def set_bad_body(self):
d = model.Document()
b = model.Body()
with self.assertRaises(ValueError):
d.set_body(b)
def test_tts_writing_extent_when_body_has_extents(self):
doc = model.ContentDocument()
body = model.Body(doc)
div = model.Div(doc)
div.set_style(
styles.StyleProperties.Extent,
get_extent_from_dimensions(123, 456, styles.LengthType.Units.px))
p = model.P(doc)
span = model.Span(doc)
text = model.Text(doc)
text.set_text("asdf")
span.push_child(text)
p.push_child(span)
div.push_child(p)
body.push_child(div)
doc.set_body(body)
r = model.Region("hello", doc)
r.set_style(
styles.StyleProperties.Extent,
get_extent_from_dimensions(123, 456, styles.LengthType.Units.px))
doc.put_region(r)
tree_from_model = imsc_writer.from_model(doc)
extent = tree_from_model.getroot().attrib.get(
f"{{{imsc_styles.StyleProperties.Extent.ns}}}{imsc_styles.StyleProperties.Extent.local_name}"
)
self.assertEqual(extent, '1920px 1080px')
def test_push_child(self):
b = model.Body()
b.push_child(model.Div())
with self.assertRaises(TypeError):
b.push_child(model.P())
def from_xml( parent_ctx: typing.Optional[TTMLElement.ParsingContext], xml_elem: et.Element ) -> typing.Optional[BodyElement.ParsingContext]: body_ctx = BodyElement.ParsingContext(BodyElement, parent_ctx, model.Body(parent_ctx.doc)) body_ctx.process(parent_ctx, xml_elem) return body_ctx
def test_default_region(self):
doc = model.ContentDocument()
b = model.Body(doc)
doc.set_body(b)
div1 = model.Div(doc)
b.push_child(div1)
p1 = model.P(doc)
div1.push_child(p1)
span1 = model.Span(doc)
span1.push_child(model.Text(doc, "hello"))
p1.push_child(span1)
isd = ISD.from_model(doc, 0)
self.assertEqual(len(isd), 1)
regions = list(isd.iter_regions())
self.assertEqual(regions[0].get_id(), ISD.DEFAULT_REGION_ID)
p = regions[0][0][0][0]
self.assertEqual(len(p), 1)
self.assertEqual(p[0][0].get_text(), "hello")
def setUp(self):
self.doc = model.ContentDocument()
r1 = model.Region("r1", self.doc)
r1.set_style(styles.StyleProperties.ShowBackground, styles.ShowBackgroundType.whenActive)
self.doc.put_region(r1)
b = model.Body(self.doc)
b.set_region(r1)
self.doc.set_body(b)
div1 = model.Div(self.doc)
b.push_child(div1)
p1 = model.P(self.doc)
p1.set_begin(1)
p1.set_end(3)
div1.push_child(p1)
span1 = model.Span(self.doc)
span1.push_child(model.Text(self.doc, "hello"))
p1.push_child(span1)
span2 = model.Span(self.doc)
span2.set_begin(1)
span2.push_child(model.Text(self.doc, "bye"))
p1.push_child(span2)
def test_compute_style_property(self):
doc = model.ContentDocument()
r1 = model.Region("r1", doc)
r1.set_style(styles.StyleProperties.FontSize, styles.LengthType(value=50, units=styles.LengthType.Units.pct))
doc.put_region(r1)
b = model.Body(doc)
b.set_style(styles.StyleProperties.FontSize, styles.LengthType(value=50, units=styles.LengthType.Units.pct))
b.set_region(r1)
doc.set_body(b)
div1 = model.Div(doc)
b.push_child(div1)
p1 = model.P(doc)
div1.push_child(p1)
span1 = model.Span(doc)
p1.push_child(span1)
t1 = model.Text(doc, "hello")
span1.push_child(t1)
isd = ISD.from_model(doc, 0)
region = list(isd.iter_regions())[0]
span = region[0][0][0][0]
fs: styles.LengthType = span.get_style(styles.StyleProperties.FontSize)
self.assertAlmostEqual(fs.value, 25 / doc.get_cell_resolution().rows)
self.assertEqual(fs.units, styles.LengthType.Units.rh)
