Querying the Database

Time to retrieve our document using AQL, ArangoDB’s query language. We candirectly look up the document we created via the id, but there are alsoother options. Click the _QUERIES menu entry to bring up the query editorand type the following (adjust the document ID to match your document):

  1. RETURN DOCUMENT("users/9883")

Then click Execute to run the query. The result appears below the query editor:

  1. [
  2. {
  3. "_key": "9883",
  4. "_id": "users/9883",
  5. "_rev": "9883",
  6. "age": 32,
  7. "name": "John Smith"
  8. }
  9. ]

As you can see, the entire document including the system attributes is returned.DOCUMENT() is a function to retrievea single document or a list of documents of which you know the _keys or _ids.We return the result of the function call as our query result, which is ourdocument inside of the result array (we could have returned more than one resultwith a different query, but even for a single document as result, we still getan array at the top level).

This type of query is called data access query. No data is created, changed ordeleted. There is another type of query called data modification query. Let’sinsert a second document using a modification query:

  1. INSERT { name: "Katie Foster", age: 27 } INTO users

The query is pretty self-explanatory: the INSERT keyword tells ArangoDB thatwe want to insert something. What to insert, a document with two attributes inthis case, follows next. The curly braces { } signify documents, or objects.When talking about records in a collection, we call them documents. Encoded asJSON, we call them objects. Objects can also be nested. Here’s an example:

  1. {
  2. "name": {
  3. "first": "Katie",
  4. "last": "Foster"
  5. }
  6. }

INTO is a mandatory part of every INSERT operation and is followed by thecollection name that we want to store the document in. Note that there are noquote marks around the collection name.

If you run above query, there will be an empty array as result because we didnot specify what to return using a RETURN keyword. It is optional inmodification queries, but mandatory in data access queries. Even with RETURN,the return value can still be an empty array, e.g. if the specified documentwas not found. Despite the empty result, the above query still created a newuser document. You can verify this with the document browser.

Let’s add another user, but return the newly created document this time:

  1. INSERT { name: "James Hendrix", age: 69 } INTO users
  2. RETURN NEW

NEW is a pseudo-variable, which refers to the document created by INSERT.The result of the query will look like this:

  1. [
  2. {
  3. "_key": "10074",
  4. "_id": "users/10074",
  5. "_rev": "10074",
  6. "age": 69,
  7. "name": "James Hendrix"
  8. }
  9. ]

Now that we have 3 users in our collection, how to retrieve them all with asingle query? The following does not work:

  1. RETURN DOCUMENT("users/9883")
  2. RETURN DOCUMENT("users/9915")
  3. RETURN DOCUMENT("users/10074")

There can only be a single RETURN statement here and a syntax error is raisedif you try to execute it. The DOCUMENT() function offers a secondary signatureto specify multiple document handles, so we could do:

  1. RETURN DOCUMENT( ["users/9883", "users/9915", "users/10074"] )

An array with the _ids of all 3 documents is passed to the function. Arraysare denoted by square brackets [ ] and their elements are separated by commas.

But what if we add more users? We would have to change the query to retrievethe newly added users as well. All we want to say with our query is: “For everyuser in the collection users, return the user document”. We can formulate thiswith a FOR loop:

  1. FOR user IN users
  2. RETURN user

It expresses to iterate over every document in users and to use user asvariable name, which we can use to refer to the current user document. It couldalso be called doc, u or ahuacatlguacamole, this is up to you. It isadvisable to use a short and self-descriptive name however.

The loop body tells the system to return the value of the variable user,which is a single user document. All user documents are returned this way:

  1. [
  2. {
  3. "_key": "9915",
  4. "_id": "users/9915",
  5. "_rev": "9915",
  6. "age": 27,
  7. "name": "Katie Foster"
  8. },
  9. {
  10. "_key": "9883",
  11. "_id": "users/9883",
  12. "_rev": "9883",
  13. "age": 32,
  14. "name": "John Smith"
  15. },
  16. {
  17. "_key": "10074",
  18. "_id": "users/10074",
  19. "_rev": "10074",
  20. "age": 69,
  21. "name": "James Hendrix"
  22. }
  23. ]

You may have noticed that the order of the returned documents is not necessarilythe same as they were inserted. There is no order guaranteed unless you explicitlysort them. We can add a SORT operation very easily:

  1. FOR user IN users
  2. SORT user._key
  3. RETURN user

This does still not return the desired result: James (10074) is returned beforeJohn (9883) and Katie (9915). The reason is that the _key attribute is a stringin ArangoDB, and not a number. The individual characters of the strings arecompared. 1 is lower than 9 and the result is therefore “correct”. If wewanted to use the numerical value of the _key attributes instead, we couldconvert the string to a number and use it for sorting. There are some implicationshowever. We are better off sorting something else. How about the age, in descendingorder?

  1. FOR user IN users
  2. SORT user.age DESC
  3. RETURN user

The users will be returned in the following order: James (69), John (32), Katie(27). Instead of DESC for descending order, ASC can be used for ascendingorder. ASC is the default though and can be omitted.

We might want to limit the result set to a subset of users, based on the ageattribute for example. Let’s return users older than 30 only:

  1. FOR user IN users
  2. FILTER user.age > 30
  3. SORT user.age
  4. RETURN user

This will return John and James (in this order). Katie’s age attribute does notfulfill the criterion (greater than 30), she is only 27 and therefore not partof the result set. We can make her age to return her user document again, usinga modification query:

  1. UPDATE "9915" WITH { age: 40 } IN users
  2. RETURN NEW

UPDATE allows to partially edit an existing document. There is also REPLACE,which would remove all attributes (except for _key and _id, which remain thesame) and only add the specified ones. UPDATE on the other hand only replacesthe specified attributes and keeps everything else as-is.

