knot-db
v1.1.12
Published
Relationship driven No-SQL database
Downloads
25
Readme
KnotDatabase
Get started
Installation
Run the following command to install the package
npm install -g knot-db
Then to start the database, you can simply run :
knot-db
The database should now be running on the port specified in the config.json file. (9600 by default)
Sending queries
To query the database, simply send an HTTP post request at localhost:<PORT>/query
with a body like the following :
POST/query
{
"table" : "employees",
"query" : "MATCH (a:Person {name: 'Marc'}) RETURN a;"
}
where table is the name of the table to apply the query and query is the actual query to execute.
A typical response would like this :
{
"a": {
"P256": {
"id": "P256",
"type": "Person",
"attributes": {
"name": "Marc",
"surname": "Davis",
"sector": "Marketing"
},
"relationships": {
"likes": ["P965"]
}
},
"P443": {
"id": "P443",
"type": "Person",
"attributes": {
"name": "Marc",
"surname": "Jones",
"sector": "Sales",
"intern": true
},
"relationships": {}
}
}
}
Insert query
// Adds a node 'P001' of type 'Person' with name 'John Doe' who owns node 'B001'.
INSERT (P001:Person {name: 'John Doe'} {owns: ['B001']});
Valid Formats:
- INSERT ( Id : Type ) ;
- INSERT ( Id : Type { Attributes } ) ;
- INSERT ( Id : Type { Attributes } { Relations } ) ;
Id: Id of the node. Must be unique.
Upper or lower camelcase format required.
Type: Type of the node.
Upper or lower camelcase format required.
Attributes: Attributes of the node.
JSON format required.
Relations: Relations of the node.
JSON format required.
Examples:
// Inserts a node with id '10001' of type 'Person'.
INSERT (10001:Person);
// Inserts a node with id '10001' of type 'Person' with name 'Paul Jones'.
INSERT (10001:Person {name: 'Paul Jones'});
// Inserts a node with id '10001' of type 'Person' with name 'Paul Jones' who likes another node with id '10003'.
INSERT (10001:Person {name: 'Paul Jones'} {likes: ['10003']});
Insert Example
INSERT (P002:Person {name: 'Paul Jones', alive: true, geolocation: [12;23;true;'Québec']} {likes: ['P001'], owns: ['B001';'B002']});
{
"P002": {
"id": "P002",
"type": "Person",
"attributes": {
"name": "Paul Jones",
"alive": true,
"geolocation": [
12,
23,
true,
"Québec"
]
},
"relationships": {
"likes": ["P001"],
"owns": ["B001","B002"]
}
}
Match query
// Matches all persons from 'Canada' who owns a library.
MATCH (a:Person {origin: 'Canada'})=[owns]>(b:Building {type: 'library'}) RETURN a;
Node Selectors
Valid Formats:
- MATCH ( Name ) ...
- MATCH ( Name : Type ) ...
- MATCH ( Name : Type { Criterias } ) ...
Name: Name of the selector. This value is used to identify the selector.
Upper or lower camelcase format required.
Type: Type of the selector. This value is used to filter using the node type.
You can also use the wildcard operator (?) to match any type of node.
Upper or lower camelcase format required.
Criterias: Criterias for selection. This value will be used to filter using the node attributes.
JSON format required.
Examples:
// Match all of any type
MATCH (a) RETURN a;
// Match all of type 'Person'
MATCH (a:Person) RETURN a;
// Match all of type 'Person' with name 'John Doe'
MATCH (a:Person {name: 'John Doe'}) RETURN a;
// Match all of any type with location 'Québec, Canada'
MATCH (a:? {location: 'Québec, Canada'}) RETURN a;
Node Relations
Valid Formats:
- MATCH SourceSelector =[ Name ]> TargetSelector ...
- MATCH TargetSelector <[ Name ]= SourceSelector ...
SourceSelector: Source node selector.
See NodeSelectors section above for more details.
Name: Name of the relation. This will be used to filter with the node relationships.
Upper or lower camelcase format required.
TargetSelector: Target node selector.
See NodeSelectors section above for more details.
Examples:
// Match all of type 'Person' type who 'likes' another 'Person'.
MATCH (a:Person)=[likes]>(b:Person) RETURN a;
// Match all of type 'Building' type who are owned by a 'Person'.
MATCH (a:Building)<[owns]=(b:Person) RETURN a;
// Match all of type 'Person' type who 'likes' another 'Person' who owns a 'Building'.
MATCH (a:Person)=[likes]>(b:Person)=[owns]>(c:Building) RETURN a;
Returns Statements
Valid Formats:
- MATCH ... RETURN Values ;
Values: Return values. If there is more than one return value, values are separated by a comma.
Values must be equal to one of the NodeSelectors Name to be valid.
Examples:
// Returns the dataset with the selector name 'a'.
MATCH (a:Person) RETURN a;
// Returns the datasets with the selector names 'a' and 'b'.
MATCH (a:Person)=[owns]>(b:Building) RETURN a,b;
Update query
// Update name of node 'P001' to value 'John Doe'
UPDATE (P001:Person => {name: 'John Doe'});
Valid Formats:
- UPDATE ( Id : Type => { UpdateValue } ) ;
- UPDATE ( ? : Type => { UpdateValue } ) ;
- UPDATE ( ? : Type { Criterias } => { UpdateValue } ) ;
Id: Id of the selector. If the value is '?', the query will update all nodes of fitting the Type and Criterias.
Upper or lower camelcase format required.
Type: Type of the selector. This value is used to filter with the node type.
Upper or lower camelcase format required.
Criterias: Criterias for selection. This value will be used to filter with the node attributes.
JSON format required.
UpdateValue: Properties of the node to update.
JSON format required.
Examples:
// Updates name of node 'P001' to value 'Paul Jones'.
UPDATE (P001:Person => {name: 'Paul Jones'});
// Updates name of all nodes of type 'Person' to value 'Paul Jones'.
UPDATE (?:Person => {name: 'Paul Jones'});
// Updates name of all nodes of type 'Person' with name 'Marie Jones' to value 'Paul Jones'.
UPDATE (?:Person {name: 'Marie Jones'} => {name: 'Paul Jones'});