def test_xpath_text(self):
testxml = '''<root>
<div>
<p>Foo</p>
</div>
</root>'''
doc = xml.etree.ElementTree.fromstring(testxml)
self.assertEqual(xpath_text(doc, 'div/p'), 'Foo')
self.assertTrue(xpath_text(doc, 'div/bar') is None)
self.assertRaises(ExtractorError, xpath_text, doc, 'div/bar', fatal=True)
def test_xpath_text(self):
testxml = '''<root>
<div>
<p>Foo</p>
</div>
</root>'''
doc = xml.etree.ElementTree.fromstring(testxml)
self.assertEqual(xpath_text(doc, 'div/p'), 'Foo')
self.assertTrue(xpath_text(doc, 'div/bar') is None)
self.assertRaises(ExtractorError, xpath_text, doc, 'div/bar', fatal=True)
def test_xpath_text(self):
testxml = """<root>
<div>
<p>Foo</p>
</div>
</root>"""
doc = compat_etree_fromstring(testxml)
self.assertEqual(xpath_text(doc, "div/p"), "Foo")
self.assertEqual(xpath_text(doc, "div/bar", default="default"), "default")
self.assertTrue(xpath_text(doc, "div/bar") is None)
self.assertRaises(ExtractorError, xpath_text, doc, "div/bar", fatal=True)
def run(self, dl_url: str, add_webcam: bool, add_annotations: bool,
add_cursor, keep_tmp_files: bool, filename: str):
m_obj = re.match(self._VALID_URL, dl_url)
video_id = m_obj.group('id')
video_website = m_obj.group('website')
self.to_screen("Downloading meta informations")
# Make sure the lesson exists
self._download_webpage(dl_url, video_id)
self._create_tmp_dir(video_id)
# Extract basic metadata
metadata_url = video_website + '/presentation/' + video_id + '/metadata.xml'
metadata = self._download_xml(metadata_url, video_id)
shapes_url = video_website + '/presentation/' + video_id + '/shapes.svg'
shapes = self._download_xml(shapes_url, video_id)
cursor_url = video_website + '/presentation/' + video_id + '/cursor.xml'
cursor_infos = self._download_xml(cursor_url, video_id)
# Parse metadata.xml
meta = metadata.find('./meta')
start_time = xpath_text(metadata, 'start_time')
recording_duration = float(xpath_text(
metadata, './playback/duration')) / 1000.0 # in seconds
title = xpath_text(meta, 'meetingName')
try:
bbb_origin_version = xpath_text(meta, 'bbb-origin-version')
if bbb_origin_version is not None:
bbb_version = bbb_origin_version.split(' ')[0]
self.to_screen("BBB version: " + bbb_version)
except Exception:
pass
# Downloading Slides
images = list()
self.xml_find_rec(shapes, _s('svg:image'), images)
# images = shapes.findall(_s("./svg:image[@class='slide']"))
slides_infos = []
bonus_images = []
img_path_to_filename = {}
counter = 0
for image in images:
img_path = image.get(_x('xlink:href'))
image_url = video_website + '/presentation/' + video_id + '/' + img_path
image_width = int(float(image.get('width')))
image_height = int(float(image.get('height')))
if not image.get('class') or image.get('class') != 'slide':
image_filename = image_url.split('/')[-1]
image_path = video_id + '/' + image_filename
bonus_images.append(
BonusImage(
image_url,
image_filename,
image_path,
image_width,
image_height,
))
continue
image_id = image.get('id')
slide_annotations = shapes.find(
_s("./svg:g[@image='{}']".format(image_id)))
if img_path.endswith('deskshare.png'):
image_url = video_website + '/presentation/' + video_id + '/deskshare/deskshare.webm'
slide_filename = 'deskshare.webm'
else:
if img_path not in img_path_to_filename:
slide_filename = 'slide-{:03d}'.format(
counter) + '.' + determine_ext(img_path)
img_path_to_filename[img_path] = slide_filename
counter += 1
else:
