我们支持3种路由模式
普通模式
_a=$app, _u=$ctl.$act
最简单的方式,专注实现业务$act函数,不需要再写额外代码
为什么参数名前面要加下划线就不解释了
easy模式
_easy=$app.$tpl.$ctl.$act
_easy=$app.$ctl.$act
在web开发中,通常我们在$act中输出一个前端页面,
easy模式下,如果未实现$act处理函数,会自动寻找并显示对应的前端模板文件。
对于简单的展示页面适合使用这种路由模式
url重写模式(需要nginx或apache配置)
apache: ^rewrite[\.\/](.*)$ /index.php?_rewrite=$1 [R,QSA]
nginx: rewrite ^/rewrite[\.\/](.*)$ /index.php?_rewrite=$1 last;
rewrite.{$app}.{$ctl}.{$act}.{$params}.html
或更加优雅的目录式访问方式
rewrite/{$app}/{$ctl}/{$act}/{$params}.html
其中$params为选填参数部分.格式为urlencode后的参数列表
如果想传递sp_uid=1&d=1.2&p=sb, 那么$params = sp_uid%3D1%26d%3D1.2%26p%3Ds%2Fb
或sp_uid/1/d/1.2/p/sb
在某些要求url中不能带?&特殊字符的场景下可以使用这种模式
1. 为了能通过qq oauth2登陆验证,需要配置重写规则
rewrite.thirdlogin.index.qqcallback.sp_uid%3D1.php
2. 资源静态化
rewrite.upload.index.out.uidm%3D310ef4b.png
3. 支付回调
rewrite.pay.weixin.native2_notify.php
4. 微信开放平台授权回调
rewrite/web/component/message/_app_id/xxxxxxx.php
部分实现代码
12345678910111213 | $a = (! empty ( $_REQUEST [ '_a' ]) && is_string ( $_REQUEST [ '_a' ])) ? $_REQUEST [ '_a' ] : 'web' ; if (!preg_match( '/^[\w\.]+$/' , $a )) { exit ( 'invalid _app name! ' . htmlspecialchars( $a )); } $GLOBALS [ '_UCT' ][ 'APP' ] = ! empty ( $a ) ? strtolower ( $a ) : 'web' ; $u = (! empty ( $_REQUEST [ '_u' ]) && is_string ( $_REQUEST [ '_u' ])) ? $_REQUEST [ '_u' ] : 'index.index' ; if (!preg_match( '/^[\w\.]+$/' , $u )) { exit ( 'invalid _url name! ' . htmlspecialchars( $u )); } $u = explode ( '.' , $u , 2); $GLOBALS [ '_UCT' ][ 'CTL' ] = ! empty ( $u [ '0' ]) ? strtolower ( $u [ '0' ]) : 'index' ; $GLOBALS [ '_UCT' ][ 'ACT' ] = ! empty ( $u [ '1' ]) ? strtolower ( $u [ '1' ]) : 'index' ; |
easy模式
1234567891011121314151617181920212223242526 | //easy 模式直接访问模板tpl if (! empty ( $_REQUEST [ '_easy' ]) && is_string ( $_REQUEST [ '_easy' ])) { $easy = explode ( '.' , $_REQUEST [ '_easy' ]); switch ( count ( $easy )) { case 4: $_GET [ '_u' ] = $_REQUEST [ '_u' ] = $easy [2] . '.' . $easy [3]; if (preg_match( '/^[\w\.]+$/' , $easy [1])) { $GLOBALS [ '_UCT' ][ 'TPL' ] = $easy [1]; } $_GET [ '_a' ] = $_REQUEST [ '_a' ] = $easy [0]; break ; case 3: $_GET [ '_u' ] = $_REQUEST [ '_u' ] = $easy [1] . '.' . $easy [2]; $_GET [ '_a' ] = $_REQUEST [ '_a' ] = $easy [0]; break ; case 2: $_GET [ '_u' ] = $_REQUEST [ '_u' ] = $easy [1]; $_GET [ '_a' ] = $_REQUEST [ '_a' ] = $easy [0]; break ; case 1: $_GET [ '_a' ] = $_REQUEST [ '_a' ] = $easy [0]; break ; default : exit ( 'invalid _easy param! ' . htmlspecialchars( $_REQUEST [ '_easy' ])); } } |
rewrite模式
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051 | //url重写模式 if (! empty ( $_REQUEST [ '_rewrite' ]) && is_string ( $_REQUEST [ '_rewrite' ])) { //1.支持apache 重写模式下?后的参数缺失的情况 if ( stripos ( $_SERVER [ 'SERVER_SOFTWARE' ], 'nginx' ) === false) { $_REQUEST [ '_rewrite' ] = urldecode( substr ( $_SERVER [ 'QUERY_STRING' ], strlen ( '_rewrite=' ))); } //2. 丢弃_rewrite中的后缀名 $rewrite = substr ( $_REQUEST [ '_rewrite' ], 0, strrpos ( $_REQUEST [ '_rewrite' ], '.' )); //3. 支持/作为分隔符 $sp = '.' ; for ( $i = 0; $i < strlen ( $rewrite ); $i ++) { if (in_array( $rewrite [ $i ], array ( '.' , '/' ))) { $sp = $rewrite [ $i ]; break ; } } $rewrite = explode ( $sp , $rewrite , 4); //最后1段是必填后缀名 switch ( count ( $rewrite )) { case 3: case 4: { $_GET [ '_a' ] = $_REQUEST [ '_a' ] = $rewrite [0]; $_GET [ '_u' ] = $_REQUEST [ '_u' ] = $rewrite [1]. '.' . $rewrite [2]; if (! empty ( $rewrite [3])) { if ( strpos ( $rewrite [3], '/' )) { $params = explode ( '/' , $rewrite [3]); for ( $i =0; $i +1< count ( $params ); $i +=2) { $_REQUEST [urldecode( $params [ $i ])] = urldecode( $params [ $i +1]); } } else { foreach ( explode ( '&' , $rewrite [3]) as $p ) { list( $k , $v ) = explode ( '=' , $p , 2); $_REQUEST [urldecode( $k )] = urldecode( $v ); } } } break ; } case 2: $_GET [ '_a' ] = $_REQUEST [ '_a' ] = $rewrite [0]; $_GET [ '_u' ] = $_REQUEST [ '_u' ] = $rewrite [1]; break ; case 1: $_GET [ '_a' ] = $_REQUEST [ '_a' ] = $rewrite [0]; break ; default : break ; } } |