Checkpoint-Golang/utils/minifier.go

package utils

import (
	"crypto/sha512"
	"encoding/base64"
	"fmt"
	"io"
	"net/http"
	"os"
	"path/filepath"
	"regexp"
	"strings"
	"sync"
	"time"

	"github.com/tdewolff/minify/v2"
	"github.com/tdewolff/minify/v2/css"
	"github.com/tdewolff/minify/v2/html"
	"github.com/tdewolff/minify/v2/js"
)

// MinifierOptions configuration for the asset minification process
type MinifierOptions struct {
	// Maximum number of goroutines to use for concurrent processing
	MaxWorkers int
	// Skip files that haven't changed since last run
	SkipUnchanged bool
	// Special file mappings (source path -> destination path)
	SpecialMappings map[string]string
	// Whether to remove comments from all file types
	RemoveComments bool
	// Whether to keep conditional comments (<!--[if IE]>, etc.)
	KeepConditionalComments bool
	// Whether to keep special comments (like license comments starting with /*!*)
	KeepSpecialComments bool
	// Whether to add integrity hashes to externally loaded scripts
	AddIntegrityHashes bool
}

// DefaultMinifierOptions returns the default options for asset minification
func DefaultMinifierOptions() MinifierOptions {
	return MinifierOptions{
		MaxWorkers:              10,
		SkipUnchanged:           true,
		RemoveComments:          true,
		KeepConditionalComments: true,
		KeepSpecialComments:     true,
		AddIntegrityHashes:      true,
		SpecialMappings: map[string]string{
			"develop/static/error.html":            "public/html/error.html",
			"develop/static/pow-interstitial.html": "public/html/pow-interstitial.html",
		},
	}
}

// WorkItem represents a file to be processed by the minifier
type WorkItem struct {
	SourcePath string
	DestPath   string
	MimeType   string
	Data       []byte
}

// FileResult contains the result of processing a single file
type FileResult struct {
	MimeType     string
	OriginalSize int64
	MinifiedSize int64
}

// TypeStats tracks statistics for a particular file type
type TypeStats struct {
	FileCount    int
	OriginalSize int64
	MinifiedSize int64
}

// Helper function to download a file from a URL
func downloadFile(url string) ([]byte, error) {
	resp, err := http.Get(url)
	if err != nil {
		return nil, err
	}
	defer resp.Body.Close()

	if resp.StatusCode != http.StatusOK {
		return nil, fmt.Errorf("bad status: %s", resp.Status)
	}

	return io.ReadAll(resp.Body)
}

// Calculates SHA-384 hash for SRI integrity attribute
func calculateIntegrity(data []byte) string {
	hash := sha512.Sum384(data)
	return "sha384-" + base64.StdEncoding.EncodeToString(hash[:])
}

// Adds integrity attributes to external scripts and stylesheets in HTML
func addIntegrityAttributes(html []byte) ([]byte, error) {
	content := string(html)

	// Create a map to cache downloaded content by URL
	urlContent := make(map[string][]byte)

	// Create pattern to normalize URLs (remove query parameters, etc)
	normalizeURL := func(url string) string {
		// For CDN URLs, you might want to remove query parameters that don't affect content
		if idx := strings.IndexByte(url, '?'); idx > 0 {
			return url[:idx]
		}
		return url
	}

	// Download content if not already cached
	getContent := func(url string) ([]byte, error) {
		normalizedURL := normalizeURL(url)
		if data, exists := urlContent[normalizedURL]; exists {
			return data, nil
		}

		fmt.Printf("Downloading resource: %s\n", url)
		data, err := downloadFile(url)
		if err != nil {
			return nil, err
		}

		// Cache the content
		urlContent[normalizedURL] = data
		return data, nil
	}

	// Calculate integrity for a URL
	getIntegrityForURL := func(url string) (string, error) {
		data, err := getContent(url)
		if err != nil {
			return "", err
		}
		return calculateIntegrity(data), nil
	}

	// Add integrity to a tag
	addIntegrityToTag := func(tag, url string, insertPos int) (string, bool) {
		if strings.Contains(tag, "integrity=") {
			return tag, false
		}

		integrity, err := getIntegrityForURL(url)
		if err != nil {
			fmt.Printf("Failed to generate integrity for %s: %v\n", url, err)
			return tag, false
		}

