////
Licensed to the Apache Software Foundation (ASF) under one or more
contributor license agreements.  See the NOTICE file distributed with
this work for additional information regarding copyright ownership.
The ASF licenses this file to You under the Apache License, Version 2.0
(the "License"); you may not use this file except in compliance with
the License.  You may obtain a copy of the License at

  http://www.apache.org/licenses/LICENSE-2.0

Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
////

= TinkerPop 3.2.0

image::https://raw.githubusercontent.com/apache/tinkerpop/master/docs/static/images/nine-inch-gremlins.png[width=225]

*Nine Inch Gremlins*

== TinkerPop 3.2.11

*Release Date: January 2, 2019*

Please see the link:https://github.com/apache/tinkerpop/blob/3.2.11/CHANGELOG.asciidoc#release-3-2-11[changelog] for a complete list of all the modifications that are part of this release.

== TinkerPop 3.2.10

*Release Date: October 15, 2018*

Please see the link:https://github.com/apache/tinkerpop/blob/3.2.10/CHANGELOG.asciidoc#release-3-2-10[changelog] for a complete list of all the modifications that are part of this release.

=== Upgrading for Users

==== SASL in Gremlin.Net

The Gremlin Javascript Driver now supports SASL Plain Text authentication against a Gremlin Server.

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1977[TINKERPOP-1977],
link:https://tinkerpop.apache.org/docs/3.2.10/reference#gremlin-javascript[Reference Documentation - Gremlin Javascript]

==== SSL Security

TinkerPop improves its security posture by removing insecure defaults and adding forward-looking standards support.

Gremlin Server no longer supports automatically creating self-signed certificates.
Self-signed certificates can still be created manually outside of Gremlin Server.
If ssl is enabled, a key store must be configured.

Cluster client no longer trusts all certs by default as this is an insecure configuration.
Instead, if no trust store is configured, Cluster will use the default CA certs.
To revert to the previous behavior and accept all certs, it must be explicitly configured.

This release introduces JKS and PKCS12 support. JKS is the legacy Java Key Store. PKCS12 has better cross-platform support and is gaining in adoption.
Be aware that JKS is the default on Java 8.  Java 9 and higher use PKCS12 as the default. Both Java keytool and OpenSSL tools can create, read, update PKCS12 files.

Other new features include specifying SSL protocols and cipher suites.
The packaged `*-secure.yaml` files now restrict the protocol to `TLSv1.2` by default.

PEM-based configurations are deprecated and may be removed in a future release.

See also https://tinkerpop.apache.org/docs/current/reference/#_configuration[Connecting via Java Configuration],
https://tinkerpop.apache.org/docs/current/reference/#_configuring_2[Gremlin Server Configuration].

link:https://issues.apache.org/jira/browse/TINKERPOP-2022[TINKERPOP-2022]
link:https://issues.apache.org/jira/browse/TINKERPOP-2023[TINKERPOP-2023]

==== Bulk Import and Export

TinkerPop has provided some general methods for importing and exporting data, but more and more graph providers are
producing their own bulk import/export facilities and they are more efficient and easier to use than TinkerPop's
methods. As a result, TinkerPop will now refer users to the bulk import/export features of individual graph providers
and as such, has deprecated `BulkLoaderVertexProgram` as of this release.

As part of this change, the `BulkDumperVertexProgram` has been renamed to `CloneVertexProgram` with the former being
deprecated. `CloneVertexProgram` is more aptly named, as it essentially copies a graph from a graph `InputFormat`
to a graph `OutputFormat`.

link:https://issues.apache.org/jira/browse/TINKERPOP-1985[TINKERPOP-1985]

==== Docker Images

Docker images are now available on link:https://hub.docker.com/u/tinkerpop/[Docker Hub] for Gremlin Console and Gremlin
Server.

[source,text]
----
$ docker run -it tinkerpop/gremlin-console:x.y.z
$ docker run tinkerpop/gremlin-server:x.y.z
----

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1897[TINKERPOP-1897],
link:https://tinkerpop.apache.org/docs/current/reference#gremlin-console-docker-image[Reference Documentation - Gremlin Console],
link:https://tinkerpop.apache.org/docs/current/reference#gremlin-server-docker-image[Reference Documentation - Gremlin Server],

==== TimedInterruptCustomizerProvider

In Gremlin Server, it is best not to use 'TimedInterruptCustomizerProvider' because it can compete with the
'scriptEvaluationTimeout' setting and produce a different error path. Simply rely on 'scriptEvaluationTimeout'
as it covers both script evaluation time and result iteration time.

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1778[TINKERPOP-1778]

==== TinkerFactory.createGratefulDead()

The Grateful Dead dataset has been with TinkerPop since the early days of 1.x. It has always been available as a
packaged dataset that needed to be loaded through the various IO options available, while other toy graphs had the
benefit of `TinkerFactory` to help get them bootstrapped. For 3.2.10, Grateful Dead is now more conveniently loaded
via that same method as the other toy graphs with `TinkerFactory.createGratefulDead()`.

==== Gremlin Javascript Script Submission

Gremlin Javascript can now submit script, with optional bindings, using the `Client` class:

[source,javascript]
----
const gremlin = require('gremlin');
const client = new gremlin.driver.Client('ws://localhost:8182/gremlin', { traversalSource: 'g' });

client.submit('g.V().tail()')
    .then(result => {
        console.log(result.length);
        console.log(result.toArray()[0]);
    });

client.submit('g.V(vid)', { vid: 1 })
    .then(result => {
        console.log(result.length);
        // Get the first item
        console.log(result.first());
    });
----

and also allows translation of bytecode steps into script:

[source,javascript]
----
const gremlin = require('gremlin');
const graph = new gremlin.process.Graph();
const client = new gremlin.driver.Client('ws://localhost:8182/gremlin', { traversalSource: 'g' });
const translator = new gremlin.process.Translator('g');

const g = graph.traversal();
const script = translator.translate(g.V().tail().getBytecode());

client.submit(script)
    .then(result => {
        console.log(result.length);
        console.log(result.first());
    });
----

=== Upgrading for Providers

==== Graph Database Providers

===== Bulk Import and Export

As noted in the user section, TinkerPop has deprecated its bulk loading feature in `BulkLoaderVertexProgram` and will
refer TinkerPop users who need bulk import/export capabilities to the native tools of the graph database they have
chosen. If a graph database provider does not have any bulk loading tools it can choose to build graph `InputFormat`
and `OutputFormat` implementations which can be used by `CloneVertexProgram` (formerly `BulkDumperVertexProgram`) as
a easy way to get such a feature.

link:https://issues.apache.org/jira/browse/TINKERPOP-1985[TINKERPOP-1985]

== TinkerPop 3.2.9

*Release Date: May 8, 2018*

Please see the link:https://github.com/apache/tinkerpop/blob/3.2.9/CHANGELOG.asciidoc#release-3-2-9[changelog] for a complete list of all the modifications that are part of this release.

=== Upgrading for Users

==== Lambda Construction

It was realized quite shortly after release of 3.2.8 that there was a bug in construction of `Lambda` instances:

[source,text]
----
gremlin> org.apache.tinkerpop.gremlin.util.function.Lambda.function("{ it.get() }")
(class: org/apache/tinkerpop/gremlin/util/function/Lambda$function, method: callStatic signature: (Ljava/lang/Class;[Ljava/lang/Object;)Ljava/lang/Object;) Illegal type in constant pool
Type ':help' or ':h' for help.
Display stack trace? [yN]n
----

The problem was related to a bug in Groovy 2.4.14 and was fixed in 2.4.15.

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1953[TINKERPOP-1953]

== TinkerPop 3.2.8

*Release Date: April 2, 2018*

Please see the link:https://github.com/apache/tinkerpop/blob/3.2.8/CHANGELOG.asciidoc#release-3-2-8[changelog] for a complete list of all the modifications that are part of this release.

=== Upgrading for Users

==== Improved Connection Monitoring

Gremlin Server now has two new settings: `idleConnectionTimeout` and `keepAliveInterval`. The `keepAliveInterval` tells
Gremlin Server how long it should wait between writes to a client before it issues a "ping" to that client to see if
it is still present. The `idleConnectionTimeout` represents how long Gremlin Server should wait between requests from
a client before it closes the connection on the server side. By default, these two configurations are set to zero,
meaning that they are both disabled.

This change should help to alleviate issues where connections are left open on the server longer than they should be
by clients that might mysteriously disappear without properly closing their connections.

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1726[TINKERPOP-1726]

==== Gremlin.Net Lambdas

Gremlin.Net now has a `Lambda` class that can be used to construct Groovy or Java lambdas which will be evaluated on the
server.

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1854[TINKERPOP-1854], link:https://tinkerpop.apache.org/docs/3.2.8/reference/#_the_lambda_solution_3[Reference Documentation - Gremlin.Net - The Lambda Solution].

==== Gremlin.Net Tokens Improved

The various Gremlin tokens (e.g. `T`, `Order`, `Operator`, etc.) that were implemented as Enums before in Gremlin.Net
are now implemented as classes. This mainly allows them to implement interfaces which their Java counterparts already
did. `T` for example now implements the new interface `IFunction` which simply mirrors its Java counterpart `Function`.
Steps that expect objects for those interfaces as arguments now explicitly use the interface. Before, they used just
`object` as the type for these arguments which made it hard for users to know what kind of `object` they can use.
However, usage of these tokens themselves shouldn't change at all (e.g. `T.Id` is still `T.Id`).

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1901[TINKERPOP-1901]

==== Gremlin.Net: Traversal Predicate Classes Merged

Gremlin.Net used two classes for traversal predicates: `P` and `TraversalPredicate`. Steps that worked with traversal
predicates expected objects of type `TraversalPredicate`, but they were constructed from the `P` class
(e.g. `P.Gt(1)` returned a `TraversalPredicate`). Merging these two classes into the `P` class should avoid unnecessary
confusion. Most users should not notice this change as predicates can still be constructed exactly as before, e.g.,
`P.Gt(1).And(P.Lt(3))` still works without any modifications.
Only users that implemented their own predicates and used `TraversalPredicate` as the base class need to change their
implementation to now use `P` as the new base class.

