From cba78821ed13d5a92c74d092914fbad4c842f889 Mon Sep 17 00:00:00 2001
From: Titus Wormer <tituswormer@gmail.com>
Date: Thu, 9 Jun 2022 14:03:04 +0200
Subject: Add temporary support for stripping whitespace

---
 readme.md               |  1 +
 src/compiler.rs         | 12 ++++++++----
 tests/code_indented.rs  | 11 +++++------
 tests/heading_atx.rs    | 11 +++++------
 tests/html_flow.rs      | 11 +++++------
 tests/thematic_break.rs | 11 +++++------
 6 files changed, 29 insertions(+), 28 deletions(-)

diff --git a/readme.md b/readme.md
index 1dc9932..15edf87 100644
--- a/readme.md
+++ b/readme.md
@@ -70,6 +70,7 @@ cargo doc --document-private-items
 
 ### Small things
 
+- [ ] (1) Parse initial and final whitespace of paragraphs (in text)
 - [ ] (3) Clean compiler
 - [ ] (1) Optionally remove dangerous protocols when compiling
 - [ ] (1) Use preferred line ending style in markdown
diff --git a/src/compiler.rs b/src/compiler.rs
index 4f362b8..3632d29 100644
--- a/src/compiler.rs
+++ b/src/compiler.rs
@@ -280,11 +280,15 @@ pub fn compile(events: &[Event], codes: &[Code], options: &CompileOptions) -> St
                 }
                 // To do: `ContentPhrasing` should be parsed as phrasing first.
                 // This branch below currently acts as the resulting `data` tokens.
-                TokenType::ContentChunk
+                // To do: initial and final whitespace should be handled in `text`.
+                TokenType::ContentChunk => {
+                    // last_was_tag = false;
+                    buf_tail_mut(buffers).push(encode(
+                        slice_serialize(codes, &get_span(events, index), false).trim(),
+                    ));
+                }
                 // To do: `ChunkString` does not belong here. Remove it when subtokenization is supported.
-                | TokenType::ChunkString
-                | TokenType::Data
-                | TokenType::CharacterEscapeValue => {
+                TokenType::ChunkString | TokenType::Data | TokenType::CharacterEscapeValue => {
                     // last_was_tag = false;
                     buf_tail_mut(buffers).push(encode(&slice_serialize(
                         codes,
diff --git a/tests/code_indented.rs b/tests/code_indented.rs
index 80dae4b..5967cb3 100644
--- a/tests/code_indented.rs
+++ b/tests/code_indented.rs
@@ -40,12 +40,11 @@ fn code_indented() {
         "should support blank lines in indented code (3)"
     );
 
-    // To do: strip whitespace.
-    // assert_eq!(
-    //     micromark("Foo\n    bar"),
-    //     "<p>Foo\nbar</p>",
-    //     "should not support interrupting paragraphs"
-    // );
+    assert_eq!(
+        micromark("Foo\n    bar"),
+        "<p>Foo\nbar</p>",
+        "should not support interrupting paragraphs"
+    );
 
     assert_eq!(
         micromark("    foo\nbar"),
diff --git a/tests/heading_atx.rs b/tests/heading_atx.rs
index 7a830fe..defc77f 100644
--- a/tests/heading_atx.rs
+++ b/tests/heading_atx.rs
@@ -99,12 +99,11 @@ fn heading_atx() {
         "should not support four initial spaces"
     );
 
-    // To do: strip whitespace.
-    // assert_eq!(
-    //     micromark("foo\n    # bar"),
-    //     "<p>foo\n# bar</p>",
-    //     "should not support four initial spaces when interrupting"
-    // );
+    assert_eq!(
+        micromark("foo\n    # bar"),
+        "<p>foo\n# bar</p>",
+        "should not support four initial spaces when interrupting"
+    );
 
     assert_eq!(
         micromark("## foo ##"),
diff --git a/tests/html_flow.rs b/tests/html_flow.rs
index 18eff2d..0a5ec72 100644
--- a/tests/html_flow.rs
+++ b/tests/html_flow.rs
@@ -814,12 +814,11 @@ fn html_flow_7_complete() {
         "should not support a line ending directly after a tag name"
     );
 
-    // To do: trim trailing whitespace.
-    // assert_eq!(
-    //     micromark_with_options("<x ", DANGER),
-    //     "<p>&lt;x</p>",
-    //     "should not support an eof after a space directly after a tag name"
-    // );
+    assert_eq!(
+        micromark_with_options("<x ", DANGER),
+        "<p>&lt;x</p>",
+        "should not support an eof after a space directly after a tag name"
+    );
 
     assert_eq!(
         micromark_with_options("<x/", DANGER),
diff --git a/tests/thematic_break.rs b/tests/thematic_break.rs
index 1e3fb1e..fa86835 100644
--- a/tests/thematic_break.rs
+++ b/tests/thematic_break.rs
@@ -63,12 +63,11 @@ fn thematic_break() {
         "should not support thematic breaks w/ 4 spaces"
     );
 
-    // To do: trim initial whitespace.
-    // assert_eq!(
-    //     micromark("Foo\n    ***"),
-    //     "<p>Foo\n***</p>",
-    //     "should not support thematic breaks w/ 4 spaces as paragraph continuation"
-    // );
+    assert_eq!(
+        micromark("Foo\n    ***"),
+        "<p>Foo\n***</p>",
+        "should not support thematic breaks w/ 4 spaces as paragraph continuation"
+    );
 
     assert_eq!(
         micromark("_____________________________________"),
-- 
cgit