Skip to content

Standing Queries

A standing query is a feature unique in Quine to incrementally MATCH a subgraph as new data enters the graph. Then, when a full subgraph matches, the result is further processed to modify the graph, produce output as a data source, or take additional action.

Standing Query Structure

Standing queries have two parts: a pattern query and outputs destinations. The pattern query defines the structure of what we're looking for, and output destinations specify actions for each result produced.

Consider the pattern query as a coarse filter that is specific enough to MATCH a model event but not so specific that it would match a unique event. A query in an output destination can process the result further with a more expressive Cypher query to then update the graph, act as an event source, generate metrics, and much more.

Cypher within a pattern query must only contain a MATCH and a RETURN, with an optional WHERE, while Cypher within output destinations is unconstrained, allowing for more expressive queries.

Standing Query Model

As described in the Create Standing Query API documentation, a standing query is created via POST to the /api/v1/query/standing/{standing-query-name} endpoint.

POST /api/v1/query/standing/STANDING-1
{
  "pattern": {
    "type": "Cypher",
    "query": "MATCH (n) RETURN DISTINCT id(n) AS id"
  },
  "outputs": {
    "destination": {
      "type": "CypherQuery",
      "query": "MATCH (n) WHERE id(n) = $that.data.id RETURN n.line",
      "andThen": {
        "type": "PrintToStandardOut"
      }
    }
  }
}

Or in YAML if you are writing a recipe.

standingQueries:
  - pattern:
      type: Cypher
      query: MATCH (n) RETURN DISTINCT id(n) AS id
    outputs:
      destination:
        type: CypherQuery
        query: >-
          MATCH (n)
          WHERE id(n) = $that.data.id
          RETURN n.line
        andThen:
          type: PrintToStandardOut

This structure ensures that the set of positive matches, minus the set of negative matches (also referred to as cancellations) produced by a standing query are like the results produced if the same Cypher query had been issued in a batch fashion after all data has been written into the graph.

Pattern Match Query

The pattern query in a standing query is a declarative graph pattern expressed using a subset of the Cypher query language containing a MATCH and a RETURN, with an optional WHERE.

Distinct ID Pattern Queries

Quine has two modes available for writing pattern queries, DistinctId, and MultipleValues. The mode is set to DistinctId by default unless you explicitly set the mode to MultipleValues within your pattern query. Quine can process standing queries that contain either mode within the same runtime or recipe.

The following constraints apply to Cypher contained in the pattern query string when mode is set to the default DistinctId:

  • Each node identified by the MATCH shall have the following:
  • Node variable name
  • Label (optional but not more than one)
  • Optional map of literal property values to match
  • Nodes in the MATCH must form a connected graph.
  • Nodes in the MATCH must not contain any cycles. In other words, the pattern must be either linear or tree-shaped.
  • Only node variables can be bound in the query MATCH. Edges cannot be aliased to a variable, and path expressions cannot be used (so -[:HAS_FATHER]-> is fine, but -[e:HAS_FATHER]-> is not).
  • Edges in the MATCH must be directed, have exactly one edge label, and cannot be variable-length.
  • Constraints inside the WHERE clause must be AND-ed together and of one of the following forms:

    • `nodeName.property = 1* - the property has the literal value on the right
    • `nodeName.property <> 1* - the property must exist but be different than the literal value on the right
    • nodeName.property IS NOT NULL - the property must exist
    • nodeName.property IS NULL - the property must not exist
    • nodeName.property =~ "regex" - the property must be a string matching the regex
    • `id(nodeName) = 12* - the ID of the node must be exactly the literal value on the right
    • id(nodeName) = idFrom('values', 'to', 'hash') - the ID of the node must match exactly the idFrom() computed from the literal values on the right
  • Exactly one value must be returned, and it must be either the DISTINCT id or strId of a node bound in the MATCH.

For example, RETURN DISTINCT strId(n) or RETURN DISTINCT id(n) as nId are OK, but not RETURN n.name or RETURN id(n) AS nId. The node whose id is returned is the root node - the location in the graph from which the pattern starts being incrementally matched.

MultipleValue Pattern Queries (Beta)

Beta Feature

This feature is in the beta phase of development.

Pattern queries with MultipleValues mode could:

  • Require syntax changes when the feature releases as GA
  • Use more RAM
  • Consume more disk space
  • Differ in performance from Distinct ID pattern query queries

MultipleValue mode pattern query queries relax some of the constraints imposed by DistinctId. In particular, the WHERE and RETURN portions of the query allow Cypher expressions to be much more expressive.

The syntax and structure of this mode is designed to supersede the DistinctId mode. Thus, any DistinctId standing query pattern is a valid MultipleValues standing query, though not the other way around.