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1919[TINKERPOP-1919]

=== Upgrading for Providers

==== Graph System Providers

===== Kitchen Sink Test Graph

The "Kitchen Sink" test graph has been added to the `gremlin-test` module. It contains (or will contain) various
disconnected subgraphs of that offer unique structures (e.g. a self-loop) for specific test cases. Graph systems that
use the test suite should not have to make any changes to account for this new graph unless that system performs some
form or special pre-initialization of their system in preparation for loading (e.g. requires a schema) or does the
loading of the graph test data outside of the standard method in which TinkerPop provides.

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1877[TINKERPOP-1877]

== TinkerPop 3.2.7

*Release Date: December 17, 2017*

Please see the link:https://github.com/apache/tinkerpop/blob/3.2.7/CHANGELOG.asciidoc#release-3-2-7[changelog] for a complete list of all the modifications that are part of this release.

=== Upgrading for Users

==== Gremlin-Python Core Types
With the addition of `UUID`, `Date`, and `Timestamp`, Gremlin-Python now implements serializers for all core GraphSON types. Users
that were using other types to represent this data can now use the Python classes `datetime.datetime` and`uuid.UUID` in GLV traversals.
Since Python does not support a native `Timestamp` object, Gremlin-Python now offers a dummy class `Timestamp`, which allows
users to wrap a float and submit it to the Gremlin Server as a `Timestamp` GraphSON type. `Timestamp` can be found in
`gremlin_python.statics`.

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1807[TINKERPOP-1807]

==== EventStrategy Detachment

`EventStrategy` forced detachment of mutated elements prior to raising them in events. While this was a desired
outcome, it may not have always fit every use case. For example, a user may have wanted a reference element or the
actual element itself. As a result, `EventStrategy` has changed to allow it to be constructed with a `detach()`
option, where it is possible to specify any of the following: `null` for no detachment, `DetachedFactory` for the
original behavior, and `ReferenceFactory` for detachment that returns reference elements.

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1829[TINKERPOP-1829]

==== Embedded Remote Connection

As Gremlin Language Variants (GLVs) expand their usage and use of `withRemote()` becomes more common, the need to mock
the "remote" in unit tests increases. To simplify mocking in Java, the new `EmbeddedRemoteConnection` provides a
simple way to provide a "remote" that is actually local to the same JVM.

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1756[TINKERPOP-1756]

==== DSL Type Specification

Prior to this version, the Java annotation processor for Gremlin DSLs has tried to infer the appropriate type
specifications when generating anonymous methods. It largely performed this inference on simple conventions in the
DSL method's template specification and there were times where it would fail. For example, a method like this:

[source,java]
----
public default GraphTraversal<S, E> person() {
    return hasLabel("person");
}
----

would generate an anonymous method like:

[source,java]
----
public static <S> SocialGraphTraversal<S, E> person() {
    return hasLabel("person");
}
----

and, of course, generate a compile error and `E` was not recognized as a symbol. The preferred generation would likely
be:

[source,java]
----
public static <S> SocialGraphTraversal<S, S> person() {
    return hasLabel("person");
}
----

To remedy this situation, a new annotation has been added which allows the user to control the type specifications
more directly providing a way to avoid/override the inference system:

[source,java]
----
@GremlinDsl.AnonymousMethod(returnTypeParameters = {"A", "A"}, methodTypeParameters = {"A"})
public default GraphTraversal<S, E> person() {
    return hasLabel("person");
}
----

which will then generate:

[source,java]
----
public static <A> SocialGraphTraversal<A, A> person() {
    return hasLabel("person");
}
----

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1791[TINKERPOP-1791]

==== Specify a Cluster Object

The `:remote connect` command can now take a pre-defined `Cluster` object as its argument as opposed to a YAML
configuration file.

[source,text]
----
gremlin> cluster = Cluster.open()
==>localhost/127.0.0.1:8182
gremlin> :remote connect tinkerpop.server cluster
==>Configured localhost/127.0.0.1:8182
----

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1787[TINKERPOP-1787]

==== Remote Traversal Timeout

There was limited support for "timeouts" with remote traversals (i.e. those traversals executed using the `withRemote()`
option) prior to 3.2.7. Remote traversals will now interrupt on the server using the `scriptEvaluationTimeout`
setting in the same way that normal script evaluations would. As a reminder, interruptions for traversals are always
considered "attempts to interrupt" and may not always succeed (a graph database implementation might not respect the
interruption, for example).

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1770[TINKERPOP-1770]

==== Modifications to match()

The `match()`-step has been generalized to support the local scoping of all barrier steps, not just reducing barrier steps.
Previously, the `order().limit()` clause would have worked globally yielding:

[source,groovy]
----
gremlin> g.V().match(
......1>   __.as('a').outE('created').order().by('weight',decr).limit(1).inV().as('b'),
......2>   __.as('b').has('lang','java')
......3> ).select('a','b').by('name')
==>[a:marko,b:lop]
----

However, now, `order()` (and all other barriers) are treated as local computations to the pattern and thus, the result set is:

[source,groovy]
----
gremlin> g.V().match(
......1>   __.as('a').outE('created').order().by('weight',decr).limit(1).inV().as('b'),
......2>   __.as('b').has('lang','java')
......3> ).select('a','b').by('name')
==>[a:marko,b:lop]
==>[a:josh,b:ripple]
==>[a:peter,b:lop]
----

Note that this is not that intense of a breaking change as all of the reducing barriers behaved in this manner previously.
This includes steps like `count()`, `min()`, `max()`, `sum()`, `group()`, `groupCount()`, etc. This update has now
generalized this behavior to all barriers and thus, adds `aggregate()`, `dedup()`, `range()`, `limit()`, `tail()`, and `order()`
to the list of locally computed clauses.

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1764[TINKERPOP-1764]

==== Clone a Graph

In `gremlin-test` there is a new `GraphHelper` class that has a `cloneElements()` method. It will clone elements from
the first graph to the second - `GraphHelper.cloneElements(Graph original, Graph clone)`. This helper method is
primarily intended for use in tests.

==== MutationListener Changes

The `MutationListener` has a method called `vertexPropertyChanged` which gathered callbacks when a property on a vertex
was modified. The method had an incorrect signature though using `Property` instead of `VertexProperty`. The old method
that used `Property` has now been deprecated and a new method added that uses `VertexProperty`. This new method has a
default implementation that calls the old method, so this change should not cause breaks in compilation on upgrade.
Internally, TinkerPop no longer calls the old method except by way of that proxy. Users who have `MutationListener`
implementations can simply add the new method and override its behavior. The old method can thus be ignored completely.

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1798[TINKERPOP-1798]

=== Upgrading for Providers

==== Direction.BOTH Requires Duplication of Self-Edges

Prior to this release, there was no semantic check to determine whether a self-edge (e.g. `e[1][2-self->2]`) would be returned
twice on a `BOTH`. The semantics have been specified now in the test suite where the edge should be returned twice as it
is both an incoming edge and an outgoing edge.

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1821[TINKERPOP-1821]

== TinkerPop 3.2.6


*Release Date: August 21, 2017*

=== Upgrading for Users

Please see the link:https://github.com/apache/tinkerpop/blob/3.2.6/CHANGELOG.asciidoc#release-3-2-6[changelog] for a complete list of all the modifications that are part of this release.

==== Deprecated useMapperFromGraph

The `userMapperFromGraph` configuration option for the Gremlin Server serializers has been deprecated. Change
configuration files to use the `ioRegistries` option instead. The `ioRegistries` option is not a new feature, but
it has not been promoted as the primary way to add `IoRegistry` instances to serializers.

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1694[TINKERPOP-1694]

==== WsAndHttpChannelizer

The `WsAndHttpChannelizer` has been added to allow for processing both WebSocket and HTTP requests on the same
port and gremlin server. The `SaslAndHttpBasicAuthenticationHandler` has also been added to service
authentication for both protocols in conjunction with the `SimpleAuthenticator`.

See: link:https://issues.apache.org/jira/browse/TINKERPOP-915[TINKERPOP-915]

=== Upgrading for Providers

==== ReferenceVertex Label

`ReferenceVertex.label()` was hard coded to return `EMPTY_STRING`. At some point, `ReferenceElements` were suppose to
return labels and `ReferenceVertex` was never updated as such. Note that `ReferenceEdge` and `ReferenceVertexProperty`
work as expected. However, given a general change at `ReferenceElement`, the Gryo serialization of `ReferenceXXX` is
different. If the vertex does not have a label `Vertex.DEFAULT_LABEL` is assumed.

See: https://issues.apache.org/jira/browse/TINKERPOP-1789[TINKERPOP-1789]

== TinkerPop 3.2.5

*Release Date: June 12, 2017*

Please see the link:https://github.com/apache/tinkerpop/blob/3.2.5/CHANGELOG.asciidoc#release-3-2-5[changelog] for a complete list of all the modifications that are part of this release.

=== Upgrading for Users

==== DSL Support

It has always been possible to construct Domain Specific Languages (DSLs) with Gremlin, but the approach has required
a somewhat deep understanding of the TinkerPop code base and it is not something that has had a recommended method
for implementation. With this release, TinkerPop simplifies DSL development and provides the best practices for their
implementation.

[source,java]
----
// standard Gremlin
g.V().hasLabel('person').
  where(outE("created").count().is(P.gte(2))).count()

// the same traversal as above written as a DSL
social.persons().where(createdAtLeast(2)).count()
----

See: link:https://issues.apache.org/jira/browse/TINKERPOP-786[TINKERPOP-786],
link:https://tinkerpop.apache.org/docs/3.2.5/reference/#dsl[Reference Documentation]

==== GraphSON Path Serialization

Serialization of `Path` with GraphSON was inconsistent with Gryo in that all the properties on any elements of
the `Path` were being included. With Gryo that, correctly, was not happening as that could be extraordinarily
expensive. GraphSON serialization has now been modified to properly not include properties. That change can cause
breaks in application code if that application code tries to access properties on elements in a `Path` as they
will no longer be there. Applications that require the properties will need to alter their Gremlin to better
restrict the data they want to retrieve.

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1676[TINKERPOP-1676]

==== Authentication Configuration

The server settings previously used `authentication.className` to set an authenticator for the the two provided
authentication handler and channelizer classes to use. This has been deprecated in favor of `authentication.authenticator`.
A class that extends `AbstractAuthenticationHandler` may also now be provided as `authentication.authenticationHandler`
to be used in either of the provided channelizer classes to handle the provided authenticator

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1657[TINKERPOP-1657]

==== Default Maximum Parameters

It was learned that compilation for scripts with large numbers of parameters is more expensive than those with less
parameters. It therefore becomes possible to make some mistakes with how Gremlin Server is used. A new setting on
the `StandardOpProcessor` and `SessionOpProcessor` called `maxParameters` controls the number of parameters that can
be passed in on a request. This setting is defaulted to sixteen.

