diff --git a/docs/conf.py b/docs/conf.py index 3d8b4b2a..6e7c1281 100644 --- a/docs/conf.py +++ b/docs/conf.py @@ -53,9 +53,9 @@ # built documents. # # The short X.Y version. -version = '2.2.4' +version = '2.2.5' # The full version, including alpha/beta/rc tags. -release = '2.2.4' +release = '2.2.5' # The language for content autogenerated by Sphinx. Refer to documentation # for a list of supported languages. diff --git a/pycaption/scc/__init__.py b/pycaption/scc/__init__.py index 222c3bc6..5b15f59a 100644 --- a/pycaption/scc/__init__.py +++ b/pycaption/scc/__init__.py @@ -301,21 +301,27 @@ def _translate_line(self, line): self.time_translator.start_at(parts[0][0]) # loop through each word - for word in parts[0][2].split(' '): - # ignore empty results or invalid commands - word = word.strip() - if len(word) == 4: - self._translate_word(word) - - def _translate_word(self, word): - if self._handle_double_command(word): + words = [word.strip() for word in parts[0][2].split(' ') if len(word) == 4] + + for idx, word in enumerate(words): + self._translate_word(word, words, idx) + + @staticmethod + def get_command(commands, idx): + try: + return commands[idx] + except IndexError: + return None + + def _translate_word(self, word, words, idx): + if self._skip_double_command(word, words, idx): # count frames for timing self.time_translator.increment_frames() return # first check if word is a command # TODO - check that all the positioning commands are here, or use # some other strategy to determine if the word is a command. - if word in COMMANDS or _is_pac_command(word): + if word in COMMANDS or _is_pac_command(word) or word in PAC_TAB_OFFSET_COMMANDS: self._translate_command(word) # second, check if word is a special character @@ -332,31 +338,39 @@ def _translate_word(self, word): # count frames for timing only after processing a command self.time_translator.increment_frames() - def _handle_double_command(self, word): + def _skip_double_command(self, word, words, idx): # If the caption is to be broadcast, each of the commands are doubled # up for redundancy in case the signal is garbled in transmission. # The decoder is programmed to ignore a second command when it is the # same as the first. # Also like codes, Special Characters are always doubled up, # with only one member of each pair being displayed. + next_command = self.get_command(words, idx + 1) + second_next = self.get_command(words, idx + 2) + if word in COMMANDS or _is_pac_command(word) or word in SPECIAL_CHARS or word in EXTENDED_CHARS: - if word == self.last_command: + # skip duplicates, execute the last occurrence if not a positioning command + if word == self.last_command and not _is_pac_command(word): self.last_command = '' return True - # Fix for the - # repetition - elif _is_pac_command(word) and word in self.last_command: + # skip consecutive positioning commands, execute the last one + elif _is_pac_command(word) and _is_pac_command(next_command): self.last_command = '' return True + # Fix for the repetition + # execute the last positioning command + elif _is_pac_command(word) and next_command in PAC_TAB_OFFSET_COMMANDS and _is_pac_command(second_next): + self.last_command = '' + return True + # execute offset commands only if previous command is PAC and next is not PAC elif word in PAC_TAB_OFFSET_COMMANDS: - if _is_pac_command(self.last_command): - self.last_command += f" {word}" + if _is_pac_command(self.last_command) and not _is_pac_command(next_command): + self.last_command = word return False else: return True - - self.last_command = word - return False + self.last_command = word + return False def _translate_special_char(self, word): self.buffer.add_chars(SPECIAL_CHARS[word]) diff --git a/pycaption/scc/constants.py b/pycaption/scc/constants.py index 6aa25c45..979b603f 100644 --- a/pycaption/scc/constants.py +++ b/pycaption/scc/constants.py @@ -987,6 +987,8 @@ def _restructure_bytes_to_position_map(byte_to_pos_map): HEADER = 'Scenarist_SCC V1.0' +UNHANDLED_COMMANDS = ["9120", "91ae", "912f", "91a1"] + # taken from # http://www.theneitherworld.com/mcpoodle/SCC_TOOLS/DOCS/CC_CHARS.HTML INCONVERTIBLE_TO_ASCII_EXTENDED_CHARS_ASSOCIATION = { diff --git a/pycaption/scc/specialized_collections.py b/pycaption/scc/specialized_collections.py index 5680869e..7738ba4c 100644 --- a/pycaption/scc/specialized_collections.py +++ b/pycaption/scc/specialized_collections.py @@ -8,7 +8,8 @@ ) from .constants import ( PAC_BYTES_TO_POSITIONING_MAP, COMMANDS, PAC_TAB_OFFSET_COMMANDS, - MICROSECONDS_PER_CODEWORD, INCONVERTIBLE_TO_ASCII_EXTENDED_CHARS_ASSOCIATION + MICROSECONDS_PER_CODEWORD, UNHANDLED_COMMANDS, + INCONVERTIBLE_TO_ASCII_EXTENDED_CHARS_ASSOCIATION ) PopOnCue = collections.namedtuple("PopOnCue", "buffer, start, end") @@ -342,10 +343,9 @@ def interpret_command(self, command): :type command: str """ - self._update_positioning(command) - + if command not in UNHANDLED_COMMANDS: + self._update_positioning(command) text = COMMANDS.get(command, '') - if 'italic' in text: if 'end' not in text: self._collection.append( diff --git a/setup.py b/setup.py index 7c1c3858..7a20a670 100644 --- a/setup.py +++ b/setup.py @@ -24,7 +24,7 @@ setup( name='pycaption', - version='2.2.4', + version='2.2.5.dev', description='Closed caption converter', long_description=open(README_PATH).read(), author='Joe Norton', diff --git a/tests/conftest.py b/tests/conftest.py index 2e361fb8..26dae1e5 100644 --- a/tests/conftest.py +++ b/tests/conftest.py @@ -61,6 +61,8 @@ sample_scc_duplicate_tab_offset, sample_scc_duplicate_special_characters, sample_scc_tab_offset, sample_scc_with_unknown_commands, sample_scc_special_and_extended_characters, + sample_scc_with_consecutive_pac_commands, + sample_scc_special_and_extended_characters, sample_scc_with_line_too_long ) from tests.fixtures.srt import ( # noqa: F401 diff --git a/tests/fixtures/dfxp.py b/tests/fixtures/dfxp.py index 714e5d52..901e9f12 100644 --- a/tests/fixtures/dfxp.py +++ b/tests/fixtures/dfxp.py @@ -920,10 +920,10 @@ def sample_dfxp_from_scc_output(): - + - + diff --git a/tests/fixtures/scc.py b/tests/fixtures/scc.py index 23661d36..8c3f8d77 100644 --- a/tests/fixtures/scc.py +++ b/tests/fixtures/scc.py @@ -24,6 +24,33 @@ def sample_scc_created_dfxp_with_wrongly_closing_spans(): """ +@pytest.fixture(scope="session") +def sample_scc_with_consecutive_pac_commands(): + return """\ +Scenarist_SCC V1.0 + +00:00:00;15 942c + +00:11:45;10 9420 94d0 94ce 5b20 cec1 5252 c154 4f52 205d 9470 946e cd4f cecb 45d9 d320 4c4f d645 2054 c849 cec7 d320 54c8 c154 2046 4cd9 ae80 942c 8080 8080 942f + +00:11:47;28 9420 9454 9723 d9c1 d9a1 94f4 9723 5b20 c84f 4f54 49ce c720 5d80 942c 8080 8080 942f + +00:11:50;08 9420 94d0 94ce 45d3 d045 4349 c14c 4cd9 2049 4620 54c8 45d9 a752 4520 54c8 4520 4fce 45d3 9470 946e 57c8 4f20 c745 5420 544f 2046 4cd9 2054 c845 cdae 942c 8080 8080 942f + +00:11:54;06 942c + +00:23:00;13 9420 1370 136e 5b20 43c8 494c c420 5d80 94d0 94ce c745 4f52 c745 20cd c1c4 4520 c120 cdc1 43c8 49ce 4580 9470 946e 464f 5220 c84f 5749 4520 544f 942c 8080 8080 942f + +00:23:02;04 9420 91d0 91ce 544f 2046 49ce c420 43d5 5249 4fd5 d320 c745 4f52 c745 9170 916e c1ce c420 c849 d320 4652 4945 cec4 d380 92d0 92ce 45d6 4552 d920 c4c1 d920 4fce 4c49 ce45 2c80 942c 8080 8080 942f + +00:23:05;00 9420 9152 91ae d357 49ce c720 c2d9 20d0 c2d3 cb49 c4d3 ae4f 52c7 9170 916e 544f 20d0 4cc1 d920 46d5 ce20 c7c1 cd45 d320 