我有两个单独的集合,假设news
和pictures
。两个集合都有一个creationDate
字段,其中包含datetime
个记录创建时间。
我想要的是从两个集合中选择最后10条记录,加入所有记录(现在列表有20条记录),然后按creationDate
字段对所有记录进行排序。它是如何在Mongoose中实现的?
答案 0 :(得分:1)
你所要求的实际上是一个客户端操作,你真的只想要最多20个记录进行排序,这应该不是问题。
async.concat方法应该对此有所帮助:
async.concat([News,Picture],function(model,callback) {
// Get the last 10 results from each collection
var query = model.find({}).sort({ "creationDate": -1 }).limit(10);
query.exec(function(err,docs) {
if (err) throw err;
callback(err,docs);
});
},
function(err,result) {
if (err) throw err;
// results are merged, now sort by date
result = result.sort(function(a,b) {
return (a.creationDate < b.creationDate)
? 1 : (a.creationDate > b.creationDate) ? -1 : 0;
});
console.log(result);
});
所以你真的只是&#34;合并&#34;两个数组,但以这种方式编码简化了事情。
端到端列表以澄清此处的用法。这会在运行代码之前为每个模型设置一些文档,如上所示:
var async = require('async'),
mongoose = require('mongoose'),
Schema = mongoose.Schema;
mongoose.connect('mongodb://localhost/series');
var newsSchema = new Schema({
type: { type: String, default: "News" },
index: Number,
creationDate: { type: Date, default: Date.now }
});
var pictureSchema = new Schema({
type: { type: String, default: "Picture" },
index: Number,
creationDate: { type: Date, default: Date.now }
});
var News = mongoose.model( "News", newsSchema );
var Picture = mongoose.model( "Picture", pictureSchema );
async.series(
[
// Clean existing
function(callback) {
async.each([News,Picture],function(model,callback) {
model.remove(function(err) {
if (err) throw err;
callback();
});
},function(err) {
callback();
});
},
// Insert 20 of each
function(callback) {
console.log("inserting");
var count = 0;
async.whilst(
function() { return count < 20 },
function(callback) {
count++;
async.eachSeries([News,Picture],function(model,callback) {
var doc = new model({ index: count });
setTimeout(function() {
doc.save(function(err) {
if (err) throw err;
callback();
});
}, 20);
},function(err) {
callback();
});
},
function(err) {
callback();
}
);
}
],
function(err) {
console.log("listing");
// Get the last 10 of each
async.concat([News,Picture],function(model,callback) {
var query = model.find({}).sort({ "creationDate": -1 }).limit(10);
query.exec(function(err,docs) {
if (err) throw err;
callback(err,docs);
});
},
function(err,result) {
if (err) throw err;
// result is merged via "concat" now sort it.
result = result.sort(function(a,b) {
return (a.creationDate < b.creationDate)
? 1 : (a.creationDate > b.creationDate)
? -1 : 0;
});
console.log(result);
mongoose.disconnect();
});
}
);
输出就在这些方面:
[ { _id: 53c79f4b7daf2d676ff0f185,
index: 20,
__v: 0,
creationDate: Thu Jul 17 2014 20:02:51 GMT+1000 (EST),
type: 'Picture' },
{ _id: 53c79f4b7daf2d676ff0f184,
index: 20,
__v: 0,
creationDate: Thu Jul 17 2014 20:02:51 GMT+1000 (EST),
type: 'News' },
{ _id: 53c79f4b7daf2d676ff0f183,
index: 19,
__v: 0,
creationDate: Thu Jul 17 2014 20:02:51 GMT+1000 (EST),
type: 'Picture' },
{ _id: 53c79f4b7daf2d676ff0f182,
index: 19,
__v: 0,
creationDate: Thu Jul 17 2014 20:02:51 GMT+1000 (EST),
type: 'News' },
{ _id: 53c79f4b7daf2d676ff0f181,
index: 18,
__v: 0,
creationDate: Thu Jul 17 2014 20:02:51 GMT+1000 (EST),
type: 'Picture' },
{ _id: 53c79f4b7daf2d676ff0f180,
index: 18,
__v: 0,
creationDate: Thu Jul 17 2014 20:02:51 GMT+1000 (EST),
type: 'News' },
{ _id: 53c79f4b7daf2d676ff0f17f,
index: 17,
__v: 0,
creationDate: Thu Jul 17 2014 20:02:51 GMT+1000 (EST),
type: 'Picture' },
{ _id: 53c79f4b7daf2d676ff0f17e,
index: 17,
__v: 0,
creationDate: Thu Jul 17 2014 20:02:51 GMT+1000 (EST),
type: 'News' },
{ _id: 53c79f4b7daf2d676ff0f17d,
index: 16,
__v: 0,
creationDate: Thu Jul 17 2014 20:02:51 GMT+1000 (EST),
type: 'Picture' },
{ _id: 53c79f4b7daf2d676ff0f17c,
index: 16,
__v: 0,
creationDate: Thu Jul 17 2014 20:02:51 GMT+1000 (EST),
type: 'News' } ]