I169 trying nimble parsec

.formatter.exs

@@ -1,3 +1,14 @@
+locals_without_parens = [
+  defparsec: 2,
+  defparsec: 3,
+  defparsecp: 2,
+  defparsecp: 3,
+  defcombinator: 2,
+  defcombinator: 3,
+  defcombinatorp: 2,
+  defcombinatorp: 3
+]
+
 [
   force_do_end_blocks: true,
   inputs: ["{mix,.formatter}.exs", "{config,lib,test}/**/*.{ex,exs}"],

lib/earmark_parser/nimble_parsers.ex

@@ -0,0 +1,9 @@
+defmodule EarmarkParser.NimbleParsers do
+  @moduledoc ~S"""
+  coming soon
+  """
+
+  defdelegate parse_html_atts(input), to: __MODULE__.HtmlAttsParser
+end
+
+# SPDX-License-Identifier: AGPL-3.0-or-later

lib/earmark_parser/nimble_parsers/html_atts_parser.ex

@@ -0,0 +1,40 @@
+defmodule EarmarkParser.NimbleParsers.HtmlAttsParser do
+  @moduledoc ~S"""
+  Parses an HTML tag
+  """
+  import NimbleParsec
+  alias EarmarkParser.NimbleParsers.StringParser
+
+  html_att_name =
+    ascii_string([?a..?z, ?A..?z, ?-..?-], min: 1)
+
+  html_att =
+    html_att_name
+    |> choice([
+      "=" |> string() |> ignore() |> parsec({StringParser, :string_value}),
+      empty()
+    ])
+    |> reduce(:reduce_att)
+
+  html_att_end =
+    string(">")
+
+  html_atts =
+    html_att
+    |> repeat(" " |> string() |> times(min: 1) |> ignore() |> concat(html_att))
+
+  defparsec(:parse_html_atts, html_atts |> optional() |> ignore(html_att_end))
+
+  @doc false
+  def reduce_att(att_ast)
+
+  def reduce_att([name]) do
+    {name, true}
+  end
+
+  def reduce_att([name | values]) do
+    {name, IO.chardata_to_string(values)}
+  end
+end
+
+# SPDX-License-Identifier: AGPL-3.0-or-later

lib/earmark_parser/nimble_parsers/html_tag_parser.ex

@@ -0,0 +1,10 @@
+defmodule EarmarkParser.NimbleParsers.HtmlTagParser do
+
+  @moduledoc ~S"""
+  Parses an HTML tag
+  """
+  import NimbleParsec
+  alias EarmarkParser.NimbleParsers.HtmlAttsParser
+
+end
+# SPDX-License-Identifier: Apache-2.0

lib/earmark_parser/nimble_parsers/string_parser.ex

@@ -0,0 +1,38 @@
+defmodule EarmarkParser.NimbleParsers.StringParser do
+  @moduledoc ~S"""
+  String combinator
+  """
+
+  import NimbleParsec
+
+  inner_string = fn combinator, ch ->
+    combinator
+    |> repeat(
+      lookahead_not(ascii_char([ch]))
+      |> choice([
+        IO.chardata_to_string(["\\", ch])
+        |> string()
+        |> replace(ch),
+        utf8_char([])
+      ])
+    )
+  end
+
+  quoted_string = fn ch ->
+    empty()
+    |> ascii_char([ch])
+    |> ignore()
+    |> inner_string.(ch)
+    |> ignore(ascii_char([ch]))
+  end
+
+  defcombinator(
+    :string_value,
+    choice([
+      quoted_string.(?"),
+      quoted_string.(?')
+    ])
+  )
+end
+
+# SPDX-License-Identifier: AGPL-3.0-or-later

lib/earmark_parser/options.ex

@@ -104,8 +104,8 @@ defmodule EarmarkParser.Options do
   Use normalize before passing it into any API function
 
         iex(1)> options = normalize(annotations: "%%")
-        ...(1)> options.annotations
-        ~r{\A(.*)(%%.*)}
+        ...(1)> options.annotations.source
+        "\\A(.*)(%%.*)"
   """
   @spec normalize(t() | keyword()) :: t()
   def normalize(options)

mix.exs

@@ -1,10 +1,14 @@
 defmodule EarmarkParser.MixProject do
   use Mix.Project
 
-  @version "1.4.44"
+  @version "1.4.45"
   @url "https://github.com/RobertDober/earmark_parser"
 
   @deps [
+    # production environnement
+    {:nimble_parsec, "~> 1.4.2", runtime: false},
+
+    # dev and test environnements
     {:benchee, "~> 1.3.1", only: [:dev]},
     # {:credo, "~> 1.7.5", only: [:dev]},
     {:dialyxir, "~> 1.4.5", only: [:dev], runtime: false},
@@ -14,6 +18,11 @@ defmodule EarmarkParser.MixProject do
     {:floki, "~> 0.36", only: [:dev, :test]}
   ]
 
