PHP智能识别收货地址信息实例


Posted in PHP onJanuary 05, 2019

功能需求:用户输入混合的收货地址,能智能识别出地址,手机,姓名

准备:需要两张表,一张地区表和一张姓氏表 (地区表得到应该不难,姓氏表我是搜索中国姓氏自制的哈,底部会附上表结构)

思路:主要思路分两种,一种是用户正常输入全地址,则顺序按地区等级匹配地址;另一种用户非正常输入(省市区有缺少的),则全面模糊搜索表,再根据结果对比原地址。

提醒:手机可以根据自己需求修改正则;

名字只匹配中文,可以根据自己的需求修改姓氏表以及正则

地址匹配暂无发现问题

效果图:

PHP智能识别收货地址信息实例

代码:

<?php
class DistinguishAddress {
/**
 * 类的入口方法
 * 传入地址信息自动识别,并返回最高匹配结果
 * 如果地址新增,则需要删除缓存文件重新缓存
 * @param $address
 **/
function getAddressResult($address){
 // 优先第一种方法
 $result = $this->getAddressArrar($address);
 // 如果结果不理想,再模糊去匹配
 if($result['level'] != 3){
  $result_sub = $this->addressVague($address);
  // 只有全匹配对才替换,否则不做任何改变
  if($result_sub['level'] == 3){
   $result = $result_sub;
  }
 }
 // 联系方式-优先匹配电话
 if(preg_match('/1\d{10}/', $address, $mobiles)){ // 手机
  $result['mobile'] = $mobiles[0];
 } else if(preg_match('/(\d{3,4}-)?\d{7,8}/', $address, $mobiles)){ // 固定电话
  $result['mobile'] = $mobiles[0];
 }
 // 识别姓名-必须空格分享的--概率
 preg_match_all('/[\x{4e00}-\x{9fa5}]{2,}/iu', $address,$names);
 if($names){
  $name_where = '';
  foreach ($names[0] as $name){
   // 必须是大于1个字符且小于5个字符的
   if(1 < mb_strlen($name,'utf-8') && mb_strlen($name, 'utf-8') < 5){
    $sub_name = mb_substr($name, 0, 1, 'utf-8');
    $name_where .= "name like '{$sub_name}%' or ";
   }
  }
  if(!empty($name_where)){
   $name_where = substr($name_where, 0, -3);
   $names_sql = "select name from surname where {$name_where} order by sort desc";
   $list = Db::getInstance('DbTrade')->getAll($names_sql);
   // 统计有多少种可能性-姓名
   $result['name_num'] = count($list);
   if($list) {
    $name_first = $list[0]['name'];
    foreach ($names[0] as $name){
     $len = mb_strlen($name_first, 'utf-8');
     if (mb_substr($name, 0, $len, 'utf-8') == $name_first){
      $result['name'] = $name;
     }
    }
   }
  }
 }
 // 去掉详细里面的姓名和电话
 $result['info'] = str_replace($result['mobile'], '', $result['info']);
 $result['info'] = str_replace($result['name'], '', $result['info']);
 $result['info'] = $result['province']['region_name'] . $result['city']['region_name'] . $result['district']['region_name'] . $result['info'];
 return $this->getCityLevelList($result);
}
/**
 * 获取对应城市等级列表
 **/
function getCityLevelList($result){
 // 获取所有地址递归列表
 $regions = $this->getRegionTreeList();
 // 获取省份列表- 只有存在值才返回对应列表
 $province_id = $result['province']['region_id'];
 if ($province_id) {
  foreach ($regions as $region){
   unset($region['childs']);
   $result['province_list'][] = $region;
  }
 }
 // 获取城市列表- 只有存在值才返回对应列表
 $city_id = $result['city']['region_id'];
 if ($city_id) {
  foreach ($regions[$province_id]['childs'] as $region){
   unset($region['childs']);
   $result['city_list'][] = $region;
  }
 }
 // 获取地区列表- 只有存在值才返回对应列表
 $district_id = $result['district']['region_id'];
 if ($district_id) {
  foreach ($regions[$province_id]['childs'][$city_id]['childs'] as $region){
   unset($region['childs']);
   $result['district_list'][] = $region;
  }
 }
 return $result;
}
/**
 * 获取所有地址递归列表
 **/
function getRegionTreeList(){
 // IO
 $file_name = 'regions.json';
 if(is_file($file_name)){
  $regions = file_get_contents($file_name);
  $regions = json_decode($regions, true);
 } else {
  $region_sql = "select region_id,region_name,parent_id from region";
