srt_reader: hide extended formatting tags behind a configuration flag

#490

Author: palemieux

README.md

@@ -201,6 +201,18 @@ Default: `23`
 
 Default: `true`
 
+### SRT Reader configuration (`"srt_reader"`)
+
+#### extended_tags
+
+`"extended_tag" : true | false`
+
+If `true`, the following extended formatting tags are supported: `{bold}`,
+`<bold>`, `{b}`, `{italic}`, `<italic>`, `{i}`, `{underline}`, `<underline>` and
+`{u}`.
+
+Default: `false`
+
 ### VTT Writer configuration (`"vtt_writer"`)
 
 #### line_position

src/main/python/ttconv/srt/config.py

@@ -30,6 +30,17 @@
 from dataclasses import dataclass, field
 from ttconv.config import ModuleConfiguration
 
+@dataclass
+class SRTReaderConfiguration(ModuleConfiguration):
+  """SRT reader configuration"""
+
+  @classmethod
+  def name(cls):
+    return "srt_reader"
+
+  # enables support for extended tags: {b}, {bold}, <bold> (and italic/underline equivalents)
+  extended_tags: bool = field(default=False, metadata={"decoder": bool})
+
 @dataclass
 class SRTWriterConfiguration(ModuleConfiguration):
   """SRT writer configuration"""

src/main/python/ttconv/srt/reader.py

@@ -36,6 +36,7 @@
 from ttconv import model
 from ttconv import style_properties as styles
 from ttconv.utils import parse_color
+from ttconv.srt.config import SRTReaderConfiguration
 
 LOGGER = logging.getLogger(__name__)
 
@@ -61,11 +62,11 @@ def handle_starttag(self, tag, attrs):
     self.parent.push_child(span)
     self.parent = span
 
-    if tag.lower() in ("b", "bold"):
+    if tag.lower() in ("b"):
       span.set_style(styles.StyleProperties.FontWeight, styles.FontWeightType.bold)
-    elif tag.lower() in ("i", "italic"):
+    elif tag.lower() in ("i"):
       span.set_style(styles.StyleProperties.FontStyle, styles.FontStyleType.italic)
-    elif tag.lower() in ("u", "underline"):
+    elif tag.lower() in ("u"):
       span.set_style(styles.StyleProperties.TextDecoration, styles.TextDecorationType(underline=True))
     elif tag.lower() == "font":
       for attr in attrs:
@@ -116,9 +117,11 @@ class _State(Enum):
 _DEFAULT_OUTLINE_COLOR = styles.NamedColors.black.value
 _DEFAULT_LINE_HEIGHT = styles.LengthType(125, styles.LengthType.Units.pct)
 
-def to_model(data_file: typing.IO, _config = None, progress_callback=lambda _: None):
+def to_model(data_file: typing.IO, _config: SRTReaderConfiguration = None, progress_callback=lambda _: None):
   """Converts an SRT document to the data model"""
 
+  extended_tags = _config.extended_tags if isinstance(_config, SRTReaderConfiguration) else False
+
   doc = model.ContentDocument()
 
   region = model.Region(_DEFAULT_REGION_ID, doc)
@@ -236,14 +239,28 @@ def to_model(data_file: typing.IO, _config = None, progress_callback=lambda _: N
     if state in (_State.TEXT, _State.TEXT_MORE):
 
