Improvements

- Lowercase method atoms - Use binaries both for header names and values - Introduce `raw_headers` which preserve headers case and order. [erlang/otp#2466](erlang/otp#2466) allows to keep headers case - Some documentation updates - Update deps - Update CI to run tests inside docker container
tank-bohr · Apr 3, 2021 · f885ff5 · f885ff5
1 parent a205a44
commit f885ff5
Show file tree

Hide file tree

Showing 12 changed files with 205 additions and 92 deletions.
diff --git a/.github/workflows/erlang.yml b/.github/workflows/erlang.yml
@@ -8,19 +8,19 @@ on:
 jobs:
 
   tests:
+    name: OTP ${{ matrix.otp }}
 
     runs-on: ubuntu-latest
-    name: OTP ${{ matrix.otp }}
+
+    container:
+      image: erlang:${{ matrix.otp }}
+
     strategy:
       matrix:
-        otp: [20.3, 21.3, 22.2]
+        otp: ['20.3', '21.3', '22.3', '23.3', '24.0-rc2']
 
     steps:
-      - uses: actions/checkout@v2
-
-      - uses: gleam-lang/[email protected]
-        with:
-          otp-version: ${{ matrix.otp }}
+      - uses: actions/[email protected]
 
       - name: Run tests
         run: rebar3 do eunit, ct, dialyzer

diff --git a/README.md b/README.md
@@ -2,9 +2,9 @@
 
 # Bookish spork #
 
-Copyright (c) 2018-2020 Alexey Nikitin
+Copyright (c) 2018-2021 Alexey Nikitin
 
-__Version:__ 0.3.6
+__Version:__ 0.4.0
 
 __Authors:__ Alexey Nikitin ([`[email protected]`](mailto:[email protected])) (_web site:_ [`https://twitter.com/tank_bohr`](https://twitter.com/tank_bohr)).
 
@@ -52,7 +52,7 @@ First step: add to your rebar config
 {profiles, [
     {test, [
         {deps, [
-            {bookish_spork, "0.3.6"}
+            {bookish_spork, "0.4.0"}
         ]}
     ]}
 ]}.
@@ -111,13 +111,15 @@ It returns you an opaque structure of the request. You can inspect it with
 
 #### <a name="Bypass_comparision">Bypass comparision</a> ####
 
-An elixir library [bypass](https://github.com/PSPDFKit-labs/bypass) does pretty much the same. And illustrates the same approach. It starts a cowboy web-server to replace a real service for test
+An elixir library [bypass](https://github.com/PSPDFKit-labs/bypass) does pretty much the same. And illustrates the same approach. It starts a cowboy web-server to replace a real service for test. It's a beautiful library with great API, documentation, and very concise source code. If you are an elixir developer, most likely, it will be a good fit for you.
 
-But bookish_spork has some advantages:
+But nevertheless bookish_spork has some advantages:
 
-* Bypass depends on `cowboy` and `plug`. Bookish spork has zero dependencies
-* Bookish spork works seamlessly with both erlang and elixir. Bypass is supposed to be an elixir only library
-* Bookish spork much simpler (I believe)
+* Bypass depends on `cowboy` and `plug`. Bookish spork has zero dependencies.
+* Bookish spork works seamlessly with both erlang and elixir. Bypass is supposed to be an elixir only library.
+* <strike>Bookish spork much simpler (I believe)</strike>
+ (not any more).
+* Bookish spork allows you to inspect the request very deeply and accurate. For example take a look at [`bookish_spork_request:raw_headers/1`](http://github.com/tank-bohr/bookish_spork/blob/master/doc/bookish_spork_request.md#raw_headers-1) and [`bookish_spork_request:ssl_info/1`](http://github.com/tank-bohr/bookish_spork/blob/master/doc/bookish_spork_request.md#ssl_info-1) and [`bookish_spork_request:tls_ext/1`](http://github.com/tank-bohr/bookish_spork/blob/master/doc/bookish_spork_request.md#tls_ext-1). It can be useful for HTTP clients testing.
 
