def test_clip_for_element(self):
tree = cElementTree.parse(FCP7_XML_EXAMPLE_PATH)
# Use clipitem-3 because it's self-contained and doesn't reference
# other elements
sequence_elem = tree.find(".//clipitem[@id='clipitem-3']/../../../..")
clip_elem = tree.find(".//clipitem[@id='clipitem-3']")
context = self.adapter._Context(sequence_elem)
# Make a parser
parser = self.adapter.FCP7XMLParser(tree)
clip, time_range = parser.item_and_timing_for_element(
clip_elem,
head_transition=None,
tail_transition=None,
context=context,
)
self.assertEqual(clip.name, "sc01_sh020_anim.mov")
expected_range = opentime.TimeRange(
start_time=opentime.RationalTime(165, 30),
duration=opentime.RationalTime(157, 30),
)
self.assertEqual(time_range, expected_range)
expected_range = opentime.TimeRange(
start_time=opentime.RationalTime(0, 30),
duration=opentime.RationalTime(157, 30),
)
self.assertEqual(clip.source_range, expected_range)
def test_transition_cut_point(self):
transition_element = cElementTree.fromstring("""
<transitionitem>
<start>538</start>
<end>557</end>
<alignment>end-black</alignment>
<cutPointTicks>160876800000</cutPointTicks>
<rate>
<timebase>30</timebase>
<ntsc>FALSE</ntsc>
</rate>
<effect>
<name>Cross Dissolve</name>
<effectid>Cross Dissolve</effectid>
<effectcategory>Dissolve</effectcategory>
<effecttype>transition</effecttype>
<mediatype>video</mediatype>
<wipecode>0</wipecode>
<wipeaccuracy>100</wipeaccuracy>
<startratio>0</startratio>
<endratio>1</endratio>
<reverse>FALSE</reverse>
</effect>
</transitionitem>
""")
alignment_element = transition_element.find("./alignment")
track_element = cElementTree.fromstring("""
<track>
<rate>
<timebase>30</timebase>
<ntsc>FALSE</ntsc>
</rate>
</track>
""")
context = self.adapter._Context(track_element)
cut_point = self.adapter._transition_cut_point(transition_element,
context)
self.assertEqual(cut_point, opentime.RationalTime(557, 30))
alignment_element.text = "end-black"
cut_point = self.adapter._transition_cut_point(transition_element,
context)
self.assertEqual(cut_point, opentime.RationalTime(557, 30))
for alignment in ("start", "start-black"):
alignment_element.text = alignment
cut_point = self.adapter._transition_cut_point(
transition_element, context)
self.assertEqual(cut_point, opentime.RationalTime(538, 30))
# TODO: Mathematically, this cut point falls at 547.5, is the rounding
# down behavior "correct"?
alignment_element.text = "center"
cut_point = self.adapter._transition_cut_point(transition_element,
context)
self.assertEqual(cut_point, opentime.RationalTime(547, 30))
def test_timeline_for_sequence(self):
tree = cElementTree.parse(FCP7_XML_EXAMPLE_PATH)
# Get the test sequence and pare out the track definitions to keep this
# test simple.
seq_elem = tree.find("sequence")
seq_elem.find("./media").clear()
seq_elem.find("./timecode/string").text = "01:00:00:00"
seq_elem.find("./timecode/frame").text = "108000"
parser = self.adapter.FCP7XMLParser(tree)
context = self.adapter._Context()
timeline = parser.timeline_for_sequence(seq_elem, context)
# Spot-check the sequence
self.assertEqual(timeline.name, "sc01_sh010_layerA")
self.assertEqual(timeline.global_start_time,
opentime.RationalTime(108000, 30))
# Spot check that metadata translated with a tag and a property
adapter_metadata = timeline.metadata["fcp_xml"]
self.assertEqual(adapter_metadata["labels"]["label2"], "Forest")
self.assertEqual(
adapter_metadata["@MZ.Sequence.VideoTimeDisplayFormat"], "104")
# make sure the media and name tags were not included in the metadata
for k in {"name", "media"}:
with self.assertRaises(KeyError):
adapter_metadata[k]
def test_missing_media_reference_from_element(self):
file_element = cElementTree.fromstring("""
<file id="101_021_0030_FG01">
<name>101_021_0030_FG01</name>
<duration>155</duration>
<rate>
<ntsc>FALSE</ntsc>
<timebase>24</timebase>
</rate>
<timecode>
<rate>
<ntsc>FALSE</ntsc>
<timebase>24</timebase>
</rate>
<frame>1308828</frame>
<displayformat>NDF</displayformat>
<string>15:08:54:12</string>
<reel>
<name>A173C021_181204_R207</name>
</reel>
</timecode>
</file>
""")
parser = self.adapter.FCP7XMLParser(file_element)
context = self.adapter._Context()
ref = parser.media_reference_for_file_element(
file_element,
context=context,
)
self.assertTrue(isinstance(ref, schema.MissingReference))