Users upgrading to this version may notice errors in their applications if they use more than sixteen parameters. To
fix this problem simply reconfigure Gremlin Server with a configuration as follows:

[source,yaml]
----
processors:
  - { className: org.apache.tinkerpop.gremlin.server.op.session.SessionOpProcessor, config: { maxParameters: 64 }}
  - { className: org.apache.tinkerpop.gremlin.server.op.standard.StandardOpProcessor, config: { maxParameters: 64 }}
----

The above configuration allows sixty-four parameters to be passed on each request.

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1663[TINKERPOP-1663]

==== GremlinScriptEngine Metrics

The `GremlinScriptEngine` has a number of new metrics about its cache size and script compilation times which should
be helpful in understanding usage problems. As `GremlinScriptEngine` instances are used in Gremlin Server these metrics
are naturally exposed as part of the standard link:https://tinkerpop.apache.org/docs/current/reference/#_metrics[metrics]
set. Note that metrics are captured for both sessionless requests as well as for each individual session that is opened.

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1644[TINKERPOP-1644]

==== Additional Error Information

Additional information on error responses from Gremlin Server should help make debugging errors easier. Error responses
now have both the exception hierarchy and the stack trace that was generated on the server. In this way, receiving an
error on a client doesn't mean having to rifle through Gremlin Server logs to try to find the associated error.

This change has been applied to all Gremlin Server protocols. For the binary protocol and the Java driver this change
means that the `ResponseException` thrown from calls to `submit()` requests to the server now have the following
methods:

[source,java]
----
public Optional<String> getRemoteStackTrace()

public Optional<List<String>> getRemoteExceptionHierarchy()
----

The HTTP protocol has also been updated and returns both `exceptions` and `stackTrace` fields in the response:

[source,js]
----
{
	"message": "Division by zero",
	"Exception-Class": "java.lang.ArithmeticException",
	"exceptions": ["java.lang.ArithmeticException"],
	"stackTrace": "java.lang.ArithmeticException: Division by zero\n\tat java.math.BigDecimal.divide(BigDecimal.java:1742)\n\tat org.codehaus.groovy.runtime.typehandling.BigDecimalMath.divideImpl(BigDecimalMath.java:68)\n\tat org.codehaus.groovy.runtime.typehandling.IntegerMath.divideImpl(IntegerMath.java:49)\n\tat org.codehaus.groovy.runtime.dgmimpl.NumberNumberDiv$NumberNumber.invoke(NumberNumberDiv.java:323)\n\tat org.codehaus.groovy.runtime.callsite.PojoMetaMethodSite.call(PojoMetaMethodSite.java:56)\n\tat org.codehaus.groovy.runtime.callsite.CallSiteArray.defaultCall(CallSiteArray.java:48)\n\tat org.codehaus.groovy.runtime.callsite.AbstractCallSite.call(AbstractCallSite.java:113)\n\tat org.codehaus.groovy.runtime.callsite.AbstractCallSite.call(AbstractCallSite.java:125)\n\tat Script4.run(Script4.groovy:1)\n\tat org.apache.tinkerpop.gremlin.groovy.jsr223.GremlinGroovyScriptEngine.eval(GremlinGroovyScriptEngine.java:834)\n\tat org.apache.tinkerpop.gremlin.groovy.jsr223.GremlinGroovyScriptEngine.eval(GremlinGroovyScriptEngine.java:547)\n\tat javax.script.AbstractScriptEngine.eval(AbstractScriptEngine.java:233)\n\tat org.apache.tinkerpop.gremlin.groovy.engine.ScriptEngines.eval(ScriptEngines.java:120)\n\tat org.apache.tinkerpop.gremlin.groovy.engine.GremlinExecutor.lambda$eval$2(GremlinExecutor.java:314)\n\tat java.util.concurrent.FutureTask.run(FutureTask.java:266)\n\tat java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142)\n\tat java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617)\n\tat java.lang.Thread.run(Thread.java:745)\n"
}
----

Note that the `Exception-Class` which was added in a previous version has been deprecated and replaced by these new
fields.

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1044[TINKERPOP-1044]

==== Gremlin Console Scripting

The `gremlin.sh` command has two flags, `-i` and `-e`, which are used to pass a script and arguments into the Gremlin
Console for execution. Those flags now allow for passing multiple scripts and related arguments to be supplied which
can yield greater flexibility in automation tasks.

[source,bash]
----
$ bin/gremlin.sh -i y.groovy 1 2 3 -i x.groovy
$ bin/gremlin.sh -e y.groovy 1 2 3 -e x.groovy
----

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1653[TINKERPOP-1653]

==== Path support for by()-, from()-, to()-modulation

It is now possible to extract analyze sub-paths using `from()` and `to()` modulations with respective, path-based steps.
Likewise, `simplePath()` and `cyclicPath()` now support, along with `from()` and `to()`, `by()`-modulation so the cyclicity
is determined by projections of the path data. This extension is fully backwards compatible.

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1387[TINKERPOP-1387]

==== GraphManager versus DefaultGraphManager
Gremlin Server previously implemented its own final `GraphManager` class. Now, the `GraphManager` has been changed to
an interface, and users can supply their own `GraphManager` implementations in their YAML. The previous `GraphManager`
class was meant be used by classes internal to Gremlin Server, but it was public so if it was used for some reason by
users then then a compile error can be expected. To correct this problem, which will likely manifest as a compile error
when trying to create a `new GraphManager()` instance, simply change the code to `new DefaultGraphManager(Settings)`.

In addition to the change mentioned above, several methods on `GraphManager` were deprecated:

* `getGraphs()` should be replaced by the combination of `getGraphNames()` and then `getGraph(String)`
* `getTraversalSources()` is similarly replaced and should instead use a combination of `getTraversalSourceNames()` and
`getTraversalSource(String)`

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1438[TINKERPOP-1438]

==== Gremlin-Python Driver
Gremlin-Python now offers a more complete driver implementation that uses connection pooling and
the Python `concurrent.futures` module to provide asynchronous I/0 using threading. The default underlying
WebSocket client implementation is still provided by Tornado, but it is trivial to plug in another client by
defining the `Transport` interface.

Using the `DriverRemoteConnection` class is the exact same as in previous versions; however,
`DriverRemoteConnection` now uses the new `Client` class to submit messages to the server.

The `Client` class implementation/interface is based on the Java Driver, with some restrictions.
Most notably, Gremlin-Python does not yet implement the `Cluster` class. Instead, `Client` is
instantiated directly. Usage is as follows:

[source,python]
----
from gremlin_python.driver import client

client = client.Client('ws://localhost:8182/gremlin', 'g')
result_set = client.submit('1 + 1')
future_results = result_set.all()  # returns a concurrent.futures.Future
results = future_results.result()  # returns a list
assert results == [2]
client.close()  # don't forget to close underlying connections
----

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1599[TINKERPOP-1599]

=== Upgrading for Providers

IMPORTANT: It is recommended that providers also review all the upgrade instructions specified for users. Many of the
changes there may prove important for the provider's implementation.

==== Graph Database Providers

===== SimplePathStep and CyclicPathStep now PathFilterStep

The Gremlin traversal machine use to support two step instructions: `SimplePathStep` and `CyclicPathStep`. These have
been replaced by a high-level instruction called `PathFilterStep` which is boolean configured for simple or cyclic paths.
Furthermore, `PathFilterStep` also support `from()`-, `to()`-, and `by()`-modulation.

===== LazyBarrierStrategy No Longer End Appends Barriers

`LazyBarrierStrategy` was trying to do to much by considering `Traverser` effects on network I/O by appending an
`NoOpBarrierStrategy` to the end of the root traversal. This should not be accomplished by `LazyBarrierStrategy`,
but instead by `RemoteStrategy`. `RemoteStrategy` now tries to barrier-append. This may effect the reasoning logic in
some `ProviderStrategies`. Most likely not, but just be aware.

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1627[TINKERPOP-1627]

== TinkerPop 3.2.4

*Release Date: February 8, 2017*

Please see the link:https://github.com/apache/tinkerpop/blob/3.2.4/CHANGELOG.asciidoc#release-3-2-4[changelog] for a complete list of all the modifications that are part of this release.

=== Upgrading for Users

==== TinkerGraph Deserialization

A TinkerGraph deserialized from Gryo or GraphSON is now configured with multi-properties enabled. This change allows
TinkerGraphs returned from Gremlin Server to properly return multi-properties, which was a problem seen when
subgraphing a graph that contained properties with a setting other than `Cardinality.single`.

This change could be considered breaking in the odd chance that a TinkerGraph returned from Gremlin Server was later
mutated, because calls to `property(k,v)` would default to `Cardinality.list` instead of `Cardinality.single`. In the
event that this is a problem, simple change calls to `property(k,v)` to `property(Cardinality.single,k,v)` and
explicitly set the `Cardinality`.

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1587[TINKERPOP-1587]

==== Traversal Promises

The `Traversal` API now has a new `promise()` method. These methods return a promise in the form of a
`CompleteableFuture`. Usage is as follows:

[source,groovy]
----
gremlin> promise = g.V().out().promise{it.next()}
==>java.util.concurrent.CompletableFuture@4aa3d36[Completed normally]
gremlin> promise.join()
==>v[3]
gremlin> promise.isDone()
==>true
gremlin> g.V().out().promise{it.toList()}.thenApply{it.size()}.get()
==>6
----

At this time, this method is only used for traversals that are configured using `withRemote()`.

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1490[TINKERPOP-1490]

==== If/Then-Semantics with Choose Step

Gremlin's `choose()`-step supports if/then/else-semantics. Thus, to effect if/then-semantics, `identity()` was required.
Thus, the following two traversals below are equivalent with the later being possible in this release.

[source,groovy]
----
g.V().choose(hasLabel('person'),out('created'),identity())
g.V().choose(hasLabel('person'),out('created'))
----

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1508[TINKERPOP-1508]

==== FastNoSuchElementException converted to regular NoSuchElementException

Previously, a call to `Traversal.next()` that did not have a result would throw a `FastNoSuchElementException`.
This has been changed to a regular `NoSuchElementException` that includes the stack trace. Code that explicitly catches
`FastNoSuchElementException` should be converted to check for the more general class of `NoSuchElementException`.

