HTML reader/writer: Better handling of <q> with cite attribute (#5837)

* HTML reader: Handle cite attribute for quotes.  If a `<q>` tag has a `cite` attribute, we interpret it as a Quoted element with an inner Span.  Closes #5798

* Refactor url canonicalization into a helper function

* Modify HTML writer to handle quote with cite.

[0]: https://developer.mozilla.org/en-US/docs/Web/HTML/Element/q

3 files changed, 25 insertions, 0 deletions

diff --git a/test/Tests/Writers/HTML.hs b/test/Tests/Writers/HTML.hs
index de8b1ef17..94549e0d8 100644
--- a/test/Tests/Writers/HTML.hs
+++ b/test/Tests/Writers/HTML.hs
@@ -13,6 +13,11 @@ import Text.Pandoc.Builder
 html :: (ToPandoc a) => a -> String
 html = unpack . purely (writeHtml4String def{ writerWrapText = WrapNone }) . toPandoc
 
+htmlQTags :: (ToPandoc a) => a -> String
+htmlQTags = unpack
+  . purely (writeHtml4String def{ writerWrapText = WrapNone, writerHtmlQTags = True })
+  . toPandoc
+
 {-
   "my test" =: X =?> Y
 
@@ -48,4 +53,16 @@ tests = [ testGroup "inline code"
             definitionList [(mempty, [para $ text "foo bar"])]
             =?> "<dl><dt></dt><dd><p>foo bar</p></dd></dl>"
           ]
+        , testGroup "quotes"
+          [ "quote with cite attribute (without q-tags)" =:
+            doubleQuoted (spanWith ("", [], [("cite", "http://example.org")]) (str "examples"))
+            =?> "“<span cite=\"http://example.org\">examples</span>”"
+          , tQ "quote with cite attribute (with q-tags)" $
+            doubleQuoted (spanWith ("", [], [("cite", "http://example.org")]) (str "examples"))
+            =?> "<q cite=\"http://example.org\">examples</q>"
+          ]
         ]
+        where
+          tQ :: (ToString a, ToPandoc a)
+               => String -> (a, String) -> TestTree
+          tQ = test htmlQTags
diff --git a/test/html-reader.html b/test/html-reader.html
index a2bca5d2c..ae937af82 100644
--- a/test/html-reader.html
+++ b/test/html-reader.html
@@ -81,6 +81,10 @@ span.pandocNoteMarker { }
 </blockquote>
 <p>And a following paragraph.</p>
 <hr />
+<h1>Inline quotes</h1>
+<p>Normal text but then a <q cite="https://www.imdb.com/title/tt0062622/quotes/qt0396921">inline quote</q>.</p>
+<p><q>Missing a cite attribute means its just normal text</q></p>
+<hr />
 <h1>Code Blocks</h1>
 <p>Code:</p>
 <pre><code>---- (should be four hyphens)
diff --git a/test/html-reader.native b/test/html-reader.native
index acd0087ef..5643fb73f 100644
--- a/test/html-reader.native
+++ b/test/html-reader.native
@@ -51,6 +51,10 @@ Pandoc (Meta {unMeta = fromList [("generator",MetaInlines [Str "pandoc"]),("titl
   [Para [Str "Don't",Space,Str "quote",Space,Str "me."]]]
 ,Para [Str "And",Space,Str "a",Space,Str "following",Space,Str "paragraph."]
 ,HorizontalRule
+,Header 1 ("inline-quotes",[],[]) [Str "Inline",Space,Str "quotes"]
+,Para [Str "Normal",Space,Str "text",Space,Str "but",Space,Str "then",Space,Str "a",Space,Quoted DoubleQuote [Span ("",[],[("cite","https://www.imdb.com/title/tt0062622/quotes/qt0396921")]) [Str "inline",Space,Str "quote"]],Str "."]
+,Para [Quoted DoubleQuote [Str "Missing",Space,Str "a",Space,Str "cite",Space,Str "attribute",Space,Str "means",Space,Str "its",Space,Str "just",Space,Str "normal",Space,Str "text"]]
+,HorizontalRule
 ,Header 1 ("code-blocks",[],[]) [Str "Code",Space,Str "Blocks"]
 ,Para [Str "Code:"]
 ,CodeBlock ("",[],[]) "---- (should be four hyphens)\n\nsub status {\n    print \"working\";\n}\n\nthis code block is indented by one tab"