self.assertEqual(ref.name, "101_021_0030_FG01")
self.assertEqual(
ref.available_range,
opentime.TimeRange(
start_time=opentime.RationalTime(1308828, 24),
duration=opentime.RationalTime(155, 24),
))
# Spot-check a metadata field
reelname = ref.metadata["fcp_xml"]["timecode"]["reel"]["name"]
self.assertEqual(reelname, "A173C021_181204_R207")
def test_build_empty_file(self):
media_ref = schema.MissingReference(
name="test_clip_name",
available_range=opentime.TimeRange(
opentime.RationalTime(820489, 24),
opentime.RationalTime(2087, 24),
),
metadata={
"fcp_xml": {
"timecode": {
"rate": {
"ntsc": "FALSE",
"timebase": "24"
},
"displayformat": "NDF",
"reel": {
"name": "test_reel_name",
},
}
}
},
)
file_element = self.adapter._build_empty_file(
media_ref,
media_ref.available_range.start_time,
br_map={},
)
self.assertEqual(file_element.find("./name").text, "test_clip_name")
self.assertEqual(file_element.find("./duration").text, "2087")
rate_element = file_element.find("./rate")
self.assertEqual(rate_element.find("./ntsc").text, "FALSE")
self.assertEqual(rate_element.find("./timebase").text, "24")
tc_element = file_element.find("./timecode")
self.assertEqual(tc_element.find("./rate/ntsc").text, "FALSE")
self.assertEqual(tc_element.find("./rate/timebase").text, "24")
self.assertEqual(tc_element.find("./string").text, "09:29:47:01")
self.assertEqual(tc_element.find("./reel/name").text, "test_reel_name")
def test_marker_for_element(self):
marker_element = cElementTree.fromstring("""
<marker>
<comment>so, this happened</comment>
<name>My MArker 1</name>
<in>113</in>
<out>-1</out>
</marker>
""")
marker = self.adapter.marker_for_element(marker_element, 30)
self.assertEqual(marker.name, "My MArker 1")
self.assertEqual(
marker.marked_range,
opentime.TimeRange(
start_time=opentime.RationalTime(113, 30),
duration=opentime.RationalTime(0, 30),
))
self.assertEqual(marker.metadata["fcp_xml"]["comment"],
"so, this happened")
with self.assertRaises(KeyError):
marker.metadata["fcp_xml"]["name"]
def test_time_from_timecode_element_ntsc_non_drop_frame(self):
tc_element = cElementTree.fromstring("""
<timecode>
<rate>
<timebase>30</timebase>
<ntsc>TRUE</ntsc>
</rate>
<string>00:59:56:12</string>
<displayformat>NDF</displayformat>
</timecode>
""")
time = self.adapter._time_from_timecode_element(tc_element)
self.assertEqual(time, opentime.RationalTime(107892, (30000 / 1001.0)))
def test_time_from_timecode_element(self):
tc_element = cElementTree.fromstring("""
<timecode>
<rate>
<timebase>30</timebase>
<ntsc>FALSE</ntsc>
</rate>
<string>01:00:00:00</string>
<frame>108000</frame>
<displayformat>NDF</displayformat>
</timecode>
""")
time = self.adapter._time_from_timecode_element(tc_element)
self.assertEqual(time, opentime.RationalTime(108000, 30))
def test_time_from_timecode_element_drop_frame(self):
tc_element = cElementTree.fromstring("""
<timecode>
<rate>
<timebase>30</timebase>
<ntsc>TRUE</ntsc>
</rate>
<string>10:03:00;05</string>
<frame>1084319</frame>
<displayformat>DF</displayformat>
</timecode>
""")
time = self.adapter._time_from_timecode_element(tc_element)
self.assertEqual(time, opentime.RationalTime(1084319,
(30000 / 1001.0)))
def test_hiero_flavored_xml(self):
timeline = adapters.read_from_file(HIERO_XML_PATH)
self.assertTrue(len(timeline.tracks), 1)
self.assertTrue(timeline.tracks[0].name == 'Video 1')
clips = [c for c in timeline.tracks[0].each_clip()]
self.assertTrue(len(clips), 2)
self.assertTrue(clips[0].name == 'A160C005_171213_R0MN')
self.assertTrue(clips[1].name == '/')
self.assertTrue(
isinstance(clips[0].media_reference, schema.ExternalReference))
self.assertTrue(
isinstance(clips[1].media_reference, schema.MissingReference))
source_range = opentime.TimeRange(
start_time=opentime.RationalTime(1101071, 24),
duration=opentime.RationalTime(1055, 24))
self.assertTrue(clips[0].source_range == source_range)
available_range = opentime.TimeRange(
start_time=opentime.RationalTime(1101071, 24),
duration=opentime.RationalTime(1055, 24))
self.assertTrue(clips[0].available_range() == available_range)
clip_1_range = clips[1].available_range()
self.assertEqual(
clip_1_range,
opentime.TimeRange(
opentime.RationalTime(),
opentime.RationalTime(1, 24),
))
# Test serialization
tmp_path = tempfile.mkstemp(suffix=".xml", text=True)[1]
adapters.write_to_file(timeline, tmp_path)
