Support ignoring whitespace in diffs

Author: dl-alexandre

assets/css/app.css

@@ -395,6 +395,16 @@ table.package-list .button {
   font-weight: 600;
 }
 
+.diff-stats-header .whitespace-toggle {
+  color: #ddd;
+  font-family: 'SFMono-Regular', 'Consolas', 'Liberation Mono', 'Menlo', monospace;
+  font-size: 14px;
+}
+
+.diff-stats-header .whitespace-toggle:hover {
+  color: white;
+}
+
 .loading-spinner {
   height: 60px;
   width: 100%;

lib/diff/hex/behaviour.ex

@@ -1,4 +1,6 @@
 defmodule Diff.Hex.Behaviour do
   @callback diff(package :: String.t(), from :: String.t(), to :: String.t()) ::
               {:ok, Enumerable.t()} | :error
+  @callback diff(package :: String.t(), from :: String.t(), to :: String.t(), Keyword.t()) ::
+              {:ok, Enumerable.t()} | :error
 end

lib/diff/hex/hex.ex

@@ -102,7 +102,9 @@ defmodule Diff.Hex do
     end
   end
 
-  def diff(package, from, to) do
+  def diff(package, from, to), do: diff(package, from, to, [])
+
+  def diff(package, from, to, opts) do
     with {:ok, tarball_from} <- get_tarball(package, from),
          path_from = Diff.TmpDir.tmp_dir("package-#{package}-#{from}"),
          :ok <- unpack_tarball(tarball_from, path_from),
@@ -131,7 +133,7 @@ defmodule Diff.Hex do
 
           with {_, true} <- {:file_size_old, file_size_check?(path_old)},
                {_, true} <- {:file_size_new, file_size_check?(path_new)},
-               {_, {:ok, output}} <- {:git_diff, git_diff(path_old, path_new)} do
+               {_, {:ok, output}} <- {:git_diff, git_diff(path_old, path_new, opts)} do
             if output do
               [{:ok, {output, path_from, path_to}}]
             else
@@ -157,24 +159,29 @@ defmodule Diff.Hex do
     end
   end
 
-  defp git_diff(path_from, path_to) do
-    case System.cmd("git", [
-           "-c",
-           "core.quotepath=false",
-           "-c",
-           "diff.algorithm=histogram",
-           "diff",
-           "--no-index",
-           "--no-color",
-           path_from,
-           path_to
-         ]) do
+  defp git_diff(path_from, path_to, opts) do
+    args =
+      [
+        "-c",
+        "core.quotepath=false",
+        "-c",
+        "diff.algorithm=histogram",
+        "diff",
+        "--no-index",
+        "--no-color"
+      ] ++ ignore_whitespace_args(opts) ++ [path_from, path_to]
+
+    case System.cmd("git", args) do
       {"", 0} -> {:ok, nil}
       {output, 1} -> {:ok, output}
       other -> {:error, other}
     end
   end
 
+  defp ignore_whitespace_args(opts) do
+    if Keyword.get(opts, :ignore_whitespace, false), do: ["-w"], else: []
+  end
+
   defp file_size_check?(path) do
     File.stat!(path).size <= @max_file_size
   end

lib/diff/storage/gcs.ex

@@ -5,8 +5,8 @@ defmodule Diff.Storage.GCS do
 
   @gs_xml_url "https://storage.googleapis.com"
 
-  def get_diff(package, from_version, to_version, diff_id) do
-    with {:ok, hash} <- combined_checksum(package, from_version, to_version),
+  def get_diff(package, from_version, to_version, diff_id, opts \\ []) do
+    with {:ok, hash} <- combined_checksum(package, from_version, to_version, opts),
          url = url(diff_key(package, from_version, to_version, hash, diff_id)),
          {:ok, 200, _headers, body} <-
            Diff.HTTP.retry("gs", fn -> Diff.HTTP.get(url, headers()) end) do
@@ -25,8 +25,8 @@ defmodule Diff.Storage.GCS do
     end
   end
 
-  def put_diff(package, from_version, to_version, diff_id, diff_data) do
-    with {:ok, hash} <- combined_checksum(package, from_version, to_version),
+  def put_diff(package, from_version, to_version, diff_id, diff_data, opts \\ []) do
+    with {:ok, hash} <- combined_checksum(package, from_version, to_version, opts),
          url = url(diff_key(package, from_version, to_version, hash, diff_id)),
          {:ok, 200, _headers, _body} <-
            Diff.HTTP.retry("gs", fn -> Diff.HTTP.put(url, headers(), diff_data) end) do
@@ -42,10 +42,10 @@ defmodule Diff.Storage.GCS do
     end
   end
 
-  def list_diffs(package, from_version, to_version) do
-    case get_metadata(package, from_version, to_version) do
+  def list_diffs(package, from_version, to_version, opts \\ []) do
+    case get_metadata(package, from_version, to_version, opts) do
       {:ok, %{total_diffs: total_diffs}} ->
-        diff_ids = 0..(total_diffs - 1) |> Enum.map(&"diff-#{&1}")
+        diff_ids = diff_ids(total_diffs)
         {:ok, diff_ids}
 