  $regions = Db::getInstance('DbTrade')->getAll($region_sql);
  $regions = $this->arrayKey($regions);
  file_put_contents($file_name, json_encode($regions));
 }
 return $regions;
}
/**
 * 第一种方法
 * 根据地址列表递归查找准确地址
 * @param $address
 * @return array
 **/
function getAddressArrar($address){
 // 获取所有地址递归列表
 $regions = $this->getRegionTreeList();
 // 初始化数据
 $province = $city = $district = array();
 // 先查找省份-第一级地区
 $province = $this->checkAddress($address, $regions);
 if($province){
  // 查找城市-第二级地区
  $city = $this->checkAddress($address, $province['list']);
  if($city){
   // 查找地区-第三级地区
   // 西藏自治区那曲市色尼区辽宁南路西藏公路 第三个参数因为这个地址冲突取消强制
   $district = $this->checkAddress($address, $city['list']);
  }
 }
 return $this->getAddressInfo($address, $province, $city, $district);
}
 /**
  * 第二种方法
  * 地址模糊查找
  **/
function addressVague($address){
 $res = preg_match_all('/\S{2}[自市区镇县乡岛州]/iu', $address,$arr);
 if(!$res) return false;
 $where = ' where ';
 foreach ($arr[0] as $value){
  if(strpos($value, '小区') === false && strpos($value, '开发区') === false){
   $where .= "region_name like '%{$value}' or ";
  }
 }
 $where = substr($where,0,-3);
 $region_sql = "select region_id,region_name,parent_id,region_type from region " . $where;
 $citys = $GLOBALS['db']->getAll($region_sql);
 // 匹配所有地址
 $result = array();
 foreach ($citys as &$city){
  // 所有相关联的地区id
  $city_ids = array();
  if($city['region_type'] == 2) {
   $city_ids = array($city['parent_id'], $city['region_id']);
   // 尝试能不能匹配第三级
   $region_sql = "select region_id,region_name,parent_id,region_type,left(region_name,2) as ab_name from region where parent_id='{$city['region_id']}'" ;
   $areas = $GLOBALS['db']->getAll($region_sql);
   foreach ($areas as $row){
    if(mb_strpos($address,$row['ab_name'])){
     $city_ids[] = $row['region_id'];
    }
   }
  } else if($city['region_type'] == 3){
   $region_sql = "select parent_id from region where region_id='{$city['parent_id']}'" ;
   $city['province_id'] = $GLOBALS['db']->getOne($region_sql);
   $city_ids = array($city['parent_id'], $city['region_id'], $city['province_id']);
  }
  // 查找该单词所有相关的地区记录
  $where = " where region_id in(" . join(',', $city_ids) . ")";
  $region_sql = "select region_id,region_name,parent_id,region_type,left(region_name,2) as ab_name from region " . $where . ' order by region_id asc';
  $city_list = $GLOBALS['db']->getAll($region_sql);
  sort($city_ids);
  $key = array_pop($city_ids);
  $result[$key] = $city_list;
  sort($result);
 }
 if($result){
  list($province, $city, $area) = $result[0];
  return $this->getAddressInfo($address, $province, $city, $area);
 }
 return false;
}
/**
 * 匹配正确的城市地址
 * @param $address
 * @param $city_list
 * @param int $force
 * @param int $str_len
 * @return array
 **/
function checkAddress($address, $city_list, $force=false, $str_len=2){
 $num = 0;
 $list = array();
 $result = array();
 // 遍历所有可能存在的城市
 foreach ($city_list as $city_key=>$city){
  $city_name = mb_substr($city['region_name'], 0, $str_len,'utf-8');
  // 判断是否存包含当前地址字符
  $city_arr = explode($city_name, $address);
  // 如果存在相关字眼,保存该地址的所有子地址
  if(count($city_arr) >= 2){
   // 必须名称长度同时达到当前比对长度
   if(strlen($city['region_name']) < $str_len){
    continue;
   }
   $num ++;
   $list = $list + $city['childs'];