# Similar to the test_roundtrip_disk2mem2disk above
# the track name element among others will not be present in a new xml.
with open(HIERO_XML_PATH, "r") as original_file:
with open(tmp_path, "r") as output_file:
self.assertNotEqual(original_file.read(), output_file.read())
def test_roundtrip_mem2disk2mem(self):
timeline = schema.Timeline('test_timeline')
RATE = 48.0
video_reference = schema.ExternalReference(
target_url="/var/tmp/test1.mov",
available_range=opentime.TimeRange(
opentime.RationalTime(value=100, rate=RATE),
opentime.RationalTime(value=1000, rate=RATE)))
video_reference.name = "test_vid_one"
audio_reference = schema.ExternalReference(
target_url="/var/tmp/test1.wav",
available_range=opentime.TimeRange(
opentime.RationalTime(value=0, rate=RATE),
opentime.RationalTime(value=1000, rate=RATE)),
)
audio_reference.name = "test_wav_one"
generator_reference = schema.GeneratorReference(
name="Color",
metadata={
"fcp_xml": {
"effectid": "Color",
"effectcategory": "Matte",
"effecttype": "generator",
"mediatype": "video",
"parameter": {
"@authoringApp": "PremierePro",
"parameterid": "fillcolor",
"name": "Color",
"value": {
"alpha": "0",
"red": "255",
"green": "255",
"blue": "255",
},
},
},
},
)
v0 = schema.Track(kind=schema.track.TrackKind.Video)
v1 = schema.Track(kind=schema.track.TrackKind.Video)
timeline.tracks.extend([v0, v1])
a0 = schema.Track(kind=schema.track.TrackKind.Audio)
timeline.tracks.append(a0)
v0.extend([
schema.Clip(name='test_clip1',
media_reference=video_reference,
source_range=opentime.TimeRange(
opentime.RationalTime(value=112, rate=RATE),
opentime.RationalTime(value=40, rate=RATE))),
schema.Gap(source_range=opentime.TimeRange(
duration=opentime.RationalTime(value=60, rate=RATE))),
schema.Clip(name='test_clip2',
media_reference=video_reference,
source_range=opentime.TimeRange(
opentime.RationalTime(value=123, rate=RATE),
opentime.RationalTime(value=260, rate=RATE))),
schema.Clip(name='test_generator_clip',
media_reference=generator_reference,
source_range=opentime.TimeRange(
opentime.RationalTime(value=292, rate=24.0),
opentime.RationalTime(value=183, rate=24.0))),
])
v1.extend([
schema.Gap(source_range=opentime.TimeRange(
duration=opentime.RationalTime(value=500, rate=RATE))),
schema.Clip(name='test_clip3',
media_reference=video_reference,
source_range=opentime.TimeRange(
opentime.RationalTime(value=112, rate=RATE),
opentime.RationalTime(value=55, rate=RATE)))
])
a0.extend([
schema.Gap(source_range=opentime.TimeRange(
duration=opentime.RationalTime(value=10, rate=RATE))),
schema.Clip(
name='test_clip4',
media_reference=audio_reference,
source_range=opentime.TimeRange(
opentime.RationalTime(value=152, rate=RATE),
opentime.RationalTime(value=248, rate=RATE)),
)
])
timeline.tracks.markers.append(
schema.Marker(name='test_timeline_marker',
marked_range=opentime.TimeRange(
opentime.RationalTime(123, RATE)),
metadata={'fcp_xml': {
'comment': 'my_comment'
}}))
v1[1].markers.append(
schema.Marker(name='test_clip_marker',
marked_range=opentime.TimeRange(
opentime.RationalTime(125, RATE)),
metadata={'fcp_xml': {
'comment': 'my_comment'
}}))
# make sure that global_start_time.rate survives the round trip
timeline.global_start_time = opentime.RationalTime(100, RATE)
result = adapters.write_to_string(timeline, adapter_name='fcp_xml')
new_timeline = adapters.read_from_string(result,
adapter_name='fcp_xml')
# Since FCP XML's "sequence" is a marriage of the timeline and the
# main tracks stack, the tracks stack loses its name
new_timeline.tracks.name = timeline.tracks.name