       {:error, :not_found} ->
@@ -56,8 +56,8 @@ defmodule Diff.Storage.GCS do
     end
   end
 
-  def get_metadata(package, from_version, to_version) do
-    with {:ok, hash} <- combined_checksum(package, from_version, to_version),
+  def get_metadata(package, from_version, to_version, opts \\ []) do
+    with {:ok, hash} <- combined_checksum(package, from_version, to_version, opts),
          url = url(metadata_key(package, from_version, to_version, hash)),
          {:ok, 200, _headers, body} <-
            Diff.HTTP.retry("gs", fn -> Diff.HTTP.get(url, headers()) end) do
@@ -79,8 +79,8 @@ defmodule Diff.Storage.GCS do
     end
   end
 
-  def put_metadata(package, from_version, to_version, metadata) do
-    with {:ok, hash} <- combined_checksum(package, from_version, to_version),
+  def put_metadata(package, from_version, to_version, metadata, opts \\ []) do
+    with {:ok, hash} <- combined_checksum(package, from_version, to_version, opts),
          url = url(metadata_key(package, from_version, to_version, hash)),
          {:ok, json} <- Jason.encode(metadata),
          {:ok, 200, _headers, _body} <-
@@ -106,12 +106,18 @@ defmodule Diff.Storage.GCS do
     [{"authorization", "#{token.type} #{token.token}"}]
   end
 
-  def combined_checksum(package, from, to) do
+  def combined_checksum(package, from, to, opts \\ []) do
     with {:ok, checksums} <- Diff.Hex.get_checksums(package, [from, to]) do
-      {:ok, :erlang.phash2({Application.get_env(:diff, :cache_version), checksums})}
+      {:ok, :erlang.phash2(Diff.Storage.cache_key(checksums, opts))}
     end
   end
 
+  defp diff_ids(0), do: []
+
+  defp diff_ids(total_diffs) do
+    0..(total_diffs - 1) |> Enum.map(&"diff-#{&1}")
+  end
+
   defp diff_key(package, from_version, to_version, hash, diff_id) do
     "diffs/#{package}-#{from_version}-#{to_version}-#{hash}-#{diff_id}.json"
   end

lib/diff/storage/local.ex

@@ -3,8 +3,8 @@ defmodule Diff.Storage.Local do
 
   @behaviour Diff.Storage
 
-  def get_diff(package, from_version, to_version, diff_id) do
-    case combined_checksum(package, from_version, to_version) do
+  def get_diff(package, from_version, to_version, diff_id, opts \\ []) do
+    case combined_checksum(package, from_version, to_version, opts) do
       {:ok, hash} ->
         filename = diff_key(package, from_version, to_version, hash, diff_id)
         path = Path.join([dir(), package, filename])
@@ -20,8 +20,8 @@ defmodule Diff.Storage.Local do
     end
   end
 
-  def put_diff(package, from_version, to_version, diff_id, diff_data) do
-    with {:ok, hash} <- combined_checksum(package, from_version, to_version),
+  def put_diff(package, from_version, to_version, diff_id, diff_data, opts \\ []) do
+    with {:ok, hash} <- combined_checksum(package, from_version, to_version, opts),
          filename = diff_key(package, from_version, to_version, hash, diff_id),
          path = Path.join([dir(), package, filename]),
          :ok <- File.mkdir_p(Path.dirname(path)) do
@@ -34,10 +34,10 @@ defmodule Diff.Storage.Local do
     end
   end
 
-  def list_diffs(package, from_version, to_version) do
-    case get_metadata(package, from_version, to_version) do
+  def list_diffs(package, from_version, to_version, opts \\ []) do
+    case get_metadata(package, from_version, to_version, opts) do
       {:ok, %{total_diffs: total_diffs}} ->
-        diff_ids = 0..(total_diffs - 1) |> Enum.map(&"diff-#{&1}")
+        diff_ids = diff_ids(total_diffs)
         {:ok, diff_ids}
 
       {:error, :not_found} ->
@@ -48,8 +48,8 @@ defmodule Diff.Storage.Local do
     end
   end
 
-  def get_metadata(package, from_version, to_version) do
-    case combined_checksum(package, from_version, to_version) do
+  def get_metadata(package, from_version, to_version, opts \\ []) do
+    case combined_checksum(package, from_version, to_version, opts) do
       {:ok, hash} ->
         filename = metadata_key(package, from_version, to_version, hash)
         path = Path.join([dir(), package, filename])
@@ -74,8 +74,8 @@ defmodule Diff.Storage.Local do
     end
   end
 