   $result[] = array(
    'region_id' => $city['region_id'],
    'region_name' => $city['region_name'],
    'list' =>$list,
   );
  }
 }
 // 如果有多个存在,则加大字符匹配长度
 if($num > 1 || $force){
  $region_name1 = $result[0]['region_name'];
  $region_name2 = $result[1]['region_name'];

  if(strlen($region_name1) == strlen($region_name2) && strlen($region_name1) == $str_len){
   $region_id1 = $result[0]['region_id'];
   $region_id2 = $result[1]['region_id'];
   $index = $region_id1 > $region_id2 ? 1 : 0;
   $result = $result[$index];
   return $result;
  }
  return $this->checkAddress($address, $city_list, $force, $str_len+1);
 } else {
  $result[0]['list'] = $list;
  return $result[0];
 }
}
/**
 * 根据原地址返回详细信息
 * @param $address
 * @param $province
 * @param $city
 * @param $area
 * @return array
 **/
function getAddressInfo($address, $province, $city, $district){
 // 查找最后出现的地址 - 截取详细信息
 $find_str = '';
 if($province['region_name']){
  $find_str = $province['region_name'];
  if($city['region_name']){
   $find_str = $city['region_name'];
   if($district['region_name']){
    $find_str = $district['region_name'];
   }
  }
 }
 // 截取详细的信息
 $find_str_len = mb_strlen($find_str,'utf-8');
 for($i=0; $i<$find_str_len-1; $i++){
  $substr = mb_substr($find_str,0,$find_str_len - $i, 'utf-8');
  $end_index = mb_strpos($address, $substr);
  if ($end_index){
   $address = mb_substr($address, $end_index + mb_strlen($substr) , mb_strlen($address) - $end_index);
  }
 }
 !empty($find_str) && $find_str = '|\S*' . $find_str;
 $area['info'] = preg_replace("/\s*|,|,|:|:{$find_str}/i", '', $address);
 $level = 0;
 if($district['region_name']){
  $level = 3;
 } else if($city['region_name']){
  $level = 2;
 } else if ($province['region_name']) {
  $level = 1;
 }
 return array(
  'province' => array('region_id'=>$province['region_id'], 'region_name'=>$province['region_name']),
  'city'  => array('region_id'=>$city['region_id'], 'region_name'=>$city['region_name']),
  'district'  => array('region_id'=>$district['region_id'], 'region_name'=>$district['region_name']),
  'info'  => $area['info'],
  'level'  => $level,
 );
}
/**
 * 递归所有地址成无限分类数组
 * @param $data
 * @param int $region_id
 * @return array
 **/
function arrayKey($data, $region_id=1){
 $result = array();
 foreach ($data as $row){
  if($region_id == $row['parent_id']){
   $key = $row['region_id'];
   $row['childs'] = $this->arrayKey($data, $row['region_id']);
   $result[$key] = $row;
  }
 }
 return $result;
}
}
?>

姓氏surname表(id,姓,优先匹配顺序)

DROP TABLE IF EXISTS `surname`;
CREATE TABLE `surname` (
 `id` int(11) NOT NULL AUTO_INCREMENT,
 `name` char(10) CHARACTER SET utf8 COLLATE utf8_general_ci NULL DEFAULT NULL,
 `sort` int(11) NULL DEFAULT NULL,
 PRIMARY KEY (`id`) USING BTREE,
 INDEX `name`(`name`) USING BTREE,
 INDEX `sort`(`sort`) USING BTREE
) ENGINE = InnoDB AUTO_INCREMENT = 481 CHARACTER SET = utf8 COLLATE = utf8_general_ci COMMENT = '姓氏表' ROW_FORMAT = Compact;

地址region表()

CREATE TABLE `region` (
 `region_id` smallint(5) UNSIGNED NOT NULL AUTO_INCREMENT,
 `parent_id` smallint(5) UNSIGNED NOT NULL DEFAULT 0,
 `region_name` varchar(120) CHARACTER SET utf8 COLLATE utf8_general_ci NOT NULL DEFAULT '',
 `region_type` tinyint(1) NOT NULL DEFAULT 2,
 `agency_id` smallint(5) UNSIGNED NOT NULL DEFAULT 0,
 PRIMARY KEY (`region_id`) USING BTREE,
 INDEX `parent_id`(`parent_id`) USING BTREE,
 INDEX `region_type`(`region_type`) USING BTREE,
) ENGINE = InnoDB AUTO_INCREMENT = 3956 AVG_ROW_LENGTH = 44 CHARACTER SET = utf8 COLLATE = utf8_general_ci ROW_FORMAT = Compact;

