def up_richard(self, ep):
host = pw.richard[ep.show.client.richard_id]
endpoint = "http://{hostname}/api/v1".format(hostname=host["host"])
api = API(endpoint)
vid = ep.public_url.split("/video/")[1].split("/")[0]
response = api.video(vid).get(username=host["user"], api_key=host["api_key"])
video_data = get_content(response)
video_data["state"] = 1
try:
update_video(endpoint, host["user"], host["api_key"], vid, video_data)
except MissingRequiredData, e:
# this shouldn't happen, prolly debugging something.
import code
code.interact(local=locals())
def up_richard(self, ep):
host = pw.richard[ep.show.client.richard_id]
endpoint = 'http://{hostname}/api/v1'.format(hostname=host['host'])
api = API(endpoint)
# vid = ep.public_url.split('/video/')[1].split('/')[0]
vid = get_video_id(ep.public_url)
response = api.video(vid).get(
username=host['user'], api_key=host['api_key'])
video_data = get_content(response)
video_data['state'] = 1
try:
update_video(endpoint, host['user'], host['api_key'],
vid, video_data)
except MissingRequiredData, e:
# this shouldn't happen, prolly debugging something.
import code
code.interact(local=locals())
def process_ep(self, ep):
""" adds Episode to pyvideo
:arg ep: Episode to add to pyvideo
:returns: different things based on what is happening
a video dictionary for updated videos
a pyvideo url for newly added videos
False if there was an exception during processing
"""
# pyvideo categories are stored in Client and Show
# chipy is an example of something that uses Client,
# pycon 2014 and 300SoC are in Show.
self.category_key = ep.show.category_key \
or ep.show.client.category_key
self.host = pw.richard[ep.show.client.richard_id]
self.pyvideo_endpoint = \
'http://{hostname}/api/v1'.format(hostname=self.host['host'])
self.api = API(self.pyvideo_endpoint)
if self.options.verbose:
print self.pyvideo_endpoint, self.host['user'], self.host['api_key'], {'category_key': self.category_key}
# FIXME using chatty hack due to problems with category handling
if not self.options.test:
create_category_if_missing( self.pyvideo_endpoint,
self.host['user'], self.host['api_key'],
{'title': self.category_key,
'description':ep.show.description})
video_data = self.create_pyvideo_episode_dict(ep, state=2)
# perhaps we could just update the dict based on scraped_meta
# scraped_metadata = self.get_scrapevideo_metadata(ep.host_url)
# video_data['thumbnail_url'] = scraped_metadata.get('thumbnail_url','')
if ep.host_url is None:
pass
# half baked idea:
# use title slide as place holder image until video is produced.
# video_data['thumbnail_url'] = "http://veyepar.nextdayvideo.com/static/%s/%s/titles/%s.png" % ( ep.show.client.slug, ep.show.slug, ep.slug )
elif "youtube" in ep.host_url:
scraped_metadata = self.get_scrapevideo_metadata(ep.host_url)
video_data['thumbnail_url'] = scraped_metadata.get('thumbnail_url','')
video_data['embed'] = \
scraped_metadata.get('object_embed_code','')
elif "vimeo" in ep.host_url:
# video_data['embed'] = scraped_metadata.get('embed_code','')
params = {'vimeo_id': ep.host_url.split('/')[-1]}
video_data['embed'] ="""<object width="640" height="480"><param name="allowfullscreen" value="true"><param name="allowscriptaccess" value="always"><param name="movie" value="http://vimeo.com/moogaloop.swf?show_byline=1&fullscreen=1&clip_id=%(vimeo_id)s&color=&server=vimeo.com&show_title=1&show_portrait=0"><embed src="http://vimeo.com/moogaloop.swf?show_byline=1&fullscreen=1&clip_id=%(vimeo_id)s&color=&server=vimeo.com&show_title=1&show_portrait=0" allowscriptaccess="always" height="480" width="640" allowfullscreen="true" type="application/x-shockwave-flash"></embed></object>""" % params
if self.options.verbose: pprint.pprint(video_data)
if \
not video_data['video_mp4_url'] \
or not video_data['source_url'] \
or not video_data['embed']:
import code
# code.interact(local=locals())
try:
if self.is_already_in_pyvideo(ep):
vid_id = ep.public_url.split('/video/')[1].split('/')[0]
print 'updating episode in pyvideo', ep.public_url, vid_id
if not self.options.test:
ret = self.update_pyvideo(vid_id, video_data)
