Multimodal support in cli (#269)

* feat: add multimodal support for image processing in CLI

* test: add tests for prompt cleaning with multiple image inputs

* Apply suggestion from @Copilot

Co-authored-by: Copilot <[email protected]>

* Apply suggestion from @Copilot

Co-authored-by: Copilot <[email protected]>

* feat: enhance image path extraction to support quoted paths and spaces

---------

Co-authored-by: Copilot <[email protected]>

Author: ilopezluna

cmd/cli/commands/images.go

@@ -0,0 +1,160 @@
+package commands
+
+import (
+	"encoding/base64"
+	"errors"
+	"fmt"
+	"io"
+	"net/http"
+	"os"
+	"regexp"
+	"slices"
+	"strings"
+)
+
+// MaxImageSizeBytes is the maximum allowed size for image files (100MB)
+const MaxImageSizeBytes int64 = 100 * 1024 * 1024
+
+// extractImagePaths finds image file paths in the input string using regex
+// Matches paths like: /path/to/file.jpg, ./image.png, C:\photos\pic.webp
+// Also handles quoted paths with spaces: "/path/to/my file.jpg" or '/path/with spaces/image.png'
+// Unquoted paths with spaces are also supported for better UX
+func extractImagePaths(input string) []string {
+	// Regex to match file paths:
+	// - Quoted paths (double or single quotes) - can contain spaces
+	// - Unix absolute: /path/to/file.jpg (with or without spaces)
+	// - Unix relative: ./path/to/file.jpg (with or without spaces)
+	// - Windows absolute: C:\path\to\file.jpg or C:/path/to/file.jpg (with or without spaces)
+	// - Windows relative: \path\to\file.jpg (with or without spaces)
+	// Pattern explanation:
+	//   1. "([^"]+?\.(?i:jpg|jpeg|png|webp))" - Double-quoted paths
+	//   2. '([^']+?\.(?i:jpg|jpeg|png|webp))' - Single-quoted paths
+	//   3. (?:[a-zA-Z]:[/\\]|[/\\]|\./)[^\n"']*?\.(?i:jpg|jpeg|png|webp)\b - Unquoted paths
+	//      - Matches from path start to image extension
+	//      - [^\n"'] allows spaces but stops at newlines or quotes
+	//      - Non-greedy *? ensures we stop at first valid extension
+	//      - \b word boundary ensures clean extension match
+	regexPattern := `"([^"]+?\.(?i:jpg|jpeg|png|webp))"|'([^']+?\.(?i:jpg|jpeg|png|webp))'|(?:[a-zA-Z]:[/\\]|[/\\]|\./)[^\n"']*?\.(?i:jpg|jpeg|png|webp)\b`
+	re := regexp.MustCompile(regexPattern)
+	matches := re.FindAllStringSubmatch(input, -1)
+
+	var paths []string
+	for _, match := range matches {
+		// match[0] is the full match
+		// match[1] is the double-quoted path (if matched)
+		// match[2] is the single-quoted path (if matched)
+		// If neither capture group matched, match[0] is the unquoted path
+		if match[1] != "" {
+			paths = append(paths, match[1]) // double-quoted path
+		} else if match[2] != "" {
+			paths = append(paths, match[2]) // single-quoted path
+		} else {
+			paths = append(paths, match[0]) // unquoted path
+		}
+	}
+
+	return paths
+}
+
+// normalizeFilePath handles escaped characters in file paths
+// Converts escaped spaces, parentheses, brackets, etc. to their literal form
+func normalizeFilePath(filePath string) string {
+	return strings.NewReplacer(
+		"\\ ", " ", // Escaped space
+		"\\(", "(", // Escaped left parenthesis
+		"\\)", ")", // Escaped right parenthesis
+		"\\[", "[", // Escaped left square bracket
+		"\\]", "]", // Escaped right square bracket
+		"\\{", "{", // Escaped left curly brace
+		"\\}", "}", // Escaped right curly brace
+		"\\$", "$", // Escaped dollar sign
+		"\\&", "&", // Escaped ampersand
+		"\\;", ";", // Escaped semicolon
+		"\\'", "'", // Escaped single quote
+		"\\\\", "\\", // Escaped backslash
+		"\\*", "*", // Escaped asterisk
+		"\\?", "?", // Escaped question mark
+		"\\~", "~", // Escaped tilde
+	).Replace(filePath)
+}
+
+// encodeImageToDataURL reads an image file, validates it, and encodes it to a base64 data URL
+// Returns a data URL like: data:image/jpeg;base64,/9j/4AAQ...
+func encodeImageToDataURL(filePath string) (string, error) {
+	file, err := os.Open(filePath)
+	if err != nil {
+		return "", err
+	}
+	defer file.Close()
+
+	// Read first 512 bytes to detect content type
+	buf := make([]byte, 512)
+	_, err = file.Read(buf)
+	if err != nil {
+		return "", err
+	}
+
+	contentType := http.DetectContentType(buf)
+	allowedTypes := []string{"image/jpeg", "image/png", "image/webp"}
+	if !slices.Contains(allowedTypes, contentType) {
+		return "", fmt.Errorf("invalid image type: %s", contentType)
+	}
+
+	info, err := file.Stat()
+	if err != nil {
+		return "", err
+	}
+
+	// Check if the file size exceeds the maximum limit
+	if info.Size() > MaxImageSizeBytes {
+		return "", fmt.Errorf("file size exceeds maximum limit (%d MB)", MaxImageSizeBytes/(1024*1024))
+	}
+
+	// Read entire file
+	buf = make([]byte, info.Size())
+	_, err = file.Seek(0, 0)
+	if err != nil {
+		return "", err
+	}
+
+	_, err = io.ReadFull(file, buf)
+	if err != nil {
+		return "", err
+	}
+
+	// Encode to base64
+	encoded := base64.StdEncoding.EncodeToString(buf)
+
+	// Create data URL
+	dataURL := fmt.Sprintf("data:%s;base64,%s", contentType, encoded)
+
+	return dataURL, nil
+}
+
+// processImagesInPrompt extracts images from the prompt, encodes them to data URLs,
+// and returns the cleaned prompt text and list of image data URLs
+func processImagesInPrompt(prompt string) (string, []string, error) {
+	imagePaths := extractImagePaths(prompt)
+	var imageDataURLs []string
+
+	for _, filePath := range imagePaths {
+		nfp := normalizeFilePath(filePath)
+		dataURL, err := encodeImageToDataURL(nfp)
+		if errors.Is(err, os.ErrNotExist) {
+			// Skip non-existent files (might be false positive from regex)
+			continue
+		} else if err != nil {
+			return "", nil, fmt.Errorf("couldn't process image %q: %w", nfp, err)
+		}
+
+		// Remove the image path from the prompt text
+		prompt = strings.ReplaceAll(prompt, "'"+nfp+"'", "")
+		prompt = strings.ReplaceAll(prompt, "'"+filePath+"'", "")
+		prompt = strings.ReplaceAll(prompt, nfp, "")
+		prompt = strings.ReplaceAll(prompt, filePath, "")
+
+		imageDataURLs = append(imageDataURLs, dataURL)
+	}
+
+	return strings.TrimSpace(prompt), imageDataURLs, nil
+}