fish03andClaude Opus 5.5
bef9f8b4ed
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
91 lines
2.4 KiB
Go
91 lines
2.4 KiB
Go
package router
|
|
|
|
import (
|
|
"html"
|
|
"net/http"
|
|
"net/url"
|
|
"os"
|
|
"path/filepath"
|
|
"regexp"
|
|
"strings"
|
|
|
|
"github.com/gin-gonic/gin"
|
|
)
|
|
|
|
const sitemapDir = "./templates/page/sitemaps"
|
|
|
|
var sitemapLocPattern = regexp.MustCompile(`<loc>([^<]+)</loc>`)
|
|
|
|
// registerSitemapRoutes 暴露 hgdj_server app 生成的 sitemap 索引及分片文件。
|
|
func registerSitemapRoutes(r *gin.Engine) {
|
|
r.StaticFile("/sitemap.xml", filepath.Join(sitemapDir, "sitemap.xml"))
|
|
r.GET("/sitemaps/:name", serveSitemapShard)
|
|
}
|
|
|
|
func serveSitemapShard(c *gin.Context) {
|
|
name := c.Param("name")
|
|
if !isValidSitemapShardName(name) {
|
|
renderNotFound(c)
|
|
return
|
|
}
|
|
raw, err := os.ReadFile(filepath.Join(sitemapDir, name))
|
|
if err != nil {
|
|
renderNotFound(c)
|
|
return
|
|
}
|
|
c.Data(http.StatusOK, "application/xml; charset=utf-8", normalizeSitemapPageURLs(raw))
|
|
}
|
|
|
|
// normalizeSitemapPageURLs 让运行时生成的旧 sitemap 与当前 canonical 路由保持一致。
|
|
// sitemap 索引和静态文件地址不经过这里，仍保留文件型 URL（无尾斜杠）。
|
|
func normalizeSitemapPageURLs(raw []byte) []byte {
|
|
return sitemapLocPattern.ReplaceAllFunc(raw, func(match []byte) []byte {
|
|
parts := sitemapLocPattern.FindSubmatch(match)
|
|
if len(parts) != 2 {
|
|
return match
|
|
}
|
|
loc := html.UnescapeString(strings.TrimSpace(string(parts[1])))
|
|
u, err := url.Parse(loc)
|
|
if err != nil || u.Path == "" {
|
|
return match
|
|
}
|
|
u.Path = canonicalSitemapPagePath(u.Path)
|
|
return []byte("<loc>" + html.EscapeString(u.String()) + "</loc>")
|
|
})
|
|
}
|
|
|
|
func canonicalSitemapPagePath(path string) string {
|
|
switch {
|
|
case path == "/more":
|
|
path = "/topic"
|
|
case strings.HasPrefix(path, "/more/"):
|
|
path = "/topic/" + strings.TrimPrefix(path, "/more/")
|
|
case path == "/tag-list":
|
|
path = "/tag"
|
|
case strings.HasPrefix(path, "/tag-list/"):
|
|
path = "/tag/" + strings.TrimPrefix(path, "/tag-list/")
|
|
}
|
|
if path != "/" && !strings.HasSuffix(path, "/") {
|
|
path += "/"
|
|
}
|
|
return path
|
|
}
|
|
|
|
// isValidSitemapShardName 校验分片文件名，防止路径遍历。
|
|
// 合法示例：sitemap-pages.xml、sitemap-playletdetail-0.xml
|
|
func isValidSitemapShardName(name string) bool {
|
|
if !strings.HasPrefix(name, "sitemap-") || !strings.HasSuffix(name, ".xml") {
|
|
return false
|
|
}
|
|
base := strings.TrimSuffix(name, ".xml")
|
|
if len(base) <= len("sitemap-") {
|
|
return false
|
|
}
|
|
for _, ch := range base[len("sitemap-"):] {
|
|
if !((ch >= 'a' && ch <= 'z') || (ch >= '0' && ch <= '9') || ch == '-') {
|
|
return false
|
|
}
|
|
}
|
|
return true
|
|
}
|