       if line is None or _EMPTY_RE.fullmatch(line):
-        subtitle_text = subtitle_text.strip('\r\n')\
-          .replace(r"\n\r", "\n")\
-          .replace(r"{bold}", r"<bold>")\
-          .replace(r"{/bold}", r"</bold>")\
-          .replace(r"{italic}", r"<italic>")\
-          .replace(r"{/italic}", r"</italic>")\
-          .replace(r"{underline}", r"<underline>")\
-          .replace(r"{/underline}", r"</underline>")
+        subtitle_text = subtitle_text.strip('\r\n').replace(r"\n\r", "\n")
+
+        if extended_tags:
+          subtitle_text = subtitle_text\
+            .replace(r"{b}", r"<b>")\
+            .replace(r"{/b}", r"</b>")\
+            .replace(r"{bold}", r"<b>")\
+            .replace(r"{/bold}", r"</b>")\
+            .replace(r"<bold>", r"<b>")\
+            .replace(r"</bold>", r"</b>")\
+            .replace(r"{i}", r"<i>")\
+            .replace(r"{/i}", r"</i>")\
+            .replace(r"{italic}", r"<i>")\
+            .replace(r"{/italic}", r"</i>")\
+            .replace(r"<italic>", r"<i>")\
+            .replace(r"</italic>", r"</i>")\
+            .replace(r"{u}", r"<u>")\
+            .replace(r"{/u}", r"</u>")\
+            .replace(r"{underline}", r"<u>")\
+            .replace(r"{/underline}", r"</u>")\
+            .replace(r"<underline>", r"<u>")\
+            .replace(r"</underline>", r"</u>")
 
         parser = _TextParser(current_p, line_index)
         parser.feed(subtitle_text)

src/main/python/ttconv/tt.py

@@ -52,15 +52,16 @@
 from ttconv.isd import ISDConfiguration
 from ttconv.scc.config import SccReaderConfiguration, SccWriterConfiguration
 from ttconv.stl.config import STLReaderConfiguration
-from ttconv.srt.config import SRTWriterConfiguration
+from ttconv.srt.config import SRTReaderConfiguration, SRTWriterConfiguration
 
 LOGGER = logging.getLogger("ttconv")
 
 CONFIGURATIONS = [
   GeneralConfiguration,
   IMSCWriterConfiguration,
   ISDConfiguration,
-  SccReaderConfiguration
+  SccReaderConfiguration,
+  SRTReaderConfiguration,
 ]
 
 
@@ -335,12 +336,16 @@ def convert(args):
       model = stl_reader.to_model(f, reader_config, progress_callback_read)
 
   elif reader_type is FileTypes.SRT:
+    #
+    # Read the config
+    #
+    reader_config = read_config_from_json(SRTReaderConfiguration, json_config_data)
 
     #
     # Open the file and pass it to the reader
     #
     with open(inputfile, "r", encoding="utf-8") as f:
-      model = srt_reader.to_model(f, None, progress_callback_read)
+      model = srt_reader.to_model(f, reader_config, progress_callback_read)
 
   elif reader_type is FileTypes.VTT:
 

