(VTK-3344) Support stripping style

pycaption/base.py

@@ -425,6 +425,23 @@ def remove_empty_captions(self):
                     out_captions.append(caption)
             self.set_captions(lang, out_captions)
 
+    def remove_styling(self):
+        """
+        Removes structured styling from all captions in all languages.
+
+        Unlike ``strip_html_tags`` (which only removes literal markup found
+        in TEXT node content), this removes styling that readers store
+        structurally: ``CaptionNode.STYLE`` nodes and the per-caption
+        ``style`` attribute. Writers generate style tags (e.g. ``<i>`` in
+        WebVTT) from these at write time, so this is what actually prevents
+        formats like SCC from producing styled output.
+        """
+        for lang in self.get_languages():
+            captions = self.get_captions(lang)
+            for caption in captions:
+                caption.style = {}
+                caption.nodes = [node for node in caption.nodes if node.type_ != CaptionNode.STYLE]
+
     def remove_layout_info(self):
         """
         Removes layout info from all captions and nodes in all languages.

tests/test_base.py

@@ -1,6 +1,6 @@
 import pytest
 
-from pycaption.base import CaptionList, Caption
+from pycaption.base import CaptionList, Caption, CaptionNode, CaptionSet
 
 
 class TestCaption:
@@ -56,3 +56,39 @@ def test_add_two_caption_lists(self):
 
         with pytest.raises(ValueError):
             newcaps = self.caps + CaptionList([4], layout_info="Other Layout")
+
+
+class TestCaptionSetRemoveStyling:
+    def setup_method(self):
+        nodes = [
+            CaptionNode.create_style(True, {"italics": True}),
+            CaptionNode.create_text("hello"),
+            CaptionNode.create_style(False, {"italics": True}),
+            CaptionNode.create_break(),
+            CaptionNode.create_text("world"),
+        ]
+        self.caption = Caption(0, 1000, nodes, style={"italics": True})
+        self.caption_set = CaptionSet({"en": CaptionList([self.caption])})
+
+    def test_removes_style_nodes(self):
+        self.caption_set.remove_styling()
+
+        node_types = [node.type_ for node in self.caption.nodes]
+        assert CaptionNode.STYLE not in node_types
+
+    def test_clears_caption_style_attribute(self):
+        self.caption_set.remove_styling()
+
+        assert self.caption.style == {}
+
+    def test_preserves_text_and_break_nodes(self):
+        self.caption_set.remove_styling()
+
+        remaining = [
+            (node.type_, node.content) for node in self.caption.nodes
+        ]
+        assert remaining == [
+            (CaptionNode.TEXT, "hello"),
+            (CaptionNode.BREAK, None),
+            (CaptionNode.TEXT, "world"),
+        ]