- db.help()
- db.COLLCECTION_NAME.find(**).count()
- db.stats()
- show statistics of current database
- help
- help keys
- db
- show current database name
- db.COLLCECTION_NAME.findOne()
- Returns one document that satisfies the specified query criteria. If multiple documents satisfy the query, this method returns the first document according to the natural order
- avoid print result, add null; after the statement;
- > use mydb; > db.dropDatabase();
- mongoimport –type csv –headerline weather.csv -d weather -c data
- db.example.update({},{$unset: {words:1}},false,true)
- delete word field from doc db.collection.update( criteria, objNew, upsert, multi )
- totalIndex
- Define document’s style, like struct in C link
- name the schema in databse, like typedef, can pre-initialize value. link
- statics
- unique index
- save VS insert
save insert or update a document. insert does only an insertion
- get result from mongoose query
mongoose promise
- search more than one field : User.find({name:’wil’, score: 80})
- Remove documents using Nodejs
- FindOneAndRemove
- MyModel.remove({search: criteria}, function() {}
- one single document should be less than 16M(BSON document size)
- upsert (if exists, update, else create)
- enable upsert by pass {upsert:true} in update
- denormalization (in RDBS) improve database reading performance by add redunct data
- durability http://www.techopedia.com/definition/27416/durability-databases
- Capped Collection are fixed-size collections, once a collection fills its allocated space, it makes room for new documents by overwriting the oldest documents in the collection.
- db is the variable that references the current database. The variable is automatically set to the default database test or is set when you use the use <db> to switch current database.
- Cursor: A pointer to the result set of a query. Clients can iterate through a cursor to retrieve results.
- GridFS
- you cannot have parrael array for multi-key
- compound index : used in sort and find, leftmost
for example, {a:1, b:1, c:1}, find({a:1}), find({a:1, b:1}) use compound index, find({b:1, c:1}) not use, but find({b:1,c:1}).sort({a:1,b:1}) use it
- Index Selectivity: 最有区分度的，变化最多的，作为选择
- hint : Tell mongodb which index to use
- find and sort canot use sparse index(就是不是所有document都有的index)
- _id can be a document
- NumberInt
- NumberLong
- new Date()
- mongodb comparision will not cross types
- $exists
- find whether a field is exist
- $type
- specify whether the value match a type
- $regex: Perl style regex (optimization problem)
- Dot Notation, find a part of document
- $group
- $match
- $project
- $pop:{array_name: -1} :: pop from left
- db.COLLCECTION_NAME.drop() is more efficient than db.COLLCECTION_NAME.remove({})
- save :: if this doc _id exists, update, otherwise, create one.
- db.data.find({“Wind Direction”: {$gte: 180, $lte: 360}}, {“State”:1, “Temperature”:1}).sort({“Temperature”:1}).limit(1)
- db.students.update({}, {$push: {scores: { $each: [], $sort:{type: 1, score: 1} }}}, {multi: true})
- db.students.update({}, {$unset:{‘scores.1’:1}}, {multi:true})
- db.students.update({}, {$pull:{‘scores’:null}}, {multi:true})
- homepage: db.posts.ensureIndex({ date: -1})
- perlink: db.posts.ensureIndex({ permalink: 1}, {unique: true})
- pertag: db.posts.ensureIndex({ tags: 1})
- db.profile.find({op: ‘query’, ns: ‘school2.students’}, {millis: 1, _id: 0}).sort({millis:-1}).limit(1)
- db.messages.ensureIndex({From:1, To:1})
- db.messages.find({ “headers.From”: “[email protected]”, “headers.To”: “[email protected]”}).count()
-
- B Tree
- document-level locking 2.8
- aggregation framework
- sort memory, limit
- unwind will generate same _id
- assert: command failed: {
“errmsg” : “exception: Exceeded memory limit for $group, but didn’t allow external sort. Pass allowDiskUse:true to opt in.”, “code” : 16945, “ok” : 0 } : aggregate failed
- index can be use for sort
- index has to be maintain, will slow the insert
- w = 0, j = 0