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1330[TINKERPOP-1330]

==== ScriptEngine support in gremlin-core

`ScriptEngine` and `GremlinPlugin` infrastructure has been moved from gremlin-groovy to gremlin-core to allow for
better re-use across different Gremlin Language Variants. At this point, this change is non-breaking as it was
implemented through deprecation.

The basic concept of a `ScriptEngine` has been replaced by the notion of a `GremlinScriptEngine` (i.e. a
"ScriptEngine" that is specifically tuned for executing Gremlin-related scripts). "ScriptEngine" infrastructure has
been developed to help support this new interface, specifically `GremlinScriptEngineFactory` and
`GremlinScriptEngineManager`. Prefer use of this infrastructure when instantiating a `GremlinScriptEngine` rather
than trying to instantiate directly.

For example, rather than instantiate a `GremlinGroovyScriptEngine` with the constructor:

[source,java]
----
GremlinScriptEngine engine = new GremlinGroovyScriptEngine();
----

prefer to instantiate it as follows:

[source,java]
----
GremlinScriptEngineManager manager = new CachedGremlinScriptEngineManager();
GremlinScriptEngine engine = manager.getEngineByName("gremlin-groovy");
----

Related to the addition of `GremlinScriptEngine`, `org.apache.tinkerpop.gremlin.groovy.plugin.GremlinPlugin` in
gremlin-groovy has been deprecated and then replaced by `org.apache.tinkerpop.gremlin.jsr223.GremlinPlugin`. The new
version of `GremlinPlugin` is similar but does carry some new methods to implement that involves the new `Customizer`
interface. The `Customizer` interface is the way in which `GremlinScriptEngine` instance can be configured with
imports, initialization scripts, compiler options, etc.

Note that a `GremlinPlugin` can be applied to a `GremlinScriptEngine` by adding it to the `GremlinScriptEngineManager`
that creates it.

[source,java]
----
GremlinScriptEngineManager manager = new CachedGremlinScriptEngineManager();
manager.addPlugin(ImportGremlinPlugin.build().classImports(java.awt.Color.class).create());
GremlinScriptEngine engine = manager.getEngineByName("gremlin-groovy");
----

All of this new infrastructure is currently optional on the 3.2.x line of code. More detailed documentation will for
these changes will be supplied as part of 3.3.0 when these features become mandatory and the deprecated code is
removed.

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1562[TINKERPOP-1562]

==== SSL Client Authentication

Added new server configuration option `ssl.needClientAuth`.

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1602[TINKERPOP-1602]

=== Upgrading for Providers

IMPORTANT: It is recommended that providers also review all the upgrade instructions specified for users. Many of the
changes there may prove important for the provider's implementation.

==== Graph Database Providers

===== CloseableIterator

Prior to TinkerPop 3.x, Blueprints had the notion of a `CloseableIterable` which exposed a way for Graph Providers
to offer a way to release resources that might have been opened when returning vertices and edges. That interface was
never exposed in TinkerPop 3.x, but has now been made available via the new `CloseableIterator`. Providers may choose
to use this interface or not when returning values from `Graph.vertices()` and `Graph.edges()`.

It will be up to users to know whether or not they need to call `close()`. Of course, users should typically not be
operating with the Graph Structure API, so it's unlikely that they would be calling these methods directly in the
first place. It is more likely that users will be calling `Traversal.close()`. This method will essentially iterate
the steps of the `Traversal` and simply call `close()` on any steps that implement `AutoCloseable`. By default,
`GraphStep` now implements `AutoCloseable` which most Graph Providers will extend upon (as was done with TinkerGraph's
`TinkerGraphStep`), so the integration should largely come for free if the provider simply returns a
`CloseableIterator` from `Graph.vertices()` and `Graph.edges()`.

See: https://issues.apache.org/jira/browse/TINKERPOP-1589[TINKERPOP-1589]

===== HasContainer AndP Splitting

Previously, `GraphTraversal` made it easy for providers to analyze `P`-predicates in `HasContainers`, but always
splitting `AndP` predicates into their component parts. This helper behavior is no longer provided because,
1.) `AndP` can be inserted into a `XXXStep` in other ways, 2.) the providers `XXXStep` should process `AndP`
regardless of `GraphTraversal` helper, and 3.) the `GraphTraversal` helper did not recursively split.
A simple way to split `AndP` in any custom `XXXStep` that implements `HasContainerHolder` is to use the following method:

[source,java]
----
@Override
public void addHasContainer(final HasContainer hasContainer) {
  if (hasContainer.getPredicate() instanceof AndP) {
    for (final P<?> predicate : ((AndP<?>) hasContainer.getPredicate()).getPredicates()) {
      this.addHasContainer(new HasContainer(hasContainer.getKey(), predicate));
    }
  } else
    this.hasContainers.add(hasContainer);
}
----

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1482[TINKERPOP-1482],
link:https://issues.apache.org/jira/browse/TINKERPOP-1502[TINKERPOP-1502]

===== Duplicate Multi-Properties

Added `supportsDuplicateMultiProperties` to `VertexFeatures` so that graph provider who only support unique values as
multi-properties have more flexibility in describing their graph capabilities.

See: link:https://issues.apache.org/jira/browse/TINKERPOP-919[TINKERPOP-919]

===== Deprecated OptIn

In 3.2.1, all `junit-benchmark` performance tests were deprecated. At that time, the `OptIn` representations of these
tests should have been deprecated as well, but they were not. That omission has been remedied now. Specifically, the
following fields were deprecated:

* `OptIn.SUITE_GROOVY_ENVIRONMENT_PERFORMANCE`
* `OptIn.SUITE_PROCESS_PERFORMANCE`
* `OptIn.SUITE_STRUCTURE_PERFORMANCE`

As of 3.2.4, the following test suites were also deprecated:

* `OptIn.SUITE_GROOVY_PROCESS_STANDARD`
* `OptIn.SUITE_GROOVY_PROCESS_COMPUTER`
* `OptIn.SUITE_GROOVY_ENVIRONMENT`
* `OptIn.SUITE_GROOVY_ENVIRONMENT_INTEGRATE`

Future testing of `gremlin-groovy` (and language variants in general) will be handled differently and will not require
a Graph Provider to validate its operations with it. Graph Providers may now choose to remove these tests from their
test suites, which should reduce the testing burden.

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1610[TINKERPOP-1610]

===== Deprecated getInstance()

TinkerPop has generally preferred static `instance()` methods over `getInstance()`, but `getInstance()` was used in
some cases nonetheless. As of this release, `getInstance()` methods have been deprecated in favor of `instance()`.
Of specific note, custom `IoRegistry` (as related to IO in general) and `Supplier<ClassResolver>` (as related to
Gryo serialization in general) now both prefer `instance()` over `getInstance()` given this deprecation.

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1530[TINKERPOP-1530]

==== Drivers Providers

===== Force Close

Closing a session will first attempt a proper close of any open transactions. A problem can occur, however, if there is
a long run job (e.g. an OLAP-based traversal) executing, as that job will block the calls to close the transactions.
By exercising the option to a do a "forced close" the session will skip trying to close the transactions and just
attempt to interrupt the long run job. By not closing transactions, the session leaves it up to the underlying graph
database to sort out how it will deal with those orphaned transactions. On the positive side though (for those graphs
which do that well) , long run jobs have the opportunity to be cancelled without waiting for a timeout of the job itself
which will allow resources to be released earlier.

The "force" argument is passed on the "close" message and is a boolean value. This is an optional argument to "close"
and defaults to `false`.

See: link:https://issues.apache.org/jira/browse/TINKERPOP-932[TINKERPOP-932],
link:https://tinkerpop.apache.org/docs/current/dev/provider/#_session_opprocessor[Provider Documentation - Session OpProcessor]

===== SASL Authentication

Gremlin Supports SASL based authentication. The server accepts either a byte array or Base64 encoded String as the in
the `sasl` argument on the `RequestMessage`, however it sends back a byte array only. Some serializers or serializer
configurations don't work well with that approach (specifically the "toString" configuration on the Gryo serializer) as
the byte array is returned in the `ResponseMessage` result. In the case of the "toString" serializer the byte array
gets "toString'd" and the can't be read by the client.

In 3.2.4, the byte array is still returned in the `ResponseMessage` result, but is also returned in the status
attributes under a `sasl` key as a Base64 encoded string. In this way, the client has options on how it chooses to
process the authentication response and the change remains backward compatible. Drivers should upgrade to using the
Base64 encoded string however as the old approach will likely be removed in the future.

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1600[TINKERPOP-1600]

== TinkerPop 3.2.3

*Release Date: October 17, 2016*

Please see the link:https://github.com/apache/tinkerpop/blob/3.2.3/CHANGELOG.asciidoc#release-3-2-3[changelog] for a complete list of all the modifications that are part of this release.

=== Upgrading for Users

==== Renamed Null Result Preference

In 3.2.2, the Gremlin Console introduced a setting called `empty.result.indicator`, which controlled the output that
was presented when no result was returned. For consistency, this setting has been renamed to `result.indicator.null`
and can be set as follows:

[source,text]
----
gremlin> graph = TinkerGraph.open()
==>tinkergraph[vertices:0 edges:0]
gremlin> graph.close()
==>null
gremlin> :set result.indicator.null nil
gremlin> graph = TinkerGraph.open()
==>tinkergraph[vertices:0 edges:0]
gremlin> graph.close()
==>nil
gremlin> :set result.indicator.null ""
gremlin> graph = TinkerGraph.open()
==>tinkergraph[vertices:0 edges:0]
gremlin> graph.close()
gremlin>
----

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1409[TINKERPOP-1409]

==== Java Driver Keep-Alive

The Java Driver now has a `keepAliveInterval` setting, which controls the amount of time in milliseconds it should wait
on an inactive connection before it sends a message to the server to keep the connection maintained. This should help
environments that use a load balancer in front of Gremlin Server by ensuring connections are actively maintained even
during periods of inactivity.

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1249[TINKERPOP-1249]

==== Where Step Supports By-Modulation

It is now possible to use `by()` with `where()` predicate-based steps. Previously, without using `match()`, if you wanted
to know who was older than their friend, the following traversal would be used.