		// Check if it already has crossorigin
		hasAttr := strings.Contains(tag, "crossorigin=")
		if hasAttr {
			return tag[:insertPos] + fmt.Sprintf(` integrity="%s"`, integrity) + tag[insertPos:], true
		} else {
			return tag[:insertPos] + fmt.Sprintf(` integrity="%s" crossorigin="anonymous"`, integrity) + tag[insertPos:], true
		}
	}

	// Helper function to update content with changed tag
	updateContent := func(fullMatch string, newTag string, matchStart, matchEnd int) {
		content = content[:matchStart] + newTag + content[matchEnd:]
	}

	// First, collect all script and preload URLs to ensure we download them once
	// and calculate the same integrity hash for both uses

	// Process script tags
	scriptRegex := regexp.MustCompile(`<script[^>]+src=["'](https?:\/\/[^"']+)["'][^>]*>`)
	scriptMatches := scriptRegex.FindAllStringSubmatchIndex(content, -1)

	// Process in reverse order to avoid invalidating offsets
	for i := len(scriptMatches) - 1; i >= 0; i-- {
		match := scriptMatches[i]
		urlStart, urlEnd := match[2], match[3]
		url := content[urlStart:urlEnd]

		// Skip non-external URLs
		if !strings.HasPrefix(url, "http") {
			continue
		}

		// Get the script tag
		tagStart, tagEnd := match[0], match[1]
		tag := content[tagStart:tagEnd]

		// Skip if already has integrity
		if strings.Contains(tag, "integrity=") {
			continue
		}

		// Add integrity attribute
		newTag, modified := addIntegrityToTag(tag, url, tagEnd-tagStart-1)
		if modified {
			updateContent(tag, newTag, tagStart, tagEnd)
		}
	}

	// Process link[rel=stylesheet] tags
	cssRegex := regexp.MustCompile(`<link[^>]+rel=["']stylesheet["'][^>]*href=["'](https?:\/\/[^"']+)["'][^>]*>|<link[^>]+href=["'](https?:\/\/[^"']+)["'][^>]*rel=["']stylesheet["'][^>]*>`)
	cssMatches := cssRegex.FindAllStringSubmatchIndex(content, -1)

	for i := len(cssMatches) - 1; i >= 0; i-- {
		match := cssMatches[i]

		// Determine which group captured the URL
		var url string
		if match[2] != -1 {
			url = content[match[2]:match[3]]
		} else if match[4] != -1 {
			url = content[match[4]:match[5]]
		} else {
			continue
		}

		// Skip non-external URLs
		if !strings.HasPrefix(url, "http") {
			continue
		}

		// Get the tag
		tagStart, tagEnd := match[0], match[1]
		tag := content[tagStart:tagEnd]

		// Skip if already has integrity
		if strings.Contains(tag, "integrity=") {
			continue
		}

		// Add integrity attribute
		newTag, modified := addIntegrityToTag(tag, url, tagEnd-tagStart-1)
		if modified {
			updateContent(tag, newTag, tagStart, tagEnd)
		}
	}

	// Process link[rel=preload] tags for scripts and styles
	preloadRegex := regexp.MustCompile(`<link[^>]+rel=["']preload["'][^>]*href=["'](https?:\/\/[^"']+)["'][^>]*as=["'](?:script|style)["'][^>]*>|<link[^>]+href=["'](https?:\/\/[^"']+)["'][^>]*rel=["']preload["'][^>]*as=["'](?:script|style)["'][^>]*>|<link[^>]+as=["'](?:script|style)["'][^>]*href=["'](https?:\/\/[^"']+)["'][^>]*rel=["']preload["'][^>]*>`)
	preloadMatches := preloadRegex.FindAllStringSubmatchIndex(content, -1)

	for i := len(preloadMatches) - 1; i >= 0; i-- {
		match := preloadMatches[i]

		// Determine which group captured the URL
		var url string
		if match[2] != -1 {
			url = content[match[2]:match[3]]
		} else if match[4] != -1 {
			url = content[match[4]:match[5]]
		} else if match[6] != -1 {
			url = content[match[6]:match[7]]
		} else {
			continue
		}

		// Skip non-external URLs
		if !strings.HasPrefix(url, "http") {
			continue
		}

		// Get the tag
		tagStart, tagEnd := match[0], match[1]
		tag := content[tagStart:tagEnd]

		// Skip if already has integrity
		if strings.Contains(tag, "integrity=") {
			continue
		}

