•

modifying documents

•

API for updating can update in various ways:

•

wholesale replacement

•

replacement of fields

•

upserts

•

updates take two arguments

•

everything you learned about find is applicable to updates

•

wholesale replacement

•

•

warning: the other fields will be deleted, it disregards what the datastore has and saves only what the application has

•

individual field replacement

•

need to add an age field

•

you need to know all the other fields

•

$set will add or modify a field

•

$inc will increment field

•

if doesn't have this field, then it will create it with the increment step

•

removing fields

•

•

manipulating arrays in documents

•

to change value, "a.2" = third element

•

$push adds to end of array

•

$pop to add to end or beginning

•

also $pushAll will add an array

•

$pull will remove the value from array no matter where it is

•

$pullAll pulls all values in an array from document array

•

$addToSet if value exists, it does nothing, if not, then it does a push

•

upserts

•

update document or, if does not exist, create a document

•

commonly used, e.g. merging data in from a data vendor

•

more than one update at a time

•

empty document matches all, but you need the option multi : true, otherwise it will only update one, and it is difficult to predict which one it will update

•

yielding means that concurrent write operations pause for other writes to take place, but no write operating will see a document half-updated

•

update all sores less tha 70 by 20

•

`db.scores.update( { 'score' : { $lt : 70 }}, {$inc : {score : 20}}, {multi:true})

•

deleting

•

delete one or many

•

if you specify an empty document, it will remove all (!)

•

•

each individual remove operation is atomic, no remove operation will see another operation have finished

•

drop collection

•

•

to find out if command succeeded

•

•

will show you if information about commands succeeded and failed

•

node.js driver and CRUD operations

•

we have been using the Mongo shell

•

first we need to import data

•

•

example using findOne with the node.js driver

•

then node app.js to run it

•

notice the _id looks different as a JSON object as when we see it in the shell

•

to get multiple documents, use toArray

•

you can also get an array with a cursor, you get the cursor which does not have the array in it, you just call .each on an object that describes your query

•

you could also call .toArray on the object

•

the response from MongoDB is not the entire result set, only a certain batch size

•

find using field projection

•

you send in the project as a second argument

•

this way we are only sending data over the wire that we care about

•

with greater than and less then

•

dollar sign signifies a query operator

•

importing from Reddit

•

we will analyze data without necessarily knowing the structure of the data

•

in reddit, if you add /.json at the end you get a json document (actually didn't work)

•

tbe require('request') is to get data from other sites

•

regex queries via node

•

search for any document that has a title which contains "NSA"

•

•

then project out part of it

•

•

full code to print out these titles

•

dot notation in MongoDB queries

•

how to find embedded nodes in a JSON document

•

note that you can make queries even though you don't know if all the fields exist in the documents

•

sort, skip, limit (in that order)

•

the order will alway be in that order no matter how you add them in the code

•

sort with two fields

•

the order that they are in the array is the order they will sort

•

this is why you have an array instead of an object, since properties on an object do not have a guaranteed order

•

you can also use an option object:

•

inserting

•

insert is straightforward:

•

you can also store an a name in the _id field, which causes he driver not to generate it, but if you try to add it twice, you will get an error

•

you can also add numerous documents:

•

in MongoDB keys are case-sensitive so here two records will be inserted:

•

update

•

to update you need to get the document you want, change it, and then update it

•

notice we are doing two queries: between the two queries the document might have changed, you have to use the _id in the second query to make sure it is the same one

•

update with $set

•

multiple updates

•

upserts

•

how to upsert:

•

upserts can also be used with $set

•

notice $set sets two fields

•

save

•

find original document

•

the save function

•

findAndModify

•

atomically updates so that there isn't a wait period between finding and modifying

•

example to increment counter

•

we need sort to give us more control over which document we are updating, since there may be multiple documents, but if not, then leave sort empty

•

options specifies to pass the new, not the original, document back

•

remove

•

straightforward:

•

building a blog with node.js

•

starts node.js as web server

•

•

shows blog project

•

logs in / logs out

•

creates entry

•

creates comment

•

building a blog on node.js

•

app.js

•

requires

•

routes = require('./routes')

•

local file

•

UsersDAO and SessinosDAO are used to access MongoDB