wait4xandGLM-5.3
3b65cc47ec
v2 残留：([[:space:]];|$) 被解析为空格+分号序列而非字符类，-x 兜底 吃行首缩进——admin location 两轮未落。统一 (^|[[:space:];]) 词边界 并显式排除 git 子域。 验证: bash -n 过；判定正则对 /etc/nginx/sites-enabled/default 实文件 干跑（博客行命中/git 行排除双验） Spec: none (脚本热修 v3 补) Co-authored-by: GLM-5.3
93 lines
4.5 KiB
Bash
Executable File
93 lines
4.5 KiB
Bash
Executable File
#!/usr/bin/env bash
|
||
# SEO 解封脚本（Phase 5f T1 操作单自动化）
|
||
# 作用：①剥共享 snippet 的全站 noindex ②内部服务 vhost 补回 ③博客 vhost /admin 局部 noindex
|
||
# ④nginx -t 后重载，失败自动回滚 ⑤终验三条 curl
|
||
# ESA 控制台三动作（purge/TTL/绕缓存）无法脚本化，结尾打印手动步骤。
|
||
# 用法：bash scripts/ops/seo-unblock.sh （需要 sudo，全程有预览和确认）
|
||
set -euo pipefail
|
||
SNIPPET=/etc/nginx/snippets/esa-origin-ssl.conf
|
||
BAK="${SNIPPET}.bak-$(date +%s)"
|
||
BAKDIR=/tmp/seo-unblock-bak-$(date +%s)
|
||
SITES=/etc/nginx/sites-enabled
|
||
|
||
echo "==== ① 现状 ===="
|
||
sudo grep -n "X-Robots-Tag" "$SNIPPET" || echo "（snippet 已无 noindex——可能已执行过）"
|
||
|
||
echo
|
||
echo "==== ② 变更计划（DRY-RUN）===="
|
||
echo " [snippet] ${SNIPPET}：删除 X-Robots-Tag noindex 行"
|
||
for f in "$SITES"/*; do
|
||
names=$(sudo grep -h "server_name" "$f" 2>/dev/null || true)
|
||
[ -z "$names" ] && continue
|
||
# 内部服务判定：server_name 含内部子域且不含主站裸域
|
||
if sudo grep -q "include snippets/esa-origin-ssl" "$f" && ! echo "$names" | grep -qE "(^|[[:space:];])(www\.)?wait4x\.online([[:space:];]|$)" && ! echo "$names" | grep -qE "(^|[[:space:]])git\.wait4x\.online"; then
|
||
sudo grep -q "X-Robots-Tag" "$f" || echo " [补回] $f：server 块加 add_header X-Robots-Tag noindex always;"
|
||
fi
|
||
# 博客主站判定：server_name 恰含裸域
|
||
if echo "$names" | grep -qE "(^|[[:space:];])(www\.)?wait4x\.online([[:space:];]|$)" && ! echo "$names" | grep -qE "(^|[[:space:]])git\.wait4x\.online"; then
|
||
sudo grep -q "location /admin/" "$f" || echo " [新增] $f：location /admin/ { noindex }"
|
||
fi
|
||
done
|
||
|
||
echo
|
||
read -rp ">> 按计划执行？(y/N) " ok; [ "$ok" = "y" ] || { echo "已取消（未做任何修改）"; exit 0; }
|
||
|
||
echo
|
||
echo "==== ③ 执行 ===="
|
||
sudo mkdir -p "$BAKDIR"
|
||
sudo cp "$SNIPPET" "$BAK" && echo " 备份：$BAK"
|
||
sudo sed -i '/X-Robots-Tag noindex/d' "$SNIPPET" && echo " snippet 已剥 noindex"
|
||
|
||
for f in "$SITES"/*; do
|
||
names=$(sudo grep -h "server_name" "$f" 2>/dev/null || true)
|
||
[ -z "$names" ] && continue
|
||
if sudo grep -q "include snippets/esa-origin-ssl" "$f" && ! echo "$names" | grep -qE "(^|[[:space:];])(www\.)?wait4x\.online([[:space:];]|$)" && ! echo "$names" | grep -qE "(^|[[:space:]])git\.wait4x\.online"; then
|
||
if ! sudo grep -q "X-Robots-Tag" "$f"; then
|
||
sudo cp "$f" "$BAKDIR/$(basename "$f")"
|
||
# 在首个 server_name 行后插入（server 块内、顶层指令位置合法）
|
||
sudo sed -i "0,/server_name/s/\(server_name.*\)/\1\n add_header X-Robots-Tag noindex always;/" "$f"
|
||
echo " 补回：$f"
|
||
fi
|
||
fi
|
||
if echo "$names" | grep -qE "(^|[[:space:];])(www\.)?wait4x\.online([[:space:];]|$)" && ! echo "$names" | grep -qE "(^|[[:space:]])git\.wait4x\.online"; then
|
||
if ! sudo grep -q "location /admin/" "$f"; then
|
||
sudo cp "$f" "$BAKDIR/$(basename "$f")"
|
||
sudo sed -i "0,/server_name/s/\(server_name.*\)/\1\n location \/admin\/ { add_header X-Robots-Tag noindex always; }/" "$f"
|
||
echo " admin 局部：$f"
|
||
fi
|
||
fi
|
||
done
|
||
|
||
echo
|
||
echo "==== ④ 语法检查+重载（失败自动回滚）===="
|
||
if ! sudo nginx -t; then
|
||
echo "!! nginx -t 失败，回滚全部改动"
|
||
sudo cp "$BAK" "$SNIPPET"
|
||
for f in "$BAKDIR"/*; do
|
||
[ -f "$f" ] && sudo cp "$f" "$SITES/$(basename "$f")"
|
||
done
|
||
sudo rm -rf "$BAKDIR"
|
||
sudo rm -f "$SITES"/*.bak-seo # v1 遗留备份也在加载路径上，一并清掉
|
||
sudo nginx -t && sudo systemctl reload nginx
|
||
echo "已回滚，请把以上报错发给 research 人工处理"; exit 1
|
||
fi
|
||
sudo systemctl reload nginx && echo " nginx 已重载"
|
||
|
||
echo
|
||
echo "==== ⑤ 终验（对源站直测）===="
|
||
H=$(curl -sI --noproxy '*' -m 10 https://wait4x.online/ | grep -ci x-robots-tag || true)
|
||
A=$(curl -sI --noproxy '*' -m 10 https://wait4x.online/admin/ | grep -ci noindex || true)
|
||
echo " 主站 noindex 头：$H（期望 0） /admin/ noindex：$A（期望 1）"
|
||
if [ "$H" = "0" ] && [ "$A" = "1" ]; then echo " ✅ 源站解封成功"; else echo " ⚠️ 不符预期——把上面两行发给 research"; fi
|
||
|
||
cat <<'EOF'
|
||
|
||
==== ⑥ 剩余手动步骤：ESA 控制台（无法脚本化）====
|
||
1. 全站 purge：wait4x.online（不做的话 noindex 头在边缘再活 30 天）
|
||
2. HTML 边缘缓存 TTL 调 ≤1 小时（当前 30 天）
|
||
3. 加规则：/artalk/api/* 路径绕过缓存（pv 数字不再冻结）
|
||
4. 顺带查 git.wait4x.online 为何 522（源站健康/回源配置）
|
||
|
||
完成后告诉 research：将跑生产 Lighthouse 入档（预期 SEO 60→95+）。
|
||
EOF
|