def process(context, inherited_space, inherited_lang, ttml_element):
element = model.Body(context.doc)
# process attributes
element.set_space(
XMLSpaceAttribute.extract(ttml_element) or inherited_space)
element.set_lang(
XMLLangAttribute.extract(ttml_element) or inherited_lang)
ContentElement.process_region_property(context, ttml_element, element)
ContentElement.process_style_properties(context, ttml_element, element)
# process children elements
for ttml_child_element in ttml_element:
child_element = ContentElement.process(context,
element.get_space(),
element.get_lang(),
ttml_child_element)
if child_element is not None:
if not isinstance(child_element, model.Div):
LOGGER.error("Children of body must be div instances")
else:
element.push_child(child_element)
return element
def set_body(self):
d = model.Document()
b = model.Body(d)
d.set_body(b)
self.assertEqual(d.get_body(), b)
d.set_body(None)
self.assertIsNone(d.get_body())
def test_text_decoration_inheritance(self):
doc = model.ContentDocument()
r1 = model.Region("r1", doc)
r1.set_style(
styles.StyleProperties.TextDecoration,
styles.TextDecorationType(
line_through=False,
underline=True,
overline=True
)
)
doc.put_region(r1)
b = model.Body(doc)
b.set_style(
styles.StyleProperties.TextDecoration,
styles.TextDecorationType(
overline=False
)
)
b.set_region(r1)
doc.set_body(b)
div1 = model.Div(doc)
b.push_child(div1)
p1 = model.P(doc)
div1.push_child(p1)
span1 = model.Span(doc)
p1.push_child(span1)
t1 = model.Text(doc, "hello")
span1.push_child(t1)
isd = ISD.from_model(doc, 0)
region = list(isd.iter_regions())[0]
span = region[0][0][0][0]
self.assertEqual(
span.get_style(styles.StyleProperties.TextDecoration),
styles.TextDecorationType(
line_through=False,
underline=True,
overline=False
)
)
def test_body_only(self):
doc = model.ContentDocument()
body = model.Body(doc)
div = model.Div(doc)
p = model.P(doc)
span = model.Span(doc)
text = model.Text(doc)
text.set_text("asdf")
span.push_child(text)
p.push_child(span)
div.push_child(p)
body.push_child(div)
doc.set_body(body)
# write the document out to a file
imsc_writer.from_model(doc).write('build/BodyElement.out.ttml',
encoding='utf-8',
xml_declaration=True)
def test_tts_writing_no_extent_when_body_has_no_extents(self):
doc = model.ContentDocument()
body = model.Body(doc)
div = model.Div(doc)
p = model.P(doc)
span = model.Span(doc)
text = model.Text(doc)
text.set_text("asdf")
span.push_child(text)
p.push_child(span)
div.push_child(p)
body.push_child(div)
doc.set_body(body)
tree_from_model = imsc_writer.from_model(doc)
extent = tree_from_model.getroot().attrib.get(
f"{{{imsc_styles.StyleProperties.Extent.ns}}}{imsc_styles.StyleProperties.Extent.local_name}"
)
self.assertEqual(extent, None)
def to_model(data_file: typing.IO, _config = None, progress_callback=lambda _: None):
"""Converts an SRT document to the data model"""
doc = model.ContentDocument()
region = model.Region(_DEFAULT_REGION_ID, doc)
region.set_style(
styles.StyleProperties.Origin,
styles.CoordinateType(
x=styles.LengthType(5, styles.LengthType.Units.pct),
y=styles.LengthType(5, styles.LengthType.Units.pct)
)
)
region.set_style(
styles.StyleProperties.Extent,
styles.ExtentType(
height=styles.LengthType(90, styles.LengthType.Units.pct),
width=styles.LengthType(90, styles.LengthType.Units.pct)
)
)
region.set_style(