		// Add integrity attribute
		newTag, modified := addIntegrityToTag(tag, url, tagEnd-tagStart-1)
		if modified {
			updateContent(tag, newTag, tagStart, tagEnd)
		}
	}

	return []byte(content), nil
}

// MinifyAssets processes all files in the develop directory and minifies them to the public directory
// using default options
func MinifyAssets() error {
	return MinifyAssetsWithOptions(DefaultMinifierOptions())
}

// MinifyAssetsWithOptions processes files with the given options
func MinifyAssetsWithOptions(opts MinifierOptions) error {
	startTime := time.Now()

	// Create a minifier instance
	m := minify.New()

	// Configure HTML minifier with consistent settings
	m.Add("text/html", &html.Minifier{
		KeepComments:            !opts.RemoveComments,
		KeepConditionalComments: opts.KeepConditionalComments,
		KeepSpecialComments:     opts.KeepSpecialComments,
		KeepWhitespace:          true,
	})

	// Configure CSS minifier with consistent settings
	m.AddFunc("text/css", css.Minify)

	// Configure JS minifier with consistent settings
	m.AddFunc("text/javascript", js.Minify)

	// Create public directories if they don't exist
	dirs := []string{"html", "css", "js", "static"}
	for _, dir := range dirs {
		publicDir := filepath.Join("public", dir)
		if err := os.MkdirAll(publicDir, 0755); err != nil {
			return fmt.Errorf("failed to create directory %s: %v", publicDir, err)
		}
	}

	// Make sure special mapping destinations have their directories
	for _, destPath := range opts.SpecialMappings {
		destDir := filepath.Dir(destPath)
		if err := os.MkdirAll(destDir, 0755); err != nil {
			return fmt.Errorf("failed to create directory %s: %v", destDir, err)
		}
	}

	// Set up worker pool and channels
	workChan := make(chan WorkItem, 100)
	resultChan := make(chan FileResult, 100)
	var wg sync.WaitGroup

	// Start workers
	maxWorkers := opts.MaxWorkers
	if maxWorkers <= 0 {
		maxWorkers = 10
	}

	// Launch workers
	for i := 0; i < maxWorkers; i++ {
		wg.Add(1)
		go func() {
			defer wg.Done()
			for item := range workChan {
				// Process HTML files to add integrity attributes if enabled
				if opts.AddIntegrityHashes && item.MimeType == "text/html" {
					modified, err := addIntegrityAttributes(item.Data)
					if err != nil {
						fmt.Printf("Failed to add integrity attributes to %s: %v\n", item.SourcePath, err)
					} else {
						item.Data = modified
					}
				}
				processFile(m, item, resultChan)
			}
		}()
	}

	// Stats tracking
	var statsWg sync.WaitGroup
	statsWg.Add(1)
	stats := make(map[string]TypeStats)

	// Start stats collector goroutine
	go func() {
		defer statsWg.Done()
		for result := range resultChan {
			typeStat, exists := stats[result.MimeType]
			if !exists {
				typeStat = TypeStats{}
			}
			typeStat.OriginalSize += result.OriginalSize
			typeStat.MinifiedSize += result.MinifiedSize
			typeStat.FileCount++
			stats[result.MimeType] = typeStat
		}
	}()

	// Collect files to process
	fileCount := 0
	skippedCount := 0
	err := filepath.Walk("develop", func(path string, info os.FileInfo, err error) error {
		if err != nil {
			return fmt.Errorf("error accessing path %s: %v", path, err)
		}

		// Skip directories
		if info.IsDir() {
			return nil
		}

		fileCount++

		// Check for special mappings first
		destPath := ""
		if specialDest, exists := opts.SpecialMappings[path]; exists {
			destPath = specialDest
		} else {
			// Get the relative path from develop
			relPath, err := filepath.Rel("develop", path)
			if err != nil {
				return fmt.Errorf("failed to get relative path: %v", err)
			}
			destPath = filepath.Join("public", relPath)
		}

		// Create destination directory if it doesn't exist
		destDir := filepath.Dir(destPath)
		if err := os.MkdirAll(destDir, 0755); err != nil {
			return fmt.Errorf("failed to create directory %s: %v", destDir, err)
		}

		// Skip unchanged files if option is enabled
		if opts.SkipUnchanged {
			srcInfo, err := os.Stat(path)
			if err == nil {
				destInfo, err := os.Stat(destPath)
				if err == nil && destInfo.ModTime().After(srcInfo.ModTime()) {
					skippedCount++
					return nil
				}
			}
		}