c1ce c420 57c1 5443 c880 92d0 9723 91ae d94f d552 2046 c1d6 4f52 4954 4520 d649 c445 4fd3 ae80 942c 8080 8080 942f + + +00:23:05;00 9420 9152 91ae d357 49ce c720 c2d9 20d0 c2d3 cb49 c4d3 ae4f 52c7 9170 916e 544f 20d0 4cc1 d920 46d5 ce20 c7c1 cd45 d320 c1ce c420 57c1 5443 c880 92d0 9723 91ae d94f d552 2046 c1d6 4f52 4954 4520 d649 c445 4fd3 ae80 942c 8080 8080 942f + +""" + + @pytest.fixture(scope="session") def scc_that_generates_webvtt_with_proper_newlines(): return """\ @@ -435,7 +462,7 @@ def sample_scc_with_line_too_long(): 00:00:00;03 942c -00:00:01;45 9420 91f4 cb45 4c4c d920 4ac1 cd45 d3ba 20c8 eff7 9254 f468 e520 7368 eff7 2073 f461 f2f4 e564 942c 8080 8080 942f +00:00:01;45 9420 91f4 cb45 4c4c d920 4ac1 cd45 d3ba 20c8 eff7 9254 f468 e520 7368 eff7d3ba 20c8 eff7 9254 f468 e520 7368 eff7 eff7 eff7 eff7 eff7 eff7 eff7 eff7 eff7 eff7 eff7 eff7 eff7 eff7 eff7 eff7 eff7 eff7 eff7 eff7 eff7 eff7 eff7 eff7 eff7 eff7 eff7 2073 f461 f2f4 e564 942c 8080 8080 942f 00:00:02;55 9420 91e0 9723 f761 7320 4361 ec20 ec20 ec20 ec20 ec20 ec20 ec20 ec20 ec20 ec20 ec20 ec20 ec20 ec20 ec20 ec20 ec20 ec20 ec20 ec20 ec20 ec20 ec20 ec20 ec20 ec20 ec20 ec20 ec20 c4e5 6ee9 73ef 6e2c 2061 20e6 f2e9 e56e 6480 9240 9723 efe6 20ef 75f2 732c 20f7 6173 2064 efe9 6e67 206d 7920 43c4 73ae 942c 8080 8080 942f diff --git a/tests/test_scc.py b/tests/test_scc.py index f5d6698b..396affe2 100644 --- a/tests/test_scc.py +++ b/tests/test_scc.py @@ -77,10 +77,10 @@ def test_positioning(self, sample_scc_multiple_positioning): ((40.0, UnitEnum.PERCENT), (53.0, UnitEnum.PERCENT)), ((70.0, UnitEnum.PERCENT), (17.0, UnitEnum.PERCENT)), ((20.0, UnitEnum.PERCENT), (35.0, UnitEnum.PERCENT)), - ((20.0, UnitEnum.PERCENT), (83.0, UnitEnum.PERCENT)), + ((25.0, UnitEnum.PERCENT), (83.0, UnitEnum.PERCENT)), ((70.0, UnitEnum.PERCENT), (11.0, UnitEnum.PERCENT)), ((40.0, UnitEnum.PERCENT), (41.0, UnitEnum.PERCENT)), - ((20.0, UnitEnum.PERCENT), (71.0, UnitEnum.PERCENT)) + ((25.0, UnitEnum.PERCENT), (71.0, UnitEnum.PERCENT)) ] actual_positioning = [ @@ -244,6 +244,38 @@ def test_flashing_cue(self, sample_scc_flashing_cue): assert exc_info.value.args[0].startswith( "Unsupported cue duration around 00:00:20.433") + def test_skip_first_pac_command(self, sample_scc_with_consecutive_pac_commands): + caption_set = SCCReader().read(sample_scc_with_consecutive_pac_commands) + caption = caption_set.get_captions('en-US') + actual_lines = [ + node.content + for cap_ in caption + for node in cap_.nodes + if node.type_ == CaptionNode.TEXT + ] + expected_lines = [ + '[ NARRATOR ]', + 'MONKEYS LOVE THINGS THAT FLY.', + 'YAY!', + '[ HOOTING ]', + "ESPECIALLY IF THEY'RE THE ONES", + 'WHO GET TO FLY THEM.', + '[ CHILD ]', + 'GEORGE MADE A MACHINE', + 'FOR HOWIE TO', + 'TO FIND CURIOUS GEORGE', + 'AND HIS FRIENDS', + 'EVERY DAY ONLINE,', + 'SWING BY PBSKIDS.ORG', + 'TO PLAY FUN GAMES AND WATCH', + 'YOUR FAVORITE VIDEOS.', + 'SWING BY PBSKIDS.ORG', + 'TO PLAY FUN GAMES AND WATCH', + 'YOUR FAVORITE VIDEOS.' + ] + # is not breaking the lines + assert expected_lines == actual_lines + def test_line_too_long(self, sample_scc_with_line_too_long): with pytest.raises(CaptionLineLengthError) as exc_info: SCCReader().read(sample_scc_with_line_too_long) @@ -345,7 +377,6 @@ def test_freeze_semicolon_spec_time(self, sample_scc_roll_up_ru2): ] actual_timings = [(c_.start, c_.end) for c_ in captions] - assert expected_timings == actual_timings def test_freeze_colon_spec_time(self, sample_scc_pop_on): @@ -584,4 +615,4 @@ def test_eoc_first_command(self, sample_scc_eoc_first_command): # just one caption, first EOC disappears num_captions = len(caption_set.get_captions('en-US')) - assert num_captions == 2 + assert num_captions == 1