styles.StyleProperties.DisplayAlign,
styles.DisplayAlignType.after
)
region.set_style(
styles.StyleProperties.TextAlign,
styles.TextAlignType.center
)
region.set_style(
styles.StyleProperties.LineHeight,
_DEFAULT_LINE_HEIGHT
)
region.set_style(
styles.StyleProperties.FontFamily,
_DEFAULT_FONT_STACK
)
region.set_style(
styles.StyleProperties.FontSize,
_DEFAULT_FONT_SIZE
)
region.set_style(
styles.StyleProperties.Color,
_DEFAULT_TEXT_COLOR
)
region.set_style(
styles.StyleProperties.TextOutline,
styles.TextOutlineType(
_DEFAULT_OUTLINE_THICKNESS,
_DEFAULT_OUTLINE_COLOR
)
)
doc.put_region(region)
body = model.Body(doc)
body.set_region(region)
doc.set_body(body)
div = model.Div(doc)
body.push_child(div)
lines : str = data_file.readlines()
state = _State.COUNTER
current_p = None
for line_index, line in enumerate(_none_terminated(lines)):
if state is _State.COUNTER:
if line is None:
break
if _EMPTY_RE.fullmatch(line):
continue
if _COUNTER_RE.search(line) is None:
LOGGER.fatal("Missing subtitle counter at line %s", line_index)
return None
progress_callback(line_index/len(lines))
state = _State.TC
continue
if state is _State.TC:
if line is None:
break
m = _TIMECODE_RE.search(line)
if m is None:
LOGGER.fatal("Missing timecode at line %s", line_index)
return None
current_p = model.P(doc)
current_p.set_begin(
int(m.group('begin_h')) * 3600 +
int(m.group('begin_m')) * 60 +
int(m.group('begin_s')) +
int(m.group('begin_ms')) / 1000
)
current_p.set_end(
int(m.group('end_h')) * 3600 +
int(m.group('end_m')) * 60 +
int(m.group('end_s')) +
int(m.group('end_ms')) / 1000
)
state = _State.TEXT
continue
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>")
parser = _TextParser(current_p, line_index)
parser.feed(subtitle_text)
parser.close()
state = _State.COUNTER
continue
if state is _State.TEXT:
div.push_child(current_p)
subtitle_text = ""
if state is _State.TEXT_MORE:
current_p.push_child(model.Br(current_p.get_doc()))
subtitle_text += line
state = _State.TEXT_MORE
continue
return doc
def setUp(self):
self.doc = model.ContentDocument()
a1 = model.DiscreteAnimationStep(
style_property=styles.StyleProperties.Color,
begin=None,
end=None,
value=styles.NamedColors.red.value
)
a2 = model.DiscreteAnimationStep(
style_property=styles.StyleProperties.Color,
begin=Fraction(1),
end=None,
value=styles.NamedColors.green.value
)
a3 = model.DiscreteAnimationStep(
style_property=styles.StyleProperties.Color,
begin=Fraction(2),
end=None,
value=styles.NamedColors.blue.value
)
r1 = model.Region("r1", self.doc)
self.doc.put_region(r1)
# r2: sig times = {2, 9}
r2 = model.Region("r2", self.doc)
r2.set_begin(Fraction(2))
r2.set_end(Fraction(9))
r2.add_animation_step(a1)
self.doc.put_region(r2)
# b: sig times = {1, 10}
b = model.Body(self.doc)
b.set_begin(Fraction(1))
b.set_end(Fraction(10))
self.doc.set_body(b)
# div1: offset = 1, sig times = {2, 4}
div1 = model.Div(self.doc)
div1.add_animation_step(a2)
div1.set_begin(Fraction(3))
div1.set_region(r1)
b.push_child(div1)
# div2: offset = 1, sig times = {10}
div2 = model.Div(self.doc)
div2.set_end(Fraction(12))
div2.set_region(r2)
b.push_child(div2)
# p1: offset = 1, sig times = {}
p1 = model.P(self.doc)
div2.push_child(p1)
# span1: offset = 1, sig times = {3}
span1 = model.Span(self.doc)
span1.add_animation_step(a3)
p1.push_child(span1)