src/test/python/test_srt_reader.py

@@ -29,6 +29,7 @@
 import unittest
 import io
 
+from ttconv.srt.config import SRTReaderConfiguration
 from ttconv.srt.reader import to_model
 import ttconv.style_properties as styles
 import ttconv.model as model
@@ -67,7 +68,7 @@ def test_sample(self):
   def test_bold(self):
     f = io.StringIO(r"""1
 00:02:16,612 --> 00:02:19,376
-Hello <bold>my</bold> name is Bob
+Hello <b>my</b> name is Bob
 """)
     doc = to_model(f)
     for e in doc.get_body().dfs_iterator():
@@ -107,17 +108,53 @@ def test_bold_alt(self):
 00:02:16,612 --> 00:02:19,376
 Hello {bold}my{/bold} name is Bob
 """)
+    doc = to_model(f, SRTReaderConfiguration(extended_tags=True))
+    for e in doc.get_body().dfs_iterator():
+      if e.get_style(styles.StyleProperties.FontWeight) == styles.FontWeightType.bold:
+        break
+    else:
+      self.fail()
+    doc = to_model(f)
+    for e in doc.get_body().dfs_iterator():
+      if e.get_style(styles.StyleProperties.FontWeight) == styles.FontWeightType.bold:
+        self.fail()
+  
+  def test_bold_alt2(self):
+    f = io.StringIO(r"""1
+00:02:16,612 --> 00:02:19,376
+Hello <bold>my</bold> name is Bob
+""")
+    doc = to_model(f, SRTReaderConfiguration(extended_tags=True))
+    for e in doc.get_body().dfs_iterator():
+      if e.get_style(styles.StyleProperties.FontWeight) == styles.FontWeightType.bold:
+        break
+    else:
+      self.fail()
     doc = to_model(f)
+    for e in doc.get_body().dfs_iterator():
+      if e.get_style(styles.StyleProperties.FontWeight) == styles.FontWeightType.bold:
+        self.fail()
+
+  def test_bold_alt3(self):
+    f = io.StringIO(r"""1
+00:02:16,612 --> 00:02:19,376
+Hello {b}my{/b} name is Bob
+""")
+    doc = to_model(f, SRTReaderConfiguration(extended_tags=True))
     for e in doc.get_body().dfs_iterator():
       if e.get_style(styles.StyleProperties.FontWeight) == styles.FontWeightType.bold:
         break
     else:
       self.fail()
+    doc = to_model(f)
+    for e in doc.get_body().dfs_iterator():
+      if e.get_style(styles.StyleProperties.FontWeight) == styles.FontWeightType.bold:
+        self.fail()
 
   def test_italic(self):
     f = io.StringIO(r"""1
 00:02:16,612 --> 00:02:19,376
-Hello <italic>my</italic> name is Bob
+Hello <i>my</i> name is Bob
 """)
     doc = to_model(f)
     for e in doc.get_body().dfs_iterator():
@@ -131,17 +168,53 @@ def test_italic_alt(self):
 00:02:16,612 --> 00:02:19,376
 Hello {italic}my{/italic} name is Bob
 """)
+    doc = to_model(f, SRTReaderConfiguration(extended_tags=True))
+    for e in doc.get_body().dfs_iterator():
+      if e.get_style(styles.StyleProperties.FontStyle) == styles.FontStyleType.italic:
+        break
+    else:
+      self.fail()
+    doc = to_model(f)
+    for e in doc.get_body().dfs_iterator():
+      if e.get_style(styles.StyleProperties.FontStyle) == styles.FontStyleType.italic:
+        self.fail()
+
+  def test_italic_alt1(self):
+    f = io.StringIO(r"""1
+00:02:16,612 --> 00:02:19,376
+Hello {i}my{/i} name is Bob
+""")
+    doc = to_model(f, SRTReaderConfiguration(extended_tags=True))
+    for e in doc.get_body().dfs_iterator():
+      if e.get_style(styles.StyleProperties.FontStyle) == styles.FontStyleType.italic:
+        break
+    else:
+      self.fail()
     doc = to_model(f)
+    for e in doc.get_body().dfs_iterator():
+      if e.get_style(styles.StyleProperties.FontStyle) == styles.FontStyleType.italic:
+        self.fail()
+
+  def test_italic_alt2(self):
+    f = io.StringIO(r"""1
+00:02:16,612 --> 00:02:19,376
+Hello <italic>my</italic> name is Bob
+""")
+    doc = to_model(f, SRTReaderConfiguration(extended_tags=True))
     for e in doc.get_body().dfs_iterator():
       if e.get_style(styles.StyleProperties.FontStyle) == styles.FontStyleType.italic:
         break
     else:
       self.fail()
+    doc = to_model(f)
+    for e in doc.get_body().dfs_iterator():
+      if e.get_style(styles.StyleProperties.FontStyle) == styles.FontStyleType.italic:
+        self.fail()
 