[source,text]
----
gremlin> g.V().as('a').out('knows').as('b').
......1>   filter(select('a','b').by('age').where('a', lt('b')))
==>v[4]
----

Now, with `where().by()` support, the above traversal can be expressed more succinctly and more naturally as follows.

[source,text]
----
gremlin> g.V().as('a').out('knows').as('b').
......1>   where('a', lt('b')).by('age')
==>v[4]
----

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1330[TINKERPOP-1330]

==== Change In has() Method Signatures

The TinkerPop 3.2.2 release unintentionally introduced a breaking change for some `has()` method overloads. In particular the
behavior for single item array arguments was changed:

[source,text]
----
gremlin> g.V().hasLabel(["software"] as String[]).count()
==>0
----

Prior this change single item arrays were treated like there was only that single item:

[source,text]
----
gremlin> g.V().hasLabel(["software"] as String[]).count()
==>2
gremlin> g.V().hasLabel("software").count()
==>2
----

TinkerPop 3.2.3 fixes this misbehavior and all `has()` method overloads behave like before, except that they no longer
support no arguments.

==== Deprecated reconnectInitialDelay

The `reconnectInitialDelay` setting on the `Cluster` builder has been deprecated. It no longer serves any purpose.
The value for the "initial delay" now comes from `reconnectInterval` (there are no longer two separate settings to
control).

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1460[TINKERPOP-1460]

==== TraversalSource.close()

`TraversalSource` now implements `AutoCloseable`, which means that the `close()` method is now available. This new
method is important in cases where `withRemote()` is used, as `withRemote()` can open "expensive" resources that need
to be released.

In the case of TinkerPop's `DriverRemoteConnection`, `close()` will destroy the `Client` instance that is created
internally by `withRemote()` as shown below:

[source,text]
----
gremlin> graph = EmptyGraph.instance()
==>emptygraph[empty]
gremlin> g = graph.traversal().withRemote('conf/remote-graph.properties')
==>graphtraversalsource[emptygraph[empty], standard]
gremlin> g.close()
gremlin>
----

Note that the `withRemote()` method will call `close()` on a `RemoteConnection` passed directly to it as well, so
there is no need to do that manually.

See: link:https://issues.apache.org/jira/browse/TINKERPOP-790[TINKERPOP-790]

==== IO Reference Documentation

There is new reference documentation for the various IO formats. The documentation provides more details and samples
that should be helpful to users and providers who intend to work directly with the TinkerPop supported serialization
formats: GraphML, GraphSON and Gryo.

See: link:https://tinkerpop.apache.org/docs/3.2.3/dev/io/[IO Reference Documentation]

=== Upgrading for Providers

IMPORTANT: It is recommended that providers also review all the upgrade instructions specified for users. Many of the
changes there may prove important for the provider's implementation.

==== Graph System Providers

===== Default LazyBarrierStrategy

`LazyBarrierStrategy` has been included as a default strategy. `LazyBarrierStrategy` walks a traversal and looks for
"flatMaps" (`out()`, `in()`, `both()`, `values()`, `V()`, etc.) and adds "lazy barriers" to dam up the stream so to
increase the probability of bulking the traversers. One of the side-effects is that:

[source,java]
g.V().out().V().has(a)

is compiled to:

[source,java]
g.V().out().barrier().V().barrier().has(a)

Given that `LazyBarrierStrategy` is an `OptimizationStrategy`, it comes before `ProviderOptimizationStrategies`.
Thus, if the provider's `XXXGraphStepStrategy` simply walks from the second `V()` looking for `has()`-only, it will not
be able to pull in the `has()` cause the `barrier()` blocks it. Please see the updates to `TinkerGraphStepStrategy` and
how it acknowledges `NoOpBarrierSteps` (i.e. `barrier()`) skipping over them and “left”-propagating labels to the
previous step.

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1488[TINKERPOP-1488]

===== Configurable Strategies

If the provider has non-configurable `TraversalStrategy` classes, those classes should expose a static `instance()`-method.
This is typical and thus, backwards compatible. However, if the provider has a `TraversalStrategy` that can be configured
(e.g. via a `Builder`), then it should expose a static `create(Configuration)`-method, where the keys of the configuration
are the method names of the `Builder` and the values are the method arguments. For instance, for Gremlin-Python to create
a `SubgraphStrategy`, it does the following:

[source,python]
----
g = Graph().traversal().withRemote(connection).
        withStrategies(SubgraphStrategy(vertices=__.hasLabel('person'),edges=__.has('weight',gt(0.5))))
----

The `SubgraphStrategy.create(Configuration)`-method is defined as:

[source,java]
----
public static SubgraphStrategy create(final Configuration configuration) {
    final Builder builder = SubgraphStrategy.build();
    if (configuration.containsKey(VERTICES))
        builder.vertices((Traversal) configuration.getProperty(VERTICES));
    if (configuration.containsKey(EDGES))
        builder.edges((Traversal) configuration.getProperty(EDGES));
    if (configuration.containsKey(VERTEX_PROPERTIES))
        builder.vertexProperties((Traversal) configuration.getProperty(VERTEX_PROPERTIES));
    return builder.create();
}
----

Finally, in order to make serialization possible from JVM-based Gremlin language variants, all strategies have a
`TraverserStrategy.getConfiguration()` method which returns a `Configuration` that can be used to `create()` the
`TraversalStrategy`.

The `SubgraphStrategy.getConfiguration()`-method is defined as:

[source,java]
----
@Override
public Configuration getConfiguration() {
    final Map<String, Object> map = new HashMap<>();
    map.put(STRATEGY, SubgraphStrategy.class.getCanonicalName());
    if (null != this.vertexCriterion)
        map.put(VERTICES, this.vertexCriterion);
    if (null != this.edgeCriterion)
            map.put(EDGES, this.edgeCriterion);
    if (null != this.vertexPropertyCriterion)
        map.put(VERTEX_PROPERTIES, this.vertexPropertyCriterion);
    return new MapConfiguration(map);
}
----

The default implementation of `TraversalStrategy.getConfiguration()` is defined as:

[source,java]
----
public default Configuration getConfiguration() {
    return new BaseConfiguration();
}
----

Thus, if the provider does not have any "builder"-based strategies, then no updates to their strategies are required.

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1455[TINKERPOP-1455]

===== Deprecated elementNotFound

Both `Graph.Exceptions.elementNotFound()` methods have been deprecated. These exceptions were being asserted in the
test suite but were not being used anywhere in `gremlin-core` itself. The assertions have been modified to simply
assert that `NoSuchElementException` was thrown, which is precisely the behavior that was being indirectly asserted
when `Graph.Exceptions.elementNotFound()` were being used.

Providers should not need to take any action in this case for their tests to pass, however, it would be wise to remove
uses of these exception builders as they will be removed in the future.

See: link:https://issues.apache.org/jira/browse/TINKERPOP-944[TINKERPOP-944]

===== Hidden Step Labels for Compilation Only

In order for `SubgraphStrategy` to work, it was necessary to have multi-level children communicate with one another
via hidden step labels. It was decided that hidden step labels are for compilation purposes only and will be removed
prior to traversal evaluation. This is a valid decision given that hidden labels for graph system providers are
not allowed to be used by users. Likewise, hidden labels for steps should not be allowed be used by
users as well.

===== PropertyMapStep with Selection Traversal

`PropertyMapStep` now supports selection of properties via child property traversal. If a provider was relying solely
on the provided property keys in a `ProviderOptimizationStrategy`, they will need to check if there is a child traversal
and if so, use that in their introspection for respective strategies. This model was created to support `SubgraphStrategy.vertexProperties()` filtering.

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1456[TINKERPOP-1456],
link:https://issues.apache.org/jira/browse/TINKERPOP-844[TINKERPOP-844]

===== ConnectiveP Nesting Inlined

There was a bug in `ConnectiveP` (`AndP`/`OrP`), where `eq(1).and(eq(2).and(eq(3)))` was `AndP(eq(1),AndP(eq(2),eq(3)))`
instead of unnested/inlined as `AndP(eq(1),eq(2),eq(3))`. Likewise, for `OrP`. If a provider was leveraging `ConnectiveP`
predicates for their custom steps (e.g. graph- or vertex-centric index lookups), then they should be aware of the inlining
and can simplify any and/or-tree walking code in their respective `ProviderOptimizationStrategy`.

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1470[TINKERPOP-1470]

== TinkerPop 3.2.2

*Release Date: September 6, 2016*

Please see the link:https://github.com/apache/tinkerpop/blob/3.2.2/CHANGELOG.asciidoc#release-3-2-2[changelog] for a complete list of all the modifications that are part of this release.

=== Upgrading for Users

==== GraphSON 2.0

GraphSON 2.0 has been introduced to improve and normalize the format of types embedded in GraphSON.

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1274[TINKERPOP-1274],
link:https://tinkerpop.apache.org/docs/3.2.2/reference/#graphson-2-0-types[Reference Documentation -
GraphSON 2.0].

==== Log4j Dependencies

There were a number of changes to the Log4j dependencies in the various modules. Log4j was formerly included as part
of the `slf4j-log4j12` in `gremlin-core`, however that "forced" use of Log4j as a logger implementation when that
really wasn't necessary or desired. If a project depended on `gremlin-core` or other TinkerPop project to get its
Log4j implementation then those applications will need to now include the dependency themselves directly.

Note that Gremlin Server and Gremlin Console explicitly package Log4j in their respective binary distributions.

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1151[TINKERPOP-1151]

==== Default for gremlinPool

The `gremlinPool` setting in Gremlin Server is now defaulted to zero. When set to zero, Gremlin Server will use the
value provided by `Runtime.availableProcessors()` to set the pool size. Note that the packaged YAML files no longer
contain the thread pool settings as all are now driven by sensible defaults. Obviously these values can be added
and overridden as needed.

See: https://issues.apache.org/jira/browse/TINKERPOP-1373[TINKERPOP-1373]

==== New Console Features

The Gremlin Console can now have its text colorized. For example, you can set the color of the Gremlin ascii art to
the more natural color of green by using the `:set` command:

[source,text]
gremlin> :set gremlin.color green

It is also possible to colorize results, like vertices, edges, and other common returns. Please see the
link:https://tinkerpop.apache.org/docs/3.2.2/reference/#console-preferences[reference documentation] for more details
on all the settings.

