u-desk/internal/common/utils.go

package common

import (
	"bytes"
	"crypto/sha256"
	"fmt"
	"io"
	"os"
	"path/filepath"
	"regexp"
	"runtime"
	"strings"
	"sync"
	"unicode/utf8"

	"golang.org/x/text/encoding/simplifiedchinese"
	"golang.org/x/text/transform"
)

// InterfaceSliceToStringSlice 将 []interface{} 安全转换为 []string
func InterfaceSliceToStringSlice(slice []interface{}) []string {
	result := make([]string, 0, len(slice))
	for _, v := range slice {
		if str, ok := v.(string); ok && str != "" {
			result = append(result, str)
		}
	}
	return result
}

// FormatBytes 格式化字节大小为人类可读格式（uint64 版本）
func FormatBytes(bytes uint64) string {
	const unit = 1024
	if bytes < unit {
		return fmt.Sprintf("%d B", bytes)
	}
	div, exp := int64(unit), 0
	for n := bytes / unit; n >= unit; n /= unit {
		div *= unit
		exp++
	}
	return fmt.Sprintf("%.2f %cB", float64(bytes)/float64(div), "KMGTPE"[exp])
}

// FormatBytesInt 格式化字节大小（int64 版本，用于文件大小）
func FormatBytesInt(bytes int64) string {
	return FormatBytes(uint64(bytes))
}

// CopyFile 复制文件内容
func CopyFile(src, dst string) error {
	in, err := os.Open(src)
	if err != nil {
		return err
	}
	defer in.Close()

	out, err := os.Create(dst)
	if err != nil {
		return err
	}
	defer out.Close()

	_, err = out.ReadFrom(in)
	return err
}

// BytesToString 智能编码转换：UTF-8 直接返回，否则尝试 GBK → UTF-8
func BytesToString(data []byte) string {
	if utf8.Valid(data) {
		return string(data)
	}
	reader := transform.NewReader(bytes.NewReader(data), simplifiedchinese.GBK.NewDecoder())
	decoded, err := io.ReadAll(reader)
	if err != nil || !utf8.Valid(decoded) {
		return string(data)
	}
	return string(decoded)
}

var (
	htmlResRegex    = regexp.MustCompile(`(?:src|href|data-src|data-url|poster)=["']([^"']+)["']`)
	htmlSrcsetRegex = regexp.MustCompile(`srcset\s*=\s*["']([^"']+)["']`)
	htmlCssUrlRe    = regexp.MustCompile(`url\(\s*["']?([^"')]+)["']?\s*\)`)
)

// parseSrcsetUrls 从 srcset 属性值中提取各 URL
// srcset 格式: "small.jpg 480w, large.jpg 1024w" → ["small.jpg", "large.jpg"]
func parseSrcsetUrls(srcset string) []string {
	var urls []string
	for _, candidate := range strings.Split(srcset, ",") {
		parts := strings.Fields(strings.TrimSpace(candidate))
		if len(parts) > 0 {
			urls = append(urls, parts[0])
		}
	}
	return urls
}

// ExtractHtmlResources 从 HTML 内容提取资源路径
func ExtractHtmlResources(html string) []string {
	seen := make(map[string]bool)
	var resources []string
	add := func(v string) {
		v = strings.TrimSpace(v)
		if v != "" && !seen[v] {
			seen[v] = true
			resources = append(resources, v)
		}
	}
	for _, m := range htmlResRegex.FindAllStringSubmatch(html, -1) {
		if len(m) > 1 {
			add(m[1])
		}
	}
	for _, m := range htmlSrcsetRegex.FindAllStringSubmatch(html, -1) {
		if len(m) > 1 {
			for _, u := range parseSrcsetUrls(m[1]) {
				add(u)
			}
		}
	}
	for _, m := range htmlCssUrlRe.FindAllStringSubmatch(html, -1) {
		if len(m) > 1 {
			add(m[1])
		}
	}
	return resources
}