   def test_underline(self):
     f = io.StringIO(r"""1
 00:02:16,612 --> 00:02:19,376
-Hello <underline>my</underline> name is Bob
+Hello <u>my</u> name is Bob
 """)
     doc = to_model(f)
     for e in doc.get_body().dfs_iterator():
@@ -156,13 +229,54 @@ def test_underline_alt(self):
 00:02:16,612 --> 00:02:19,376
 Hello {underline}my{/underline} name is Bob
 """)
+    doc = to_model(f, SRTReaderConfiguration(extended_tags=True))
+    for e in doc.get_body().dfs_iterator():
+      text_decoration = e.get_style(styles.StyleProperties.TextDecoration)
+      if text_decoration is not None and text_decoration.underline:
+        break
+    else:
+      self.fail()
+    doc = to_model(f)
+    for e in doc.get_body().dfs_iterator():
+      text_decoration = e.get_style(styles.StyleProperties.TextDecoration)
+      if text_decoration is not None and text_decoration.underline:
+        self.fail()
+
+  def test_underline_alt1(self):
+    f = io.StringIO(r"""1
+00:02:16,612 --> 00:02:19,376
+Hello {u}my{/u} name is Bob
+""")
+    doc = to_model(f, SRTReaderConfiguration(extended_tags=True))
+    for e in doc.get_body().dfs_iterator():
+      text_decoration = e.get_style(styles.StyleProperties.TextDecoration)
+      if text_decoration is not None and text_decoration.underline:
+        break
+    else:
+      self.fail()
     doc = to_model(f)
+    for e in doc.get_body().dfs_iterator():
+      text_decoration = e.get_style(styles.StyleProperties.TextDecoration)
+      if text_decoration is not None and text_decoration.underline:
+        self.fail()
+
+  def test_underline_alt2(self):
+    f = io.StringIO(r"""1
+00:02:16,612 --> 00:02:19,376
+Hello <underline>my</underline> name is Bob
+""")
+    doc = to_model(f, SRTReaderConfiguration(extended_tags=True))
     for e in doc.get_body().dfs_iterator():
       text_decoration = e.get_style(styles.StyleProperties.TextDecoration)
       if text_decoration is not None and text_decoration.underline:
         break
     else:
       self.fail()
+    doc = to_model(f)
+    for e in doc.get_body().dfs_iterator():
+      text_decoration = e.get_style(styles.StyleProperties.TextDecoration)
+      if text_decoration is not None and text_decoration.underline:
+        self.fail()
 
   def test_blue(self):
     f = io.StringIO(r"""1
@@ -180,8 +294,8 @@ def test_blue(self):
   def test_multiline_tags(self):
     f = io.StringIO(r"""1
 00:02:16,612 --> 00:02:19,376
-Hello <bold>my
-</bold> name is Bob
+Hello <b>my
+</b> name is Bob
 """)
     doc = to_model(f)
     for e in doc.get_body().dfs_iterator():

src/test/python/test_tt.py

@@ -230,6 +230,26 @@ def test_lcd_filter(self):
       '--filter', 'lcd',
       '--config', '{"lcd": {"bg_color":"red"}}'
       ])
+    
+  def test_srt_reader_extended_tags(self):
+    in_path = "src/test/resources/srt/extended-tags.srt"
+
+    out_path = "build/extended_tags.ttml"
+    tt.main(['convert',
+      '-i', in_path,
+      '-o', out_path,
+      '--config', '{"srt_reader": {"extended_tags": true}}'
+      ])
+    with open(out_path, encoding="utf-8") as f:
+      self.assertRegex(f.read(), "fontWeight")
+    
+    out_path = "build/no-extended_tags.ttml"
+    tt.main(['convert',
+      '-i', in_path,
+      '-o', out_path
+      ])
+    with open(out_path, encoding="utf-8") as f:
+      self.assertNotRegex(f.read(), "fontWeight")
 
   def test_imsc11filter(self):
     out_path = "build/imsc11filter.ttml"

src/test/resources/srt/extended-tags.srt

@@ -0,0 +1,3 @@
+1
+00:02:16,612 --> 00:02:19,376
+Hello <bold>my</bold> name is Bob