Nebula Graph Versions Save

A distributed, fast open-source graph database featuring horizontal scalability and high availability. This is an archived repo for v2.5 only, from 2.6.0 +, NebulaGraph switched back to https://github.com/vesoft-inc/nebula

v2.5.1

2 years ago

bug fix:

download:

v2.5.0

2 years ago

feature

bug fix

enhancement

change

  • A little bit grammar change of Subgraph
# Add the WITH PROP keyword to the output property
GET SUBGRAPH WITH PROP FROM <vids>

# The original syntax will only output the graph structure without properties
GET SUBGRAPH FROM <vids>
  • we must use the symbol $-. in ‘ORDER BY’. But in earlier versions, there is no need. Example:
Before:
(root@nebula) [basketballplayer]> LOOKUP ON player \
                               ->         YIELD player.age As playerage \
                               ->         | GROUP BY $-.playerage \
                               ->         YIELD $-.playerage as age, count(*) AS number \
                               ->         | ORDER BY number DESC, age DESC;


From v2.5.0
(czp@nebula) [basketballplayer]> LOOKUP ON player \
                               ->         YIELD player.age As playerage \
                               ->         | GROUP BY $-.playerage \
                               ->         YIELD $-.playerage as age, count(*) AS number \
                               ->         | ORDER BY $-.number DESC, $-.age DESC;

v2.0.1

3 years ago

Improvements

Bugfix

v2.0.0

3 years ago

New Features

  • vertexID supports both Integer and String.
  • New data types:
    • NULL: the property can be set to NULL. NOT NULL constraint is also supported
    • Composite types: LIST, SET, and MAP(Cannot be set as property types)
    • Temporal types: DATE and DATETIME.
    • FIXED_STRING: a fixed size String
  • Full-text indexes are supported to do prefix, wildcard, regex, and fuzzy search on a string property.
  • Explain & Profile outputs the execution plan of an nGQL statement and execution profile.
  • Subgraph to retrieve vertices and edges reachable from the start vertices.
  • Support to collect statistics of the graph space.
  • OpenCypher compatibility
    • Partially support the MATCH clause
    • Support RETURN, WITH, UNWIND, LIMIT & SKIP clauses
  • More built-in functions
    • Predicate functions
    • Scalar functions
    • List functions
    • Aggregating functions
    • Mathematical functions
    • String functions
    • Temporal functions

Improvements

  • Optimize the performance of inserting, updating, and deleting data with indexes.
  • LOOKUP ON filtering data supports OR and AND operators.
  • FIND PATH supports finding paths with or without regard to direction, and also supports excluding cycles in paths.
  • SHOW HOSTS graph/meta/storage supports to retrieve the basic information of graphd/metad/storaged hosts.

Changelog

  • The data type of vertexID must be specified when creating a graph space.
  • FETCH PROP ON returns a composite object if not specify the result set.
  • Changed the default port numbers of metad, graphd, and storaged.
  • Refactor metrics counters.

Nebula-graph Console

Supports local commands mode. :set csv outputs the query results to the console and the specified CSV file. For more information, please refer to https://github.com/vesoft-inc/nebula-console.

Clients

Support connection pool and load balance.

Nebula Graph Studio

With Studio, you can create a graph schema, load data, execute nGQL statements, and explore graphs in one stop. For more information, please refer to https://github.com/vesoft-inc/nebula-web-docker.

Known Issues

v2.0.0-rc1

3 years ago

New Features

Changelog

v2.0.0-beta

3 years ago

Nebula Graph

Note: Schema & data are not compatible with 1.0 yet. Still working on the compatibility.

Compatibility with more nGQL 1.x queries

  • DESCRIBE TAG/EDGE INDEX is to obtain information about the index.
  • REBUILD TAG/EDGE INDEX is to rebuild an index.
  • SHOW TAG/EDGE INDEXES is to list the created indexes.
  • SHOW TAG/EDGE INDEXES STATUS is to return the created index status.

New Features

  • full-text search, allows users to write queries that match terms anywhere within the indexed string properties.
  • LOOKUP ON TAG/EDGE is to get all vertices of a tag or edges of an edge type.
  • SUBMIT JOB STATS & SHOW STATS is to get all vertices and edges in the current space.
  • Support more openCypher Clauses.
    • Partially supported MATCH clause
    • RETURN & DISTINCT
    • LIMIT & SKIP
  • Add more built-in functions
    • Predicate functions
    • Scalar functions
    • List functions
    • Aggregating functions
    • Mathematical functions
    • String functions
    • Temporal functions

Clients

Nebula Graph Studio

  • Compatible with Nebula Graph 2.x，refer to Studio.

Tools

v2.0.0-alpha

3 years ago

Nebula Graph

Note: Schema & data are not compatible with 1.0 yet. Still working on the compatibility.

New Features

  • Support String vertex ID
  • New data types
    • NULL, the property for vertices or edges can be set to NULL. NOT NULL constraint is supported to ensure the property values are not NULL.
    • Composite types: list, set and map(Cannot be stored as properties).
    • Temporal types like DATE and DATETIME.
  • New Functions
    • size() returns the number of items in a list
    • time() returns the current Time
    • date() or datetime() Returns the current Date or DateTime
    • id() returns the id of a vertex
    • tags() and labels() return the tags of the vertex
    • collect(), an aggregation function, returns a list containing the values returned by an expression
    • collect_set(), aggregation function, returns a set containing the values returned by an expression
    • properties() returns a map containing all the properties of a vertex or an edge
    • type() returns the edge type of an edge
    • src() returns the src id of an edge
    • dst() returns the dst id of an edge
    • rank() returns the rank of an edge
  • New Operators
    • IN checks if a specified value matches any value in a list.
    • CONTAINS performs a case-sensitive match regardless of location within a string.
    • STARTS WITH performs a case-sensitive match for the beginning of a string.
    • ENDS WITH performs a case-sensitive match for the ending of a string.
  • Explain & Profile. EXPLAIN displays the execution plan of a nGQL statement without executing the statement. PROFILE executes the statement, then outputs the execution plan as well as the execution profile.
  • Subgraph. GET SUBGRAPH statement retrieves information of vertices and edges reachable from the start vertices over the specified edges to the maximum number of hops.
  • Partially supported MATCH statement.
  • A DELETE statement can have an input from the pipeline |.

Features different from 1.0

  • CREATE SPACE supports specifying the type of vertex ID. The type fixed_string( int64 will be supported in the next release).
  • Simplify UPDATE/UPSERT statement.
  • UPSERT ... SET the unspecified properties will be set to the default value when inserting a new vertex/edge. If no default value is set, they will be set to NULL. If the NOT NULL constraint is set on the property, an error will return.
  • $- in the pipeline statement is not equal to $-.id anymore.
  • The expression specified in the default value will be computed when the data inserted instead of schema defined.

Upcoming Soon (Work in process)

  • DESCRIBE TAG/EDGE INDEX
  • REBUILD TAG/EDGE INDEX
  • SHOW TAG/EDGE INDEXES
  • SHOW TAG/EDGE INDEX STATUS
  • FETCH PROP ON multiple tags
  • Compatible with 1.0 Integer vertex ID
  • Support more OpenCypher statements
  • Backup & Restore
  • Fulltext search

Console

  • Nebula-console supports local commands now. For example, :set csv outputs the query results to the console and the specified CSV file. For more information, please refer to nebula-console.

Clients

  • Java
  • Python
  • Golang

Clients support connection pooling and load balancing. For more information, please refer to nebula-clients.

Studio

  • Not Compatible with Nebula Graph 2.0 yet (Working in progress).

Document