 
 #### <a name="Elli_comparision">Elli comparision</a> ####
@@ -227,9 +229,9 @@ defmodule ChuckNorrisApiTest do
   use ExUnit.Case
   doctest ChuckNorrisApi
 
-  setup_all do
-    {:ok, _} = :bookish_spork.start_server
-    {:ok, %{}}
+  setup do
+    {:ok, _} = :bookish_spork.start_server()
+    on_exit(fn -> :bookish_spork.stop_server() end)
   end
 
   test "retrieves a random joke" do
@@ -238,8 +240,8 @@ defmodule ChuckNorrisApiTest do
     }"])
     assert ChuckNorrisApi.random == "Chuck norris tried to crank that soulja boy but it wouldn't crank up"
 
-    {:ok, request} = :bookish_spork.capture_request
-    assert request.uri == '/jokes/random'
+    {:ok, request} = :bookish_spork.capture_request()
+    assert request.uri === "/jokes/random"
   end
 end
 

diff --git a/doc/README.md b/doc/README.md
@@ -2,9 +2,9 @@
 
 # Bookish spork #
 
-Copyright (c) 2018-2020 Alexey Nikitin
+Copyright (c) 2018-2021 Alexey Nikitin
 
-__Version:__ 0.3.6
+__Version:__ 0.4.0
 
 __Authors:__ Alexey Nikitin ([`[email protected]`](mailto:[email protected])) (_web site:_ [`https://twitter.com/tank_bohr`](https://twitter.com/tank_bohr)).
 
@@ -52,7 +52,7 @@ First step: add to your rebar config
 {profiles, [
     {test, [
         {deps, [
-            {bookish_spork, "0.3.6"}
+            {bookish_spork, "0.4.0"}
         ]}
     ]}
 ]}.
@@ -111,13 +111,15 @@ It returns you an opaque structure of the request. You can inspect it with
 
 #### <a name="Bypass_comparision">Bypass comparision</a> ####
 
-An elixir library [bypass](https://github.com/PSPDFKit-labs/bypass) does pretty much the same. And illustrates the same approach. It starts a cowboy web-server to replace a real service for test
+An elixir library [bypass](https://github.com/PSPDFKit-labs/bypass) does pretty much the same. And illustrates the same approach. It starts a cowboy web-server to replace a real service for test. It's a beautiful library with great API, documentation, and very concise source code. If you are an elixir developer, most likely, it will be a good fit for you.
 
-But bookish_spork has some advantages:
+But nevertheless bookish_spork has some advantages:
 
-* Bypass depends on `cowboy` and `plug`. Bookish spork has zero dependencies
-* Bookish spork works seamlessly with both erlang and elixir. Bypass is supposed to be an elixir only library
-* Bookish spork much simpler (I believe)
+* Bypass depends on `cowboy` and `plug`. Bookish spork has zero dependencies.
+* Bookish spork works seamlessly with both erlang and elixir. Bypass is supposed to be an elixir only library.
+* <strike>Bookish spork much simpler (I believe)</strike>
+ (not any more).
+* Bookish spork allows you to inspect the request very deeply and accurate. For example take a look at [`bookish_spork_request:raw_headers/1`](bookish_spork_request.md#raw_headers-1) and [`bookish_spork_request:ssl_info/1`](bookish_spork_request.md#ssl_info-1) and [`bookish_spork_request:tls_ext/1`](bookish_spork_request.md#tls_ext-1). It can be useful for HTTP clients testing.
 
 
 #### <a name="Elli_comparision">Elli comparision</a> ####
@@ -227,9 +229,9 @@ defmodule ChuckNorrisApiTest do
   use ExUnit.Case
   doctest ChuckNorrisApi
 
-  setup_all do
-    {:ok, _} = :bookish_spork.start_server
-    {:ok, %{}}
+  setup do
+    {:ok, _} = :bookish_spork.start_server()
+    on_exit(fn -> :bookish_spork.stop_server() end)
   end
 
   test "retrieves a random joke" do
@@ -238,8 +240,8 @@ defmodule ChuckNorrisApiTest do
     }"])
     assert ChuckNorrisApi.random == "Chuck norris tried to crank that soulja boy but it wouldn't crank up"
 
-    {:ok, request} = :bookish_spork.capture_request
-    assert request.uri == '/jokes/random'
+    {:ok, request} = :bookish_spork.capture_request()
+    assert request.uri === "/jokes/random"
   end
 end
 

diff --git a/doc/bookish_spork_request.md b/doc/bookish_spork_request.md
@@ -22,7 +22,7 @@ __abstract datatype__: `t()`
 ## Function Index ##
 
