Refactoring tutorial #440
Refactoring tutorial #440
Conversation
|
This PR includes documentation updates New pages: |
|
|
||
| The answer for these issues is, therefore, to refactor the property `languages` into a node and connect it to the `Movie` nodes with a new relationship. | ||
|
|
||
| == Eliminating duplicated data |
There was a problem hiding this comment.
technically we're not deleting data in this section
rename to "reshaping the data"?
| * *In order to perform the query, all `Movie` nodes must be retrieved* -> As the graph scales, the performance of a similar query can be dimished by the way you modeled your data. | ||
| * *The name of the language is duplicated in many `Movie` nodes (in this case, all of them)* -> If many nodes share a same property value, it could be a sign that this property value could instead become a new entity, like a node or a relationship, for example. | ||
|
|
||
| The answer for these issues is, therefore, to refactor the property `languages` into a node and connect it to the `Movie` nodes with a new relationship. |
There was a problem hiding this comment.
what about a visual to represent the data model before/after?
| [source,cypher] | ||
| -- | ||
| MATCH (m:Movie) | ||
| MATCH (l:Language) |
There was a problem hiding this comment.
- this does not work, there is a where clause missing to take only the language nodes matching the movie languages
- this will work only in small graphs, mid-size graphs and above require sub-transactions
- a unique constraint is missing on the language identifier
- In practice we would do all of this in a single query for efficiency
CREATE CONSTRAINT unique_language FOR (n:Language) REQUIRE n.name IS UNIQUE
Note: untested query
:auto <1>
MATCH (m:Movie)
CALL (m) {
UNWIND m.languages AS language
MERGE (l:Language {name:language})
MERGE (m)-[:IN_LANGUAGE]->(l)
SET m.languages = null
} IN TRANSACTIONS OF 10000 rows
<1> required in neo4j browser to run nested transactions
| One way to improve your current model is to check for duplicate key values and see if you can turn them into another entity, like a node or a relationship. | ||
| In this case, both production companies are based in California, so the state could be turned into a node for `State` and be connected to the producer companies via a new relationship `LOCATED_AT`: | ||
|
|
||
| image::california.svg[The producer company nodes now have one less property for state and connect to a state node for California, role=popup] |
There was a problem hiding this comment.
For data consistency, the country property should also move to the State nodes
| RETURN m | ||
| -- | ||
|
|
||
| | How many users rated a movie? |
There was a problem hiding this comment.
A more 'graphy' alternative could be
MATCH (m:Movie)
WHERE m.title = 'Apollo 13'
RETURN COUNT {(:User)-[:RATED]->(m)} AS `Number of reviewers`
|
|
||
| This should be the result: | ||
|
|
||
| image::query-plan.png[Screenshot of Browser featuring a query plan that shows the number of database hits when you retrieve all person nodes,400,400,role=popup] |
There was a problem hiding this comment.
the data on the screenshot looks suspicious
I would not expect to have 0 rows in the execution pipeline, unless the DB is empty
No description provided.