From a361a018a068a11994cb6233724581337153b685 Mon Sep 17 00:00:00 2001
From: Marcel Otto <marcelotto@gmx.de>
Date: Wed, 4 Nov 2020 21:09:29 +0100
Subject: [PATCH] Add RDF.NQuads.Encoder.stream/2

---
 CHANGELOG.md                             |  3 +-
 lib/rdf/serializations/nquads_encoder.ex | 49 ++++++++++++++++++++----
 test/unit/nquads_encoder_test.exs        | 31 +++++++++++++++
 3 files changed, 74 insertions(+), 9 deletions(-)

diff --git a/CHANGELOG.md b/CHANGELOG.md
index 657fb80..35bedf3 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -9,7 +9,8 @@ This project adheres to [Semantic Versioning](http://semver.org/) and
 
 ### Added
 
-- serialization functions for reading from and writing to streams 
+- general serialization functions for reading from and writing to streams
+  and implementations for N-Triples and N-Quads (Turtle still to come)
 - `RDF.Dataset.prefixes/1` for getting an aggregated `RDF.PrefixMap` over all graphs
 - `RDF.PrefixMap.put/3` for adding a prefix mapping and overwrite an existing one
 - `RDF.BlankNode.value/1` for getting the internal string representation of a blank node
diff --git a/lib/rdf/serializations/nquads_encoder.ex b/lib/rdf/serializations/nquads_encoder.ex
index 25acc1d..09dbf7c 100644
--- a/lib/rdf/serializations/nquads_encoder.ex
+++ b/lib/rdf/serializations/nquads_encoder.ex
@@ -8,13 +8,20 @@ defmodule RDF.NQuads.Encoder do
   @impl RDF.Serialization.Encoder
   @callback encode(RDF.Data.t(), keyword) :: {:ok, String.t()} | {:error, any}
   def encode(data, _opts \\ []) do
-    result =
-      data
-      |> Enum.reduce([], &[statement(&1) | &2])
-      |> Enum.reverse()
-      |> Enum.join("\n")
+    {:ok,
+     data
+     |> Enum.reduce([], &[statement(&1) | &2])
+     |> Enum.reverse()
+     |> Enum.join()}
+  end
 
-    {:ok, if(result == "", do: result, else: "#{result}\n")}
+  @impl RDF.Serialization.Encoder
+  @spec stream(RDF.Data.t(), keyword) :: Enumerable.t()
+  def stream(data, opts \\ []) do
+    case Keyword.get(opts, :mode, :string) do
+      :string -> Stream.map(data, &statement(&1))
+      :iodata -> Stream.map(data, &iolist_statement(&1))
+    end
   end
 
   @spec statement(Statement.t()) :: String.t()
@@ -25,12 +32,38 @@ defmodule RDF.NQuads.Encoder do
   end
 
   def statement({subject, predicate, object, graph}) do
-    "#{term(subject)} #{term(predicate)} #{term(object)} #{term(graph)} ."
+    "#{term(subject)} #{term(predicate)} #{term(object)} #{term(graph)} .\n"
   end
 
   def statement({subject, predicate, object}) do
-    "#{term(subject)} #{term(predicate)} #{term(object)} ."
+    "#{term(subject)} #{term(predicate)} #{term(object)} .\n"
   end
 
   defdelegate term(value), to: RDF.NTriples.Encoder
+
+  @spec iolist_statement(Statement.t()) :: iolist
+  def iolist_statement(statement)
+
+  def iolist_statement({subject, predicate, object, nil}) do
+    iolist_statement({subject, predicate, object})
+  end
+
+  def iolist_statement({subject, predicate, object, graph}) do
+    [
+      iolist_term(subject),
+      " ",
+      iolist_term(predicate),
+      " ",
+      iolist_term(object),
+      " ",
+      iolist_term(graph),
+      " .\n"
+    ]
+  end
+
+  def iolist_statement({subject, predicate, object}) do
+    [iolist_term(subject), " ", iolist_term(predicate), " ", iolist_term(object), " .\n"]
+  end
+
+  defdelegate iolist_term(value), to: RDF.NTriples.Encoder
 end
diff --git a/test/unit/nquads_encoder_test.exs b/test/unit/nquads_encoder_test.exs
index 901ba23..6d1e5e5 100644
--- a/test/unit/nquads_encoder_test.exs
+++ b/test/unit/nquads_encoder_test.exs
@@ -14,6 +14,10 @@ defmodule RDF.NQuads.EncoderTest do
 
   defvocab EX, base_iri: "http://example.org/#", terms: [], strict: false
 
+  test "stream_support?/0" do
+    assert NQuads.Encoder.stream_support?()
+  end
+
   describe "serializing a graph" do
     test "an empty graph is serialized to an empty string" do
       assert NQuads.Encoder.encode!(Graph.new()) == ""
@@ -123,4 +127,31 @@ defmodule RDF.NQuads.EncoderTest do
                """
     end
   end
+
+  describe "stream/2" do
+    dataset =
+      Dataset.new([
+        {EX.S1, EX.p1(), EX.O1},
+        {EX.S2, EX.p2(), RDF.bnode("foo"), EX.G},
+        {EX.S3, EX.p3(), ~L"foo"},
+        {EX.S3, EX.p3(), ~L"foo"en, EX.G}
+      ])
+
+    expected_result = """
+    <http://example.org/#S1> <http://example.org/#p1> <http://example.org/#O1> .
+    <http://example.org/#S3> <http://example.org/#p3> "foo" .
+    <http://example.org/#S2> <http://example.org/#p2> _:foo <http://example.org/#G> .
+    <http://example.org/#S3> <http://example.org/#p3> "foo"@en <http://example.org/#G> .
+    """
+
+    assert NQuads.Encoder.stream(dataset, mode: :string)
+           |> Enum.to_list()
+           |> IO.iodata_to_binary() ==
+             expected_result
+
+    assert NQuads.Encoder.stream(dataset, mode: :iodata)
+           |> Enum.to_list()
+           |> IO.iodata_to_binary() ==
+             expected_result
+  end
 end