修复: 网站预览资源路径+七牛目录层级

internal/filesystem/site_resource.go

@@ -0,0 +1,47 @@
+package filesystem
+
+import (
+	"regexp"
+	"strings"
+)
+
+var (
+	htmlResRegex = regexp.MustCompile(`(?:src|href|data|poster)=["']([^"']+)["']`)
+	htmlCssUrlRe = regexp.MustCompile(`url\(\s*["']?([^"')]+)["']?\s*\)`)
+)
+
+// 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 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:")
+}