View Source Mongo.OrderedBulk (mongodb-driver v1.4.1)
An ordered bulk is filled in the memory with the bulk operations. If the ordered bulk is written to the database, the order is preserved. Only same types of operation are grouped and only if they have been inserted one after the other.
Example
alias Mongo.OrderedBulk
alias Mongo.BulkWrite
bulk = "bulk"
|> OrderedBulk.new()
|> OrderedBulk.insert_one(%{name: "Greta"})
|> OrderedBulk.insert_one(%{name: "Tom"})
|> OrderedBulk.insert_one(%{name: "Waldo"})
|> OrderedBulk.update_one(%{name: "Greta"}, %{"$set": %{kind: "dog"}})
|> OrderedBulk.update_one(%{name: "Tom"}, %{"$set": %{kind: "dog"}})
|> OrderedBulk.update_one(%{name: "Waldo"}, %{"$set": %{kind: "dog"}})
|> OrderedBulk.update_many(%{kind: "dog"}, %{"$set": %{kind: "cat"}})
|> OrderedBulk.delete_one(%{kind: "cat"})
|> OrderedBulk.delete_one(%{kind: "cat"})
|> OrderedBulk.delete_one(%{kind: "cat"})
BulkWrite.write(:mongo, bulk, w: 1)This example would not work by using an unordered bulk, because the OrderedBulk.update_many would executed too early.
To reduce the memory usage the ordered bulk can be used with streams as well.
Example
alias Mongo.OrderedBulk
1..1000
|> Stream.map(fn
1 -> BulkOps.get_insert_one(%{count: 1})
1000 -> BulkOps.get_delete_one(%{count: 999})
i -> BulkOps.get_update_one(%{count: i - 1}, %{"$set": %{count: i}})
end)
|> OrderedBulk.write(:mongo, "bulk", 25)
|> Stream.run()Of course, this example is a bit silly. A sequence of update operations is created that only work in the correct order.
Summary
Functions
Appends a delete operation with :limit = 0.
Appends a delete operation with :limit = 1.
Returns true, if the bulk is empty, that means it contains no inserts, updates or deletes operations
Appends an insert operation.
Creates an empty ordered bulk for a collection.
Appends a bulk operation to the ordered bulk.
Appends a replace operation with :multi = false.
Appends a update operation with :multi = true.
Appends a update operation with :multi = false.
Returns a stream chunk function that can be used with streams. The limit specifies the number
of operation hold in the memory while processing the stream inputs.
Types
@type t() :: %Mongo.OrderedBulk{coll: String.t(), ops: [Mongo.BulkOps.bulk_op()]}
Functions
@spec delete_many(t(), BSON.document()) :: t()
Appends a delete operation with :limit = 0.
Example:
Mongo.OrderedBulk.delete_many(bulk, %{name: "Waldo"})
%Mongo.OrderedBulk{coll: "bulk", ops: [delete: {%{name: "Waldo"}, [limit: 0]}]}@spec delete_one(t(), BSON.document()) :: t()
Appends a delete operation with :limit = 1.
Example:
Mongo.OrderedBulk.delete_one(bulk, %{name: "Waldo"})
%Mongo.OrderedBulk{coll: "bulk", ops: [delete: {%{name: "Waldo"}, [limit: 1]}]}Returns true, if the bulk is empty, that means it contains no inserts, updates or deletes operations
@spec insert_one(t(), BSON.document()) :: t()
Appends an insert operation.
Example:
Mongo.OrderedBulk.insert_one(bulk, %{name: "Waldo"})
%Mongo.OrderedBulk{coll: "bulk", ops: [insert: %{name: "Waldo"}]}Creates an empty ordered bulk for a collection.
Example:
Mongo.orderedBulk.new("bulk")
%Mongo.OrderedBulk{coll: "bulk", ops: []}@spec push(Mongo.BulkOps.bulk_op(), t()) :: t()
Appends a bulk operation to the ordered bulk.
@spec replace_one(t(), BSON.document(), BSON.document(), Keyword.t()) :: t()
Appends a replace operation with :multi = false.
Example:
Mongo.OrderedBulk.replace_one(bulk, %{name: "Waldo"}, %{name: "Greta", kind: "dog"})
%Mongo.OrderedBulk{
coll: "bulk",
ops: [
update: {%{name: "Waldo"}, %{kind: "dog", name: "Greta"}, [multi: false]}
]
}@spec update_many(t(), BSON.document(), BSON.document(), Keyword.t()) :: t()
Appends a update operation with :multi = true.
Example:
Mongo.OrderedBulk.update_many(bulk, %{name: "Waldo"}, %{"$set": %{name: "Greta", kind: "dog"}})
%Mongo.OrderedBulk{
coll: "bulk",
ops: [
update: {%{name: "Waldo"}, %{"$set": %{kind: "dog", name: "Greta"}},
[multi: true]}
]
}@spec update_one(t(), BSON.document(), BSON.document(), Keyword.t()) :: t()
Appends a update operation with :multi = false.
Example:
Mongo.OrderedBulk.update_one(bulk, %{name: "Waldo"}, %{"$set": %{name: "Greta", kind: "dog"}})
%Mongo.OrderedBulk{
coll: "bulk",
ops: [
update: {%{name: "Waldo"}, %{"$set": %{kind: "dog", name: "Greta"}},
[multi: false]}
]
}@spec write( Enumerable.t(), GenServer.server(), String.t(), non_neg_integer(), Keyword.t() ) :: Enumerable.t()
Returns a stream chunk function that can be used with streams. The limit specifies the number
of operation hold in the memory while processing the stream inputs.
The inputs of the stream should be Mongo.BulkOps.bulk_op. See Mongo.BulkOps