pbs · Marcelpv96 · Jul 7, 2022
@@ -65,6 +65,8 @@
 }
 
 DFXP_DEFAULT_LANGUAGE_CODE = "en"
+DFXP_DEFAULT_FRAMERATE = "30"
+DFXP_DEFAULT_FRAMERATEMULTIPLIER = "1000 1000"
 
 
 class DFXPReader(BaseReader):
@@ -81,6 +83,7 @@ def __init__(self, *args, **kw):
         self.read_invalid_positioning = (
             kw.get('read_invalid_positioning', False))
         self.nodes = []
+        self.framerate = self._get_framerate(DFXP_DEFAULT_FRAMERATE, DFXP_DEFAULT_FRAMERATEMULTIPLIER)
 
     def detect(self, content):
         if '</tt>' in content.lower():
@@ -101,6 +104,11 @@ def read(self, content):
 
         default_language = dfxp_document.tt.attrs.get('xml:lang',
                                                       DEFAULT_LANGUAGE_CODE)
+        framerate = dfxp_document.tt.attrs.get('ttp:framerate',
+                                               DFXP_DEFAULT_FRAMERATE)
+        framerate_multiplier = dfxp_document.tt.attrs.get('ttp:frameratemultiplier',
+                                                          DFXP_DEFAULT_FRAMERATEMULTIPLIER)
+        self.framerate = self._get_framerate(framerate, framerate_multiplier)
 
         # Each div represents all the captions for a single language.
         for div in dfxp_document.find_all('div'):
@@ -169,6 +177,13 @@ def _find_and_convert_times(self, p_tag):
 
         return start, end
 
+    @staticmethod
+    def _get_framerate(framerate, framerate_multiplier):
+        numerator, denominator = framerate_multiplier.split()
+        framerate_multiplier = int(numerator) / int(denominator)
+        framerate = float(framerate) * framerate_multiplier
+        return framerate
+
     def _convert_timestamp_to_microseconds(self, stamp):
         match = TIME_EXPRESSION_PATTERN.search(stamp)
         if not match:
@@ -180,8 +195,7 @@ def _convert_timestamp_to_microseconds(self, stamp):
         else:
             return self._convert_time_count_to_microseconds(match)
 
-    @staticmethod
-    def _convert_clock_time_to_microseconds(clock_time_match):
+    def _convert_clock_time_to_microseconds(self, clock_time_match):
         microseconds = int(clock_time_match.group('hours')) * \
                        MICROSECONDS_PER_UNIT["hours"]
         microseconds += int(clock_time_match.group('minutes')) * \
@@ -192,12 +206,11 @@ def _convert_clock_time_to_microseconds(clock_time_match):
             microseconds += int(clock_time_match.group('sub_frames').ljust(
                 3, '0')) * MICROSECONDS_PER_UNIT["milliseconds"]
         elif clock_time_match.group('frames'):
-            microseconds += int(clock_time_match.group('frames')) / 30 * \
+            microseconds += int(clock_time_match.group('frames')) / self.framerate * \
                             MICROSECONDS_PER_UNIT["seconds"]
         return int(microseconds)
 
-    @staticmethod
-    def _convert_time_count_to_microseconds(time_count_match):
+    def _convert_time_count_to_microseconds(self, time_count_match):
         value = float(time_count_match.group('time_count'))
         metric = time_count_match.group("metric")
         if metric == "h":
@@ -209,7 +222,7 @@ def _convert_time_count_to_microseconds(time_count_match):
         elif metric == "ms":
             microseconds = value * MICROSECONDS_PER_UNIT["milliseconds"]
         elif metric == "f":
-            microseconds = value / 30 * MICROSECONDS_PER_UNIT["seconds"]
+            microseconds = value / self.framerate * MICROSECONDS_PER_UNIT["seconds"]
         elif metric == "t":
             raise NotImplementedError("The tick metric for time count is "
                                       "not currently implemented.")