+  def cli do
+    [
+    ]
+  end
+
   def project do
     [
       app: :earmark_parser,

mix.lock

@@ -17,6 +17,7 @@
   "jason": {:hex, :jason, "1.4.4", "b9226785a9aa77b6857ca22832cffa5d5011a667207eb2a0ad56adb5db443b8a", [:mix], [{:decimal, "~> 1.0 or ~> 2.0", [hex: :decimal, repo: "hexpm", optional: true]}], "hexpm", "c5eb0cab91f094599f94d55bc63409236a8ec69a21a67814529e8d5f6cc90b3b"},
   "metrics": {:hex, :metrics, "1.0.1", "25f094dea2cda98213cecc3aeff09e940299d950904393b2a29d191c346a8486", [:rebar3], [], "hexpm", "69b09adddc4f74a40716ae54d140f93beb0fb8978d8636eaded0c31b6f099f16"},
   "mimerl": {:hex, :mimerl, "1.2.0", "67e2d3f571088d5cfd3e550c383094b47159f3eee8ffa08e64106cdf5e981be3", [:rebar3], [], "hexpm", "f278585650aa581986264638ebf698f8bb19df297f66ad91b18910dfc6e19323"},
+  "nimble_parsec": {:hex, :nimble_parsec, "1.4.2", "8efba0122db06df95bfaa78f791344a89352ba04baedd3849593bfce4d0dc1c6", [:mix], [], "hexpm", "4b21398942dda052b403bbe1da991ccd03a053668d147d53fb8c4e0efe09c973"},
   "parse_trans": {:hex, :parse_trans, "3.3.1", "16328ab840cc09919bd10dab29e431da3af9e9e7e7e6f0089dd5a2d2820011d8", [:rebar3], [], "hexpm", "07cd9577885f56362d414e8c4c4e6bdf10d43a8767abb92d24cbe8b24c54888b"},
   "ssl_verify_fun": {:hex, :ssl_verify_fun, "1.1.7", "354c321cf377240c7b8716899e182ce4890c5938111a1296add3ec74cf1715df", [:make, :mix, :rebar3], [], "hexpm", "fe4c190e8f37401d30167c8c405eda19469f34577987c76dde613e838bbc67f8"},
   "statistex": {:hex, :statistex, "1.0.0", "f3dc93f3c0c6c92e5f291704cf62b99b553253d7969e9a5fa713e5481cd858a5", [:mix], [], "hexpm", "ff9d8bee7035028ab4742ff52fc80a2aa35cece833cf5319009b52f1b5a86c27"},

test/nimble_parsers/html_att_test.exs

@@ -0,0 +1,50 @@
+defmodule Test.NimbleParsers.HtmlAttTest do
+  use Support.NimbleTestCase
+
+  describe "empty att list" do
+    test "returns an empty list if end char (>) is present" do
+      parse_html_atts(">")
+      |> parsed_ok([])
+    end
+
+    test "does not parse an empty string" do
+      parse_html_atts("")
+      |> parsed_error("expected string \">\"")
+    end
+  end
+
+  describe "boolean attribute" do
+    test "just it's presence" do
+      parse_html_atts("hidden>")
+      |> parsed_ok([{"hidden", true}])
+    end
+
+    test "two boolean attributes" do
+      parse_html_atts("hidden and-visible>")
+      |> parsed_ok([{"hidden", true}, {"and-visible", true}])
+    end
+  end
+
+  describe "a string attribute" do
+    test "elixir, what else?" do
+      parse_html_atts(~S{lang="elixir">})
+      |> parsed_ok([{"lang", "elixir"}])
+    end
+
+    test "escaped double quote" do
+      parse_html_atts(~S{lang="\"lua\"">})
+      |> parsed_ok([{"lang", "\"lua\""}])
+    end
+
+    test "single quoted string too" do
+      parse_html_atts(~S{lang="\"pt-br\"" lang='fr-fr' lang='de-\'at\''>})
+      |> parsed_ok([
+        {"lang", "\"pt-br\""},
+        {"lang", "fr-fr"},
+        {"lang", "de-'at'"}
+      ])
+    end
+  end
+end
+
+# SPDX-License-Identifier: Apache-2.0

test/nimble_parsers/html_oneline_tag_test.exs

@@ -0,0 +1,13 @@
+defmodule Test.NimbleParsers.HtmlOnelineTagTest do
+
+  use Support.NimbleTestCase
+
+  describe "no attributes" do
+    test "br" do
+      # parse_html_tag("<br/>")
+      # |> parsed_ok({"br", [], [], %{verbatim: true}})
+    end
+  end
+
+end
+# SPDX-License-Identifier: Apache-2.0

test/support/nimble_test_case.ex

@@ -0,0 +1,12 @@
+defmodule Support.NimbleTestCase do
+  defmacro __using__(_options) do
+    quote do
+      use ExUnit.Case, async: true
+
+      import EarmarkParser.NimbleParsers
+      import Support.NimbleTests
+    end
+  end
+end
+
+# SPDX-License-Identifier: Apache-2.0

test/support/nimble_tests.ex

@@ -0,0 +1,22 @@
+defmodule Support.NimbleTests do
+  @moduledoc ~S"""
+  Makes asserting on NimbleParsec results simpler
+  """
+  defmacro parsed_error(parsed, expected) do
+    quote do
+      # |> IO.inspect()
+      {:error, message, _, _, _, _} = unquote(parsed)
+      assert message == unquote(expected)
+    end
+  end
+
+  defmacro parsed_ok(parsed, expected) do
+    quote do
+      # |> IO.inspect()
+      {:ok, result, _, _, _, _} = unquote(parsed)
+      assert result == unquote(expected)
+    end
+  end
+end
+
+# SPDX-License-Identifier: Apache-2.0