def topic_media(topics_id):
if access_public_topic(topics_id):
media_list = topic_media_list(TOOL_API_KEY, topics_id, snapshots_id=None, timespans_id=None, foci_id=None, sort=None, limit=None, link_id=None)
elif is_user_logged_in():
media_list = topic_media_list(user_mediacloud_key(), topics_id)
else:
return jsonify({'status': 'Error', 'message': 'Invalid attempt'})
return jsonify(media_list)
def topic_media(topics_id):
if access_public_topic(topics_id):
media_list = apicache.topic_media_list(TOOL_API_KEY, topics_id, snapshots_id=None, timespans_id=None,
foci_id=None, sort=None, limit=None, link_id=None)
elif is_user_logged_in():
media_list = apicache.topic_media_list(user_mediacloud_key(), topics_id)
else:
return jsonify({'status': 'Error', 'message': 'Invalid attempt'})
return jsonify(media_list)
def _stream_media_list_csv(user_mc_key, filename, topics_id, **kwargs):
# Helper method to stream a list of media back to the client as a csv. Any args you pass in will be
# simply be passed on to a call to topicMediaList.
all_media = []
more_media = True
params = kwargs
params[
'limit'] = 1000 # an arbitrary value to let us page through with big pages
try:
while more_media:
page = apicache.topic_media_list(user_mediacloud_key(), topics_id,
**params)
media_list = page['media']
all_media = all_media + media_list
if 'next' in page['link_ids']:
params['link_id'] = page['link_ids']['next']
more_media = True
else:
more_media = False
return csv.download_media_csv(all_media, filename,
TOPIC_MEDIA_CSV_PROPS)
except Exception as exception:
return json.dumps({'error': str(exception)},
separators=(',', ':')), 400
def media(topics_id, media_id):
user_mc = user_admin_mediacloud_client()
combined_media_info = apicache.topic_media_list(user_mediacloud_key(), topics_id, media_id=media_id)['media'][0]
media_info = user_mc.media(media_id)
for key in list(media_info.keys()):
if key not in list(combined_media_info.keys()):
combined_media_info[key] = media_info[key]
return jsonify(combined_media_info)
def media(topics_id, media_id):
user_mc = user_admin_mediacloud_client()
combined_media_info = topic_media_list(user_mediacloud_key(), topics_id, media_id=media_id)['media'][0]
media_info = user_mc.media(media_id)
for key in media_info.keys():
if key not in combined_media_info.keys():
combined_media_info[key] = media_info[key]
return jsonify(combined_media_info)
def _generate_network_of_frames(topics_id, timespans_id, num_of_sources, out_name, top_media_sort,
remove_media_list=None, remove_word_list=[], generate_word_lists=False,
include_media_list=None, media_attribs=None, num_words=None):
if remove_media_list is None:
remove_media_list = []
# use this specify attributes on the media source that should be added to the node as attributes
# if(media_attribs == None):
# media_attribs = {}
if include_media_list is None:
media_sources_md = topic_media_list(user_mediacloud_key(), topics_id, timespans_id=timespans_id,
limit=num_of_sources + len(remove_media_list), sort=top_media_sort)['media']
else:
media_sources_md = include_media_list
if remove_media_list is not None:
for r in remove_media_list:
media_sources_md = _remove_media_source(r, media_sources_md)
top_words = _build_top_words(media_sources_md, topics_id, timespans_id, remove_word_list, num_words)
if remove_word_list is not None:
top_words = _clean_top_words(top_words, remove_word_list)
frame_network = _build_network(top_words, media_sources_md, media_attribs)
_export_gexf_network(frame_network, '%s.gexf' % out_name)
_export_d3_network(frame_network, '%s' % out_name)
if generate_word_lists:
with open('%s.txt' % out_name, 'w', encoding="utf-8") as wl:
all_words = []
media_sources = {ms['media_id']: ms['name'] for ms in media_sources_md}
# counts = {}
for ms in top_words:
# wl.write("\n\n%s (media id: %d):\n" % (media_sources[ms].encode('ascii', 'ignore'), ms))
wl.write("\n\n{} (media id: {}):\n".format(media_sources[ms], ms))
for w in top_words[ms]:
all_words.append(w['term'])
# increment count to see how many media source include each word
# counts[ms]
# wl.write("- %s (%d)\n" % (w['term'].encode('ascii', 'ignore'), w['count']))
wl.write("- {} ({})\n".format(w['term'], w['count']))
wl.write("\n")
linefeed = chr(10) # linefeed=\n
s = linefeed.join(nx.generate_gexf(frame_network)) # doctest: +SKIP
# for line in nx.generate_gexf(frame_network): # doctest: +SKIP
# logger.debug line
return s
def _stream_media_list_csv(user_mc_key, filename, topics_id, **kwargs):