		// Read the source file
		data, err := os.ReadFile(path)
		if err != nil {
			return fmt.Errorf("failed to read file %s: %v", path, err)
		}

		// Determine content type based on file extension
		var mimeType string
		ext := strings.ToLower(filepath.Ext(path))
		switch ext {
		case ".html":
			mimeType = "text/html"
		case ".css":
			mimeType = "text/css"
		case ".js":
			mimeType = "text/javascript"
		default:
			// Copy non-minifiable files as-is immediately (don't send to worker pool)
			if err := os.WriteFile(destPath, data, 0644); err != nil {
				return fmt.Errorf("failed to write file %s: %v", destPath, err)
			}
			return nil
		}

		// Queue file for processing
		workChan <- WorkItem{
			SourcePath: path,
			DestPath:   destPath,
			MimeType:   mimeType,
			Data:       data,
		}

		return nil
	})

	// Close channel and wait for all workers to finish
	close(workChan)
	wg.Wait()
	close(resultChan)
	statsWg.Wait()

	elapsedTime := time.Since(startTime)

	// Print summary by file type
	fmt.Println("\n=== Minification Summary ===")

	// Print HTML stats if available
	if htmlStats, ok := stats["text/html"]; ok && htmlStats.FileCount > 0 {
		reduction := 100.0 - (float64(htmlStats.MinifiedSize) / float64(htmlStats.OriginalSize) * 100.0)
		fmt.Printf("HTML: %.1f%% Reduction (%d files, %d KB → %d KB)\n",
			reduction,
			htmlStats.FileCount,
			htmlStats.OriginalSize/1024,
			htmlStats.MinifiedSize/1024)
	}

	// Print CSS stats if available
	if cssStats, ok := stats["text/css"]; ok && cssStats.FileCount > 0 {
		reduction := 100.0 - (float64(cssStats.MinifiedSize) / float64(cssStats.OriginalSize) * 100.0)
		fmt.Printf("CSS: %.1f%% Reduction (%d files, %d KB → %d KB)\n",
			reduction,
			cssStats.FileCount,
			cssStats.OriginalSize/1024,
			cssStats.MinifiedSize/1024)
	}

	// Print JS stats if available
	if jsStats, ok := stats["text/javascript"]; ok && jsStats.FileCount > 0 {
		reduction := 100.0 - (float64(jsStats.MinifiedSize) / float64(jsStats.OriginalSize) * 100.0)
		fmt.Printf("JavaScript: %.1f%% Reduction (%d files, %d KB → %d KB)\n",
			reduction,
			jsStats.FileCount,
			jsStats.OriginalSize/1024,
			jsStats.MinifiedSize/1024)
	}

	// Print overall totals
	totalOriginal := int64(0)
	totalMinified := int64(0)
	totalFiles := 0
	for _, typeStat := range stats {
		totalOriginal += typeStat.OriginalSize
		totalMinified += typeStat.MinifiedSize
		totalFiles += typeStat.FileCount
	}

	if totalOriginal > 0 {
		totalReduction := 100.0 - (float64(totalMinified) / float64(totalOriginal) * 100.0)
		fmt.Printf("Overall: %.1f%% Reduction (%d files, %d KB → %d KB)\n",
			totalReduction,
			totalFiles,
			totalOriginal/1024,
			totalMinified/1024)
	}

	fmt.Printf("\nProcessed %d files, skipped %d unchanged files in %v\n",
		fileCount-skippedCount, skippedCount, elapsedTime)

	return err
}

// processFile minifies a single file and sends the result to the result channel
func processFile(m *minify.M, item WorkItem, resultChan chan<- FileResult) {
	// Minify the content
	minified, err := m.Bytes(item.MimeType, item.Data)
	if err != nil {
		fmt.Printf("Failed to minify %s: %v\n", item.SourcePath, err)
		// Fall back to original data if minification fails
		minified = item.Data
	}

	// Write the minified content to the destination
	if err := os.WriteFile(item.DestPath, minified, 0644); err != nil {
		fmt.Printf("Failed to write file %s: %v\n", item.DestPath, err)
		return
	}

	// Send result for statistics
	resultChan <- FileResult{
		MimeType:     item.MimeType,
		OriginalSize: int64(len(item.Data)),
		MinifiedSize: int64(len(minified)),
	}
}