slide_filename = img_path_to_filename[img_path]
slide_path = video_id + '/' + slide_filename
slide_ts_in = float(image.get('in'))
slide_ts_out = float(image.get('out'))
slide_ts_duration = max(
0.0,
min(recording_duration - slide_ts_in,
slide_ts_out - slide_ts_in))
slides_infos.append(
Slide(
image_id,
image_url,
slide_filename,
slide_path,
image_width,
image_height,
slide_ts_in,
slide_ts_out,
slide_ts_duration,
slide_annotations,
))
# We now change the xml tree, all hrefs of all images now point to local files
for image in images:
image.attrib[_x('xlink:href')] = video_id + '/' + image.attrib[_x(
'xlink:href')].split('/')[-1]
self.to_screen("Downloading slides")
self._write_slides(slides_infos, self.ydl)
self._write_slides(bonus_images, self.ydl)
if add_annotations:
slides_infos = self._add_annotations(slides_infos)
if add_cursor:
slides_infos = self._add_cursor(slides_infos, cursor_infos)
# Downlaoding Webcam / Deskshare
video_base_url = video_website + '/presentation/' + video_id
if not self.verbose:
self.ydl.to_stderr_backup = self.ydl.to_stderr
self.ydl.to_stderr = types.MethodType(dummy_to_stderr, self.ydl)
webcams_path = video_id + '/webcams.webm'
try:
self.to_screen("Downloading webcams.webm")
webcams_dl = {
'id': video_id,
'title': title,
'url': video_base_url + '/video/webcams.webm',
'timestamp': int(start_time),
}
self.ydl.params['outtmpl'] = webcams_path
self.ydl.process_ie_result(webcams_dl)
except DownloadError:
self.to_screen(
"Downloading webcams.webm failed! Downloading webcams.mp4 instead"
)
webcams_path = video_id + '/webcams.mp4'
try:
webcams_dl = {
'id': video_id,
'title': title,
'url': video_base_url + '/video/webcams.mp4',
'timestamp': int(start_time),
}
self.ydl.params['outtmpl'] = webcams_path
self.ydl.process_ie_result(webcams_dl)
except DownloadError:
webcams_path = None
self.to_screen("Error: Downloading webcams.mp4 failed!")
deskshare_path = video_id + '/deskshare.webm'
try:
self.to_screen("Downloading deskshare.webm")
deskshare_dl = {
'id': video_id,
'title': title,
'url': video_base_url + '/deskshare/deskshare.webm',
'timestamp': int(start_time),
}
self.ydl.params['outtmpl'] = deskshare_path
self.ydl.process_ie_result(deskshare_dl)
except DownloadError:
self.to_screen(
"Downloading deskshare.webm failed! Downloading deskshare.mp4 instead"
)
deskshare_path = video_id + '/deskshare.mp4'
try:
deskshare_dl = {
'id': video_id,
'title': title,
'url': video_base_url + '/deskshare/deskshare.mp4',
'timestamp': int(start_time),
}
self.ydl.params['outtmpl'] = deskshare_path
self.ydl.process_ie_result(deskshare_dl)
except DownloadError:
deskshare_path = None
self.to_screen(
"Warning: Downloading deskshare.mp4 failed - No desk was likely shared in this session."
)
if not self.verbose:
self.ydl.to_stderr = self.ydl.to_stderr_backup
# Post processing
slideshow_w, slideshow_h = self._rescale_slides(slides_infos)
slideshow_path = self._create_slideshow(slides_infos, video_id,
slideshow_w, slideshow_h)
formatted_date = datetime.fromtimestamp(
int(start_time) / 1000).strftime('%Y-%m-%dT%H-%M-%S')
if filename is not None:
result_path = filename
else:
result_path = formatted_date + '_' + title.replace(
'/', '_', title.count('/')) + '.mp4'
self.to_screen("Mux Slideshow")
webcam_w, webcam_h = self._get_webcam_size(slideshow_w, slideshow_h)
if os.path.isfile(result_path):
self.report_warning("Final Slideshow already exists. Abort!")