# above ret= isn't working. returns None I think?
# lets hope there wasn't a problem and blaze ahead.
ret = True
else:
ret = False
else:
if not self.options.test:
self.pvo_url = self.create_pyvideo(video_data)
print 'new pyvideo url', self.pvo_url
ep.public_url = self.pvo_url
ret = self.pvo_url
else:
ret = False
except Exception as exc:
print "exc:", exc
import code
code.interact(local=locals())
raise exc
ep.save()
return ret
class Richard(Process):
ready_state = 5
# ready_state = None
def process_ep(self, ep):
""" adds Episode to pyvideo
:arg ep: Episode to add to pyvideo
:returns: different things based on what is happening
a video dictionary for updated videos
a pyvideo url for newly added videos
False if there was an exception during processing
"""
# pyvideo categories are stored in Client and Show
# chipy is an example of something that uses Client,
# pycon 2014 and 300SoC are in Show.
self.category_key = ep.show.category_key \
or ep.show.client.category_key
self.host = pw.richard[ep.show.client.richard_id]
self.pyvideo_endpoint = \
'http://{hostname}/api/v1'.format(hostname=self.host['host'])
self.api = API(self.pyvideo_endpoint)
if self.options.verbose:
print self.pyvideo_endpoint, self.host['user'], self.host['api_key'], {'category_key': self.category_key}
# FIXME using chatty hack due to problems with category handling
if not self.options.test:
create_category_if_missing( self.pyvideo_endpoint,
self.host['user'], self.host['api_key'],
{'title': self.category_key,
'description':ep.show.description})
video_data = self.create_pyvideo_episode_dict(ep, state=2)
# perhaps we could just update the dict based on scraped_meta
# scraped_metadata = self.get_scrapevideo_metadata(ep.host_url)
# video_data['thumbnail_url'] = scraped_metadata.get('thumbnail_url','')
if ep.host_url is None:
pass
# half baked idea:
# use title slide as place holder image until video is produced.
# video_data['thumbnail_url'] = "http://veyepar.nextdayvideo.com/static/%s/%s/titles/%s.png" % ( ep.show.client.slug, ep.show.slug, ep.slug )
elif "youtube" in ep.host_url:
scraped_metadata = self.get_scrapevideo_metadata(ep.host_url)
video_data['thumbnail_url'] = scraped_metadata.get('thumbnail_url','')
video_data['embed'] = \
scraped_metadata.get('object_embed_code','')
elif "vimeo" in ep.host_url:
# video_data['embed'] = scraped_metadata.get('embed_code','')
params = {'vimeo_id': ep.host_url.split('/')[-1]}
video_data['embed'] ="""<object width="640" height="480"><param name="allowfullscreen" value="true"><param name="allowscriptaccess" value="always"><param name="movie" value="http://vimeo.com/moogaloop.swf?show_byline=1&fullscreen=1&clip_id=%(vimeo_id)s&color=&server=vimeo.com&show_title=1&show_portrait=0"><embed src="http://vimeo.com/moogaloop.swf?show_byline=1&fullscreen=1&clip_id=%(vimeo_id)s&color=&server=vimeo.com&show_title=1&show_portrait=0" allowscriptaccess="always" height="480" width="640" allowfullscreen="true" type="application/x-shockwave-flash"></embed></object>""" % params
if self.options.verbose: pprint.pprint(video_data)
if \
not video_data['video_mp4_url'] \
or not video_data['source_url'] \
or not video_data['embed']:
import code
# code.interact(local=locals())
try:
if self.is_already_in_pyvideo(ep):
vid_id = ep.public_url.split('/video/')[1].split('/')[0]
print 'updating episode in pyvideo', ep.public_url, vid_id
if not self.options.test:
ret = self.update_pyvideo(vid_id, video_data)