The console also now includes better multi-line support:

[source,text]
----
gremlin> g.V().out().
......1>       has('name','josh').
......2>       out('created')
==>v[5]
==>v[3]
----

This is a nice feature in that it can help you understand if a line is incomplete and unevaluated.

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1285[TINKERPOP-1285],
link:https://issues.apache.org/jira/browse/TINKERPOP-1285[TINKERPOP-1037],
link:https://tinkerpop.apache.org/docs/3.2.2/reference/#console-preferences[Reference Documentation -
Console Preferences]

==== REST API Renamed to HTTP API
This is only a rename to clarify the design of the API. There is no change to the API itself.


=== Upgrading for Providers

IMPORTANT: It is recommended that providers also review all the upgrade instructions specified for users. Many of the
changes there may prove important for the provider's implementation.

==== Graph System Providers

===== Deprecated Io.Builder.registry()

The `Io.Builder.registry()` has been deprecated in favor of `Io.Builder.onMapper(Consumer<Mapper>)`. This change gives
the `Graph` implementation greater flexibility over how to modify the `Mapper` implementation. In most cases, the
implementation will simply add its `IoRegistry` to allow the `Mapper` access to custom serialization classes, but this
approach makes it possible to also set other specific settings that aren't generalized across all IO implementations.
A good example of this type of usage would be to provide a custom `ClassRessolver` implementation to a `GryoMapper`.

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1402[TINKERPOP-1402]

===== Log4j Dependencies

There were a number of changes to the Log4j dependencies in the various modules. Log4j was formerly included as part
of the `slf4j-log4j12` in `gremlin-core`, however that "forced" use of log4j as a logger implementation when that
really wasn't necessary or desired. The `slf4j-log4j12` dependency is now in "test" scope for most of the modules. The
exception to that rule is `gremlin-test` which prescribes it as "optional". That change means that developers
dependending on `gremlin-test` (or `gremlin-groovy-test`) will need to explicitly specify it as a dependency in their
`pom.xml` (or a different slf4j implementation if that better suits them).

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1151[TINKERPOP-1151]

==== Drivers Providers

===== GraphSON 2.0

Drivers providers can exploit the new format of typed values JSON serialization offered by GraphSON 2.0. This format
has been created to allow easy and agnostic parsing of a GraphSON payload without type loss. Drivers of non-Java
languages can then implement their own mapping of the GraphSON's language agnostic type IDs (e.g. `UUID`, `LocalDate`)
to the appropriate representation for the driver's language.

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1274[TINKERPOP-1274],
link:https://tinkerpop.apache.org/docs/3.2.2/reference/#graphson-2.0-types[Reference Documentation -
GraphSON 2.0].

===== Traversal Serialization

There was an "internal" serialization format in place for `Traversal` which allowed one to be submitted to Gremlin
Server directly over `RemoteGraph`. That format has been removed completely and is wholly replaced by the non-JVM
specific approach of serializing `Bytecode`.

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1392[TINKERPOP-1392]

== TinkerPop 3.2.1

*Release Date: July 18, 2016*

Please see the link:https://github.com/apache/tinkerpop/blob/3.2.1/CHANGELOG.asciidoc#release-3-2-1[changelog] for a complete list of all the modifications that are part of this release.

=== Upgrading for Users

==== Gephi Plugin

The Gephi Plugin has been updated to support Gephi 0.9.x. Please upgrade to this latest version to use the Gephi Plugin
for Gremlin Console.

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1297[TINKERPOP-1297]

==== GryoMapper Construction

It is now possible to override existing serializers with calls to `addCustom` on the `GryoMapper` builder. This option
allows complete control over the serializers used by Gryo. Of course, this also makes it possible to produce completely
non-compliant Gryo files. This feature should be used with caution.

==== TraversalVertexProgram

`TraversalVertexProgram` always maintained a `HALTED_TRAVERSERS` `TraverserSet` for each vertex throughout the life
of the OLAP computation. However, if there are no halted traversers in the set, then there is no point in keeping that
compute property around as without it, time and space can be saved. Users that have `VertexPrograms` that are chained off
of `TraversalVertexProgram` and have previously assumed that `HALTED_TRAVERSERS` always exists at each vertex, should no
longer assume that.

[source,java]
// bad code
TraverserSet haltedTraversers = vertex.value(TraversalVertexProgram.HALTED_TRAVERSERS);
// good code
TraverserSet haltedTraversers = vertex.property(TraversalVertexProgram.HALTED_TRAVERSERS).orElse(new TraverserSet());

==== Interrupting Traversals

Traversals now better respect calls to `Thread.interrupt()`, which mean that a running `Traversal` can now be
cancelled. There are some limitations that remain, but most OLTP-based traversals should cancel without
issue. OLAP-based traversals for Spark will also cancel and clean up running jobs in Spark itself. Mileage may vary
on other process implementations and it is possible that graph providers could potentially write custom step
implementations that prevent interruption. If it is found that there are configurations or specific traversals that
do not respect interruption, please mention them on the mailing list.

See: https://issues.apache.org/jira/browse/TINKERPOP-946[TINKERPOP-946]

==== Gremlin Console Flags

Gremlin Console had several methods for executing scripts from file at the start-up of `bin/gremlin.sh`. There were
two options:

[source,text]
bin/gremlin.sh script.groovy    <1>
bin/gremlin.sh -e script.groovy <2>

<1> The `script.groovy` would be executed as a console initialization script setting the console up for use and leaving
it open when the script completed successfully or closing it if the script failed.
<2> The `script.groovy` would be executed by the `ScriptExecutor` which meant that commands for the Gremlin Console,
such as `:remote` and `:>` would not be respected.

Changes in this version of TinkerPop have added much more flexibility here and only a minor breaking change should be
considered when using this version. First of all, recognize that hese two lines are currently equivalent:

[source,text]
bin/gremlin.sh script.groovy
bin/gremlin.sh -i script.groovy

but users should start to explicitly specify the `-i` flag as TinkerPop will eventually remove the old syntax. Despite
the one used beware of the fact that neither will close the console on script failure anymore. In that sense, this
behavior represents a breaking change to consider. To ensure the console closes on failure or success, a script will
have to use the `-e` option.

The console also has a number of new features in addition to `-e` and `-i`:

* View the available flags for the console with `-h`.
* Control console output with `-D`, `-Q` and -`V`
* Get line numbers on script failures passed to `-i` and `-e`.

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1268[TINKERPOP-1268],
link:https://issues.apache.org/jira/browse/TINKERPOP-1155[TINKERPOP-1155], link:https://issues.apache.org/jira/browse/TINKERPOP-1156[TINKERPOP-1156],
link:https://issues.apache.org/jira/browse/TINKERPOP-1157[TINKERPOP-1157],
link:https://tinkerpop.apache.org/docs/3.2.1/reference/#interactive-mode[Reference Documentation - Interactive Mode],
link:https://tinkerpop.apache.org/docs/3.2.1/reference/#execution-mode[Reference Documentation - Execution Mode]

=== Upgrading for Providers

IMPORTANT: It is recommended that providers also review all the upgrade instructions specified for users. Many of the
changes there may prove important for the provider's implementation.

==== Graph System Providers

===== VertexComputing API Change

The `VertexComputing` API is used by steps that wrap a `VertexProgram`. There is a method called
`VertexComputing.generateProgram()` that has changed which now takes a second argument of `Memory`. To  upgrade, simply
fix the method signature of your `VertexComputing` implementations. The `Memory` argument can be safely ignored to
effect the exact same semantics as prior. However, now previous OLAP job `Memory` can be leveraged when constructing
the next `VertexProgram` in an OLAP traversal chain.

===== Interrupting Traversals

Several tests have been added to the TinkerPop test suite to validate that a `Traversal` can be cancelled with
`Thread.interrupt()`. The test suite does not cover all possible traversal scenarios. When implementing custom steps,
providers should take care to not ignore an `InterruptionException` that might be thrown in their code and to be sure
to check `Thread.isInterrupted()` as needed to ensure that the step remains cancellation compliant.

See: https://issues.apache.org/jira/browse/TINKERPOP-946[TINKERPOP-946]

===== Performance Tests

All "performance" tests have been deprecated. In the previous 3.2.0-incubating release, the `ProcessPerformanceSuite`
and `TraversalPerformanceTest` were deprecated, but some other tests remained. It is the remaining tests that have
been deprecated on this release:

* `StructurePerformanceSuite
** `GraphReadPerformanceTest`
** `GraphWriterPerformanceTest`
* `GroovyEnvironmentPerformanceSuite`
** `SugarLoaderPerformanceTest`
** `GremlinExecutorPerformanceTest`
* Gremlin Server related performance tests
* TinkerGraph related performance tests

Providers should implement their own performance tests and not rely on these deprecated tests as they will be removed
in a future release along with the "JUnit Benchmarks" dependency.

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1294[TINKERPOP-1294]

==== Graph Database Providers

===== Transaction Tests

Tests and assertions were added to the structure test suite to validate that transaction status was in the appropriate
state following calls to close the transaction with `commit()` or `rollback()`. It is unlikely that this change would
cause test breaks for providers, unless the transaction status was inherently disconnected from calls to close the
transaction somehow.

In addition, other tests were added to enforce the expected semantics for threaded transactions. Threaded transactions
are expected to behave like manual transactions. They should be open automatically when they are created and once
closed should no longer be used. This behavior is not new and is the typical expected method for working with these
types of transactions. The test suite just requires that the provider implementation conform to these semantics.

See: link:https://issues.apache.org/jira/browse/TINKERPOP-947[TINKERPOP-947],
link:https://issues.apache.org/jira/browse/TINKERPOP-1059[TINKERPOP-1059]

===== GraphFilter and GraphFilterStrategy

`GraphFilter` has been significantly advanced where the determination of an edge direction/label legality is more stringent.
Along with this, `GraphFilter.getLegallyPositiveEdgeLabels()` has been added as a helper method to make it easier for `GraphComputer`
providers to know the space of labels being accessed by the traversal and thus, better enable provider-specific push-down predicates.

Note that `GraphFilterStrategy` is now a default `TraversalStrategy` registered with `GraphComputer.` If `GraphFilter` is
expensive for the underlying `GraphComputer` implementation, it can be deactivated as is done for `TinkerGraphComputer`.