self.assertEqual(new_timeline.name, 'test_timeline')
# Before comparing, scrub ignorable metadata introduced in
# serialization (things like unique ids minted by the adapter)
# Since we seeded metadata for the generator, keep that metadata
del (new_timeline.metadata["fcp_xml"])
for child in new_timeline.tracks.each_child():
try:
del (child.metadata["fcp_xml"])
except KeyError:
pass
try:
is_generator = isinstance(child.media_reference,
schema.GeneratorReference)
if not is_generator:
del (child.media_reference.metadata["fcp_xml"])
except (AttributeError, KeyError):
pass
self.assertJsonEqual(new_timeline, timeline)
def test_read(self):
timeline = adapters.read_from_file(FCP7_XML_EXAMPLE_PATH)
self.assertTrue(timeline is not None)
self.assertEqual(len(timeline.tracks), 8)
video_tracks = [
t for t in timeline.tracks if t.kind == schema.TrackKind.Video
]
audio_tracks = [
t for t in timeline.tracks if t.kind == schema.TrackKind.Audio
]
self.assertEqual(len(video_tracks), 4)
self.assertEqual(len(audio_tracks), 4)
video_clip_names = ((None, 'sc01_sh010_anim.mov'),
(None, 'sc01_sh010_anim.mov', None,
'sc01_sh020_anim.mov', 'sc01_sh030_anim.mov',
'Cross Dissolve', None, 'sc01_sh010_anim'),
(None, 'test_title'),
(None, 'sc01_master_layerA_sh030_temp.mov',
'Cross Dissolve', 'sc01_sh010_anim.mov'))
for n, track in enumerate(video_tracks):
self.assertTupleEqual(tuple(c.name for c in track),
video_clip_names[n])
audio_clip_names = ((None, 'sc01_sh010_anim.mov', None,
'sc01_sh010_anim.mov'),
(None, 'sc01_placeholder.wav', None,
'sc01_sh010_anim'), (None, 'track_08.wav'),
(None, 'sc01_master_layerA_sh030_temp.mov',
'sc01_sh010_anim.mov'))
for n, track in enumerate(audio_tracks):
self.assertTupleEqual(tuple(c.name for c in track),
audio_clip_names[n])
video_clip_durations = (((536, 30.0), (100, 30.0)),
((13, 30.0), (100, 30.0), (52, 30.0),
(157, 30.0), (235, 30.0),
((19, 30.0), (0, 30.0)), (79, 30.0),
(320, 30.0)), ((15, 30.0), (941, 30.0)),
((956, 30.0), (208, 30.0),
((12, 30.0), (13, 30.0)), (82, 30.0)))
for t, track in enumerate(video_tracks):
for c, clip in enumerate(track):
if isinstance(clip, schema.Transition):
self.assertEqual(
clip.in_offset,
opentime.RationalTime(*video_clip_durations[t][c][0]))
self.assertEqual(
clip.out_offset,
opentime.RationalTime(*video_clip_durations[t][c][1]))
else:
self.assertEqual(
clip.source_range.duration,
opentime.RationalTime(*video_clip_durations[t][c]))
audio_clip_durations = (((13, 30.0), (100, 30.0), (423, 30.0),
(100, 30.0), (423, 30.0)),
((335, 30.0), (170, 30.0), (131, 30.0),
(294, 30.0), (34, 30.0),
(124, 30.0)), ((153, 30.0), (198, 30.0)),
((956, 30.0), (221, 30.0), (94, 30.0)))
for t, track in enumerate(audio_tracks):
for c, clip in enumerate(track):
self.assertEqual(
clip.source_range.duration,
opentime.RationalTime(*audio_clip_durations[t][c]))
timeline_marker_names = ('My MArker 1', 'dsf', "")
for n, marker in enumerate(timeline.tracks.markers):
self.assertEqual(marker.name, timeline_marker_names[n])
timeline_marker_start_times = ((113, 30.0), (492, 30.0), (298, 30.0))
for n, marker in enumerate(timeline.tracks.markers):
self.assertEqual(
marker.marked_range.start_time,
opentime.RationalTime(*timeline_marker_start_times[n]))
timeline_marker_comments = ('so, this happened', 'fsfsfs', None)
for n, marker in enumerate(timeline.tracks.markers):
self.assertEqual(
marker.metadata.get('fcp_xml', {}).get('comment'),
timeline_marker_comments[n])
clip_with_marker = video_tracks[1][4]
