pycaption 2.2.13__tar.gz → 2.2.15__tar.gz

{pycaption-2.2.13/pycaption.egg-info → pycaption-2.2.15}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: pycaption
-Version: 2.2.13
+Version: 2.2.15
 Summary: Closed caption converter
 Author: Joe Norton
 Author-email: joey@nortoncrew.com

{pycaption-2.2.13 → pycaption-2.2.15}/pycaption/scc/__init__.py

@@ -367,13 +367,12 @@ class SCCReader(BaseReader):
         # doubled special characters and doubled extended characters
         # with only one member of each pair being displayed.
 
-        doubled_types = word != "94a1" and word in COMMANDS or _is_pac_command(word)
+        doubled_types = (word != "94a1" and word in COMMANDS) or _is_pac_command(word) or word in SPECIAL_CHARS
         if self.double_starter:
             doubled_types = (
                 doubled_types
                 or word in EXTENDED_CHARS
                 or word == "94a1"
-                or word in SPECIAL_CHARS
             )
 
         if word in CUE_STARTING_COMMAND and word != self.last_command:

{pycaption-2.2.13 → pycaption-2.2.15}/pycaption/scc/constants.py

@@ -988,7 +988,7 @@ HEADER = 'Scenarist_SCC V1.0'
 
 # taken from
 # http://www.theneitherworld.com/mcpoodle/SCC_TOOLS/DOCS/CC_CHARS.HTML
-INCONVERTIBLE_TO_ASCII_EXTENDED_CHARS_ASSOCIATION = {
+SUBSTITUTE_CHARACTERS = {
     '¡': ["!", "i"],   # inverted exclamation mark
      '¤': ["C"],  # currency
      '¥': ["Y"],  # yen

{pycaption-2.2.13 → pycaption-2.2.15}/pycaption/scc/specialized_collections.py

@@ -424,7 +424,7 @@ class InstructionNodeCreator:
                 # command sets a different style (underline, plain)
                 # so we need to close italics if we have an open italics tag
                 # otherwise we ignore it
-                # if break is required,add style tag then break
+                # if break is required, add style tag then break
                 if self.last_style == "italics on":
                     self._collection.append(
                         _InstructionNode.create_italics_style(
@@ -446,7 +446,7 @@ class InstructionNodeCreator:
         prev_text_node = self.get_previous_text_node()
         prev_node_is_break = prev_text_node is not None and any(
             x.is_explicit_break()
-            for x in self._collection[self._collection.index(prev_text_node) :]
+            for x in self._collection[self._collection.index(prev_text_node):]
         )
         if (
             command in MID_ROW_CODES
@@ -471,10 +471,12 @@ class InstructionNodeCreator:
 
         :type command: str
         """
+        is_offset = False
         if command in PAC_TAB_OFFSET_COMMANDS:
             prev_positioning = self._position_tracer.default
             tab_offset = PAC_TAB_OFFSET_COMMANDS[command]
             positioning = (prev_positioning[0], prev_positioning[1] + tab_offset)
+            is_offset = True
         else:
             first, second = command[:2], command[2:]
             try:
@@ -483,11 +485,25 @@ class InstructionNodeCreator:
             except KeyError:
                 # if not PAC or OFFSET we're not changing position
                 return
-        self._position_tracer.update_positioning(positioning)
+        offset_after_break = is_offset and self.has_break_before(self._collection)
+        if not offset_after_break:
+            # Tab offsets after line breaks will be ignored to avoid repositioning
+            self._position_tracer.update_positioning(positioning)
 
     def __iter__(self):
         return iter(_format_italics(self._collection))
 
+    @staticmethod
+    def has_break_before(collection):
+        if len(collection) == 0:
+            return False
+        for element in collection[::-1]:
+            if element._type == 0:
+                return False
+            elif element._type == 1:
+                return True
+        return False
+
     @classmethod
     def from_list(cls, stash_list, position_tracker):
         """Having received a list of instances of this class, creates a new