# Helper method to stream a list of media back to the client as a csv. Any args you pass in will be
# simply be passed on to a call to topicMediaList.
add_metadata = False # off for now because this is SUPER slow
all_media = []
more_media = True
params = kwargs
params[
'limit'] = 1000 # an arbitrary value to let us page through with big pages
try:
cols_to_export = TOPICS_TEMPLATE_PROPS
if not add_metadata:
cols_to_export = cols_to_export[:-4] # remove the metadata cols
while more_media:
page = apicache.topic_media_list(user_mediacloud_key(), topics_id,
**params)
media_list = page['media']
user_mc = user_admin_mediacloud_client()
if add_metadata:
for media_item in media_list:
media_info = user_mc.media(media_item['media_id'])
for eachItem in media_info['media_source_tags']:
if is_metadata_tag_set(eachItem['tag_sets_id']):
format_metadata_fields(media_item, eachItem)
all_media = all_media + media_list
if 'next' in page['link_ids']:
params['link_id'] = page['link_ids']['next']
more_media = True
else:
more_media = False
return csv.download_media_csv(all_media, filename, cols_to_export)
except Exception as exception:
return json.dumps({'error': str(exception)},
separators=(',', ':')), 400
def _stream_media_list_csv(user_mc_key, filename, topics_id, **kwargs):
# Helper method to stream a list of media back to the client as a csv. Any args you pass in will be
# simply be passed on to a call to topicMediaList.
all_media = []
more_media = True
params = kwargs
params['limit'] = 1000 # an arbitrary value to let us page through with big pages
try:
while more_media:
page = apicache.topic_media_list(user_mediacloud_key(), topics_id, **params)
media_list = page['media']
all_media = all_media + media_list
if 'next' in page['link_ids']:
params['link_id'] = page['link_ids']['next']
more_media = True
else:
more_media = False
return csv.download_media_csv(all_media, filename, TOPIC_MEDIA_CSV_PROPS)
except Exception as exception:
return json.dumps({'error': str(exception)}, separators=(',', ':')), 400
def topic_provider_media(topics_id):
media_list = apicache.topic_media_list(user_mediacloud_key(), topics_id)
return jsonify(media_list)
def _generate_network_of_frames(topics_id,
timespans_id,
num_of_sources,
out_name,
top_media_sort,
remove_media_list=None,
remove_word_list=[],
generate_word_lists=False,
include_media_list=None,
media_attribs=None,
num_words=None):
if remove_media_list is None:
remove_media_list = []
# use this specify attributes on the media source that should be added to the node as attributes
# if(media_attribs == None):
# media_attribs = {}
if include_media_list is None:
media_sources_md = topic_media_list(user_mediacloud_key(),
topics_id,
timespans_id=timespans_id,
limit=num_of_sources +
len(remove_media_list),
sort=top_media_sort)['media']
else:
media_sources_md = include_media_list
if remove_media_list is not None:
for r in remove_media_list:
media_sources_md = _remove_media_source(r, media_sources_md)
top_words = _build_top_words(media_sources_md, topics_id, timespans_id,
remove_word_list, num_words)
if remove_word_list is not None:
top_words = _clean_top_words(top_words, remove_word_list)
frame_network = _build_network(top_words, media_sources_md, media_attribs)
_export_gexf_network(frame_network, '%s.gexf' % out_name)
_export_d3_network(frame_network, '%s' % out_name)
if generate_word_lists:
with open('%s.txt' % out_name, 'w', encoding="utf-8") as wl:
all_words = []
media_sources = {
ms['media_id']: ms['name']
for ms in media_sources_md
}
# counts = {}
for ms in top_words:
# wl.write("\n\n%s (media id: %d):\n" % (media_sources[ms].encode('ascii', 'ignore'), ms))
wl.write("\n\n{} (media id: {}):\n".format(
media_sources[ms], ms))
for w in top_words[ms]:
all_words.append(w['term'])
# increment count to see how many media source include each word
# counts[ms]
# wl.write("- %s (%d)\n" % (w['term'].encode('ascii', 'ignore'), w['count']))
wl.write("- {} ({})\n".format(w['term'], w['count']))
wl.write("\n")
linefeed = chr(10) # linefeed=\n
s = linefeed.join(nx.generate_gexf(frame_network)) # doctest: +SKIP
# for line in nx.generate_gexf(frame_network): # doctest: +SKIP
# logger.debug line
return s
