Use $project aggregation to edit documents schema - MongoDB - mongodb

Question

I have this MongoDB document:
{
"_id" : 1,
title: "abc123",
isbn: "0001122223334",
author: { last: "zzz", first: "aaa" },
copies: 5
}
Using $project (aggregation operator) I have to reshape the previous schema to get:
{
"_id" : 1,
field: {key:"title", value:"abc123"},
isbn: "0001122223334",
author: { last: "zzz", first: "aaa" },
copies: 5
}
To reach my target I used the following aggregation:
db.book.aggregate([{$project: {"field": {"key":"title", "value":"$title"}}}])
but I got an error:
{
"ok" : 0,
"errmsg" : "FieldPath 'isbn' doesn't start with $",
"code" : 16873
} : aggregate failed
I don't understand why that aggregation does not work, since if I want to reshape the previous schema to get:
{
"_id" : 1,
"author" : {
"last" : "zzz",
"first" : "aaa"
},
"copies" : 5,
"fieldTitle" : {
"key" : "abc123"
}
}
I can use this aggregation (and it works):
db.book.aggregate([{$project: {_id:1, fieldTitle:{key:"$title"}, author:1, copies:1}}])

Answer 1

Use the $literal operator to return a value without parsing. It's used for values that the aggregation pipeline may interpret as an expression, like the error you are presently getting:
db.book.aggregate([
{
$project: {
"field.key": { "$literal": "title" },
"field.value": "$title",
"author": 1, "copies": 1
}
}
])
Sample Output
{
"result" : [
{
"_id" : 1,
"author" : {
"last" : "zzz",
"first" : "aaa"
},
"copies" : 5,
"field" : {
"key" : "title",
"value" : "abc123"
}
}
],
"ok" : 1
}