{pycaption-2.2.13 → pycaption-2.2.15}/pycaption/scc/state_machines.py

@@ -42,12 +42,15 @@ class _PositioningTracker:
         is_tab_offset = new_row == row and col + 1 <= new_col <= col + 3
         # One line below will be treated as line break, not repositioning
         if new_row == row + 1:
-            self._positions.append((new_row, new_col))
+            self._positions.append((new_row, col))
             self._break_required = True
             self._last_column = new_col
         # Tab offsets after line breaks will be ignored to avoid repositioning
         elif self._break_required and is_tab_offset:
             return
+        # force not to reposition on the same coordinates
+        elif positioning == current:
+            return
         else:
             # Reset the "current" position altogether.
             self._positions = [positioning]

{pycaption-2.2.13 → pycaption-2.2.15}/pycaption/webvtt.py

@@ -394,7 +394,6 @@ class WebVTTWriter(BaseWriter):
             return []
 
         current_layout = None
-        current_node = None
 
         # A list with layout groups. Since WebVTT only support positioning
         # for different cues, each layout group has to be represented in a
@@ -408,18 +407,12 @@ class WebVTTWriter(BaseWriter):
                 if s and current_layout and node.layout_info != current_layout:
                     # If the positioning changes from one text node to
                     # another, a new WebVTT cue has to be created.
-                    row, column = node.position if node.position else (0, 0)
-                    prev_row, prev_column = (
-                        current_node.position if current_node.position else (0, 0)
-                    )
-                    if row != prev_row + 1:
-                        layout_groups.append((s, current_layout))
-                        s = ""
+                    layout_groups.append((s, current_layout))
+                    s = ""
                 # ATTENTION: This is where the plain unicode node content is
                 # finally encoded as WebVTT.
                 s += self._encode_illegal_characters(node.content) or " "
                 current_layout = node.layout_info
-                current_node = node
             elif node.type_ == CaptionNode.STYLE:
                 resulting_style = self._calculate_resulting_style(
                     node.content, caption_set

{pycaption-2.2.13 → pycaption-2.2.15/pycaption.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: pycaption
-Version: 2.2.13
+Version: 2.2.15
 Summary: Closed caption converter
 Author: Joe Norton
 Author-email: joey@nortoncrew.com

{pycaption-2.2.13 → pycaption-2.2.15}/setup.py

@@ -20,7 +20,7 @@ transcript_dependencies = ["nltk==3.8.0"]
 
 setup(
     name="pycaption",
-    version="2.2.13",
+    version="2.2.15",
     description="Closed caption converter",
     long_description=open(README_PATH).read(),
     author="Joe Norton",

{pycaption-2.2.13 → pycaption-2.2.15}/tests/fixtures/scc.py

@@ -144,6 +144,14 @@ Scenarist_SCC V1.0
 
 00:00:13;07    9425 9425 94ad 94ad 9470 9470 c1c2 c3c4 c580 91bf
 
+00:00:13;15    9425 9425 94ad 94ad 9470 9470 9137 
+
+00:00:13;20    9425 9425 94ad 94ad 9470 9470 9137 9137
+
+00:00:13;25    9425 9425 94ad 94ad 9470 9470 9137 9137 9137
+
+00:00:13;40    9425 9425 94ad 94ad 9470 9470 9137 9137 9137 9137
+
 00:00:14;07    9425 9425 94ad 94ad 9470 9470 9220 9220 92a1 92a2 92a7
 
 00:00:17;01    9426 9426 94ad 94ad 9470 9470 57c8 4552 4520 d94f d5a7 5245 20d3 54c1 cec4 49ce c720 ce4f 572c

{pycaption-2.2.13 → pycaption-2.2.15}/tests/test_scc.py

@@ -419,8 +419,12 @@ class TestCoverageOnly:
             "AND IMPROVING THE LIVES OF ALL",
             "WE SERVE.",
             "®°½",
-            "®°½½",
+            "®°½",
             "ABû",
+            '♪',
+            '♪',
+            '♪♪',
+            '♪♪',
             "ÁÉÓ¡",
             "WHERE YOU'RE STANDING NOW,",
             "LOOKING OUT THERE, THAT'S AL",
@@ -430,6 +434,8 @@ class TestCoverageOnly:
             "And wildlife.",
             ">> Bike Iowa, your source for",
         ]
+        for x in actual_texts:
+            print(x)
         assert expected_texts == actual_texts
 
     def test_multiple_formats(self, sample_scc_multiple_formats):