return
if add_webcam:
self.ffmpeg.mux_slideshow_with_webcam(slideshow_path, webcams_path,
webcam_w, webcam_h,
result_path)
else:
self.ffmpeg.mux_slideshow(slideshow_path, webcams_path,
result_path)
if not keep_tmp_files:
self.to_screen("Cleanup")
self._remove_tmp_dir(video_id)
def run(self, dl_url: str, add_webcam: bool, add_annotations: bool, add_cursor, keep_tmp_files: bool, filename: str):
m_obj = self._VALID_URL_RE.match(dl_url)
video_id = m_obj.group('id')
video_website = m_obj.group('website')
self.to_screen("Downloading meta informations")
# Make sure the lesson exists
self._download_webpage(dl_url, video_id)
self._create_tmp_dir(video_id)
# Extract basic metadata
metadata_url = video_website + '/presentation/' + video_id + '/metadata.xml'
metadata = self._download_xml(metadata_url, video_id)
shapes_url = video_website + '/presentation/' + video_id + '/shapes.svg'
shapes = self._download_xml(shapes_url, video_id)
cursor_url = video_website + '/presentation/' + video_id + '/cursor.xml'
cursor_infos = self._download_xml(cursor_url, video_id)
# Parse metadata.xml
meta = metadata.find('./meta')
start_time = xpath_text(metadata, 'start_time')
title = xpath_text(meta, 'meetingName')
bbb_version = xpath_text(meta, 'bbb-origin-version').split(' ')[0]
self.to_screen("BBB version: " + bbb_version)
# Downloading Slides
images = shapes.findall(_s("./svg:image[@class='slide']"))
slides_infos = []
img_path_to_filename = {}
counter = 0
for image in images:
img_path = image.get(_x('xlink:href'))
image_id = image.get('id')
image_url = video_website + '/presentation/' + video_id + '/' + img_path
image_width = int(image.get('width'))
image_height = int(image.get('height'))
slide_annotations = shapes.find(_s("./svg:g[@image='{}']".format(image_id)))
if img_path.endswith('deskshare.png'):
image_url = video_website + '/presentation/' + video_id + '/deskshare/deskshare.webm'
slide_filename = 'deskshare.webm'
else:
if img_path not in img_path_to_filename:
slide_filename = 'slide-{:03d}'.format(counter) + '.' + determine_ext(img_path)
img_path_to_filename[img_path] = slide_filename
counter += 1
else:
slide_filename = img_path_to_filename[img_path]
slide_path = video_id + '/' + slide_filename
slide_ts_in = float(image.get('in'))
slide_ts_out = float(image.get('out'))
slides_infos.append(
Slide(
image_id,
image_url,
slide_filename,
slide_path,
image_width,
image_height,
slide_ts_in,
slide_ts_out,
max(0, slide_ts_out - slide_ts_in),
slide_annotations,
)
)
self.to_screen("Downloading slides")
self._write_slides(slides_infos, self.ydl)
if add_annotations:
slides_infos = self._add_annotations(slides_infos)
if add_cursor:
slides_infos = self._add_cursor(slides_infos, cursor_infos)
# Downlaoding Webcam / Deskshare
video_base_url = video_website + '/presentation/' + video_id
webcams_path = video_id + '/webcams.webm'
try:
self.to_screen("Downloading webcams.webm")
webcams_dl = {
'id': video_id,
'title': title,
'url': video_base_url + '/video/webcams.webm',
'timestamp': int(start_time),
}
self.ydl.params['outtmpl'] = webcams_path
self.ydl.process_ie_result(webcams_dl)
except DownloadError:
pass
deskshare_path = video_id + '/deskshare.webm'
try:
self.to_screen("Downloading deskshare.webm")
deskshare_dl = {
'id': video_id,
'title': title,
'url': video_base_url + '/deskshare/deskshare.webm',
'timestamp': int(start_time),
}
self.ydl.params['outtmpl'] = deskshare_path
self.ydl.process_ie_result(deskshare_dl)
except DownloadError:
pass
# Post processing
slideshow_w, slideshow_h = self._rescale_slides(slides_infos)
slideshow_path = self._create_slideshow(slides_infos, video_id, slideshow_w, slideshow_h)
formatted_date = datetime.fromtimestamp(int(start_time) / 1000).strftime('%Y-%m-%dT%H-%M-%S')
if filename is not None:
result_path = filename
else:
result_path = formatted_date + '_' + title.replace('/','_',title.count('/')) + '.mp4'
self.to_screen("Mux Slideshow")
webcam_w, webcam_h = self._get_webcam_size(slideshow_w, slideshow_h)
if os.path.isfile(result_path):
self.report_warning("Final Slideshow already exists. Abort!")
return
if add_webcam:
self.ffmpeg.mux_slideshow_with_webcam(slideshow_path, webcams_path, webcam_w, webcam_h, result_path)
else:
self.ffmpeg.mux_slideshow(slideshow_path, webcams_path, result_path)
if not keep_tmp_files:
self.to_screen("Cleanup")
self._remove_tmp_dir(video_id)