The UPDATE keyword is followed by the document key (or a document / objectwith a _key attribute) to identify what to modify. The attributes to updateare written as object after the WITH keyword. IN denotes in which collectionto perform this operation in, just like INTO (both keywords are actuallyinterchangeable here). The full document with the changes applied is returnedif we use the NEW pseudo-variable:

  1. [
  2. {
  3. "_key": "9915",
  4. "_id": "users/9915",
  5. "_rev": "12864",
  6. "age": 40,
  7. "name": "Katie Foster"
  8. }
  9. ]

If we used REPLACE instead, the name attribute would be gone. With UPDATE,the attribute is kept (the same would apply to additional attributes if we hadthem).

Let us run our FILTER query again, but only return the user names this time:

  1. FOR user IN users
  2. FILTER user.age > 30
  3. SORT user.age
  4. RETURN user.name

This will return the names of all 3 users:

  1. [
  2. "John Smith",
  3. "Katie Foster",
  4. "James Hendrix"
  5. ]

It is called a projection if only a subset of attributes is returned. Anotherkind of projection is to change the structure of the results:

  1. FOR user IN users
  2. RETURN { userName: user.name, age: user.age }

The query defines the output format for every user document. The user name isreturned as userName instead of name, the age keeps the attribute key inthis example:

  1. [
  2. {
  3. "userName": "James Hendrix",
  4. "age": 69
  5. },
  6. {
  7. "userName": "John Smith",
  8. "age": 32
  9. },
  10. {
  11. "userName": "Katie Foster",
  12. "age": 40
  13. }
  14. ]

It is also possible to compute new values, for example by concatenation:

  1. FOR user IN users
  2. RETURN CONCAT(user.name, "'s age is ", user.age)

CONCAT() is a function that can join elements together to a string. We use ithere to return a statement for every user. As you can see, the result set doesnot always have to be an array of objects:

  1. [
  2. "James Hendrix's age is 69",
  3. "John Smith's age is 32",
  4. "Katie Foster's age is 40"
  5. ]

Now let’s do something crazy: for every document in the users collection,iterate over all user documents again and return user pairs, e.g. John and Katie.We can use a loop inside a loop for this to get the cross product (every possiblecombination of all user records, 3 * 3 = 9). We don’t want pairings like John +John however, so let’s eliminate them with a filter condition:

  1. FOR user1 IN users
  2. FOR user2 IN users
  3. FILTER user1 != user2
  4. RETURN [user1.name, user2.name]

We get 6 pairings. Pairs like James + John and John + James are basicallyredundant, but fair enough:

  1. [
  2. [ "James Hendrix", "John Smith" ],
  3. [ "James Hendrix", "Katie Foster" ],
  4. [ "John Smith", "James Hendrix" ],
  5. [ "John Smith", "Katie Foster" ],
  6. [ "Katie Foster", "James Hendrix" ],
  7. [ "Katie Foster", "John Smith" ]
  8. ]

We could calculate the sum of both ages and compute something new this way:

  1. FOR user1 IN users
  2. FOR user2 IN users
  3. FILTER user1 != user2
  4. RETURN {
  5. pair: [user1.name, user2.name],
  6. sumOfAges: user1.age + user2.age
  7. }

We introduce a new attribute sumOfAges and add up both ages for the value:

  1. [
  2. {
  3. "pair": [ "James Hendrix", "John Smith" ],
  4. "sumOfAges": 101
  5. },
  6. {
  7. "pair": [ "James Hendrix", "Katie Foster" ],
  8. "sumOfAges": 109
  9. },
  10. {
  11. "pair": [ "John Smith", "James Hendrix" ],
  12. "sumOfAges": 101
  13. },
  14. {
  15. "pair": [ "John Smith", "Katie Foster" ],
  16. "sumOfAges": 72
  17. },
  18. {
  19. "pair": [ "Katie Foster", "James Hendrix" ],
  20. "sumOfAges": 109
  21. },
  22. {
  23. "pair": [ "Katie Foster", "John Smith" ],
  24. "sumOfAges": 72
  25. }
  26. ]

If we wanted to post-filter on the new attribute to only return pairs with asum less than 100, we should define a variable to temporarily store the sum,so that we can use it in a FILTER statement as well as in the RETURNstatement:

  1. FOR user1 IN users
  2. FOR user2 IN users
  3. FILTER user1 != user2
  4. LET sumOfAges = user1.age + user2.age
  5. FILTER sumOfAges < 100
  6. RETURN {
  7. pair: [user1.name, user2.name],
  8. sumOfAges: sumOfAges
  9. }

The LET keyword is followed by the designated variable name (sumOfAges),then there’s a = symbol and the value or an expression to define what valuethe variable is supposed to have. We re-use our expression to calculate thesum here. We then have another FILTER to skip the unwanted pairings andmake use of the variable we declared before. We return a projection with anarray of the user names and the calculated age, for which we use the variableagain:

  1. [
  2. {
  3. "pair": [ "John Smith", "Katie Foster" ],
  4. "sumOfAges": 72
  5. },
  6. {
  7. "pair": [ "Katie Foster", "John Smith" ],
  8. "sumOfAges": 72
  9. }
  10. ]

Pro tip: when defining objects, if the desired attribute key and the variableto use for the attribute value are the same, you can use a shorthand notation:{ sumOfAges } instead of { sumOfAges: sumOfAges }.

Finally, let’s delete one of the user documents:

  1. REMOVE "9883" IN users

It deletes the user John (_key: "9883"). We could also remove documents in aloop (same goes for INSERT, UPDATE and REPLACE):

  1. FOR user IN users
  2. FILTER user.age >= 30
  3. REMOVE user IN users

The query deletes all users whose age is greater than or equal to 30.