raw:     raw,
		Parsed:  msg,
	}, nil
}

// GetMessageDate extracts the date from an email message
// Falls back to current time if date cannot be parsed
func GetMessageDate(msg *mail.Message) time.Time {
	dateStr := msg.Header.Get("Date")
	if dateStr == "" {
		return time.Now()
	}

	parsed, err := mail.ParseDate(dateStr)
	if err != nil {
		return time.Now()
	}

	return parsed
}

// GetMessageSubject extracts and decodes the subject from an email message
// Returns "no_subject" if subject is empty
func GetMessageSubject(msg *mail.Message) string {
	subject := msg.Header.Get("Subject")
	if subject == "" {
		return "no_subject"
	}

	// Decode RFC 2047 encoded-words
	decoded := DecodeHeaderValue(subject)
	if decoded == "" {
		return "no_subject"
	}

	return decoded
}

// DecodeHeaderValue decodes RFC 2047 encoded-words in headers
func DecodeHeaderValue(encoded string) string {
	dec := new(mime.WordDecoder)
	decoded, err := dec.DecodeHeader(encoded)
	if err != nil {
		return encoded
	}
	return decoded
}

// ExtractAttachments extracts attachments from an email and saves them to a zip file
// Returns the number of attachments extracted
func ExtractAttachments(msg *mail.Message, emlPath string) (int, error) {
	mediaType, params, err := mime.ParseMediaType(msg.Header.Get("Content-Type"))
	if err != nil {
		// Not a multipart message or invalid content-type
		return 0, nil
	}

	if !strings.HasPrefix(mediaType, "multipart/") {
		// Not a multipart message
		return 0, nil
	}

	attachments := make([]attachment, 0)

	// Parse multipart message
	boundary := params["boundary"]
	if boundary == "" {
		return 0, nil
	}

	mr := multipart.NewReader(msg.Body, boundary)
	if err := extractPartsRecursive(mr, &attachments); err != nil {
		// Ignore errors in attachment extraction
		if len(attachments) == 0 {
			return 0, nil
		}
	}

	if len(attachments) == 0 {
		return 0, nil
	}

	// Create zip file
	zipPath := strings.TrimSuffix(emlPath, filepath.Ext(emlPath)) + ".zip"
	zipFile, err := os.Create(zipPath)
	if err != nil {
		return 0, fmt.Errorf("failed to create zip file: %w", err)
	}
	defer zipFile.Close()

	zw := zip.NewWriter(zipFile)
	defer zw.Close()

	// Track duplicate filenames
	seenNames := make(map[string]int)

	for _, att := range attachments {
		filename := att.filename

		// Handle duplicate names
		if count, exists := seenNames[filename]; exists {
			seenNames[filename]++
			ext := filepath.Ext(filename)
			name := strings.TrimSuffix(filename, ext)
			filename = fmt.Sprintf("%s_%d%s", name, count+1, ext)
		} else {
			seenNames[filename] = 0
		}

		// Write to zip
		w, err := zw.Create(filename)
		if err != nil {
			continue
		}

		if _, err := w.Write(att.data); err != nil {
			continue
		}
	}

	return len(attachments), nil
}

type attachment struct {
	filename string
	data     []byte
}

// extractPartsRecursive recursively extracts attachments from multipart message
func extractPartsRecursive(mr *multipart.Reader, attachments *[]attachment) error {
	for {
		part, err := mr.NextPart()
		if err == io.EOF {
			break
		}
		if err != nil {
			return err
		}

		// Check Content-Disposition
		disposition := part.Header.Get("Content-Disposition")
		if disposition == "" {
			// Check if it's a nested multipart
			contentType := part.Header.Get("Content-Type")
			mediaType, params, err := mime.ParseMediaType(contentType)
			if err == nil && strings.HasPrefix(mediaType, "multipart/") {
				boundary := params["boundary"]
				if boundary != "" {
					nestedMr := multipart.NewReader(part, boundary)
					extractPartsRecursive(nestedMr, attachments)
				}
			}
			part.Close()
			continue
		}

		// Check if it's an attachment or inline
		if !strings.Contains(disposition, "attachment") && !strings.Contains(disposition, "inline") {
			part.Close()
			continue
		}

		// Get filename
		filename := part.FileName()
		if filename == "" {
			part.Close()
			continue
		}

		// Decode filename if needed
		filename = DecodeHeaderValue(filename)
		filename = SanitizeFilename(filename, 100)

		// Read attachment data
		data, err := io.ReadAll(part)
		part.Close()

		if err != nil {
			continue
		}

		if len(data) > 0 {
			*attachments = append(*attachments, attachment{
				filename: filename,
				data:     data,
			})
		}
	}

	return nil
}