 
-<table width="100%" border="1" cellspacing="0" cellpadding="2" summary="function index"><tr><td valign="top"><a href="#body-1">body/1</a></td><td>request body.</td></tr><tr><td valign="top"><a href="#content_length-1">content_length/1</a></td><td>Content-Length header value as intger.</td></tr><tr><td valign="top"><a href="#from_transport-1">from_transport/1</a></td><td></td></tr><tr><td valign="top"><a href="#header-2">header/2</a></td><td>Returns a particular header from request.</td></tr><tr><td valign="top"><a href="#headers-1">headers/1</a></td><td>http headers map.</td></tr><tr><td valign="top"><a href="#is_keepalive-1">is_keepalive/1</a></td><td>tells you if the request is keepalive or not <a href="https://tools.ietf.org.md/rfc6223" target="_top"><tt>https://tools.ietf.org/html/rfc6223</tt></a></td></tr><tr><td valign="top"><a href="#method-1">method/1</a></td><td>http verb: 'GET', 'POST','PUT', 'DELETE', 'OPTIONS', ...</td></tr><tr><td valign="top"><a href="#transport-1">transport/1</a></td><td></td></tr><tr><td valign="top"><a href="#uri-1">uri/1</a></td><td>path with query string.</td></tr><tr><td valign="top"><a href="#version-1">version/1</a></td><td>http protocol version tuple.</td></tr></table>
+<table width="100%" border="1" cellspacing="0" cellpadding="2" summary="function index"><tr><td valign="top"><a href="#body-1">body/1</a></td><td>request body.</td></tr><tr><td valign="top"><a href="#content_length-1">content_length/1</a></td><td>Content-Length header value as intger.</td></tr><tr><td valign="top"><a href="#from_transport-1">from_transport/1</a></td><td></td></tr><tr><td valign="top"><a href="#header-2">header/2</a></td><td>Returns a particular header from request.</td></tr><tr><td valign="top"><a href="#headers-1">headers/1</a></td><td>HTTP headers map.</td></tr><tr><td valign="top"><a href="#is_keepalive-1">is_keepalive/1</a></td><td>tells you if the request is keepalive or not <a href="https://tools.ietf.org.md/rfc6223" target="_top"><tt>https://tools.ietf.org/html/rfc6223</tt></a></td></tr><tr><td valign="top"><a href="#method-1">method/1</a></td><td>http verb in lower case: get, post, put, delete, options, ...</td></tr><tr><td valign="top"><a href="#raw_headers-1">raw_headers/1</a></td><td>HTTP raw headers.</td></tr><tr><td valign="top"><a href="#transport-1">transport/1</a></td><td></td></tr><tr><td valign="top"><a href="#uri-1">uri/1</a></td><td>path with query string.</td></tr><tr><td valign="top"><a href="#version-1">version/1</a></td><td>http protocol version tuple.</td></tr></table>
 
 
 <a name="functions"></a>
@@ -65,11 +65,11 @@ from_transport(Transport::<a href="bookish_spork_transport.md#type-t">bookish_sp
 ### header/2 ###
 
 <pre><code>
-header(Request::<a href="#type-t">t()</a>, HeaderName::string()) -&gt; string() | nil
+header(Request::<a href="#type-t">t()</a>, HeaderName::string() | binary()) -&gt; binary() | nil
 </code></pre>
 <br />
 
-Returns a particular header from request. Header name is lowerced
+Returns a particular header from request.
 
 <a name="headers-1"></a>
 
@@ -80,7 +80,7 @@ headers(Request::<a href="#type-t">t()</a>) -&gt; map()
 </code></pre>
 <br />
 
-http headers map. Header names are normalized and lowercased
+HTTP headers map. Header names are normalized and lowercased
 
 <a name="is_keepalive-1"></a>
 
@@ -102,7 +102,18 @@ method(Request::<a href="#type-t">t()</a>) -&gt; atom()
 </code></pre>
 <br />
 
-http verb: 'GET', 'POST','PUT', 'DELETE', 'OPTIONS', ...
+http verb in lower case: get, post, put, delete, options, ...
+
+<a name="raw_headers-1"></a>
+
+### raw_headers/1 ###
+
+<pre><code>
+raw_headers(Request::<a href="#type-t">t()</a>) -&gt; <a href="proplists.md#type-proplist">proplists:proplist()</a>
+</code></pre>
+<br />
+
+HTTP raw headers. Headers order and case are preserved
 
 <a name="transport-1"></a>
 

diff --git a/doc/overview.edoc b/doc/overview.edoc
@@ -112,17 +112,19 @@ It returns you an opaque structure of the request. You can inspect it with
 
 === Bypass comparision ===
 
-An elixir library <a target="_blank" href="https://github.com/PSPDFKit-labs/bypass">bypass</a> does pretty much the same. And illustrates the same approach. It starts a cowboy web-server to replace a real service for test
+An elixir library <a target="_blank" href="https://github.com/PSPDFKit-labs/bypass">bypass</a> does pretty much the same. And illustrates the same approach. It starts a cowboy web-server to replace a real service for test. It's a beautiful library with great API, documentation, and very concise source code. If you are an elixir developer, most likely, it will be a good fit for you.
 
