Improve whitespace handling to match old library

python/acl_anthology/text/markuptext.py

@@ -103,7 +103,7 @@ def as_text(self) -> str:
             The plain text with any markup stripped. The only transformation that will be performed is replacing TeX-math expressions with their corresponding Unicode representation, if possible.
         """
         if isinstance(self._content, str):
-            return self._content
+            return remove_extra_whitespace(self._content)
         if self._text is not None:
             return self._text
         element = deepcopy(self._content)
@@ -123,7 +123,7 @@ def as_html(self, allow_url: bool = True) -> str:
                 `<a href="...">` tags, but in simply `<span>` tags.
         """
         if isinstance(self._content, str):
-            return xml_escape(self._content)
+            return xml_escape(remove_extra_whitespace(self._content))
         if self._html is not None:
             return self._html
         element = deepcopy(self._content)
@@ -153,10 +153,10 @@ def as_latex(self) -> str:
         if self._latex is not None:
             return self._latex
         if isinstance(self._content, str):
-            self._latex = latex_convert_quotes(latex_encode(self._content))
+            latex = latex_convert_quotes(latex_encode(self._content))
         else:
             latex = markup_to_latex(self._content)
-            self._latex = remove_extra_whitespace(latex)
+        self._latex = remove_extra_whitespace(latex)
         return self._latex
 
     def as_xml(self) -> str:
@@ -167,7 +167,7 @@ def as_xml(self) -> str:
         if isinstance(self._content, str):
             return xml_escape(self._content)
         if self._xml is None:
-            self._xml = remove_extra_whitespace(stringify_children(self._content))
+            self._xml = stringify_children(self._content)
         return self._xml
 
     @classmethod

python/tests/text/markuptext_test.py

@@ -65,6 +65,22 @@
             "latex": "Workshop on Topic A \\& B",
         },
     ),
+    (
+        "Title with\n\n line breaks",
+        {
+            "text": "Title with line breaks",
+            "html": "Title with line breaks",
+            "latex": "Title with line breaks",
+        },
+    ),
+    (
+        "<span>Title with\n\n line breaks</span>",
+        {
+            "text": "Title with line breaks",
+            "html": "<span>Title with line breaks</span>",
+            "latex": "Title with line breaks",
+        },
+    ),
     (
         "<fixed-case>U</fixed-case>pstream <fixed-case>M</fixed-case>itigation <fixed-case>I</fixed-case>s <i><fixed-case>N</fixed-case>ot</i> <fixed-case>A</fixed-case>ll <fixed-case>Y</fixed-case>ou <fixed-case>N</fixed-case>eed",
         {