pycaption 2.3.3__tar.gz → 2.3.5__tar.gz

{pycaption-2.3.3/pycaption.egg-info → pycaption-2.3.5}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: pycaption
-Version: 2.3.3
+Version: 2.3.5
 Summary: Closed caption converter
 Author: Joe Norton
 Author-email: joey@nortoncrew.com

{pycaption-2.3.3 → pycaption-2.3.5}/pycaption/base.py

@@ -5,12 +5,15 @@ CaptionSet -> CaptionList -> Caption -> CaptionNode.  Also provides the
 CaptionConverter orchestrator and base classes for readers/writers.
 """
 
+import logging
 import os
 from datetime import timedelta
 from numbers import Number
 
 from .exceptions import CaptionReadError, CaptionReadTimingError, InvalidInputError
 
+logger = logging.getLogger(__name__)
+
 # `und` a special identifier for an undetermined language according to ISO 639-2
 DEFAULT_LANGUAGE_CODE = os.getenv("PYCAPTION_DEFAULT_LANG", "und")
 
@@ -79,9 +82,7 @@ class BaseReader:
             try:
                 content = content.decode("utf-8-sig")
             except UnicodeDecodeError as e:
-                raise InvalidInputError(
-                    f"Content is not valid UTF-8: {e}"
-                ) from e
+                raise InvalidInputError(f"Content is not valid UTF-8: {e}") from e
         elif isinstance(content, str):
             if content.startswith(""):
                 content = content[1:]
@@ -89,8 +90,23 @@ class BaseReader:
             raise InvalidInputError(
                 "The content must be a unicode string or UTF-8 bytes."
             )
+        content = BaseReader._repair_double_encoding(content)
         return content
 
+    @staticmethod
+    def _repair_double_encoding(text):
+        """Fix double-encoded UTF-8 (bytes misread as CP-1252)."""
+        try:
+            repaired = text.encode("cp1252").decode("utf-8")
+        except (UnicodeEncodeError, UnicodeDecodeError):
+            return text
+        if repaired != text:
+            logger.warning(
+                "Detected and repaired double-encoded UTF-8 in caption content"
+            )
+            return repaired
+        return text
+
     def detect(self, content):
         """Return True if content appears to be in this reader's format.
 
@@ -396,7 +412,11 @@ class CaptionSet:
     """
 
     def __init__(
-        self, captions, styles=None, layout_info=None, regions=None,
+        self,
+        captions,
+        styles=None,
+        layout_info=None,
+        regions=None,
         visual_alignment_default=None,
     ):
         """

{pycaption-2.3.3 → pycaption-2.3.5}/pycaption/dfxp/constants.py

@@ -30,7 +30,7 @@ DFXP_DEFAULT_REGION = Layout(
     alignment=Alignment(HorizontalAlignmentEnum.START, VerticalAlignmentEnum.BOTTOM)
 )
 
-# Writer fallback alignment used when layout is None or SCC positional.
+# Writer fallback alignment used when layout is None.
 DFXP_WRITER_FALLBACK_ALIGNMENT = Alignment(
     HorizontalAlignmentEnum.CENTER, VerticalAlignmentEnum.BOTTOM
 )

{pycaption-2.3.3 → pycaption-2.3.5}/pycaption/dfxp/writer.py

@@ -294,8 +294,7 @@ class RegionCreator:
     def _collect_unique_regions(caption_set, ignore_region):
         """Collect all unique Layout objects from the caption set.
 
-        Excludes None, ignore_region (typically the default region), and
-        SCC positional layouts (which map to the default center region).
+        Excludes None and ignore_region (typically the default region).
 