@@ -24,7 +24,7 @@
     sample_dfxp_invalid_positioning_value_template,
     sample_dfxp_multiple_captions_with_the_same_timing,
     sample_dfxp_with_ampersand_character, sample_dfxp_with_nested_spans,
-    dfxp_style_region_align_conflict, dfxp_with_concurrent_captions,
+    dfxp_style_region_align_conflict, dfxp_with_concurrent_captions, sample_dfxp_framerate
 )
 from tests.fixtures.microdvd import (  # noqa: F401
     sample_microdvd, sample_microdvd_2,

@@ -1508,4 +1508,49 @@ def sample_dfxp_default_styling_p_tags():
    </p>
   </div>
  </body>
-</tt>"""
+</tt>"""
+
+
+@pytest.fixture(scope="session")
+def sample_dfxp_framerate():
+    return """\
+<?xml version="1.0" encoding="UTF-8" standalone="no"?>
+<tt xml:lang="en" xmlns="http://www.w3.org/ns/ttml" 
+    xmlns:tts="http://www.w3.org/ns/ttml#styling" 
+    xmlns:ttm="http://www.w3.org/ns/ttml#metadata" 
+        xmlns:ttp="http://www.w3.org/ns/ttml#parameter" 
+        xmlns:smpte="http://www.smpte-ra.org/schemas/2052-1/2010/smpte-tt" 
+        xmlns:m608="http://www.smpte-ra.org/schemas/2052-1/2010/smpte-tt#cea608" 
+        ttp:timeBase="media" ttp:frameRate="24" ttp:frameRateMultiplier="1000 1001">
+    <head>
+        <metadata>
+            <ttm:desc>SMPTE 2052 Timed Text Captions document created by MacCaption™ version 7.0.12</ttm:desc>
+            <smpte:information xmlns:m608="http://www.smpte-ra.org/schemas/2052-1/2010/smpte-tt#cea608" 
+            origin="http://www.smpte-ra.org/schemas/2052-1/2010/smpte-tt#cea608" 
+            mode="Preserved" m608:channel="CC1" m608:programName="Demo" m608:captionService="F1C1CC"/>
+        </metadata>
+        <styling>
+            <style xml:id="basic" tts:color="white" tts:backgroundColor="black" tts:showBackground="whenActive" 
+            tts:fontFamily="monospace" tts:fontSize="80%" tts:fontWeight="normal" tts:fontStyle="normal"/>
+        </styling>
+        <layout>
+            <region xml:id="pop1" tts:backgroundColor="transparent" tts:showBackground="whenActive"/>
+            <region xml:id="pop2" tts:backgroundColor="transparent" tts:showBackground="whenActive"/>
+            <region xml:id="pop3" tts:backgroundColor="transparent" tts:showBackground="whenActive"/>
+            <region xml:id="pop4" tts:backgroundColor="transparent" tts:showBackground="whenActive"/>
+        </layout>
+    </head>
+    <body>
+        <div>
+            <p begin="00:00:00:15" end="00:00:08:05" region="bottom" style="default">
+            MAN 2:<br/>
+            E equals m c-squared is<br/>
+            not about an old Einstein.
+           </p>
+           <p begin="00:00:08:05" end="00:00:09:16" region="bottom" style="default">
+            MAN 2:<br/>
+            It's all about an eternal Einstein.
+           </p>
+        </div>
+    </body>
+</tt>"""
@@ -40,6 +40,13 @@ def test_proper_timestamps(self, sample_dfxp):
         assert 17000000 == paragraph.start
         assert 18752000 == paragraph.end
 
+    def test_proper_timestamps_framerate(self, sample_dfxp_framerate):
+        captions = DFXPReader().read(sample_dfxp_framerate)
+        paragraph = captions.get_captions("en")[1]
+
+        assert 8208541 == paragraph.start
+        assert 9667333 == paragraph.end
+
     def test_incorrect_time_format(self, sample_dfxp_incorrect_time_format):
         with pytest.raises(CaptionReadTimingError) as exc_info:
             DFXPReader().read(sample_dfxp_incorrect_time_format)
@@ -71,6 +78,13 @@ def test_convert_timestamp_to_microseconds(self):
         with pytest.raises(NotImplementedError):
             reader._convert_timestamp_to_microseconds("2.3t")
 
+    def test_convert_timestamp_to_microseconds_framerate(self):
+        reader = DFXPReader()
+        reader.framerate = 24
+
+        assert 66666 == reader._convert_timestamp_to_microseconds("1.6f")
+        assert 625000 == reader._convert_timestamp_to_microseconds("00:00:00:15")
+
     @pytest.mark.parametrize('timestamp, microseconds', [
         ('12:23:34', 44614000000), ('23:34:45:56', 84886866666),
         ('34:45:56.7', 125156700000), ('13:24:35.67', 48275670000),