I have a graph with offers and customers. A customer can share an offer with another customer, so when this happens, I create a hyperedge.
(CustomerA)-[:SHARED_OFFER]->(newNode)
(newNode)-[:FOR_OFFER]->(offer)
(newNode)-[:SHARED_WITH]->(customers) (this can be many customers)
Now if another customer B shares the same offer with others, I would want a new node created to represent this relationship.
Is there a way to accomplish all this in one Cypher query?
I am using:
start c=node:node_auto_index(name="C1"), o=node:node_auto_index(name="Offer"), sharedCustomer=node:node_auto_index(name="C2")
create unique c-[:SHARED_OFFER]->(sharedOffer)-[:FOR_OFFER]->(o), (sharedOffer)-[:SHARED_WITH]->(sharedCustomer)
which works for the first time. See the console at: http://console.neo4j.org/r/76no2g
This query correctly created relationships when C1 shared Offer with C2.
Executing the query for the case when C2 now shares Offer with C3 causes the same node to be reused---that's not what I want. There should be a new node created from C2 with the SHARED_OFFER relationship. Here is the query:
start c=node:node_auto_index(name="C2"), o=node:node_auto_index(name="Offer"), sharedCustomer=node:node_auto_index(name="C3")
create unique c-[:SHARED_OFFER]->(sharedOffer)-[:FOR_OFFER]->(o), (sharedOffer)-[:SHARED_WITH]->(sharedCustomer)
Any help is appreciated.
Note: I'm using 1.8.1 REST, so trying to accomplish this all in one go rather than in parts.
simply don't use create unique but just create. but remember to first specify all nodes you got in the path in the start clause and leave the param sharedOffer always unspecified so the create command will create just the unspecified elements.
update use create instead of create unique and filter on existing relations (or use 2 queries - one to check whether an sharedOffer already exists with C1 and second to update sharedOffer with C3):
START c=node:node_auto_index(name="C1"), o=node:node_auto_index(name="Offer"), sharedCustomer=node:node_auto_index(name="C2")
WHERE not(c-[:SHARED_OFFER]->(sharedOffer)-[:FOR_OFFER]->(o))
CREATE c-[:SHARED_OFFER]->(sharedOffer)-[:FOR_OFFER]->(o), (sharedOffer)-[:SHARED_WITH]->(sharedCustomer)
Related
I would like to know if this is possible. I have a query that produces a nice report showing a relationship between two entities through two other nodes. There can be more than one path. I now want to create a direct relationship between those two nodes and count the number of paths and sum based upon data in the nodes in between. the report query is below.
match (bo:BuyerAgency)<-[:IS_FOR_BO]-(sol:Solicitation)-[:SELECTED]->(prop:Proposal)<-[:OWNS_BID]-(so:VendorOrg)
where sol.currStatus='Awarded'
return bo.AgencyName, count(sol.Number) as awards, so.orgName, sum(prop.finalPrice) as awardVolume;
What I want to do is similar to below which will not work.
match (bo:BuyerAgency)<-[:IS_FOR_BO]-(sol:Solicitation)-[:SELECTED]->(prop:Proposal)<-[:OWNS_BID]-(so:VendorOrg)
where sol.currStatus='Awarded'
create (bo)-[:HAS_AWARDED{awardCount: count(sol.Number), awardVolume: sum(prop.finalPrice)}]->(so);
If I remove the properties for the relationship, it works but want to add the properties without to much programing.
I am using the most recent version of Neo4j 3.2.
thanks
The problem here is you are trying to use count() and sum() functions in an invalid context. The below query should work:
match (bo:BuyerAgency)<-[:IS_FOR_BO]-(sol:Solicitation)-[:SELECTED]->(prop:Proposal)<-[:OWNS_BID]-(so:VendorOrg)
where sol.currStatus='Awarded'
with bo, so, count(sol.Number) as count_sol, sum(prop.finalPrice) as sum_finalPrice
create (bo)-[:HAS_AWARDED{awardCount: count_sol, awardVolume: sum_finalPrice}]->(so);
This query uses WITH to pass bo, so and the result of the aggregation functions count(sol.Number) and sum(prop.finalPrice) to the next context. After, these values are used to create the new relation between bo and so.
I'm doing an experiment with using a graph database (neo4j). I have two csv's that I imported into a neo4j datastore. I'm a little shakey on the neo terminology; so forgive me. Lets say I have:
Customer (AccountNumber, CustomerName) and
CustomerGroups (AccountNumber, GroupName).
I would like to create a new Node called groups which is comprised of the distinct GroupName from CustomerGroups. I'll call it Group.
I then want to create relationships "HAS_GROUP" from Customer to Group using the common AccountNumber from CustomerGroups.
Once the above is completed, I could delete CustomerGroups as its no longer needed.
I'm just stuck at the syntax. I can get the distinct groups from CustomerGroups with:
MATCH (n:CustomerGroups) distinct n.GROUP_NAME
and I get back about 50 distinct groups, but can't figure how to add the create statement to the results and CREATE g:Group {GroupName: n.GROUP_NAME}
I then know my followup question is how to do the MATCH to the new group using the old table with common account numbers.