[source,java]
----
static {
  TraversalStrategies.GlobalCache.registerStrategies(TinkerGraphComputer.class,
    TraversalStrategies.GlobalCache.getStrategies(GraphComputer.class).clone().removeStrategies(GraphFilterStrategy.class));
}
----

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1293[TINKERPOP-1293]

==== Graph Language Providers

===== VertexTest Signatures

The method signatures of `get_g_VXlistXv1_v2_v3XX_name` and `get_g_VXlistX1_2_3XX_name` of `VertexTest` were changed
to take arguments for the `Traversal` to be constructed by extending classes.

== TinkerPop 3.2.0

*Release Date: Release Date: April 8, 2016*

Please see the link:https://github.com/apache/tinkerpop/blob/3.2.0-incubating/CHANGELOG.asciidoc#tinkerpop-320-release-date-april-8-2016[changelog] for a complete list of all the modifications that are part of this release.

=== Upgrading for Users

==== Hadoop FileSystem Variable

The `HadoopGremlinPlugin` defines two variables: `hdfs` and `fs`. The first is a reference to the HDFS `FileSystemStorage`
and the latter is a reference to the local `FileSystemStorage`. Prior to 3.2.x, `fs` was called `local`. However,
there was a variable name conflict with `Scope.local`. As such `local` is now `fs`. This issue existed prior to 3.2.x,
but was not realized until this release. Finally, this only effects Gremlin Console users.

==== Hadoop Configurations

Note that `gremlin.hadoop.graphInputFormat`, `gremlin.hadoop.graphOutputFormat`, `gremlin.spark.graphInputRDD`, and
`gremlin.spark.graphOuputRDD` have all been deprecated. Using them still works, but moving forward, users only need to
leverage `gremlin.hadoop.graphReader` and `gremlin.hadoop.graphWriter`. An example properties file snippet is provided
below.

```
gremlin.graph=org.apache.tinkerpop.gremlin.hadoop.structure.HadoopGraph
gremlin.hadoop.graphReader=org.apache.tinkerpop.gremlin.hadoop.structure.io.gryo.GryoInputFormat
gremlin.hadoop.graphWriter=org.apache.tinkerpop.gremlin.hadoop.structure.io.gryo.GryoOutputFormat
gremlin.hadoop.jarsInDistributedCache=true
gremlin.hadoop.defaultGraphComputer=org.apache.tinkerpop.gremlin.spark.process.computer.SparkGraphComputer
```

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1082[TINKERPOP-1082],
link:https://issues.apache.org/jira/browse/TINKERPOP-1222[TINKERPOP-1222]

==== TraversalSideEffects Update

There were changes to `TraversalSideEffect` both at the semantic level and at the API level. Users that have traversals
of the form `sideEffect{...}` that leverage global side-effects should read the following carefully. If the user's traversals do
not use lambda-based side-effect steps (e.g. `groupCount("m")`), then the changes below will not effect them. Moreover, if user's
traversal only uses `sideEffect{...}` with closure (non-`TraversalSideEffect`) data references, then the changes below will not effect them.
If the user's traversal uses sideEffects in OLTP only, the changes below will not effect them. Finally, providers should not be
effected by the changes save any tests cases.

===== TraversalSideEffects Get API Change

`TraversalSideEffects` can now logically operate within a distributed OLAP environment. In order to make this possible,
it is necessary that each side-effect be registered with a reducing `BinaryOperator`. This binary operator will combine
distributed updates into a single global side-effect at the master traversal. Many of the methods in `TraversalSideEffect`
have been `Deprecated`, but they are backwards compatible save that `TraversalSideEffects.get()` no longer returns an `Optional`,
but instead throws an `IllegalArgumentException`. While the `Optional` semantics could have remained, it was deemed best to
directly return the side-effect value to reduce object creation costs and because all side-effects must be registered apriori,
there is never a reason why an unknown side-effect key would be used. In short:

[source,java]
----
// change
traversal.getSideEffects().get("m").get()
// to
traversal.getSideEffects().get("m")
----

===== TraversalSideEffects Registration Requirement

All `TraversalSideEffects` must be registered upfront. This is because, in OLAP, side-effects map to `Memory` compute keys
and as such, must be declared prior to the execution of the `TraversalVertexProgram`. If a user's traversal creates a
side-effect mid-traversal, it will fail. The traversal must use `GraphTraversalSource.withSideEffect()` to declare
the side-effects it will use during its execution lifetime. If the user's traversals use standard side-effect Gremlin
steps (e.g. `group("m")`), then no changes are required.

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1192[TINKERPOP-1192]

===== TraversalSideEffects Add Requirement

In a distributed environment, a side-effect can not be mutated and be expected to exist in the mutated form at the final,
aggregated, master traversal. For instance, if the side-effect "myCount" references a `Long`, the `Long` can not be updated
directly via `sideEffects.set("myCount", sideEffects.get("myCount") + 1)`. Instead, it must rely on the registered reducer
to do the merging and thus, the `Step` must do `sideEffect.add("mySet",1)`, where the registered reducer is `Operator.sum`.
Thus, the below will increment "a". If no operator was provided, then the operator is assumed `Operator.assign` and the
final result of "a" would be 1. Note that `Traverser.sideEffects(key,value)` uses `TraversalSideEffect.add()`.

[source,groovy]
----
gremlin> traversal = g.withSideEffect('a',0,sum).V().out().sideEffect{it.sideEffects('a',1)}
==>v[3]
==>v[2]
==>v[4]
==>v[5]
==>v[3]
==>v[3]
gremlin> traversal.getSideEffects().get('a')
==>6
gremlin> traversal = g.withSideEffect('a',0).V().out().sideEffect{it.sideEffects('a',1)}
==>v[3]
==>v[2]
==>v[4]
==>v[5]
==>v[3]
==>v[3]
gremlin> traversal.getSideEffects().get('a')
==>1
----

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1192[TINKERPOP-1192],
https://issues.apache.org/jira/browse/TINKERPOP-1166[TINKERPOP-1166]

==== ProfileStep Update and GraphTraversal API Change

The `profile()`-step has been refactored into 2 steps -- `ProfileStep` and `ProfileSideEffectStep`. Users who previously
used the `profile()` in conjunction with `cap(TraversalMetrics.METRICS_KEY)` can now simply omit the cap step. Users who
retrieved `TraversalMetrics` from the side-effects after iteration can still do so, but will need to specify a side-effect
key when using the `profile()`. For example, `profile("myMetrics")`.

See: link:https://issues.apache.org/jira/browse/TINKERPOP-958[TINKERPOP-958]

==== BranchStep Bug Fix

There was a bug in `BranchStep` that also rears itself in subclass steps such as `UnionStep` and `ChooseStep`.
For traversals with branches that have barriers (e.g. `count()`, `max()`, `groupCount()`, etc.), the traversal needs to be updated.
For instance, if a traversal is of the form  `g.V().union(out().count(),both().count())`, the result is now different
(the bug fix yields a different output). In order to yield the same result, the traversal should be rewritten as
`g.V().local(union(out().count(),both().count()))`. Note that if a branch does not have a barrier, then no changes are required.
For instance, `g.V().union(out(),both())` does not need to be updated. Moreover, if the user's traversal already used
the `local()`-form, then no change are required either.

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1188[TINKERPOP-1188]

==== MemoryComputeKey and VertexComputeKey

Users that have custom `VertexProgram` implementations will need to change their implementations to support the new
`VertexComputeKey` and `MemoryComputeKey` classes. In the `VertexPrograms` provided by TinkerPop, these changes were trivial,
taking less than 5 minutes to make all the requisite updates.

* `VertexProgram.getVertexComputeKeys()` returns a `Set<VertexComputeKey>`. No longer a `Set<String>`.
Use `VertexComputeKey.of(String key,boolean transient)` to generate a `VertexComputeKey`.
Transient keys were not supported in the past, so to make the implementation semantically equivalent,
the boolean transient should be false.

* `VertexProgram.getMemoryComputeKeys()` returns a `Set<MemoryComputeKey>`. No longer a `Set<String>`.
Use `MemoryComputeKey.of(String key, BinaryOperator reducer, boolean broadcast, boolean transient)` to generate a `MemoryComputeKey`.
Broadcasting and transients were not supported in the past so to make the implementation semantically equivalent,
the boolean broadcast should be true and the boolean transient should be false.

An example migration looks as follows. What might currently look like:

```
public Set<String> getMemoryComputeKeys() {
   return new HashSet<>(Arrays.asList("a","b","c"))
}
```

Should now look like:

```
public Set<MemoryComputeKey> getMemoryComputeKeys() {
  return new HashSet<>(Arrays.asList(
    MemoryComputeKey.of("a", Operator.and, true, false),
    MemoryComputeKey.of("b", Operator.sum, true, false),
    MemoryComputeKey.of("c", Operator.or, true, false)))
}
```

A similar patterns should also be used for `VertexProgram.getVertexComputeKeys()`.

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1162[TINKERPOP-1162]

==== SparkGraphComputer and GiraphGraphComputer Persistence

The `MapReduce`-based steps in `TraversalVertexProgram` have been removed and replaced using a new `Memory`-reduction model.
`MapReduce` jobs always created a persistence footprint, e.g. in HDFS. `Memory` data was never persisted to HDFS.
As such, there will be no data on the disk that is accessible. For instance, there is no more `~reducing`, `~traversers`,
and specially named side-effects such as `m` from a `groupCount('m')`. The data is still accessible via `ComputerResult.memory()`,
it simply does not have a corresponding on-disk representation.

==== RemoteGraph

`RemoteGraph` is a lightweight `Graph` implementation that acts as a proxy for sending traversals to Gremlin Server for
remote execution. It is an interesting alternative to the other methods for connecting to Gremlin Server in that all
other methods involved construction of a `String` representation of the `Traversal` which is then submitted as a script
to Gremlin Server (via driver or HTTP).

[source,groovy]
----
gremlin> graph = RemoteGraph.open('conf/remote-graph.properties')
==>remotegraph[DriverServerConnection-localhost/127.0.0.1:8182 [graph='graph]]
gremlin> g = graph.traversal()
==>graphtraversalsource[remotegraph[DriverServerConnection-localhost/127.0.0.1:8182 [graph='graph]], standard]
gremlin> g.V().valueMap(true)
==>[name:[marko], label:person, id:1, age:[29]]
==>[name:[vadas], label:person, id:2, age:[27]]
==>[name:[lop], label:software, id:3, lang:[java]]
==>[name:[josh], label:person, id:4, age:[32]]
==>[name:[ripple], label:software, id:5, lang:[java]]
==>[name:[peter], label:person, id:6, age:[35]]
----

Note that `g.V().valueMap(true)` is executing in Gremlin Server and not locally in the console.