t1 = model.Text(self.doc, "hello")
span1.push_child(t1)
def __init__(self,
gsi_block: bytes,
disable_fill_line_gap: bool = False,
disable_line_padding: bool = False,
start_tc: typing.Optional[str] = None,
font_stack: typing.Tuple[typing.Union[
str, styles.GenericFontFamilyType]] = None,
max_row_count: typing.Optional[typing.Union[int,
str]] = None):
self.gsi = _GSIBlock._make(
struct.unpack(
'3s8sc2s2s32s32s32s32s32s32s16s6s6s2s5s5s3s2s2s1s8s8s1s1s3s32s32s32s75x576s',
gsi_block))
self.doc = model.ContentDocument()
self.doc.set_cell_resolution(
model.CellResolutionType(
columns=round(100 * DEFAULT_TELETEXT_COLS /
(100 - 2 * DEFAULT_HORIZONTAL_SAFE_MARGIN_PCT)),
rows=round(100 * DEFAULT_TELETEXT_ROWS /
(100 - 2 * DEFAULT_VERTICAL_SAFE_MARGIN_PCT))))
self.doc.set_active_area(
model.ActiveAreaType(
left_offset=DEFAULT_HORIZONTAL_SAFE_MARGIN_PCT / 100,
top_offset=DEFAULT_VERTICAL_SAFE_MARGIN_PCT / 100,
width=1 - 2 * DEFAULT_HORIZONTAL_SAFE_MARGIN_PCT / 100,
height=1 - 2 * DEFAULT_VERTICAL_SAFE_MARGIN_PCT / 100))
self.body = model.Body(self.doc)
if not disable_fill_line_gap:
self.body.set_style(styles.StyleProperties.FillLineGap, True)
if not disable_line_padding:
self.body.set_style(
styles.StyleProperties.LinePadding,
styles.LengthType(LINE_PADDING_LENGTH_C,
styles.LengthType.Units.c))
if font_stack is not None:
self.body.set_style(styles.StyleProperties.FontFamily, font_stack)
else:
self.body.set_style(styles.StyleProperties.FontFamily,
DEFAULT_FONT_STACK)
self.doc.set_body(self.body)
self.sgn_to_div_map = {}
self.last_sn = None
self.is_in_extension = False
self.tti_tf = None
self.fps = _DFC_FRACTION_MAP.get(self.gsi.DFC)
if self.fps is None:
LOGGER.error("Unknown GSI DFC value %s, defaulting to 25 fps",
self.gsi.DFC)
self.fps = Fraction(25)
else:
LOGGER.debug("GSI DFC: %s", self.gsi.DFC)
self.cct = self.gsi.CCT
LOGGER.debug("GSI CCT: %s", self.gsi.CCT)
try:
self.tti_count = int(self.gsi.TNB)
LOGGER.debug("GSI TNB: %s", self.gsi.TNB)
except ValueError:
LOGGER.error("Invalid TNB field value: %s", self.gsi.TNB)
self.tti_count = sys.maxsize
self.language = _LC_BCP47_MAP.get(self.gsi.LC)
if self.language is None:
LOGGER.warning(
"Unknown LC value: %s, defaulting to 'unspecified''",
self.gsi.LC)
self.language = ""
else:
LOGGER.debug("GSI LC: %s", self.gsi.LC)
self.doc.set_lang(self.language)
if start_tc is None:
self.start_offset = 0
elif start_tc == "TCP":
try:
self.start_offset = SmpteTimeCode(
int(self.gsi.TCP[0:2]), int(self.gsi.TCP[2:4]),
int(self.gsi.TCP[4:6]), int(self.gsi.TCP[6:8]),
self.get_fps()).to_temporal_offset()
LOGGER.debug("GSI TCP: %s", self.gsi.TCP)
except ValueError:
LOGGER.error("Invalid TCP value: %s", self.gsi.tcp)
self.start_offset = 0
else:
try:
self.start_offset = SmpteTimeCode.parse(
start_tc, self.get_fps()).to_temporal_offset()
except ValueError:
LOGGER.error("Invalid start_tc value")
raise
if max_row_count is None or self.is_teletext():
self.max_row_count = DEFAULT_TELETEXT_ROWS
elif isinstance(max_row_count, str) and max_row_count == "MNR":
try:
self.max_row_count = int(self.gsi.MNR)
LOGGER.debug("GSI MNR: %s", self.gsi.MNR)
except ValueError:
LOGGER.error("Invalid MNR value: %s", self.gsi.MNR)
self.start_offset = DEFAULT_TELETEXT_ROWS
else:
self.max_row_count = max_row_count
# p_element for use across cumulative subtitles
self.cur_p_element = None