-But bookish_spork has some advantages:
+But nevertheless bookish_spork has some advantages:
 
 <ul>
 
-  <li>Bypass depends on `cowboy' and `plug'. Bookish spork has zero dependencies</li>
+  <li>Bypass depends on `cowboy' and `plug'. Bookish spork has zero dependencies.</li>
 
-  <li>Bookish spork works seamlessly with both erlang and elixir. Bypass is supposed to be an elixir only library</li>
+  <li>Bookish spork works seamlessly with both erlang and elixir. Bypass is supposed to be an elixir only library.</li>
 
-  <li>Bookish spork much simpler (I believe)</li>
+  <li><strike>Bookish spork much simpler (I believe)</strike> (not any more).</li>
+
+  <li>Bookish spork allows you to inspect the request very deeply and accurate. For example take a look at {@link bookish_spork_request:raw_headers/1} and {@link bookish_spork_request:ssl_info/1} and {@link bookish_spork_request:tls_ext/1}. It can be useful for HTTP clients testing.</li>
 
 </ul>
 
@@ -221,9 +223,9 @@ defmodule ChuckNorrisApiTest do
   use ExUnit.Case
   doctest ChuckNorrisApi
 
-  setup_all do
-    {:ok, _} = :bookish_spork.start_server
-    {:ok, %{}}
+  setup do
+    {:ok, _} = :bookish_spork.start_server()
+    on_exit(fn -> :bookish_spork.stop_server() end)
   end
 
   test "retrieves a random joke" do
@@ -232,8 +234,8 @@ defmodule ChuckNorrisApiTest do
     }"])
     assert ChuckNorrisApi.random == "Chuck norris tried to crank that soulja boy but it wouldn't crank up"
 
-    {:ok, request} = :bookish_spork.capture_request
-    assert request.uri == '/jokes/random'
+    {:ok, request} = :bookish_spork.capture_request()
+    assert request.uri === "/jokes/random"
   end
 end
 </pre>

diff --git a/elvis.config b/elvis.config
@@ -9,6 +9,12 @@
                     {elvis_style, function_naming_convention, #{
                         ignore => [bookish_spork_request],
                         regex => "^([a-z][a-z0-9]*_?)*$"
+                    }},
+                    {elvis_style, state_record_and_type, #{
+                        ignore => [
+                            bookish_spork_acceptor,
+                            bookish_spork_blocking_queue
+                        ]
                     }}
                 ],
                 ruleset => erl_files

diff --git a/rebar.config b/rebar.config
@@ -10,7 +10,7 @@
 {profiles, [
     {test, [
         {deps, [
-            {gun, "1.3.1"}
+            {gun, "1.3.3"}
         ]},
 
         {plugins, [
@@ -33,7 +33,7 @@
     ]},
     {elvis, [
         {plugins, [
-            {rebar3_lint, "0.1.11"}
+            {rebar3_lint, "0.4.0"}
         ]}
     ]}
 ]}.

diff --git a/src/bookish_spork_handler.erl b/src/bookish_spork_handler.erl
@@ -85,7 +85,8 @@ receive_request(#state{transport = Transport, request = RequestIn} = State) ->
         {ok, {http_request, Method, {abs_path, Uri}, Version}} ->
             RequestOut = bookish_spork_request:request_line(RequestIn, Method, Uri, Version),
             ?FUNCTION_NAME(State#state{request = RequestOut});
-        {ok, {http_header, _, Header, _, Value}} ->
+        {ok, {http_header, _, NormalizedHeader, PreserveCaseHeader, Value}} ->
+            Header = choose_header(NormalizedHeader, PreserveCaseHeader),
             RequestOut = bookish_spork_request:add_header(RequestIn, Header, Value),
             ?FUNCTION_NAME(State#state{request = RequestOut});
         {ok, http_eoh} ->
@@ -136,6 +137,11 @@ complete_connection(State = #state{transport = Transport, request = Request}) ->
             {halt, normal}
     end.
 
+choose_header(NormalizedHeader, undefined) ->
+    NormalizedHeader;
+choose_header(_NormalizedHeader, PreserveCaseHeader) ->
+    PreserveCaseHeader.
+
 reduce_while(State, []) ->
     {noreply, State};
 reduce_while(State, [Fun|Rest]) ->