# above ret= isn't working. returns None I think?
# lets hope there wasn't a problem and blaze ahead.
ret = True
else:
ret = False
else:
if not self.options.test:
self.pvo_url = self.create_pyvideo(video_data)
print 'new pyvideo url', self.pvo_url
ep.public_url = self.pvo_url
ret = self.pvo_url
else:
ret = False
except Exception as exc:
print "exc:", exc
import code
code.interact(local=locals())
raise exc
ep.save()
return ret
def update_pyvideo(self, vid, new_data):
""" updates a pyvideo record
:arg vid: video id for pyvideo
:arg new_data: dict of fields to update
:returns: a dict from the updated video
"""
try:
# fetch current record
response = self.api.video(vid).get(username=self.host['user'], api_key=self.host['api_key'])
video_data = get_content(response)
if self.options.verbose: pprint.pprint( video_data )
# update dict with new information
video_data.update(new_data)
if self.options.verbose: pprint.pprint( video_data )
# update in pyvideo
return update_video(self.pyvideo_endpoint, self.host['user'], self.host['api_key'], vid, video_data)
except MissingRequiredData as e:
print '#2, Missing required fields', e.errors
code.interact(local=locals())
raise e
def create_pyvideo(self, video_data):
""" creates a pyvideo record
:arg video_data: dict of video information to be added
:returns: a pyvideo url for the video
"""
try:
video_data['added'] = datetime.datetime.now().isoformat()
vid = create_video(self.pyvideo_endpoint, self.host['user'], self.host['api_key'], video_data)
return 'http://%s/video/%s/%s' % (self.host['host'], vid['id'],vid['slug'])
except MissingRequiredData as e:
print '#3, Missing required fields', e.errors
raise e
def create_pyvideo_episode_dict(self, ep, state=1):
""" create dict for pyvideo based on Episode
This creates a dict of based on information available in an Episode
object. This does not populate information that has to be derived
from scrapping.
:arg ep: Episode to convert in to a pyvideo dict
:arg state: pyvideo state. 1=live, 2=draft. defaults to 1
:returns: dict of Episode information for use by pyvideo
"""
# clean up messy Episode data
speakers = self.clean_pyvideo_speakers(ep)
tags = self.clean_pyvideo_tags(ep)
summary = self.clean_pyvideo_summary(ep)
mp4url = self.clean_archive_mp4_url(ep)
video_data = {
'state': state,
'title': ep.name,
'category': self.category_key,
'summary': summary,
'source_url': ep.host_url,
'copyright_text': ep.license,
'tags': tags,
'speakers': speakers,
'recorded': ep.start.isoformat(),
'language': 'English',
'duration': int(ep.get_minutes()),
'video_ogv_url': ep.archive_ogv_url,
'video_mp4_url': mp4url,
'video_mp4_download_only': True,
}
return video_data
def clean_archive_mp4_url(self, ep):
# strip off parameters that archive adds.
# maybe this should go into archive_uploader.py ?
if ep.archive_mp4_url:
o = urlparse(ep.archive_mp4_url)
mp4url = "%(scheme)s://%(netloc)s%(path)s" % o._asdict()
else:
mp4url = ""
return mp4url
def clean_pyvideo_summary(self, ep):
# Richard wants markdown
# so if ep data is in html or somthing, convert to markdown.
# best to get event site to provide markdown.
# desc = linebreaks(urlize(force_escape(ep.description)))
return ep.description
def clean_pyvideo_speakers(self, ep):
""" sanitize veyepar authors to create pyvideo speakers
:arg ep: Episode with authors, comma sepperated
:returns: list of speakers
"""
# speakers = [] if ep.authors is None else ep.authors.split(',')
speakers = ep.authors.split(',') if ep.authors else []
speakers = [ s.strip() for s in speakers ]
return speakers
def clean_pyvideo_tags(self, ep):
""" sanitize veyepar tags for use by pyvideo
:arg ep: Episode with veyepar tags
:returns: list of pyvideo tags
"""
# remove blacklisted tags, and tags with a / in them. and strip spaces
if ep.tags is None:
tags = ''
else:
tags = ep.tags.split(',')
tags = [t.strip() for t in tags]
tags = [t for t in tags
if t not in [
u'enthought', u'scipy_2012',
u'Introductory/Intermediate',
]
and '/' not in t
and t]
return tags
def get_scrapevideo_metadata(self, host_url):
""" scrapes metadata from the host_url of the episode
This is a wrapper around steve's scrapevideo. It preps
the host_url if necessary, and repeatedly calls scrapevideo
until no error is raised
:arg ep: Episode of video to scrape
:returns: dict of metadata, or {}
"""
if host_url is None or host_url == '':
# there's nothing to scrape
return {}
# FIXME: error handling kinda crazy here
while True:
# keep trying until it doesn't error doh!
try:
scraped_meta = scrapevideo(host_url)
break
except KeyError as e:
print "KeyError", e
if self.options.verbose:
pprint.pprint( scraped_meta )
return scraped_meta
def is_already_in_pyvideo(self, ep):
# its truthiness implies that the video already exists in pyvideo
return ep.public_url
def add_more_options(self, parser):
# oh wait.. I am not sure how to implement this...
# parser.add_option('--all', action="store_true",
# help="process all, regardless of state. (does not change state)")
pass