clip_marker = clip_with_marker.markers[0]
self.assertEqual(clip_marker.name, "")
self.assertEqual(clip_marker.marked_range.start_time,
opentime.RationalTime(73, 30.0))
self.assertEqual(
clip_marker.metadata.get('fcp_xml', {}).get('comment'), None)
def test_generator_for_element(self):
generator_element = cElementTree.fromstring("""
<generatoritem id="clipitem-29">
<name>White</name>
<enabled>TRUE</enabled>
<duration>1035764</duration>
<start>383</start>
<end>432</end>
<in>86313</in>
<out>86362</out>
<rate>
<timebase>24</timebase>
<ntsc>TRUE</ntsc>
</rate>
<effect>
<name>Color</name>
<effectid>Color</effectid>
<effectcategory>Matte</effectcategory>
<effecttype>generator</effecttype>
<mediatype>video</mediatype>
<parameter authoringApp="PremierePro">
<parameterid>fillcolor</parameterid>
<name>Color</name>
<value>
<alpha>0</alpha>
<red>255</red>
<green>255</green>
<blue>255</blue>
</value>
</parameter>
</effect>
</generatoritem>
""")
parent_context_element = cElementTree.fromstring("""
<track>
<rate>
<timebase>24</timebase>
<ntsc>TRUE</ntsc>
</rate>
</track>
""")
context = self.adapter._Context(parent_context_element)
# Make a parser
parser = self.adapter.FCP7XMLParser(generator_element)
clip, time_range = parser.item_and_timing_for_element(
generator_element,
head_transition=None,
tail_transition=None,
context=context,
)
self.assertEqual(clip.name, "White")
expected_range = opentime.TimeRange(
start_time=opentime.RationalTime(383, (24000 / 1001.0)),
duration=opentime.RationalTime(49, (24000 / 1001.0)),
)
self.assertEqual(time_range, expected_range)
expected_source_range = opentime.TimeRange(
start_time=opentime.RationalTime(86313, (24000 / 1001.0)),
duration=opentime.RationalTime(49, (24000 / 1001.0)),
)
self.assertEqual(clip.source_range, expected_source_range)
ref = clip.media_reference
self.assertTrue(isinstance(ref, schema.GeneratorReference))
self.assertEqual(ref.name, "Color")
self.assertEqual(ref.metadata["fcp_xml"]["parameter"]["value"]["red"],
"255")
def test_media_reference_from_element(self):
file_element = cElementTree.fromstring("""
<file id="file-3">
<name>sc01_sh030_anim.mov</name>
<pathurl>file:///Scratch/media/sc01_sh030_anim.2.mov</pathurl>
<rate>
<timebase>30</timebase>
<ntsc>FALSE</ntsc>
</rate>
<duration>400</duration>
<timecode>
<rate>
<timebase>30</timebase>
<ntsc>FALSE</ntsc>
</rate>
<string>01:00:00:00</string>
<frame>108000</frame>
<displayformat>NDF</displayformat>
<reel>
<name/>
</reel>
</timecode>
<media>
<video>
<samplecharacteristics>
<rate>
<timebase>30</timebase>
<ntsc>FALSE</ntsc>
</rate>
<width>1280</width>
<height>720</height>
<anamorphic>FALSE</anamorphic>
<pixelaspectratio>square</pixelaspectratio>
<fielddominance>none</fielddominance>
</samplecharacteristics>
</video>
<audio>
<samplecharacteristics>
<depth>16</depth>
<samplerate>48000</samplerate>
</samplecharacteristics>
<channelcount>2</channelcount>
</audio>
</media>
</file>
""")
parser = self.adapter.FCP7XMLParser(file_element)
context = self.adapter._Context()
ref = parser.media_reference_for_file_element(
file_element,
context=context,
)
self.assertEqual(ref.target_url,
"file:///Scratch/media/sc01_sh030_anim.2.mov")
self.assertEqual(ref.name, "sc01_sh030_anim.mov")
self.assertEqual(
ref.available_range,
opentime.TimeRange(
start_time=opentime.RationalTime(108000, 30),
duration=opentime.RationalTime(400, 30),
))
# Spot-check a metadata field
video_metadata = ref.metadata["fcp_xml"]["media"]["video"]
self.assertEqual(video_metadata["samplecharacteristics"]["height"],
"720")
