def _read_smil(self, input_file):
"""
Read from SMIL file.
Limitations:
1. parses only <par> elements, in order
2. timings must have hh:mm:ss.mmm or ss.mmm format (autodetected)
3. both clipBegin and clipEnd attributes of <audio> must be populated
"""
smil_ns = "{http://www.w3.org/ns/SMIL}"
contents = input_file.read()
root = etree.fromstring(contents.encode("utf-8"))
for par in root.iter(smil_ns + "par"):
for child in par:
if child.tag == (smil_ns + "text"):
identifier = gf.split_url(child.get("src"))[1]
elif child.tag == (smil_ns + "audio"):
begin = gf.time_from_hhmmssmmm(child.get("clipBegin"))
if begin is None:
begin = gf.time_from_ssmmm(child.get("clipBegin"))
end = gf.time_from_hhmmssmmm(child.get("clipEnd"))
if end is None:
end = gf.time_from_ssmmm(child.get("clipEnd"))
# TODO read text from additional text_file?
text = u""
text_fragment = TextFragment(identifier=identifier, lines=[text])
sm_fragment = SyncMapFragment(text_fragment, begin, end)
self.append(sm_fragment)
def parse(self, input_text, syncmap):
"""
Read from SMIL file.
Limitations:
1. parses only ``<par>`` elements, in order
2. timings must have ``hh:mm:ss.mmm`` or ``ss.mmm`` format (autodetected)
3. both ``clipBegin`` and ``clipEnd`` attributes of ``<audio>`` must be populated
"""
from lxml import etree
smil_ns = "{http://www.w3.org/ns/SMIL}"
root = etree.fromstring(gf.safe_bytes(input_text))
for par in root.iter(smil_ns + "par"):
for child in par:
if child.tag == (smil_ns + "text"):
identifier = gf.safe_unicode(gf.split_url(child.get("src"))[1])
elif child.tag == (smil_ns + "audio"):
begin_text = child.get("clipBegin")
if ":" in begin_text:
begin = gf.time_from_hhmmssmmm(begin_text)
else:
begin = gf.time_from_ssmmm(begin_text)
end_text = child.get("clipEnd")
if ":" in end_text:
end = gf.time_from_hhmmssmmm(end_text)
else:
end = gf.time_from_ssmmm(end_text)
# TODO read text from additional text_file?
self._add_fragment(
syncmap=syncmap,
identifier=identifier,
lines=[u""],
begin=begin,
end=end
)
def parse(self, input_text, syncmap):
contents_dict = json.loads(input_text)
for fragment in contents_dict["smil_data"]:
# TODO read text from additional text_file?
self._add_fragment(syncmap=syncmap,
identifier=fragment["id"],
lines=[u""],
begin=gf.time_from_ssmmm(fragment["begin"]),
end=gf.time_from_ssmmm(fragment["end"]))
def parse(self, input_text, syncmap):
contents_dict = json.loads(input_text)
for fragment in contents_dict["smil_data"]:
# TODO read text from additional text_file?
self._add_fragment(
syncmap=syncmap,
identifier=fragment["id"],
lines=[u""],
begin=gf.time_from_ssmmm(fragment["begin"]),
end=gf.time_from_ssmmm(fragment["end"])
)
def parse(self, input_text, syncmap):
contents_dict = json.loads(input_text)
for fragment in contents_dict["fragments"]:
self._add_fragment(
syncmap=syncmap,
identifier=fragment["id"],
language=fragment["language"],
lines=fragment["lines"],
begin=gf.time_from_ssmmm(fragment["begin"]),
end=gf.time_from_ssmmm(fragment["end"])
)
def _read_rbse(self, input_file):
"""
Read from RBSE file
"""
contents = input_file.read()
contents_dict = json.loads(contents)
for fragment in contents_dict["smil_data"]:
identifier = fragment["id"]
begin = gf.time_from_ssmmm(fragment["begin"])
end = gf.time_from_ssmmm(fragment["end"])
