fix: newlines in bold/italic

commonmark.go

@@ -172,9 +172,13 @@ var commonmark = []Rule{
 			if trimmed == "" {
 				return &trimmed
 			}
-			trimmed = opt.StrongDelimiter + trimmed + opt.StrongDelimiter
 
-			// always have a space to the side to recognize the delimiter
+			// If there is a newline character between the start and end delimiter
+			// the delimiters won't be recognized. Either we remove all newline characters
+			// OR on _every_ line we put start & end delimiters.
+			trimmed = delimiterForEveryLine(trimmed, opt.StrongDelimiter)
+
+			// Always have a space to the side to recognize the delimiter
 			trimmed = AddSpaceIfNessesary(selec, trimmed)
 
 			return &trimmed
@@ -193,9 +197,13 @@ var commonmark = []Rule{
 			if trimmed == "" {
 				return &trimmed
 			}
-			trimmed = opt.EmDelimiter + trimmed + opt.EmDelimiter
 
-			// always have a space to the side to recognize the delimiter
+			// If there is a newline character between the start and end delimiter
+			// the delimiters won't be recognized. Either we remove all newline characters
+			// OR on _every_ line we put start & end delimiters.
+			trimmed = delimiterForEveryLine(trimmed, opt.EmDelimiter)
+
+			// Always have a space to the side to recognize the delimiter
 			trimmed = AddSpaceIfNessesary(selec, trimmed)
 
 			return &trimmed

testdata/TestCommonmark/italic/goldmark.golden

@@ -5,3 +5,9 @@
 <p>Some <em>19,80€</em> Text</p>
 <p><em>Content</em> and no space afterward.</p>
 <p>_Not Italic_</p>
+<p><em>First</em></p>
+<p><em>Second</em></p>
+<p><em>Text und Fotos: Max Mustermann</em></p>
+<p><em>Original veröffentlicht am 01.01.2021 auf</em> <a href="http://example.com/posts/100979">Zeitung</a></p>
+<p><em>Übersetzung: Max Mustermann</em></p>
+<p><em>veröffentlicht am 02.02.2021</em></p>

testdata/TestCommonmark/italic/input.html

@@ -21,4 +21,14 @@
 
 
 <!--with markdown characters (escaping)-->
-<p>_Not Italic_</p>
+<p>_Not Italic_</p>
+
+
+<!--with br inside it-->
+<p><em>First <br /> Second</em></p>
+
+<p>
+    <em>Text und Fotos: Max Mustermann<br>Original veröffentlicht am 01.01.2021 auf </em>
+    <a href="/posts/100979">Zeitung</a><br>
+    <em>Übersetzung: Max Mustermann<br>veröffentlicht am 02.02.2021</em>
+</p>

testdata/TestCommonmark/italic/output.asterisks.golden

@@ -10,4 +10,16 @@ Some *19,80€* Text
 
 *Content* and no space afterward.
 
-\_Not Italic\_
+\_Not Italic\_
+
+*First*
+
+*Second*
+
+*Text und Fotos: Max Mustermann*
+
+*Original veröffentlicht am 01.01.2021 auf* [Zeitung](http://example.com/posts/100979)
+
+*Übersetzung: Max Mustermann*
+
+*veröffentlicht am 02.02.2021*

testdata/TestCommonmark/italic/output.underscores.golden

@@ -10,4 +10,16 @@ Some _19,80€_ Text
 
 _Content_ and no space afterward.
 
-\_Not Italic\_
+\_Not Italic\_
+
+_First_
+
+_Second_
+
+_Text und Fotos: Max Mustermann_
+
+_Original veröffentlicht am 01.01.2021 auf_ [Zeitung](http://example.com/posts/100979)
+
+_Übersetzung: Max Mustermann_
+
+_veröffentlicht am 02.02.2021_

utils.go

@@ -326,6 +326,25 @@ func getCodeContent(selec *goquery.Selection) string {
 	return content
 }
 
+// delimiterForEveryLine puts the delimiter not just at the start and end of the string
+// but if the text is divided on multiple lines, puts the delimiters on every line with content.
+//
+// Otherwise the bold/italic delimiters won't be recognized if it contains new line characters.
+func delimiterForEveryLine(text string, delimiter string) string {
+	lines := strings.Split(text, "\n")
+
+	for i, line := range lines {
+		line = strings.TrimSpace(line)
+		if line == "" {
+			// Skip empty lines
+			continue
+		}
+
+		lines[i] = delimiter + line + delimiter
+	}
+	return strings.Join(lines, "\n")
+}
+
 // isWrapperListItem returns wether the list item has own
 // content or is just a wrapper for another list.
 // e.g. "<li><ul>..."