The MultipleValues mode retains the MATCH - WHERE - RETURN shape from DistinctId mode with the addition of the constraints below.

  • Any number of results (not just one) can be returned in the RETURN, including results that aren't node IDs
  • Constraints in the WHERE clause are reduced
  • DISTINCT is required for DistinctId standing queries, but the MultipleValues mode does not support DISTINCT return values.
  • Use of variables must represent a node
  • Variable usage is ok when dereferencing node properties. For example RETURN n.name is ok but RETURN n is not.
  • WHERE and RETURN support id(n) and strId(n) but other functions are not supported.

The MATCH portion of standing queries using the MultipleValues mode removes the syntactic requirements for running in DistinctId mode with two exceptions:

  • Multiple IDs and property values from matched nodes can be returned by RETURN. For example, RETURN n.age + * strId(n) + " " + m.name is fine, but RETURN properties(n) is not.

  • Constraints in the WHERE clause must be defined in the IDs and properties of matched nodes and can not include sub-queries or procedures.

  • Can not MATCH variable length patterns
  • MATCH does not support pattern expressions

Since there isn't exactly one ID being returned, the root of the standing query pattern (the place in the pattern from which incremental matching starts) is instead set to be the first node in the MATCH pattern. This makes it possible to make any node in the pattern the "root".

Pattern Match Results

Both modes for the pattern query return a StandingQueryResult JSON object with meta and data sub-objects.

The meta JSON sub-object consists of the following:

  • isPositiveMatch: whether the result is a new match. When this value is false, it signifies that a previously matched result no longer matches

  • resultId: a UUID generated for each result. This is useful if you wish to track a result in some external system since the resultId of the result with isPositiveMatch = false will match the resultId of the original result (when isPositiveMatch = true).

The data JSON sub-object consists of the following:

  • On a positive match, the data JSON object contains results returned by the pattern query.
  • This objects keys are the names of the values returned (ex: RETURN DISTINCT strId(n) would have key "strId(n)" and RETURN DISTINCT id(n) AS theId would have key "theId").
  • Each query data returned is analogous to a row returned from a regular Cypher query - the key names match what would normally be Cypher column names.

When DistinctId mode is set, a result is emitted when a complete pattern matches or stops matching, but additional results won't be emitted if there are interim new complete pattern matches.

Example of single-result per root semantics

Consider the following query for watching friends.

// Find people with friends
MATCH (n:Person)-[:friend]->(m:Person)
RETURN DISTINCT strId(n)

If we start by creating disconnected "Peter", "John", and "James" nodes, there will be no matches.

CREATE (:Person { name: "Peter" }),
      (:Person { name: "John" }),
      (:Person { name: "James" })

Then, if we add a "friend" edge from "Peter" to "John", "Peter" will trigger a new standing query match.

MATCH (peter:Person { name: "Peter" }), (john:Person { name: "John" })
CREATE (peter)-[:friend]->(john)

However, adding a second "friend" edge from "Peter" to "James", "Peter" will not trigger a new match since he is already matching (that is, the "Peter" node is not distinct).

MATCH (peter:Person { name: "Peter" }), (james:Person { name: "James" })
CREATE (peter)-[:friend]->(james)

Note, unlike DistinctId mode queries, MultipleValues mode pattern query results can be emitted from each root node. This means that the "Find people with friends" example, if run in the MultipleValues mode, would produce two results (one for each friend) unlike the single result produced in the DistinctId mode.

Sample StandingQueryResult:

{
    "meta": {
        "resultId": "b3c35fa4-2515-442c-8a6a-35a3cb0caf6b",
        "isPositiveMatch": true
    },
    "data": {
        "strId(n)": "a0f93a88-ecc8-4bd5-b9ba-faa6e9c5f95d"
    }
}

Result Outputs

Once a full pattern match occurs, a StandingQueryResult is produced. A standing query can have any number of output destinations to route StandingQueryResults. The output destinations are processed in parallel.

Cypher Query

The Cypher query is particularly powerful, making it possible to post-process pattern query results to collect more information from the graph or to filter out matches that don't meet some requirements.

The StandingQueryResult object is passed the standing query output via the parameter $that, for use in query Cypher.

When running a Cypher query as the standing query output, the user can pass in another output step using the andThen field. This field can include any output options described here, even additional Cypher Query steps.

Be aware that non-trivial or long-running operations with StandingQueryResults will consume system resources and cause the system to backpressure and slow down other processing (like data ingest).

Drop

Drop the current StandingQueryResult output and end processing the destination.

POST to Webhook

Makes an HTTP[S] POST for each StandingQueryResult. The data in the request payload can be customized in a Cypher query preceding this step.

Publish to Slack

Sends a message to Slack via a configured Slack App webhook URL. See https://api.slack.com/messaging/webhooks.