-  def put_metadata(package, from_version, to_version, metadata) do
-    with {:ok, hash} <- combined_checksum(package, from_version, to_version),
+  def put_metadata(package, from_version, to_version, metadata, opts \\ []) do
+    with {:ok, hash} <- combined_checksum(package, from_version, to_version, opts),
          filename = metadata_key(package, from_version, to_version, hash),
          path = Path.join([dir(), package, filename]),
          :ok <- File.mkdir_p(Path.dirname(path)),
@@ -89,12 +89,18 @@ defmodule Diff.Storage.Local do
     end
   end
 
-  def combined_checksum(package, from, to) do
+  def combined_checksum(package, from, to, opts \\ []) do
     with {:ok, checksums} <- Diff.Hex.get_checksums(package, [from, to]) do
-      {:ok, :erlang.phash2({Application.get_env(:diff, :cache_version), checksums})}
+      {:ok, :erlang.phash2(Diff.Storage.cache_key(checksums, opts))}
     end
   end
 
+  defp diff_ids(0), do: []
+
+  defp diff_ids(total_diffs) do
+    0..(total_diffs - 1) |> Enum.map(&"diff-#{&1}")
+  end
+
   defp diff_key(package, from_version, to_version, hash, diff_id) do
     "diffs/#{package}-#{from_version}-#{to_version}-#{hash}-#{diff_id}.json"
   end

lib/diff/storage/storage.ex

@@ -4,6 +4,7 @@ defmodule Diff.Storage do
   @type to_version :: String.t()
   @type diff_id :: String.t()
   @type diff_html :: String.t()
+  @type diff_options :: Keyword.t()
   @type diff_metadata :: %{
           total_diffs: non_neg_integer(),
           total_additions: non_neg_integer(),
@@ -14,34 +15,59 @@ defmodule Diff.Storage do
   # New diff-level storage callbacks
   @callback get_diff(package, from_version, to_version, diff_id) ::
               {:ok, diff_html} | {:error, term}
+  @callback get_diff(package, from_version, to_version, diff_id, diff_options) ::
+              {:ok, diff_html} | {:error, term}
   @callback put_diff(package, from_version, to_version, diff_id, diff_html) ::
               :ok | {:error, term}
+  @callback put_diff(package, from_version, to_version, diff_id, diff_html, diff_options) ::
+              :ok | {:error, term}
   @callback list_diffs(package, from_version, to_version) :: {:ok, [diff_id]} | {:error, term}
+  @callback list_diffs(package, from_version, to_version, diff_options) ::
+              {:ok, [diff_id]} | {:error, term}
 
   # Metadata storage callbacks
   @callback get_metadata(package, from_version, to_version) ::
               {:ok, diff_metadata} | {:error, term}
+  @callback get_metadata(package, from_version, to_version, diff_options) ::
+              {:ok, diff_metadata} | {:error, term}
   @callback put_metadata(package, from_version, to_version, diff_metadata) :: :ok | {:error, term}
+  @callback put_metadata(package, from_version, to_version, diff_metadata, diff_options) ::
+              :ok | {:error, term}
 
   defp impl(), do: Application.get_env(:diff, :storage_impl)
 
-  def get_diff(package, from_version, to_version, diff_id) do
-    impl().get_diff(package, from_version, to_version, diff_id)
+  def get_diff(package, from_version, to_version, diff_id, opts \\ []) do
+    impl().get_diff(package, from_version, to_version, diff_id, opts)
   end
 
-  def put_diff(package, from_version, to_version, diff_id, diff_html) do
-    impl().put_diff(package, from_version, to_version, diff_id, diff_html)
+  def put_diff(package, from_version, to_version, diff_id, diff_html, opts \\ []) do
+    impl().put_diff(package, from_version, to_version, diff_id, diff_html, opts)
   end
 
-  def list_diffs(package, from_version, to_version) do
-    impl().list_diffs(package, from_version, to_version)
+  def list_diffs(package, from_version, to_version, opts \\ []) do
+    impl().list_diffs(package, from_version, to_version, opts)
   end
 
-  def get_metadata(package, from_version, to_version) do
-    impl().get_metadata(package, from_version, to_version)
+  def get_metadata(package, from_version, to_version, opts \\ []) do
+    impl().get_metadata(package, from_version, to_version, opts)
+  end
+
+  def put_metadata(package, from_version, to_version, metadata, opts \\ []) do
+    impl().put_metadata(package, from_version, to_version, metadata, opts)
+  end
+
+  def cache_key(checksums, opts) do
+    base_key = {Application.get_env(:diff, :cache_version), checksums}
+
+    case cache_options(opts) do
+      [] -> base_key
+      options -> {base_key, options}
+    end
   end
 
-  def put_metadata(package, from_version, to_version, metadata) do
-    impl().put_metadata(package, from_version, to_version, metadata)
+  defp cache_options(opts) do
+    opts
+    |> Keyword.take([:ignore_whitespace])
+    |> Enum.reject(fn {_key, value} -> value in [false, nil] end)
   end
 end