         :type caption_set: CaptionSet
         :param ignore_region: a Layout to exclude from the result
@@ -315,9 +314,6 @@ class RegionCreator:
 
         unique_regions.pop(None, None)
         unique_regions.pop(ignore_region, None)
-        for layout in list(unique_regions):
-            if layout and layout.is_positional_anchor:
-                unique_regions.pop(layout)
         return unique_regions
 
     def _create_unique_regions(self, unique_layouts, dfxp, id_factory):
@@ -432,9 +428,6 @@ class RegionCreator:
             if not layout_info:
                 layout_info = caption_set.layout_info
 
-        if layout_info and layout_info.is_positional_anchor:
-            layout_info = None
-
         region_id = self._region_map.get(layout_info)
         if not region_id:
             region_id = DFXP_DEFAULT_REGION_ID
@@ -506,8 +499,8 @@ def _convert_layout_to_attributes(layout, fallback_alignment=None):
     Maps origin, extent, padding, alignment, and writing_direction to their
     tts: namespace equivalents.
 
-    When layout is None or carries a positional anchor (SCC row/col coords),
-    uses fallback_alignment if provided (per RP 2052-10 center promotion).
+    When layout is None, uses fallback_alignment if provided
+    (per RP 2052-10 center promotion).
 
     :type layout: Layout | None
     :param fallback_alignment: Alignment to use when the layout lacks one.
@@ -521,14 +514,6 @@ def _convert_layout_to_attributes(layout, fallback_alignment=None):
             return _create_external_alignment(fallback_alignment)
         return result
 
-    if layout.is_positional_anchor:
-        if fallback_alignment:
-            result.update(_create_external_alignment(fallback_alignment))
-        writing_mode = _WRITING_DIRECTION_TO_DFXP.get(layout.writing_direction)
-        if writing_mode:
-            result["tts:writingMode"] = writing_mode
-        return result
-
     if layout.origin:
         result["tts:origin"] = layout.origin.to_xml_attribute()
 

{pycaption-2.3.3 → pycaption-2.3.5}/pycaption/geometry.py

@@ -597,7 +597,6 @@ class Layout:
         alignment=None,
         webvtt_positioning=None,
         writing_direction=None,
-        is_positional_anchor=False,
         inherit_from=None,
     ):
         """
@@ -623,11 +622,6 @@ class Layout:
         :type writing_direction: WritingDirectionEnum
         :param writing_direction: WebVTT vertical writing direction (rl or lr).
 
-        :type is_positional_anchor: bool
-        :param is_positional_anchor: True when the origin/alignment describe a
-            coordinate-system anchor point (e.g. SCC row/col positioning) rather
-            than visual text alignment.
-
         :type inherit_from: Layout
         :param inherit_from: A Layout with the positioning parameters to be
             used if not specified by the positioning arguments,
@@ -639,7 +633,6 @@ class Layout:
         self.alignment = alignment
         self.webvtt_positioning = webvtt_positioning
         self.writing_direction = writing_direction
-        self.is_positional_anchor = is_positional_anchor
 
         if inherit_from:
             for attr_name in [
@@ -716,7 +709,6 @@ class Layout:
         params = {
             "alignment": self.alignment,
             "writing_direction": self.writing_direction,
-            "is_positional_anchor": self.is_positional_anchor,
         }
         for attr_name in ["origin", "extent", "padding"]:
             attr = getattr(self, attr_name)
@@ -751,7 +743,6 @@ class Layout:
             padding=self.padding,
             alignment=self.alignment,
             writing_direction=self.writing_direction,
-            is_positional_anchor=self.is_positional_anchor,
         )
 
     def _corrected_extent(self, diff_horizontal, diff_vertical):

{pycaption-2.3.3 → pycaption-2.3.5}/pycaption/sami/writer.py

@@ -131,9 +131,6 @@ class SAMIWriter(BaseWriter):
         if not layout_info:
             return
 
-        if layout_info.is_positional_anchor:
-            return
-
         if layout_info.origin:
             if layout_info.origin.x:
                 parts.append(f"margin-left:{layout_info.origin.x};")
@@ -152,9 +149,6 @@ class SAMIWriter(BaseWriter):
         if not layout_info:
             return "center" if self._promote_center else None
 
