A, location usage summary
The location can be requested in different ways, to a different positioning approach.
1.location usage
location ~ * /js/.*/\.js begin =, exact matching; if only the root end of the directory matches the request, can not be followed with any character string. ^ ~ At the beginning, showing uri string beginning with a conventional, not a regular match beginning ~, it represents a case-sensitive match regular; to ~ * beginning, showing a case-insensitive match regular begins with a /, generic matches, If no other match, any requests are matched to the
Order matching location is the "first match regular, ordinary match again."
Correction: location of the matching sequence is actually "ordinary first match, then match the regular." I say, everyone will refute me, because by "ordinary first match, then match the regular" can not we usually customary press the "first match regular, then matched normal" experience to explain. My only temporarily explain here, the cause of this misunderstanding is: regular match will be covered by ordinary match.
Usage example 2.location
-
Exact match # /, the hostname can not take any string
location = / { [ configuration A ] }
2. # all addresses starting with /, so this rule will default to the last match request
# But regular and will give priority to the longest string match
location / { [ configuration B ] }
Example:
LOCATION / { proxy_pass HTTP: // server_pools; } # This rule does not meet the only other requirements in order to match; will be the last match to the lowest degree of matching to achieve the above functions are: for example, the site is www.blog.com; back when nothing input,
when other rules do not match, and finally to the server load balancing pool
Match any addresses / documents / beginning, after matching in line, but also to continue down search 3. #
# When only back to the regular expression does not match, this one will be using this one
location /documents/ { [ configuration C ] }
Example:
LOCATION / static / { the rewrite http://www.abc.com ^; } # achieve the above functionality: this domain is assumed www.blog.com; then the configuration above function when the input www.blog.com/static/ , regardless of what is behind the static page (the page may not exist), then the same will eventually jump to www.abc.com this site.
Match any addresses / documents / beginning, after matching in line, but also to continue down the search 4. #
# When only back to the regular expression does not match, this one will be using this one
location ~ /documents/Abc { [ configuration CC ] }
5. # matches any address / images / the beginning of the match in line after regular stop down search, the use of this one.
location ^~ /images/ { [ configuration D ] }
6. # match any gif, jpg, or jpeg end request
# However, all requests image / images / the config D will be processed, since it can not get ~ ^ a regular
location ~* \.(gif|jpg|jpeg)$ { [ configuration E ] }
Example:
7. # characters to match / images /, continue down, you will find there is ^ ~
location /images/ { [ configuration F ] }
The longest matching character to # 8 / images / abc, continue down, will find that the presence of ^ ~
# F. And the placement order is not related to G
location /images/abc { [ configuration G ] }
9. # config D is valid only removed: first the longest matching address G at the beginning of config, search continues down to this match a regular, using
location ~ /images/abc/ { [ configuration H ] }
按照上面的location写法,以下的匹配示例成立: / -> config A 精确完全匹配,即使/index.html也匹配不了 /downloads/download.html -> config B 匹配B以后,往下没有任何匹配,采用B /images/1.gif -> configuration D 匹配到F,往下匹配到D,停止往下 /images/abc/def -> config D 最长匹配到G,往下匹配D,停止往下 你可以看到 任何以/images/开头的都会匹配到D并停止,FG写在这里是没有任何意义的,H是永远轮不到的,这里只是为了说明匹配顺序 /documents/document.html -> config C 匹配到C,往下没有任何匹配,采用C /documents/1.jpg -> configuration E 匹配到C,往下正则匹配到E /documents/Abc.jpg -> config CC 最长匹配到C,往下正则顺序匹配到CC,不会往下到E
3、实际使用建议
#直接匹配网站根,通过域名访问网站首页比较频繁,使用这个会加速处理,官网如是说。
#这里是直接转发给后端应用服务器了,也可以是一个静态首页
# 第一个必选规则location = / { proxy_pass http://tomcat:8080/index } # 第二个必选规则是处理静态文件请求,这是nginx作为http服务器的强项
# 有两种配置模式,目录匹配或后缀匹配,任选其一或搭配使用
location ^~ /static/ { root /webroot/static/; } location ~* \.(gif|jpg|jpeg|png|css|js|ico)$ { root /webroot/res/; } #第三个规则就是通用规则,用来转发动态请求到后端应用服务器
#非静态文件请求就默认是动态请求,自己根据实际把握#毕竟目前的一些框架的流行,带.php,.jsp后缀的情况很少了
