If you click on a Person card anywhere on the website, you will be taken to a Person profile. This API call is the same regardless of whether the person is an actor, director, or both.
We have already implemented the methods that populate the endpoints in the MovieService to get a list of movies that the person has either acted in or directed.
But if you take a look at the PeopleService, you will see two methods that need to be implemented to complete this page:
-
findById() - should find a person by their ID.
-
getSimilarPeople() - should return a list of similar people who are commonly connected through the same movies.
findById()
The findById() method is currently hardcoded to return data by id from the people.json list.
public Map<String, Object> findById(String id) {
// TODO: Find a user by their ID
return people.stream().filter(p -> id.equals(p.get("tmdbId"))).findAny().get();
}You will update this method to run the following cypher statement in a read transaction.
MATCH (p:Person {tmdbId: $id})
RETURN p {
.*,
actedCount: count { (p)-[:ACTED_IN]->() },
directedCount: count { (p)-[:DIRECTED]->() }
} AS personThe query will return the properties for the person with the corresponding tmdbId, along with a count of the number of movies that the person has acted in and directed.
Your Task
-
Modify the
findById()method to query Neo4j and return details for the requested person. -
The returned object should include counts of the number of movies that the person has acted in or directed.
-
If not exactly one record is returned, the method will throw a
NotFoundErrorby using theresult.single()API call. -
Remember to auto-close the session and use the
toMap()function to convert the object into native Java types.
Click here to reveal the final findById() method.
public Map<String, Object> findById(String id) {
// Open a new database session
try (var session = driver.session()) {
// Get a person from the database
var person = session.executeRead(tx -> {
String query = """
MATCH (p:Person {tmdbId: $id})
RETURN p {
.*,
actedCount: count { (p)-[:ACTED_IN]->() },
directedCount: count { (p)-[:DIRECTED]->() }
} AS person
""";
var res = tx.run(query, Values.parameters("id", id));
return res.single().get("person").asMap();
});
return person;
}
}getSimilarPeople()
The getSimilarPeople() method should return a paginated list of similar people based on their second degree connections - either people who have acted in or have directed the same movies as our person.
public List<Map<String,Object>> getSimilarPeople(String id, Params params) {
// TODO: Get a list of similar people to the person by their id
return AppUtils.process(people, params);
}There could be a more clever algorithm for finding similar people by weighting the type of relationship differently, but for now, the following query will find a list of similar people based on the number of relationships in common.
MATCH (:Person {tmdbId: $id})-[:ACTED_IN|DIRECTED]->(m)<-[r:ACTED_IN|DIRECTED]-(p)
WITH p, collect(m {.tmdbId, .title, type: type(r)}) AS inCommon
RETURN p {
.*,
actedCount: count { (p)-[:ACTED_IN]->() },
directedCount: count {(p)-[:DIRECTED]->() },
inCommon: inCommon
} AS person
ORDER BY size(person.inCommon) DESC
SKIP 0
LIMIT 6Parameters in Neo4j Browser
To run this query in Neo4j Sandbox, you will have to define a parameter in your Neo4j Browser session.
To do so, you can use the :param command.
For example, to view Kevin Bacon, you can set the id parameter to "4724":
:param id: "4724"Your Task
-
Modify the
getSimilarPeople()method to query Neo4j and return a list of similar people. -
The returned objects should include a list of actors and
-
An
inCommonproperty to show how the two people are related (a list of the movies they in common, with theirtmbdId,title, and type of relationship). -
Here we order by the number of entries that two people have in common
-
The pagination information is passed as parameters
-
Remember to auto-close the session and use the
toMap()function within alist()function onResultto convert the values to native Java types.
Click here to reveal the final getSimilarPeople() method.
public List<Map<String,Object>> getSimilarPeople(String id, Params params) {
// Open a new database session
try (var session = driver.session()) {
// Get a list of similar people to the person by their id
var res = session.executeRead(tx -> tx.run("""
MATCH (:Person {tmdbId: $id})-[:ACTED_IN|DIRECTED]->(m)<-[r:ACTED_IN|DIRECTED]-(p)
RETURN p {
.*,
actedCount: count { (p)-[:ACTED_IN]->() },
directedCount: count { (p)-[:DIRECTED]->() },
inCommon: collect(m {.tmdbId, .title, type: type(r)})
} AS person
ORDER BY size(person.inCommon) DESC
SKIP $skip
LIMIT $limit
""",Values.parameters("id",id, "skip", params.skip(), "limit", params.limit()))
.list(row -> row.get("person").asMap()));
return res;
}
}Testing
To test that this functionality has been correctly implemented, run the following code in a new terminal session:
mvn test -Dtest=neoflix._15_PersonProfileTestThe test file is located at src/test/java/neoflix/_15_PersonProfileTest.java.
Are you stuck? Click here for help
If you get stuck, you can see a working solution by checking out the 15-person-profile branch by running:
git checkout 15-person-profileYou may have to commit or stash your changes before checking out this branch. You can also click here to expand the Support pane.
Verifying the Test
According to our algorithm, who is the most similar person to Francis Ford Coppola?
As part of the test suite, the final test will log the name of the most common person to Francis Ford Coppola according to our Cypher statement.
Paste the name of the person the box below without quotes or whitespace and click Check Answer.
-
✓ Frederic Forrest
Hint
You can also find the answer by running the following Cypher statement:
MATCH (:Person {tmdbId: '1776'})-[:ACTED_IN|DIRECTED]->(m)<-[r:ACTED_IN|DIRECTED]-(p)
WITH p, collect(m) AS inCommon
RETURN p.name
ORDER BY size(inCommon) DESC
LIMIT 1Copy the answer without any double quotes or whitespace.
Lesson Summary
In this Challenge, you have implemented the final methods in the PeopleService.