Moved from GitHub dgraph/4500
Posted by ChStark:
What you wanted to do
If we see the documentation, we can find an example of an upsert like this
upsert {
query {
q(func: eq(email, "user@company1.io")) {
v as uid
name
}
}
mutation{
set {
uid(v) <name> "first last" .
uid(v) <email> "user@company1.io" .
}
}
}
Wich is ok if you want to do the actual upsert, but most often than not, we need to not insert data if the data is present, according to the docs of the conditional upsert The upsert block also allows specifying conditional mutation blocks using an @if directive. The mutation is executed only when the specified condition is true. If the condition is false, the mutation is silently ignored
What you actually did
Since we can ensure to not store repeated information with the @if directive, we need to look up on the response to see if the mutation created a new node or the query to see if the query got a result.
what I did
upsert {
query {
q(func: eq(email, "user@company1.io")) {
v as uid
name
}
}
mutation @if( eq( len( v ) , 0 ) ){
set {
_:new <name> "first last" .
_:new <email> "user@company1.io" .
}
}
}
And then look on the response if data.uids.new contains a new id
Why that wasn’t great, with examples
Is not great to need to inspect for each possible outcome of the mutation, it would be great if upsert block support intentional fails, something like
upsert {
query {
q(func: eq(email, "user@company1.io")) {
v as uid
name
}
}
raise @if( lt( len( v ) , 0 ) ) {
exception("Email was already on the database") #and rollback the transaction automatically
#notice( "..." ) if it shouldn't stop transaction, but the message should be included on the response
}
mutation{ #we no longer need to add the @if directive here
set {
_:new <name> "first last" .
_:new <email> "user@company1.io" .
}
}
}
Or something like that
Any external references to support your case
We can take the RAISE keyword from plpsql docs