-        if layout_info.is_positional_anchor:
-            return "center" if self._promote_center else None
-
         if not layout_info.alignment:
             return "center" if self._promote_center else None
 

{pycaption-2.3.3 → pycaption-2.3.5}/pycaption/scc/specialized_collections.py

@@ -632,7 +632,6 @@ def _get_layout_from_tuple(position_tuple):
     return Layout(
         origin=Point(horizontal, vertical),
         alignment=Alignment(HorizontalAlignmentEnum.LEFT, VerticalAlignmentEnum.TOP),
-        is_positional_anchor=True,
     )
 
 

{pycaption-2.3.3 → pycaption-2.3.5}/pycaption/webvtt/writer.py

@@ -344,9 +344,6 @@ class WebVTTWriter(BaseWriter):
         if not layout:
             return ""
 
-        if layout.is_positional_anchor:
-            return ""
-
         if layout.webvtt_positioning:
             return f" {layout.webvtt_positioning}"
 

{pycaption-2.3.3 → pycaption-2.3.5/pycaption.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: pycaption
-Version: 2.3.3
+Version: 2.3.5
 Summary: Closed caption converter
 Author: Joe Norton
 Author-email: joey@nortoncrew.com

{pycaption-2.3.3 → pycaption-2.3.5}/pycaption.egg-info/SOURCES.txt

@@ -41,9 +41,11 @@ tests/__init__.py
 tests/conftest.py
 tests/mixins.py
 tests/test_base.py
+tests/test_bytes_input.py
 tests/test_dfxp.py
 tests/test_dfxp_conversion.py
 tests/test_dfxp_extras.py
+tests/test_double_encoding.py
 tests/test_functions.py
 tests/test_geometry.py
 tests/test_microdvd.py

{pycaption-2.3.3 → pycaption-2.3.5}/setup.py

@@ -20,7 +20,7 @@ transcript_dependencies = ["nltk==3.10.0"]
 
 setup(
     name="pycaption",
-    version="2.3.3",
+    version="2.3.5",
     description="Closed caption converter",
     long_description=open(README_PATH).read(),
     author="Joe Norton",

{pycaption-2.3.3 → pycaption-2.3.5}/tests/fixtures/webvtt.py

@@ -317,7 +317,7 @@ def sample_webvtt_from_scc_properly_writes_newlines_output():
     return """\
 WEBVTT
 
-00:21:30.000 --> 00:21:34.000
+00:21:30.000 --> 00:21:34.000 align:left position:20% line:83% size:70%
 aa
 bb
 """