Slack limits the rate of messages which can be posted; limited to 1 message per second at the time of writing. Quine matches that rate-limit internally by aggregating StandingQueryResults together if they arrive faster than can be published to Slack. The aggregated results will be published as a single message to Slack when the rate-limit allows.

Log JSON to Standard Out

Prints each result as a single-line JSON object to standard output on the Quine server.

This output type can be configured with Complete to print a line for every StandingQueryResult, backpressuring and slowing down the stream as needed to print every result.

Or it can be configured with FastSampling to log results in a best effort, by dropping some results to avoid slowing down the stream.

Note that neither option changes the behavior of other StandingQueryResultOutputs registered on the same standing query.

Log JSON to a File

Write each StandingQueryResult as a single-line JSON object to a file on the local filesystem.

Publish to Kafka Topic

Publishes a record for each StandingQueryResult to the provided Apache Kafka topic. Records can be serialized as JSON or Protocol Buffers before being published to Kafka.

Publish to Kinesis Stream

Publishes a record for each StandingQueryResult to the provided Kinesis stream. Records can be serialized as JSON or Protocol Buffers before being published to Kinesis.

Publish to SNS Topic

Publishes an AWS SNS record to the provided topic containing JSON for each result. For the format of the result, see "Standing Query Result Output".

Ensure your credentials and topic ARN are correct. If writing to SNS fails, the write will be retried indefinitely. If the error is not fixable (e.g.: the topic or credentials cannot be found), the outputs will never be emitted and the Standing Query this output is attached to could stop running.

Inspecting Running Queries

Since pattern query queries use a subset of regular Cypher query syntax, the match query itself can be run as a regular query either to see what data already in the graph would have been matched by the query or to understand why a particular node in the graph is not a match.

When doing so, you should constrain the starting points of the query if there is already a large amount of data in the system (see Using IDs in a Query).

In addition, Quine provides ways to stream results as they are being produced and inspect them live.

standing.wiretap Cypher procedure

From the Exploration UI, the standing.wiretap Cypher procedure can issue a query that will incrementally return results. Since this is just a regular, Cypher procedure, it can feed its outputs automatically into another query too.

For example:

// Wiretap "hasMaternalGrandpaJoe" and return properties of matching nodes
CALL standing.wiretap({ name: "hasMaternalGrandpaJoe" }) YIELD meta, data
WHERE meta.isPositiveMatch
MATCH (n) WHERE id(n) = data.id
RETURN properties(n)

Then, you will see results incrementally appear as they match. When you are satisfied, you can cancel the query.

Warning

The standing.wiretap procedure only stops running if the standing query is canceled (since otherwise, it can never be certain that there won't be more forthcoming match results). This means that it is risky to use the procedure in the Cypher REST API or in other places where results are not reported incrementally and queries cannot be canceled.

wiretap screencap

Results API Endpoint

Quine can produce wiretap results outside of the Exploration UI (and without going through the standing.wiretap Cypher procedure) by using the server side events (SSE) endpoint /api/v1/query/standing/{standingQueryName}/results.

The endpoint will surface new matches as they are produced. Connecting a Chrome web browser, for example, will append new results to the bottom of the page as they become available.

Alternatively, curl will print out new results as they arrive in a terminal window.

$ curl http://localhost:8080/api/v1/query/standing/hasMaternalGrandpaJoe/results
data:

data:

data:{"data":{"id":"2756309260014435"},"meta":{"isInitialResult":true,"ispositivematch":true,"resultId":"8f408026-8fb3-3955-c81a-7259175f41b8"}}
event:result
id:8f408026-8fb3-3955-c81a-7259175f41b8

data:{"data":{"id":"7945274922095468"},"meta":{"isInitialResult":true,"isPositiveMatch":true,"resultId":"6a83dda3-08a1-e085-ee7d-14138398f336"}}
event:result
id:6a83dda3-08a1-e085-ee7d-14138398f336

data:{"data":{"id":"6994090876991233"},"meta":{"isInitialResult":true,"isPositiveMatch":true,"resultId":"59b215b4-4084-b5bb-379d-9654bb2a7c83"}}
event:result
id:59b215b4-4084-b5bb-379d-9654bb2a7c83

data:

Using the output above, it is possible to query the matching nodes directly with a Cypher query.

For instance, we can go look for current children of some of the matches the SSE output above tells us we found:

// Query for children of nodes with IDs from the SSE endpoint above
UNWIND [2756309260014435, 7945274922095468, 6994090876991233] AS personId
MATCH (person)<-[:HAS_MOTHER|:HAS_FATHER]-(child) WHERE id(person) = personId
RETURN person.name, child.name, child.yearBorn