// ShouldSkipResource 判断资源路径是否应跳过
func ShouldSkipResource(p string) bool {
	return strings.HasPrefix(p, "data:") ||
		strings.HasPrefix(p, "http://") ||
		strings.HasPrefix(p, "https://") ||
		strings.HasPrefix(p, "//") ||
		strings.HasPrefix(p, "#") ||
		strings.HasPrefix(p, "javascript:") ||
		strings.HasPrefix(p, "mailto:") ||
		strings.HasPrefix(p, "blob:")
}

// Contains 检查切片是否包含元素
func Contains[T comparable](slice []T, item T) bool {
	for _, s := range slice {
		if s == item {
			return true
		}
	}
	return false
}

// Difference 返回在 a 中但不在 b 中的元素
func Difference[T comparable](a, b []T) []T {
	mb := make(map[T]struct{}, len(b))
	for _, x := range b {
		mb[x] = struct{}{}
	}
	var diff []T
	for _, x := range a {
		if _, found := mb[x]; !found {
			diff = append(diff, x)
		}
	}
	return diff
}

// IsWindows 判断是否为Windows系统
func IsWindows() bool {
	return runtime.GOOS == "windows"
}

// siteCacheDir 返回站点预览的确定性缓存目录路径
func SiteCacheDir(transport, connID, remotePath string, fileSize int64, modTime string) string {
	h := sha256.New()
	h.Write([]byte(fmt.Sprintf("site:%s:%s:%s:%d:%s", transport, connID, remotePath, fileSize, modTime)))
	hash := fmt.Sprintf("%x", h.Sum(nil))[:16]
	return filepath.Join(os.TempDir(), "u-desk-cache", "site-"+hash)
}

// 站点预览补充下载限制（sftp 与 ossdrv 共用）
const (
	MaxSupplementFiles     = 50              // 每目录补充下载文件数上限
	MaxSupplementFileSize  = 5 << 20         // 单文件补充下载大小上限（5MB）
	SiteDownloadConcurrency = 8              // 相对路径资源并行下载并发数
)

// RunConcurrent 以最多 limit 个并发执行 tasks，全部完成后返回
// fn 在独立 goroutine 中运行，需自行处理错误与共享状态的加锁
func RunConcurrent[T any](tasks []T, limit int, fn func(t T)) {
	if len(tasks) == 0 || limit <= 0 {
		return
	}
	if limit > len(tasks) {
		limit = len(tasks)
	}
	sem := make(chan struct{}, limit)
	var wg sync.WaitGroup
	for _, task := range tasks {
		sem <- struct{}{}
		wg.Add(1)
		go func(t T) {
			defer func() { <-sem; wg.Done() }()
			fn(t)
		}(task)
	}
	wg.Wait()
}

// UniqueName 返回不与现有条目冲突的文件名：重名时追加「 - 副本」「 - 副本 (2)」后缀（扩展名前插入）。
// exists 报告候选文件名是否已被占用（本地文件系统与 SFTP 各自提供实现）。
func UniqueName(name string, exists func(candidate string) (bool, error)) (string, error) {
	taken, err := exists(name)
	if err != nil {
		return "", err
	}
	if !taken {
		return name, nil
	}
	ext := filepath.Ext(name)
	base := strings.TrimSuffix(name, ext)
	for i := 1; ; i++ {
		suffix := " - 副本"
		if i > 1 {
			suffix = fmt.Sprintf(" - 副本 (%d)", i)
		}
		candidate := base + suffix + ext
		taken, err := exists(candidate)
		if err != nil {
			return "", err
		}
		if !taken {
			return candidate, nil
		}
	}
}

// CountingReader 包装读取器，按实际读取字节数回调 OnN（n > 0 时）
type CountingReader struct {
	R   io.Reader
	OnN func(n int64)
}

func (cr *CountingReader) Read(buf []byte) (int, error) {
	n, err := cr.R.Read(buf)
	if n > 0 {
		cr.OnN(int64(n))
	}
	return n, err
}