pycaption-2.3.5/tests/test_bytes_input.py

@@ -0,0 +1,161 @@
+"""Tests for bytes input support across all readers.
+
+All readers should accept bytes (raw file content) and decode as UTF-8
+internally, preventing the double-encoding gibberish that occurs when
+callers decode with the wrong system encoding (e.g., cp1252).
+"""
+
+from pycaption import (
+    DFXPReader,
+    MicroDVDReader,
+    SAMIReader,
+    SCCReader,
+    SRTReader,
+    WebVTTReader,
+)
+
+
+class TestSRTReaderBytes:
+    def setup_class(self):
+        self.reader = SRTReader()
+
+    def test_read_bytes_produces_same_result_as_str(self, sample_srt):
+        from_str = self.reader.read(sample_srt)
+        from_bytes = self.reader.read(sample_srt.encode("utf-8"))
+        str_captions = from_str.get_captions("en-US")
+        bytes_captions = from_bytes.get_captions("en-US")
+        assert len(str_captions) == len(bytes_captions)
+        for s, b in zip(str_captions, bytes_captions):
+            assert s.get_text() == b.get_text()
+            assert s.start == b.start
+            assert s.end == b.end
+
+    def test_read_bytes_preserves_music_notes(self, sample_srt):
+        captions = self.reader.read(sample_srt.encode("utf-8"))
+        texts = [c.get_text() for c in captions.get_captions("en-US")]
+        assert any("♪" in t for t in texts)
+
+    def test_read_bytes_with_bom(self, sample_srt):
+        content_with_bom = b"\xef\xbb\xbf" + sample_srt.encode("utf-8")
+        captions = self.reader.read(content_with_bom)
+        assert len(captions.get_captions("en-US")) == 7
+
+    def test_detect_bytes(self, sample_srt):
+        assert self.reader.detect(sample_srt.encode("utf-8")) is True
+
+
+class TestWebVTTReaderBytes:
+    def setup_class(self):
+        self.reader = WebVTTReader()
+
+    def test_read_bytes_produces_same_result_as_str(self, sample_webvtt):
+        from_str = self.reader.read(sample_webvtt)
+        from_bytes = self.reader.read(sample_webvtt.encode("utf-8"))
+        str_captions = from_str.get_captions("en-US")
+        bytes_captions = from_bytes.get_captions("en-US")
+        assert len(str_captions) == len(bytes_captions)
+        for s, b in zip(str_captions, bytes_captions):
+            assert s.get_text() == b.get_text()
+
+    def test_read_bytes_with_bom(self, sample_webvtt):
+        content_with_bom = b"\xef\xbb\xbf" + sample_webvtt.encode("utf-8")
+        captions = self.reader.read(content_with_bom)
+        assert len(captions.get_captions("en-US")) > 0
+
+    def test_detect_bytes(self, sample_webvtt):
+        assert self.reader.detect(sample_webvtt.encode("utf-8")) is True
+
+
+class TestDFXPReaderBytes:
+    def setup_class(self):
+        self.reader = DFXPReader()
+
+    def test_read_bytes_produces_same_result_as_str(self, sample_dfxp):
+        from_str = self.reader.read(sample_dfxp)
+        from_bytes = self.reader.read(sample_dfxp.encode("utf-8"))
+        for lang in from_str.get_languages():
+            str_captions = from_str.get_captions(lang)
+            bytes_captions = from_bytes.get_captions(lang)
+            assert len(str_captions) == len(bytes_captions)
+            for s, b in zip(str_captions, bytes_captions):
+                assert s.get_text() == b.get_text()
+
+    def test_read_bytes_with_bom(self, sample_dfxp):
+        content_with_bom = b"\xef\xbb\xbf" + sample_dfxp.encode("utf-8")
+        captions = self.reader.read(content_with_bom)
+        assert not captions.is_empty()
+
+    def test_detect_bytes(self, sample_dfxp):
+        assert self.reader.detect(sample_dfxp.encode("utf-8")) is True
+
+
+class TestSAMIReaderBytes:
+    def setup_class(self):
+        self.reader = SAMIReader()
+
+    def test_read_bytes_produces_same_result_as_str(self, sample_sami):
+        from_str = self.reader.read(sample_sami)
+        from_bytes = self.reader.read(sample_sami.encode("utf-8"))
+        for lang in from_str.get_languages():
+            str_captions = from_str.get_captions(lang)
+            bytes_captions = from_bytes.get_captions(lang)
+            assert len(str_captions) == len(bytes_captions)
+            for s, b in zip(str_captions, bytes_captions):
+                assert s.get_text() == b.get_text()
+
+    def test_read_bytes_preserves_music_notes(self, sample_sami):
+        captions = self.reader.read(sample_sami.encode("utf-8"))
+        langs = list(captions.get_languages())
+        texts = [c.get_text() for c in captions.get_captions(langs[0])]
+        assert any("♪" in t for t in texts)
+
+    def test_read_bytes_with_bom(self, sample_sami):
+        content_with_bom = b"\xef\xbb\xbf" + sample_sami.encode("utf-8")
+        captions = self.reader.read(content_with_bom)
+        assert not captions.is_empty()
+
+    def test_detect_bytes(self, sample_sami):
+        assert self.reader.detect(sample_sami.encode("utf-8")) is True
+
+
+class TestSCCReaderBytes:
+    def test_read_bytes_produces_same_result_as_str(self, sample_scc_pop_on):
+        from_str = SCCReader().read(sample_scc_pop_on)
+        from_bytes = SCCReader().read(sample_scc_pop_on.encode("utf-8"))
+        str_captions = from_str.get_captions("en-US")
+        bytes_captions = from_bytes.get_captions("en-US")
+        assert len(str_captions) == len(bytes_captions)
+        for s, b in zip(str_captions, bytes_captions):
+            assert s.get_text() == b.get_text()
+
+    def test_read_bytes_with_bom(self, sample_scc_pop_on):
+        content_with_bom = b"\xef\xbb\xbf" + sample_scc_pop_on.encode("utf-8")
+        captions = SCCReader().read(content_with_bom)
+        assert len(captions.get_captions("en-US")) > 0
+
+    def test_detect_bytes(self, sample_scc_pop_on):
+        assert SCCReader().detect(sample_scc_pop_on.encode("utf-8")) is True
+
+
+class TestMicroDVDReaderBytes:
+    def setup_class(self):
+        self.reader = MicroDVDReader()
+
+    def test_read_bytes_produces_same_result_as_str(self, sample_microdvd):
+        from_str = self.reader.read(sample_microdvd)
+        from_bytes = self.reader.read(sample_microdvd.encode("utf-8"))
+        str_captions = from_str.get_captions("und")
+        bytes_captions = from_bytes.get_captions("und")
+        assert len(str_captions) == len(bytes_captions)
+        for s, b in zip(str_captions, bytes_captions):
+            assert s.get_text() == b.get_text()
+
+    def test_read_bytes_with_bom(self, sample_microdvd):
+        content_with_bom = b"\xef\xbb\xbf" + sample_microdvd.encode("utf-8")
+        captions = self.reader.read(content_with_bom)
+        assert not captions.is_empty()
+
+    def test_detect_bytes(self, sample_microdvd):
+        assert self.reader.detect(sample_microdvd.encode("utf-8")) is True
+
+

