MongoDB查找符合条件的嵌套对象

本文介绍了MongoDB查找符合条件的嵌套对象的处理方法，对大家解决问题具有一定的参考价值，需要的朋友们下面随着小编来一起学习吧！ 问题描述

我有一个MongoDB文档，其结构类似于以下结构.我正在根据people.search_columns.surname和people.columns.givenname进行搜索.因此，例如，当我搜索给定的名称"Valentine"时，我想找回该文档，但不应包括Nicholas Barsaloux.

I have a MongoDB document that is structured similar to the structure below follows. I am searching based on people.search_columns.surname and people.columns.givenname. So for example, when I search for the given name of "Valentine", I want to get the document back, but Nicholas Barsaloux should not be included.

数据结构:

[_id] => MongoId Object ( [$id] => 53b1b1ab72f4f852140dbdc9 ) [name] => People From 1921 [people] => Array ( [0] => Array ( [name] => Barada, Valentine [search_columns] => Array ( [surname] => Array ( [0] => Mardan, [1] => Barada ) [givenname] => Array ( [0] => Valentine ) ) ) [1] => Array ( [name] => Barsaloux, Nicholas [search_columns] => Array ( [surname] => Array ( [1] => Barsaloux ) [givenname] => Array ( [0] => Nicholas ) [place] => Array ( ) ) ) )

这是我正在处理的代码:

Here is the code I was working on:

$criteria = array("people" => array( '$elemMatch' => array("givenname" => "Valentine") )); $projection = array("people" => true); $documents_with_results = $db->genealogical_data->find($criteria, $projection)->skip(0)->limit(5);

当前该代码返回零结果.

Currently that code returns zero results.

推荐答案

由于数组是嵌套的，因此无法像find一样使用基本投影.同样，为了从文档中过滤"数组内容，您需要首先展开"数组内容.为此，您可以使用聚合框架:

Since the arrays are nested you cannot use basic projection as you can with find. Also in order to "filter" the array contents from a document you need to "unwind" the array content first. For this you use the aggregation framework:

$results = $db->genealogical_data->aggregate(array( array( '$match' => array( 'people.search_columns.givenname' => 'Valentine' )), array( '$unwind' => '$people' ), array( '$match' => array( 'people.search_columns.givenname' => 'Valentine' )), array( '$group' => array( '_id' => '$id', 'name' => array( '$first' => '$name' ), 'people' => array( '$push' => '$people' ) )) ));

第一个 $match 的点阶段是减少可能符合您条件的文档.第二次是在 $unwind ，从结果中过滤"文档中的实际数组"项.

The point of the first $match stage is to reduce the documents that possibly match your criteria. The second time is done after the $unwind, where the actual "array" items in the document are "filtered" from the results.

最终的 $group 将原始数组恢复为正常，减去不符合条件的项目.

The final $group puts the original array back to normal, minus the items that do not match the criteria.