ArangoDB 2.8: New Features and Enhancements

We welcome 2016 with our first big news yet – the release of ArangoDB 2.8!

Now you can use new AQL keywords to traverse a graph even more convenient – a big deal for those who like to get the maximum out of their connected data. ArangoDB is getting faster with every iteration, in this release we have implemented several AQL functions and arithmetic operations in super-fast C++ code, optimizer rules and indexing improved further to help you getting things done faster. Download ArangoDB 2.8 here.

Array Indexes

The added Array Indexes are a major improvement to ArangoDB that you will love and never want to miss again. Hash indexes and skiplist indexes can now be defined for array values as well, so it’s freaking fast to access documents by individual array values. Let assume you want to retrieve articles that are tagged with “graphdb”, you can now use an index on the tags array:

  { 
    text: "Here's what I want to retrieve...",
    tags: [ "graphdb", "ArangoDB", "multi-model" ] 
  }

An added hash-index on tags (ensureHashIndex("tags[*]")) can be used for finding all documents having "graphdb" somewhere in their tags array using the following AQL query:

  FOR doc IN documents 
    FILTER "graphdb" IN doc.tags[*] 
    RETURN doc

Have fun with these new indexes!

AQL Graph Traversal

Next, the mentioned AQL graph traversals. The query language AQL adds the keywords GRAPH, OUTBOUND, INBOUND and ANY for use in graph traversals. Using plain AQL in ArangoDB 2.8 you can create a shopping list for your friends birthday gifts, related to products they already own and up to 5 ideas ordered by price.

FOR friend IN OUTBOUND @me isFriendOf
  LET toBuy = (
  FOR bought IN OUTBOUND friend hasBought     
    FOR combinedProduct IN OUTBOUND bought combinedProducts
      SORT combinedProduct.price
      LIMIT 5
      RETURN combinedProduct
)
RETURN { friend, toBuy }

You can improve this list by limiting the result to friends that have a birthday within the next 2 months (assuming

birthday: "1970-01-15").

  LET maxDate = DATE_ADD(DATE_NOW(), 2, 'months')
  ...
  FILTER DATE_ISO8601(DATE_YEAR(DATE_NOW()),DATE_MONTH(friend.birthday),DATE_DAY(friend.birthday)) < maxDate 

Using Dave as a bind parameter for @me, we get the following result for our shopping tour:

[
  {
    "friend": {
      "name": "Julia",
      "_id": "users/Julia",
      "_rev": "1868379126",
      "_key": "Julia"
    },
    "toBuy": [
      {
        "price": 12,
        "name": "SanDisk Extreme SDHC UHS-I/U3 16GB Memory Card",
        "_id": "products/SanDisk16",
        "_rev": "2012820470",
        "_key": "SanDisk16"
      },
      {
        "price": 21,
        "name": "Lightweight Tripod 60-Inch with Bag",
        "_id": "products/Tripod",
        "_rev": "2003514358",
        "_key": "Tripod"
      },
      {
        "price": 99,
        "name": "Apple Pencil",
        "_id": "products/ApplePencil",
        "_rev": "2019177462",
        "_key": "ApplePencil"
      },
      {
        "price": 169,
        "name": "Smart Keyboard",
        "_id": "products/SmartKeyboard",
        "_rev": "2020160502",
        "_key": "SmartKeyboard"
      }
    ]
  },
  {
    "friend": {
      "name": "Debby",
      "city": "Dallas",
      "_id": "users/Debby",
      "_rev": "1928803318",
      "_key": "Debby"
    },
    "toBuy": [
      {
        "price": 12,
        "name": "Lixada Bag for Self Balancing Scooter",
        "_id": "products/LixadaScooterBag",
        "_rev": "2018194422",
        "_key": "LixadaScooterBag"
      }
    ]
  }
]

Usage of these new keywords as collection names, variable names or attribute names in AQL queries will not be possible without quoting. For example, the following AQL query will still work as it uses a quoted collection name and a quoted attribute name:

FOR doc IN `OUTBOUND`
  RETURN doc.`any`

Please have a look in the documentation for further details.

Syntax for managed graphs:

FOR vertex[, edge[, path]] IN MIN [..MAX] OUTBOUND|INBOUND|ANY startVertex GRAPH graphName

Working on collection sets:

FOR vertex[, edge[, path]] IN MIN[..MAX] OUTBOUND|INBOUND|ANY startVertex edgeCollection1, .., edgeCollectionN

AQL COLLECT … AGGREGATE

Additional, there is a cool new aggregation feature that was added after the beta releases. AQL introduces the keyword AGGREGATE for use in AQL COLLECT statements.

Using AGGREGATE allows more efficient aggregation (incrementally while building the groups) than previous versions of AQL, which built group aggregates afterwards from the total of all group values.

AGGREGATE can be used inside a COLLECT statement only. If used, it must follow the declaration of grouping keys:

FOR doc IN collection
  COLLECT gender = doc.gender AGGREGATE minAge = MIN(doc.age), maxAge = MAX(doc.age)
  RETURN { gender, minAge, maxAge }

or, if no grouping keys are used, it can follow the COLLECT keyword:

FOR doc IN collection
  COLLECT AGGREGATE minAge = MIN(doc.age), maxAge = MAX(doc.age)
  RETURN { minAge, maxAge }

Only specific expressions are allowed on the right-hand side of each AGGREGATE assignment:

• on the top level the expression must be a call to one of the supported aggregation functions LENGTH, MIN, MAX, SUM, AVERAGE, STDDEV_POPULATION, STDDEV_SAMPLE, VARIANCE_POPULATION, or VARIANCE_SAMPLE
• the expression must not refer to variables introduced in the COLLECT itself

Within the last weeks we have already published blog posts on several new features and enhancements in ArangoDB 2.8. So have a look at AQL function speedups, automatic deadlock detection (which is backported to 2.7.5 as well). The blog post about using multiple indexes per collection is worth to read, as well as the index speedups article. In the web interface you can now use bind parameters in the AQL editor.

There is a lot more to read in the changelog of ArangoDB 2.8 and we will proceed with the presentation of some features in detailed blog posts. You can find the latest documentation on docs.arangodb.com.