location / { proxy_pass http://tomcat:8080/ } http://tengine.taobao.org/book/chapter_02.html http://nginx.org/en/docs/http/ngx_http_rewrite_module.html
二、Rewrite用法总结
1.rewrite的定义
rewrite只能放在 server{}, location{}, if{}中,并且只能对域名后边的除去传递的参数外的字符串起作用。 例如 http://seanlook.com/a/we/index.php?id=1&u=str 只对/a/we/index.php重写。
2.rewirte的 语法
1 执行server块的rewrite指令 2 执行location匹配 3 执行选定的location中的rewrite指令
flag标志位
- last : 相当于Apache的[L]标记,表示完成rewrite
- break : 停止执行当前虚拟主机的后续rewrite指令集
- redirect : 返回302临时重定向,地址栏会显示跳转后的地址
- permanent : 返回301永久重定向,地址栏会显示跳转后的地址
- last一般写在server和if中,而break一般使用在location中
- last不终止重写后的url匹配,即新的url会再从server走一遍匹配流程,而break终止重写后的匹配
- break和last都能组织继续执行后面的rewrite指令
3.rewrite常用正则
- . : 匹配除换行符以外的任意字符
- ? : 重复0次或1次
- + : 重复1次或更多次
- * : 重复0次或更多次
- \d :匹配数字
- ^ : 匹配字符串的开始
- $ : 匹配字符串的结束
- {n} : 重复n次
- {n,} : 重复n次或更多次
- [c] : 匹配单个字符c
- [a-z] : 匹配a-z小写字母的任意一个
rewrite实例
1 例1: 2 http { 3 # 定义image日志格式 4 log_format imagelog '[$time_local] ' $image_file ' ' $image_type ' ' $body_bytes_sent ' ' $status; 5 # 开启重写日志 6 rewrite_log on; 7 8 server { 9 root /home/www; 10 11 location / { 12 # 重写规则信息 13 error_log logs/rewrite.log notice; 14 # 注意这里要用‘’单引号引起来,避免{} 15 rewrite '^/images/([a-z]{2})/([a-z0-9]{5})/(.*)\.(png|jpg|gif)$' /data?file=$3.$4; 16 # 注意不能在上面这条规则后面加上“last”参数,否则下面的set指令不会执行 17 set $image_file $3; 18 set $image_type $4; 19 } 20 21 location /data { 22 # 指定针对图片的日志格式,来分析图片类型和大小 23 access_log logs/images.log mian; 24 root /data/images; 25 # 应用前面定义的变量。判断首先文件在不在,不在再判断目录在不在,如果还不在就跳转到最后一个url里 26 try_files /$arg_file /image404.html; 27 } 28 location = /image404.html { 29 # 图片不存在返回特定的信息 30 return 404 "image not found\n"; 31 } 32 } 33 34 对形如/images/ef/uh7b3/test.png的请求,重写到/data?file=test.png,于是匹配到location /data,先看/data/images/test.png文件存不存在,如果存在则正常响应,如果不存在则重写tryfiles到新的image404 location,直接返回404状态码。 35 36 例2: 37 rewrite ^/images/(.*)_(\d+)x(\d+)\.(png|jpg|gif)$ /resizer/$1.$4?width=$2&height=$3? last; 38 对形如/images/bla_500x400.jpg的文件请求,重写到/resizer/bla.jpg?width=500&height=400地址,并会继续尝试匹配location。
if指令与全局变量
if判断指令语法
当表达式只是一个变量时,如果值为空或任何以0开头的字符串都会当做false 直接比较变量和内容时,使用=或!= ~ 正则表达式匹配 ~* 不区分大小写的匹配 !~ 区分大小写的不匹配 -f和!-f 用来判断是否存在文件 -d和!-d 用来判断是否存在目录 -e和!-e 用来判断是否存在文件或目录 -x和!-x 用来判断文件是否可执行
1 如果用户设备为IE浏览器的时候,重定向 2 if ($http_user_agent ~ MSIE) { 3 rewrite ^(.*)$ /msie/$1 break; 4 } //如果UA包含"MSIE",rewrite请求到/msid/目录下 5 6 if ($http_cookie ~* "id=([^;]+)(?:;|$)") { 7 set $id $1; 8 } //如果cookie匹配正则,设置变量$id等于正则引用部分 9 10 if ($request_method = POST) { 11 return 405; 12 } //如果提交方法为POST,则返回状态405(Method not allowed)。return不能返回301,302 13 14 if ($slow) { 15 limit_rate 10k; 16 } //限速,$slow可以通过 set 指令设置 17 18 if (!-f $request_filename){ 19 break; 20 proxy_pass http://127.0.0.1; 21 } //如果请求的文件名不存在,则反向代理到localhost 。这里的break也是停止rewrite检查 22 23 if ($args ~ post=140){ 24 rewrite ^ http://example.com/ permanent; 25 } //如果query string中包含"post=140",永久重定向到example.com 26 27 location ~* \.(gif|jpg|png|swf|flv)$ { 28 valid_referers none blocked www.jefflei.comwww.leizhenfang.com; 29 if ($invalid_referer) { 30 return 404; 31 } //防盗链 32 }
全局变量
- $args : #这个变量等于请求行中的参数,同$query_string
- $content_length : 请求头中的Content-length字段。
- $content_type : 请求头中的Content-Type字段。
- $document_root : 当前请求在root指令中指定的值。
- $host : 请求主机头字段,否则为服务器名称。
- $http_user_agent : 客户端agent信息
- $http_cookie : 客户端cookie信息
- $limit_rate : 这个变量可以限制连接速率。
- $request_method : 客户端请求的动作,通常为GET或POST。
- $remote_addr : 客户端的IP地址。
- $remote_port : 客户端的端口。
- $remote_user : 已经经过Auth Basic Module验证的用户名。
- $ Request_filename: the file path of the current request, requested by the instruction root URI or alias generated.
- $ Scheme: HTTP method (such as http, https).
- $ Server_protocol: protocol requests, usually HTTP / 1.0 or HTTP / 1.1.
- $ Server_addr: server address, after the completion of a system call can determine this value.
- $ Server_name: server name.
- $ Server_port: request reaches the server's port number.
- $ Request_uri: contains the original request URI parameters, it does not include the host name, such as: "/ foo / bar.php arg = baz?".
- $ Uri: with no request parameters of the current URI, $ uri does not contain a host name, such as "/foo/bar.html".
- $ Document_uri: the same as the $ uri.
Example:
1 http://localhost:88/test1/test2/test.php 2 $host:localhost 3 $server_port:88 4 $request_uri:http://localhost:88/test1/test2/test.php 5 $document_uri:/test1/test2/test.php 6 $document_root:/var/www/html 7 $request_filename:/var/www/html/test1/test2/test.php