regex - 如何在Nginx服务器上区分大小写的正则表达式？

将服务器从Ubuntu迁移到Debian后发生了一个严重的问题。 Debian不允许两个文件(例如“a.html”和“A.html”)位于同一目录中。

我的服务器收到三种类型的请求，这是当前状态:

文件提供了/archive/2014/www.Test.com之类的请求:/archive/2014/blank.html
文件/archive/2015/Test.com提供了诸如/archive/2015/www.Test.com和/archive/2015/T.html之类的请求

文件/archive/2015/test.com提供了诸如/archive/2015/www.test.com和/archive/2015/t.html之类的请求

我希望后两种请求在两种情况下都以不区分大小写的方式提供文件/archive/2015/t.html。

我怎么能达到这个结果？

当前服务器设置为:

server {
    listen   127.0.0.1:80;
    server_name 127.0.0.1;

    access_log /srv/siteone/logs/access.log;
    error_log /srv/siteone/logs/error.log error;

    location / {
        root   /srv/siteone/html;
        index  index.html index.htm;
        expires 1d;
    }

    rewrite ^/archive/2014/(.+)$ /archive/2014/blank.html last;
    rewrite ^/archive/2015/(www\.)*(.)(.+)$ /archive/2015/$2.html last;

    error_page  403  /403.html;
    error_page  404  /404.html;

}

最佳答案

有很多方法可以解决此问题。

由于您只需要将一个特定的字母更改为小写，因此可以对不区分大小写的正则表达式使用“map”:

map $request $letter {
    "~*^/archive/[0-9]{4}/(www\.)?a(.*)?$"  a;
    "~*^/archive/[0-9]{4}/(www\.)?b(.*)?$"  b;
    "~*^/archive/[0-9]{4}/(www\.)?c(.*)?$"  c;
    "~*^/archive/[0-9]{4}/(www\.)?d(.*)?$"  d;
    "~*^/archive/[0-9]{4}/(www\.)?e(.*)?$"  e;
    "~*^/archive/[0-9]{4}/(www\.)?f(.*)?$"  f;
    "~*^/archive/[0-9]{4}/(www\.)?g(.*)?$"  g;
    "~*^/archive/[0-9]{4}/(www\.)?h(.*)?$"  h;
    "~*^/archive/[0-9]{4}/(www\.)?i(.*)?$"  i;
    "~*^/archive/[0-9]{4}/(www\.)?j(.*)?$"  j;
    "~*^/archive/[0-9]{4}/(www\.)?k(.*)?$"  k;
    "~*^/archive/[0-9]{4}/(www\.)?l(.*)?$"  l;
    "~*^/archive/[0-9]{4}/(www\.)?m(.*)?$"  m;
    "~*^/archive/[0-9]{4}/(www\.)?n(.*)?$"  n;
    "~*^/archive/[0-9]{4}/(www\.)?o(.*)?$"  o;
    "~*^/archive/[0-9]{4}/(www\.)?p(.*)?$"  p;
    "~*^/archive/[0-9]{4}/(www\.)?q(.*)?$"  q;
    "~*^/archive/[0-9]{4}/(www\.)?r(.*)?$"  r;
    "~*^/archive/[0-9]{4}/(www\.)?s(.*)?$"  s;
    "~*^/archive/[0-9]{4}/(www\.)?t(.*)?$"  t;
    "~*^/archive/[0-9]{4}/(www\.)?u(.*)?$"  u;
    "~*^/archive/[0-9]{4}/(www\.)?v(.*)?$"  v;
    "~*^/archive/[0-9]{4}/(www\.)?w(.*)?$"  w;
    "~*^/archive/[0-9]{4}/(www\.)?x(.*)?$"  x;
    "~*^/archive/[0-9]{4}/(www\.)?y(.*)?$"  y;
    "~*^/archive/[0-9]{4}/(www\.)?z(.*)?$"  z;
}

server {
    listen   127.0.0.1:80;
    server_name 127.0.0.1;

    access_log /srv/siteone/logs/access.log;
    error_log /srv/siteone/logs/error.log error;

    root   /srv/siteone/html;

    location / {
        index  index.html index.htm;
        expires 1d;
    }

    rewrite ^/archive/2014/(.+)$ /archive/2014/blank.html last;
    rewrite ^/archive/2015/(www\.)?(.)(.+)$ /archive/2015/$letter.html last;

    error_page  403  /403.html;
    error_page  404  /404.html;
}

如果您安装了嵌入式Perl模块(sudo apt-get install nginx-extras)，则可以使用Perl将请求行转换为小写:

perl_set $uri_lowercase 'sub {
    my $r = shift;
    return lc($r->uri);
}';

server {
    listen   127.0.0.1:80;
    server_name 127.0.0.1;

    access_log /srv/siteone/logs/access.log;
    error_log /srv/siteone/logs/error.log error;

    root   /srv/siteone/html;

    location / {
        index  index.html index.htm;
        expires 1d;
    }

    rewrite ^/archive/2014/(.+)$ /archive/2014/blank.html last;
    rewrite ^/archive/2015/(www\.)?(.)(.+)$ $uri_lowercase;
    rewrite ^/archive/2015/(www\.)?(.)(.+)$ /archive/2015/$2.html last;

    error_page  403  /403.html;
    error_page  404  /404.html;
}

如果在Perl之前更喜欢Lua，则可以对Lua进行相同的操作(同样，您将需要安装nginx-extras):

server {
    listen   127.0.0.1:80;
    server_name 127.0.0.1;

    access_log /srv/siteone/logs/access.log;
    error_log /srv/siteone/logs/error.log error;

    root   /srv/siteone/html;

    location / {
        index  index.html index.htm;
        expires 1d;
    }

    rewrite ^/archive/2014/(.+)$ /archive/2014/blank.html last;
    rewrite_by_lua 'ngx.req.set_uri(string.lower(ngx.var.uri), false)';
    rewrite ^/archive/2015/(www\.)?(.)(.+)$ /archive/2015/$2.html last;

    error_page  403  /403.html;
    error_page  404  /404.html;
}

如果您不喜欢以上所有内容，那么总会有一些黑暗的Nginx技巧可能会有所帮助(但我真的不建议这样做):

server {
    listen   127.0.0.1:8484;

    access_log off;

    rewrite ^.*$ /archive/2015/$host.html;

    root /srv/siteone/html;

    location / {
        index  index.html index.htm;
        expires 1d;
    }
}

server {
    listen   127.0.0.1:80;
    server_name 127.0.0.1;

    access_log /srv/siteone/logs/access.log;
    error_log /srv/siteone/logs/error.log error;

    root   /srv/siteone/html;

    location / {
        index  index.html index.htm;
        expires 1d;
    }

    location ~* ^/archive/2015/(?<letter>[A-Z])\.html$ {
        proxy_set_header Host $letter;
        proxy_pass http://127.0.0.1:8484;
    }

    rewrite ^/archive/2014/(.+)$ /archive/2014/blank.html last;
    rewrite ^/archive/2015/(www\.)?(.)(.+)$ /archive/2015/$2.html last;

    error_page  403  /403.html;
    error_page  404  /404.html;
}

关于regex - 如何在Nginx服务器上区分大小写的正则表达式？，我们在Stack Overflow上找到一个类似的问题：https://stackoverflow.com/questions/29945344/