总结

以上就是这篇文章的全部内容了,希望本文的内容对大家的学习或者工作具有一定的参考学习价值,谢谢大家对三水点靠木的支持。如果你想了解更多相关内容请查看下面相关链接

PHP 相关文章推荐
PHP5 面向对象程序设计
Feb 13 PHP
色色整理的PHP面试题集锦
Mar 08 PHP
利用浏览器的Javascript控制台调试PHP程序
Jan 08 PHP
Smarty变量调节器失效的解决办法
Aug 20 PHP
ThinkPHP无限级分类原理实现留言与回复功能实例
Oct 31 PHP
linux下php上传文件注意事项
Jun 11 PHP
PHP时间戳格式全部汇总 (获取时间、时间戳)
Jun 13 PHP
PHP常用的三种设计模式汇总
Aug 28 PHP
PHP的mysqli_set_charset()函数讲解
Jan 23 PHP
Codeigniter里的无刷新上传的实现代码
Apr 14 PHP
PHP中散列密码的安全性分析
Jul 26 PHP
tp5框架前台无限极导航菜单类实现方法分析
Mar 29 PHP
PHP数字金额转换成中文大写显示
Jan 05 #PHP
PHP yield关键字功能与用法分析
Jan 03 #PHP
PHP获取对象属性的三种方法实例分析
Jan 03 #PHP
PHP获取HTTP body内容的方法
Dec 31 #PHP
php两点地理坐标距离的计算方法
Dec 29 #PHP
php如何计算两坐标点之间的距离
Dec 29 #PHP
PHP命名空间简单用法示例
Dec 28 #PHP
You might like
实现树状结构的两种方法
2006/10/09 PHP
用PHP+MySql编写聊天室
2006/10/09 PHP
Linux下编译redis和phpredis的方法
2016/04/07 PHP
php curl获取到json对象并转成数组array的方法
2018/05/31 PHP
JQuery表格拖动调整列宽效果(自己动手写的)
2014/09/01 Javascript
angularjs的一些优化小技巧
2014/12/06 Javascript
javascript实现获取字符串hash值
2015/05/10 Javascript
jquery实现的树形目录实例
2015/06/26 Javascript
JavaScript比较当前时间是否在指定时间段内的方法
2016/08/02 Javascript
JavaScript 深层克隆对象详解及实例
2016/11/03 Javascript
Angular2 路由问题修复详解
2017/03/01 Javascript
如何编写jquery插件
2017/03/29 jQuery
ionic+AngularJs实现获取验证码倒计时按钮
2017/04/22 Javascript
JavaScript模拟文件拖选框样式v1.0的实例
2017/08/04 Javascript
ES6使用Set数据结构实现数组的交集、并集、差集功能示例
2017/10/31 Javascript
node简单实现一个更改头像功能的示例
2017/12/29 Javascript
基于JavaScript实现简单的音频播放功能
2018/01/07 Javascript
深入浅出vue图片路径的实现
2019/09/04 Javascript
浅谈监听单选框radio改变事件(和layui中单选按钮改变事件)
2019/09/10 Javascript
jquery实现垂直手风琴菜单
2020/03/04 jQuery
微信公众号中的JSSDK接入及invalid signature等常见错误问题分析(全面解析)
2020/04/11 Javascript
Python实现模拟登录及表单提交的方法
2015/07/25 Python
十条建议帮你提高Python编程效率
2016/02/16 Python
用Python3创建httpServer的简单方法
2018/06/04 Python
python求质数的3种方法
2018/09/28 Python
Python文件读写w+和r+区别解析
2020/03/26 Python
Python scrapy爬取小说代码案例详解
2020/07/09 Python
html5使用canvas实现弹幕功能示例
2017/09/11 HTML / CSS
实习生自荐信范文分享
2013/11/27 职场文书
财务专业大学生职业生涯规划范文
2013/12/30 职场文书
促销活动策划方案
2014/01/12 职场文书
毕业生如何写自荐信
2014/03/26 职场文书
学生保证书范文
2014/04/28 职场文书
学习演讲稿范文
2014/05/10 职场文书
MySQL 重写查询语句的三种策略
2021/05/10 MySQL
使用Redis实现实时排行榜功能
2021/07/02 Redis