The most direct way to use Cypher programmatically is to execute a HTTP POST operation against the transactional Cypher endpoint.
You can send a large number of statements with parameters to the server with each request.
For immediate execution you can use the /db/data/transaction/commit endpoint with a JSON payload like this:
curl -i -H accept:application/json -H content-type:application/json -XPOST http://localhost:7474/db/data/transaction/commit \
-d '{"statements":[{"statement":"CREATE (p:Person {name:{name},born:{born}}) RETURN p","parameters":{"name":"Keanu Reeves","born":1964}}]}'
The above command results in:
{"results":[{"columns":["p"],"data":[{"row":[{"name":"Keanu Reeves","born":1964}]}]}],"errors":[]}
You can add as many "statement" objects in the "statements" list as you want.
For larger use-cases that span multiple requests but whose read-write-read-write operations should be executed within the same transactional scope you’d use the /db/data/transaction endpoint.
This will give you a transaction URL as the Location header, which you can continue to write to and read from.
At the end you either commit the whole transaction by POSTing to the (also returned) commit URL or by issuing a DELETE request against the transaction URL.
curl -i -H accept:application/json -H content-type:application/json -XPOST http://localhost:7474/db/data/transaction \
-d '{"statements":[{"statement":"CREATE (p:Person {name:{name},born:{born}}) RETURN p","parameters":{"name":"Clint Eastwood","born":1930}}]}'
The above command results in:
HTTP/1.1 201 Created
Location: http://localhost:7474/db/data/transaction/261
{"commit":"http://localhost:7474/db/data/transaction/261/commit","transaction":{"expires":"Wed, 03 Sep 2014 23:26:51 +0000"},"errors":[],
"results":[{"columns":["p"],"data":[{"row":[{"name":"Clint Eastwood","born":1930}]}]}]}
See Section 20.1, “Transactional Cypher HTTP endpoint” for more information.