pycaption-2.3.5/tests/test_double_encoding.py

@@ -0,0 +1,89 @@
+import logging
+
+import pytest
+
+from pycaption import SAMIReader, SRTReader
+from pycaption.base import BaseReader
+from pycaption.dfxp import DFXPReader
+
+
+def _double_encode(text):
+    """Simulate double-encoding: UTF-8 bytes misread as CP-1252, re-encoded."""
+    return text.encode("utf-8").decode("cp1252")
+
+
+ORIGINAL_CHARS = ["♪", "—", "’", "é"]
+
+
+class TestRepairDoubleEncoding:
+    @pytest.mark.parametrize("original", ORIGINAL_CHARS)
+    def test_fixes_double_encoded(self, original):
+        garbled = _double_encode(original)
+        assert BaseReader._repair_double_encoding(garbled) == original
+
+    def test_leaves_clean_utf8_alone(self):
+        clean = "♪ This is — perfectly fine é text"
+        assert BaseReader._repair_double_encoding(clean) == clean
+
+    def test_logs_warning_on_repair(self, caplog):
+        garbled = _double_encode("♪")
+        with caplog.at_level(logging.WARNING, logger="pycaption.base"):
+            BaseReader._repair_double_encoding(garbled)
+        assert "double-encoded" in caplog.text.lower()
+
+    def test_no_warning_for_clean_input(self, caplog):
+        with caplog.at_level(logging.WARNING, logger="pycaption.base"):
+            BaseReader._repair_double_encoding("♪ Music ♪")
+        assert caplog.text == ""
+
+
+class TestDoubleEncodingEndToEnd:
+    def test_srt_reader(self):
+        garbled_note = _double_encode("♪")
+        content = (
+            "1\n"
+            "00:00:01,000 --> 00:00:02,000\n"
+            f"{garbled_note} Music {garbled_note}\n"
+        )
+        captions = SRTReader().read(content)
+        nodes = captions.get_captions("en-US")[0].nodes
+        text = "".join(n.content for n in nodes)
+        assert "♪" in text
+        assert garbled_note not in text
+
+    def test_dfxp_reader(self):
+        garbled = _double_encode("élève")
+        content = (
+            '<?xml version="1.0" encoding="UTF-8"?>\n'
+            '<tt xml:lang="en" xmlns="http://www.w3.org/ns/ttml">\n'
+            "  <body><div>\n"
+            f'    <p begin="00:00:01.000" end="00:00:02.000">'
+            f"{garbled}</p>\n"
+            "  </div></body>\n"
+            "</tt>\n"
+        )
+        captions = DFXPReader().read(content)
+        nodes = captions.get_captions("en")[0].nodes
+        text = "".join(n.content for n in nodes)
+        assert "élève" in text
+        assert garbled not in text
+
+    def test_sami_reader(self):
+        garbled_dash = _double_encode("—")
+        content = (
+            "<SAMI>\n"
+            "<Body>\n"
+            "  <Sync Start=1000>\n"
+            f"    <P Class=enCC>{garbled_dash} Hello</P>\n"
+            "  </Sync>\n"
+            "  <Sync Start=2000>\n"
+            "    <P Class=enCC>&nbsp;</P>\n"
+            "  </Sync>\n"
+            "</Body>\n"
+            "</SAMI>\n"
+        )
+        captions = SAMIReader().read(content)
+        lang = list(captions.get_languages())[0]
+        nodes = captions.get_captions(lang)[0].nodes
+        text = "".join(n.content for n in nodes)
+        assert "—" in text