FYI: I've indexed the AccountNumber in both Nodes. Both Customer and CustomerGroups have over 5 Million nodes. Not bad for a laptop (2 min to import using neo4j-import). I was impressed!
Thanks for any help you can give!
Instead of creating a CustomerGroups label and creating nodes for that, you should be able to define relationships that you would like to create in your neo4j-import. It would certainly be a lot faster too. See:
http://neo4j.com/docs/stable/import-tool-header-format.html
To your question, you could probably do something like:
MATCH (cg:CustomerGroup)
MATCH (customer:Customer {AccountNumber: cg. AccountNumber}), (group:Group {GroupName: cg.GroupName})
CREATE (customer)-[:IN_GROUP]->(group)
You'd definitely want to make sure you have indexes on :Customer(AccountNumber) and :Group(GroupName) first. But even then it would still be much slower than doing it as part of your initial import.
Also, you may or may not want MERGE instead of CREATE
I am trying to create a graph using some nodes and relationships using py2noe. say, I'm creating a family tree.
I am creating nodes using get_or_create() so that my script doesn't create duplicates if I supply the same value again.
How can I do the same for a relationship? I can't find any reference to a get_or_create() like function for a relationship.
I want to publish (Joe)-[:son]->(John)
the first time it creates 2 nodes joe and john and a link between them.
if I re run my script, as the nodes are unique, they aren't published but a new relationship is created.
This gives me a graph with 2 nodes and n relationships where n is the number of times I run the script.
I also tried using cypher and I get the same issue. It keeps on creating relationships.
Can anyone suggest me a way to solve this problem?
Thanks.
I don't know py2neo (so there may be a wrapper for this function), but the way to achieve that would be using a Cypher MERGE which it looks like you have to run using the raw Cypher statement:
cypher_merge_result = neo4j.CypherQuery(graph_db,
"MERGE (s:Person{name:Joe})-[:SON]->(f:Person{name:John})")
That will create 2 Person nodes and 1 SON relationship, no matter how many times that you run it.
Neo4J documentation is here and it is important to understand how it works as partial matches in the MERGE statement will cause the whole pattern to be created. So if your Person nodes already exist you should match them in advance to avoid duplicate Persons being created. i.e
cypher_merge_result = neo4j.CypherQuery(graph_db,
"MATCH (s:Person{name:Joe}), (f:Person{name:John}) MERGE (s)-[:SON]->(f)")
I'm trying to write an importer script that will take a MySQL resultset and add nodes and relationships in my Neo4J DB. To simplify things, my resultset has the following fields:
application_id
amount
application_date
account_id
I want to create a node with Application label with the application_id, amount, application_date fields and another node with Account label with account_id field and a relationship between them.
My SQL query is from the applications table so I'm not afraid of dups there, but an account_id can appear more than once and I obviously don't want to create multiple nodes for that.
I'm using neography (but willing to switch if there is something simpler). What would be the best (easiest) way to achieve that?
My script will drop the database before it starts so no leftovers to take care of.
Should I create an index before and use create_unique_node?
I think I can do what I want in cypher using "MATCH .. CRAETE UNIQUE..", what's the equivalent of that in neography? I don't understand how index_name gets into the equation...
Should I or should I not define the constraint on Account?
Many thanks,
It's my first time with graph DBs so apologies if I miss a concept here..
From what you describe, it looks like you should use the constraints that come with Neo4j 2.0 : http://docs.neo4j.org/chunked/milestone/query-constraints.html#constraints-create-uniqueness-constraint
CREATE CONSTRAINT ON (account:ACCOUNT) ASSERT account.account_id IS UNIQUE
Then you can use the MATCH .. CREATE UNIQUE clauses for all your inserts. You can use neography to submit the cypher queries, see the examples here: https://github.com/maxdemarzi/neography/wiki/Scripts-and-queries
We have selected neo4j as the DB for our web application. The user has a large number of relations and connected nodes. As of now there are about 20 relations for a user. One of the features is a newsfeed feature. If i want to delete a user completely, is the cypher query the best way to delete or is there any other alternative?
Since we are still planning to add new features, the relationships and nodes connected to the user also will increase. So if we use cypher query, the query has to be modified for every new relationship added. Please advise.
Thanks,
Pavan
Yes, you can use Cypher to remove a user. Of course, there are alternative methods, depending on the language or framework you're using with your web application. If you like to have advise on that, please specifiy how you're using Neo4j in detail.
Note that you have to remove all relationships (outgoing and incoming) first in order to be able to remove the node.
Example:
START n = node(3)
MATCH n-[r]-()
DELETE n, r
This example was taken from the official manual: http://docs.neo4j.org/chunked/milestone/query-delete.html
As of Neo4j 2.3 there is another way to do this:
MATCH (n { name:'Andres' })
DETACH DELETE n
I found this example in the documentation at: http://neo4j.com/docs/stable/query-delete.html
An alternative could be to write a gremlin script that traverses your graph starting with your user and is putting in two collection the relationships and the nodes that you intend to delete. If you want to delete everything, perhaps you can implement your depth first traversal in Gremlin and delete while traversing.