See: link:https://issues.apache.org/jira/browse/TINKERPOP-575[TINKERPOP-575],
link:https://tinkerpop.apache.org/docs/3.2.0-incubating/reference/#connecting-via-remotegraph[Reference Documentation - Remote Graph]

=== Upgrading for Providers

IMPORTANT: It is recommended that providers also review all the upgrade instructions specified for users. Many of the
changes there may prove important for the provider's implementation.

==== Graph System Providers

===== GraphStep Compilation Requirement

OLTP graph providers that have a custom `GraphStep` implementation should ensure that `g.V().hasId(x)` and `g.V(x)` compile
to the same representation. This ensures a consistent user experience around random access of elements based on ids
(as opposed to potentially the former doing a linear scan). A static helper method called `GraphStep.processHasContainerIds()`
has been added. `TinkerGraphStepStrategy` was updated as such:

```
((HasContainerHolder) currentStep).getHasContainers().forEach(tinkerGraphStep::addHasContainer);
```

is now

```
((HasContainerHolder) currentStep).getHasContainers().forEach(hasContainer -> {
  if (!GraphStep.processHasContainerIds(tinkerGraphStep, hasContainer))
    tinkerGraphStep.addHasContainer(hasContainer);
});
```

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1219[TINKERPOP-1219]

===== Step API Update

The `Step` interface is fundamental to Gremlin. `Step.processNextStart()` and `Step.next()` both returned `Traverser<E>`.
We had so many `Traverser.asAdmin()` and direct typecast calls throughout (especially in `TraversalVertexProgram`) that
it was deemed prudent to have `Step.processNextStart()` and `Step.next()` return `Traverser.Admin<E>`. Moreover it makes
sense as this is internal logic where `Admins` are always needed. Providers with their own step definitions will simply
need to change the method signatures of `Step.processNextStart()` and `Step.next()`. No logic update is required -- save
that `asAdmin()` can be safely removed if used. Also, `Step.addStart()` and `Step.addStarts()` take `Traverser.Admin<S>`
and `Iterator<Traverser.Admin<S>>`, respectively.

===== Traversal API Update

The way in which `TraverserRequirements` are calculated has been changed (for the better). The ramification is that post
compilation requirement additions no longer make sense and should not be allowed. To enforce this,
`Traversal.addTraverserRequirement()` method has been removed from the interface. Moreover, providers/users should never be able
to add requirements manually (this should all be inferred from the end compilation). However, if need be, there is always
`RequirementStrategy` which will allow the provider to add a requirement at strategy application time
(though again, there should not be a reason to do so).

===== ComparatorHolder API Change

Providers that either have their own `ComparatorHolder` implementation or reason on `OrderXXXStep` will need to update their code.
`ComparatorHolder` now returns `List<Pair<Traversal,Comparator>>`. This has greatly reduced the complexity of comparison-based
steps like `OrderXXXStep`. However, its a breaking API change that is trivial to update to, just some awareness is required.

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1209[TINKERPOP-1209]

===== GraphComputer Semantics and API

Providers that have a custom `GraphComputer` implementation will have a lot to handle. Note that if the graph system
simply uses `SparkGraphComputer` or `GiraphGraphComputer` provided by TinkerPop, then no updates are required. This
only effects providers that have their own custom `GraphComputer` implementations.

`Memory` updates:

* Any `BinaryOperator` can be used for reduction and is made explicit in the `MemoryComputeKey`.
* `MemoryComputeKeys` can be marked transient and must be removed from the resultant `ComputerResult.memory()`.
* `MemoryComputeKeys` can be specified to not broadcast and thus, must not be available to workers to read in `VertexProgram.execute()`.
* The `Memory` API has been changed. No more `incr()`, `and()`, etc. Now its just `set()` (setup/terminate) and `add()` (execute).

`VertexProgram` updates:

* `VertexComputeKeys` can be marked transient and must be removed from the resultant `ComputerResult.graph()`.

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1166[TINKERPOP-1166],
link:https://issues.apache.org/jira/browse/TINKERPOP-1164[TINKERPOP-1164],
link:https://issues.apache.org/jira/browse/TINKERPOP-951[TINKERPOP-951]

Operational semantic test cases have been added to `GraphComputerTest` to ensure that all the above are implemented correctly.

===== Barrier Step Updates

The `Barrier` interface use to simply be a marker interface. Now it has methods and it is the primary means by which
distributed steps across an OLAP job are aggregated and distributed. It is unlikely that `Barrier` was ever used
directly by a provider's custom step. Instead, a provider most likely extended `SupplyingBarrierStep`, `CollectingBarrierStep`,
and/or `ReducingBarrierStep`.

Providers that have custom extensions to these steps or that use `Barrier` directly will need to adjust their implementation slightly to
accommodate a new API that reflects the `Memory` updates above. This should be a simple change. Note that `FinalGet`
no longer exists and such post-reduction processing is handled by the reducing step (via the new `Generating` interface).

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1164[TINKERPOP-1164]

===== Performance Tests

The `ProcessPerformanceSuite` and `TraversalPerformanceTest` have been deprecated.  They are still available, but going forward,
providers should implement their own performance tests and not rely on the built-in JUnit benchmark-based performance test suite.

==== Graph Processor Providers

===== GraphFilter and GraphComputer

The `GraphComputer` API has changed with the addition of `GraphComputer.vertices(Traversal)` and `GraphComputer.edges(Traversal)`.
These methods construct a `GraphFilter` object which is also new to TinkerPop 3.2.0. `GraphFilter` is a "push-down predicate"
used to selectively retrieve subgraphs of the underlying graph to be OLAP processed.

* If the graph system provider relies on an existing `GraphComputer` implementations such as `SparkGraphComputer` and/or `GiraphGraphComputer`,
then there is no immediate action required on their part to remain TinkerPop-compliant. However, they may wish to update
their `InputFormat` or `InputRDD` implementation to be `GraphFilterAware` and handle the `GraphFilter` filtering at the disk/database
level. It is advisable to do so in order to reduce OLAP load times and memory/GC usage.

* If the graph system provider has their own `GraphComputer` implementation, then they should implement the two new methods
and ensure that `GraphFilter` is processed correctly. There is a new test case called `GraphComputerTest.shouldSupportGraphFilter()`
which ensures the semantics of `GraphFilter` are handled correctly. For a "quick and easy" way to move forward, look to
`GraphFilterInputFormat` as a way of wrapping an existing `InputFormat` to do filtering prior to `VertexProgram` or `MapReduce`
execution.

NOTE: To quickly move forward, the `GraphComputer` implementation can simply set `GraphComputer.Features.supportsGraphFilter()`
to `false` and ensure that `GraphComputer.vertices()` and `GraphComputer.edges()` throws `GraphComputer.Exceptions.graphFilterNotSupported()`.
This is not recommended as its best to support `GraphFilter`.

See: link:https://issues.apache.org/jira/browse/TINKERPOP-962[TINKERPOP-962]

===== Job Chaining and GraphComputer

TinkerPop 3.2.0 has integrated `VertexPrograms` into `GraphTraversal`. This means, that a single traversal can compile to multiple
`GraphComputer` OLAP jobs. This requires that `ComputeResults` be chainable. There was never any explicit tests to verify if a
provider's `GraphComputer` could be chained, but now there are. Given a reasonable implementation, it is likely that no changes
are required of the provider. However, to ensure the implementation is "reasonable" `GraphComputerTests` have been added.

* For providers that support their own `GraphComputer` implementation, note that there is a new `GraphComputerTest.shouldSupportJobChaining()`.
This tests verifies that the `ComputerResult` output of one job can be fed into the input of a subsequent job. Only linear chains are tested/required
currently. In the future, branching DAGs may be required.

* For providers that support their own `GraphComputer` implementation, note that there is a new `GraphComputerTest.shouldSupportPreExistingComputeKeys()`.
When chaining OLAP jobs together, if an OLAP job requires the compute keys of a previous OLAP job, then the existing compute keys must be accessible.
A simple 2 line change to `SparkGraphComputer` and `TinkerGraphComputer` solved this for TinkerPop. `GiraphGraphComputer` did not need an update as
this feature was already naturally supported.

See: link:https://issues.apache.org/jira/browse/TINKERPOP-570[TINKERPOP-570]

==== Graph Language Providers

===== ScriptTraversal

Providers that have custom Gremlin language implementations (e.g. Gremlin-Scala), there is a new class called `ScriptTraversal`
which will handle script-based processing of traversals. The entire `GroovyXXXTest`-suite was updated to use this new class.
The previous `TraversalScriptHelper` class has been deprecated so immediate upgrading is not required, but do look into
`ScriptTraversal` as TinkerPop will be using it as a way to serialize "String-based traversals" over the network moving forward.

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1154[TINKERPOP-1154]

===== ByModulating and Custom Steps

If the provider has custom steps that leverage `by()`-modulation, those will now need to implement `ByModulating`.
Most of the methods in `ByModulating` are `default` and, for most situations, only `ByModulating.modulateBy(Traversal)`
needs to be implemented. Note that this method's body will most like be identical the custom step's already existing
`TraversalParent.addLocalChild()`. It is recommended that the custom step not use `TraversalParent.addLocalChild()`
as this method may be deprecated in a future release. Instead, barring any complex usages, simply rename the
`CustomStep.addLocalChild(Traversal)` to `CustomStep.modulateBy(Traversal)`.

See: link:https://issues.apache.org/jira/browse/TINKERPOP-1153[TINKERPOP-1153]

===== TraversalEngine Deprecation and GraphProvider

The `TraversalSource` infrastructure has been completely rewritten. Fortunately for users, their code is backwards compatible.
Unfortunately for graph system providers, a few tweaks to their implementation are in order.

* If the graph system supports more than `Graph.compute()`, then implement `GraphProvider.getGraphComputer()`.
* For custom `TraversalStrategy` implementations, change `traverser.getEngine().isGraphComputer()` to `TraversalHelper.onGraphComputer(Traversal)`.
* For custom `Steps`, change `implements EngineDependent` to `implements GraphComputing`.

See: link:https://issues.apache.org/jira/browse/TINKERPOP-971[TINKERPOP-971]