# TODO read text from additional text_file?
text = u""
text_fragment = TextFragment(identifier=identifier, lines=[text])
sm_fragment = SyncMapFragment(text_fragment, begin, end)
self.append(sm_fragment)
def parse(self, input_text, syncmap):
from lxml import etree
# namespaces
xsi = "http://www.w3.org/2001/XMLSchema-instance"
ns_map = {"xsi": xsi}
# get root
root = etree.fromstring(gf.safe_bytes(input_text))
# get time slots
time_slots = dict()
for ts in root.iter("TIME_SLOT"):
time_slots[ts.get("TIME_SLOT_ID")] = gf.time_from_ssmmm(
ts.get("TIME_VALUE")) / 1000
# parse annotations
for alignable in root.iter("ALIGNABLE_ANNOTATION"):
identifier = gf.safe_unicode(alignable.get("ANNOTATION_ID"))
begin = time_slots[alignable.get("TIME_SLOT_REF1")]
end = time_slots[alignable.get("TIME_SLOT_REF2")]
lines = []
for value in alignable.iter("ANNOTATION_VALUE"):
lines.append(gf.safe_unicode(value.text))
self._add_fragment(syncmap=syncmap,
identifier=identifier,
lines=lines,
begin=begin,
end=end)
def parse(self, input_text, syncmap):
from lxml import etree
# namespaces
xsi = "http://www.w3.org/2001/XMLSchema-instance"
ns_map = {"xsi": xsi}
# get root
root = etree.fromstring(gf.safe_bytes(input_text))
# get time slots
time_slots = dict()
for ts in root.iter("TIME_SLOT"):
time_slots[ts.get("TIME_SLOT_ID")] = gf.time_from_ssmmm(ts.get("TIME_VALUE")) / 1000
# parse annotations
for alignable in root.iter("ALIGNABLE_ANNOTATION"):
identifier = gf.safe_unicode(alignable.get("ANNOTATION_ID"))
begin = time_slots[alignable.get("TIME_SLOT_REF1")]
end = time_slots[alignable.get("TIME_SLOT_REF2")]
lines = []
for value in alignable.iter("ANNOTATION_VALUE"):
lines.append(gf.safe_unicode(value.text))
self._add_fragment(
syncmap=syncmap,
identifier=identifier,
lines=lines,
begin=begin,
end=end
)
def _read_xml(self, input_file):
"""
Read from XML file
"""
contents = input_file.read()
root = etree.fromstring(contents.encode("utf-8"))
for frag in root:
identifier = frag.get("id")
begin = gf.time_from_ssmmm(frag.get("begin"))
end = gf.time_from_ssmmm(frag.get("end"))
lines = []
for child in frag:
if child.tag == "line":
lines.append(child.text)
text_fragment = TextFragment(identifier=identifier, lines=lines)
sm_fragment = SyncMapFragment(text_fragment, begin, end)
self.append(sm_fragment)
def _read_json(self, input_file):
"""
Read from JSON file
"""
contents = input_file.read()
contents_dict = json.loads(contents)
for fragment in contents_dict["fragments"]:
identifier = fragment["id"]
language = fragment["language"]
begin = gf.time_from_ssmmm(fragment["begin"])
end = gf.time_from_ssmmm(fragment["end"])
lines = []
for line in fragment["lines"]:
lines.append(line)
text_fragment = TextFragment(identifier=identifier, language=language, lines=lines)
sm_fragment = SyncMapFragment(text_fragment, begin, end)
self.append(sm_fragment)
def parse(self, input_text, syncmap):
from lxml import etree
root = etree.fromstring(gf.safe_bytes(input_text))
for frag in root:
for child in frag:
if child.tag == "identifier":
identifier = gf.safe_unicode(child.text)
elif child.tag == "start":
begin = gf.time_from_ssmmm(child.text)
elif child.tag == "end":
end = gf.time_from_ssmmm(child.text)
# TODO read text from additional text_file?
self._add_fragment(syncmap=syncmap,
identifier=identifier,
lines=[u""],
begin=begin,
end=end)
def parse(self, input_text, syncmap):
from lxml import etree
root = etree.fromstring(gf.safe_bytes(input_text))
for frag in root:
identifier = gf.safe_unicode(frag.get("id"))
begin = gf.time_from_ssmmm(frag.get("begin"))
end = gf.time_from_ssmmm(frag.get("end"))
lines = []
for child in frag:
if child.tag == "line":
lines.append(gf.safe_unicode(child.text))
self._add_fragment(
syncmap=syncmap,
identifier=identifier,
lines=lines,
begin=begin,
end=end
)
def _read_xml_legacy(self, input_file):
"""
Read from XML file (legacy format)
"""
contents = input_file.read()
root = etree.fromstring(contents.encode("utf-8"))
for frag in root:
for child in frag:
if child.tag == "identifier":
identifier = child.text
elif child.tag == "start":
begin = gf.time_from_ssmmm(child.text)
elif child.tag == "end":
end = gf.time_from_ssmmm(child.text)
# TODO read text from additional text_file?
text = ""
text_fragment = TextFragment(identifier=identifier, lines=[text])
sm_fragment = SyncMapFragment(text_fragment, begin, end)
self.append(sm_fragment)
def parse(self, input_text, syncmap):
from lxml import etree
root = etree.fromstring(gf.safe_bytes(input_text))
for frag in root:
for child in frag:
if child.tag == "identifier":
identifier = gf.safe_unicode(child.text)
elif child.tag == "start":
begin = gf.time_from_ssmmm(child.text)
elif child.tag == "end":
end = gf.time_from_ssmmm(child.text)
# TODO read text from additional text_file?
self._add_fragment(
syncmap=syncmap,
identifier=identifier,
lines=[u""],
begin=begin,
end=end
)
def test_time_from_ssmmm(self):
tests = [
[None, 0],
["", 0],
["0", 0],
["000", 0],
["1", 1],
["001", 1],
["1.234", 1.234],
["001.234", 1.234],
]
for test in tests:
self.assertEqual(gf.time_from_ssmmm(test[0]), test[1])
def test_time_from_ssmmm(self):
tests = [
(None, TimeValue("0")),
("", TimeValue("0")),
("0", TimeValue("0")),
("000", TimeValue("0")),
("1", TimeValue("1")),
("001", TimeValue("1")),
("1.234", TimeValue("1.234")),
("001.234", TimeValue("1.234")),
]
for test in tests:
self.assertEqual(gf.time_from_ssmmm(test[0]), test[1])
def test_time_from_ssmmm(self):
tests = [
[None, 0],
["", 0],
["0", 0],
["000", 0],
["1", 1],
["001", 1],
["1.234", 1.234],
["001.234", 1.234],
]
for test in tests:
self.assertEqual(gf.time_from_ssmmm(test[0]), test[1])
def test_time_from_ssmmm(self):
tests = [
(None, TimeValue("0")),
("", TimeValue("0")),
("0", TimeValue("0")),
("000", TimeValue("0")),
("1", TimeValue("1")),
("001", TimeValue("1")),
("1.234", TimeValue("1.234")),
("001.234", TimeValue("1.234")),
]
for test in tests:
self.assertEqual(gf.time_from_ssmmm(test[0]), test[1])