{pycaption-2.3.3 → pycaption-2.3.5}/tests/test_sami_conversion.py

@@ -211,10 +211,10 @@ class TestWebVTTtoSAMIStyles(SAMITestingMixIn):
 
 
 class TestSCCtoSAMI(SAMITestingMixIn):
-    def test_scc_to_sami_center_aligned(self, sample_scc_pop_on):
+    def test_scc_to_sami_preserves_positioning(self, sample_scc_pop_on):
         caption_set = SCCReader().read(sample_scc_pop_on)
         result = SAMIWriter().write(caption_set)
 
-        assert "text-align:center;" in result
-        assert "margin-left" not in result
-        assert "margin-top" not in result
+        assert "text-align:left;" in result
+        assert "margin-left:" in result
+        assert "margin-top:" in result

{pycaption-2.3.3 → pycaption-2.3.5}/tests/test_scc_conversion.py

@@ -1,7 +1,5 @@
 import re
 
-import pytest
-
 from pycaption import (
     DFXPWriter,
     SCCReader,
@@ -42,9 +40,8 @@ class TestSCCtoDFXP:
         caption_set = SCCReader().read(sample_scc_multiple_positioning)
         dfxp = DFXPWriter(relativize=False, fit_to_screen=False).write(caption_set)
 
-        assert 'tts:textAlign="center"' in dfxp
-        assert 'tts:textAlign="left"' not in dfxp
-        assert "tts:origin" not in dfxp
+        assert 'tts:textAlign="left"' in dfxp
+        assert "tts:origin" in dfxp
         assert "abab" in dfxp
         assert "ghgh" in dfxp
 
@@ -58,8 +55,7 @@ class TestSCCtoDFXP:
 
         dfxp = DFXPWriter().write(caption_set)
 
-        assert 'tts:textAlign="center"' in dfxp
-        assert 'tts:textAlign="left"' not in dfxp
+        assert 'tts:textAlign="left"' in dfxp
         assert 'tts:fontStyle="italic"' in dfxp
         from bs4 import BeautifulSoup
 
@@ -72,7 +68,7 @@ class TestSCCtoDFXP:
 
         dfxp = DFXPWriter().write(caption_set)
 
-        assert 'tts:textAlign="center"' in dfxp
+        assert 'tts:textAlign="left"' in dfxp
         assert "&" in dfxp
         from bs4 import BeautifulSoup