def _topic_media_link_list_by_page_as_csv_row(user_mc_key, topics_id, props,
**kwargs):
local_mc = user_admin_mediacloud_client(
user_mc_key
) #having issues with calling apicache call.. so trying directly
yield u','.join(props) + u'\n' # first send the column names
all_media = []
more_media = True
link_id = 0
params = kwargs
params[
'limit'] = 1000 # an arbitrary value to let us page through with big pages
while more_media:
media_link_page = apicache.topic_media_link_list_by_page(
TOOL_API_KEY, topics_id, link_ids=link_id, **kwargs)
media_list = media_link_page['links']
media_src_ids = [
str(s['source_media_id']) for s in media_link_page['links']
]
media_ref_ids = [
str(s['ref_media_id']) for s in media_link_page['links']
]
media_src_ids = media_src_ids + media_ref_ids
# user_mc_key isn't working
#
for m in media_link_page['links']:
q = "media_id:{[61164, 4434, 18380]}"
params['q'] = q
# TODO - CSB this currently doesn't work the way we need it to. waiting for hal to respond
media_info = local_mc.topicMediaList(topics_id, **params)
for m_info in media_info['media']:
if m['source_media_id'] == m_info['media_id']:
m['source_info'] = m_info
if m['ref_media_id'] == m_info['media_id']:
m['ref_info'] = m_info
if 'next' in media_link_page['link_ids']:
link_id = media_link_page['link_ids']['next']
else:
more_media = False
for s in media_link_page['links']:
cleaned_source_info = csv.dict2row(TOPIC_MEDIA_CSV_PROPS,
s['source_info'])
cleaned_ref_info = csv.dict2row(TOPIC_MEDIA_CSV_PROPS,
s['ref_info'])
row_string = u','.join(cleaned_source_info) + ',' + u','.join(
cleaned_ref_info) + u'\n'
yield row_string
def _story_list_by_page_as_csv_row(api_key, q, fq, stories_per_page, sort, page_limit, props):
yield ','.join(props) + '\n' # first send the column names
for page in _story_list_by_page(api_key, q, fq, stories_per_page, sort, page_limit):
for story in page:
cleaned_row = csv.dict2row(props, story)
row_string = ','.join(cleaned_row) + '\n'
yield row_string
def _story_list_by_page_as_csv_row(q, fq, stories_per_page, sort, page_limit, props):
yield ','.join(props) + '\n' # first send the column names
for page in _story_list_by_page(q, fq, stories_per_page, sort, page_limit):
for story in page:
cleaned_row = csv.dict2row(props, story)
row_string = ','.join(cleaned_row) + '\n'
yield row_string
def _topic_story_link_list_by_page_as_csv_row(user_key, topics_id, props,
**kwargs):
local_mc = user_admin_mediacloud_client(user_key)
yield u','.join(props) + u'\n' # first send the column names
link_id = 0
more_pages = True
while more_pages:
story_link_page = topic_story_link_list_by_page(user_key,
topics_id,
link_ids=link_id,
**kwargs)
story_src_ids = [
str(s['source_stories_id']) for s in story_link_page['links']
]
story_ref_ids = [
str(s['ref_stories_id']) for s in story_link_page['links']
]
story_src_ids = story_src_ids + story_ref_ids
# TODO there is a cached topic story list... but paging is different...
storiesInfoList = local_mc.topicStoryList(topics_id,
stories_id=story_src_ids)
# get all source and ref story link ids and fetch them with topicStoryList
if 'next' in story_link_page['link_ids']:
link_id = story_link_page['link_ids']['next']
else:
more_pages = False
for s in storiesInfoList['stories']:
cleaned_row = csv.dict2row(props, s)
row_string = u','.join(cleaned_row) + u'\n'
yield row_string
def _topic_media_link_list_by_page_as_csv_row(user_mc_key, topics_id, props,
**kwargs):
yield ','.join(props) + '\n' # first send the column names
more_media = True
link_id = 0
basic_media_props = ['media_id', 'name', 'url']
while more_media:
# fetch one page of data
media_link_page = apicache.topic_media_link_list_by_page(
TOOL_API_KEY, topics_id, link_id, **kwargs)
# get the media info for all the media sources
media_src_ids = [
str(s['source_media_id']) for s in media_link_page['links']
]
media_ref_ids = [
str(s['ref_media_id']) for s in media_link_page['links']
]
media_src_ids = set(media_src_ids + media_ref_ids) # make it distinct
# TODO: PARALLELIZE - can't use executor here, because we are out of the context?
media_lookup = {
int(mid): _media_info_worker({
'key': user_mc_key,
'media_id': mid
})
for mid in media_src_ids
}
# connect the link data to the media info data
for link_pair in media_link_page['links']:
link_pair['source_info'] = media_lookup[int(
link_pair['source_media_id'])]
link_pair['ref_info'] = media_lookup[int(
link_pair['ref_media_id'])]
# stream this page's results to the client
for s in media_link_page['links']:
cleaned_source_info = csv.dict2row(basic_media_props,
s['source_info'])
cleaned_ref_info = csv.dict2row(basic_media_props, s['ref_info'])
row_string = ','.join(cleaned_source_info) + ',' + ','.join(
cleaned_ref_info) + '\n'
yield row_string
# set up to grab the next page
if 'next' in media_link_page['link_ids']:
link_id = media_link_page['link_ids']['next']
else:
more_media = False
def _topic_media_link_list_by_page_as_csv_row(user_mc_key, topics_id, props, **kwargs):
yield ','.join(props) + '\n' # first send the column names
more_media = True
use_pool = True
link_id = 0
basic_media_props = ['media_id', 'name', 'url']
while more_media:
# fetch one page of data
media_link_page = apicache.topic_media_link_list_by_page(TOOL_API_KEY, topics_id, link_id, **kwargs)
# get the media info for all the media sources
media_src_ids = [str(s['source_media_id']) for s in media_link_page['links']]
media_ref_ids = [str(s['ref_media_id']) for s in media_link_page['links']]
media_src_ids = set(media_src_ids + media_ref_ids) # make it distinct
if use_pool:
# for editing users, add in last scrape and active feed count (if requested)
jobs = [{'key': user_mc_key, 'media_id': mid} for mid in media_src_ids]
pool = Pool(processes=15)
page_media_info = pool.map(_media_info_worker, jobs) # blocks until they are all done
media_lookup = {int(m['media_id']): m for m in page_media_info}
else:
media_lookup = {int(mid): _media_info_worker({'key': user_mc_key, 'media_id': mid})
for mid in media_src_ids}
# connect the link data to the media info data
for link_pair in media_link_page['links']:
link_pair['source_info'] = media_lookup[int(link_pair['source_media_id'])]
link_pair['ref_info'] = media_lookup[int(link_pair['ref_media_id'])]
# stream this page's results to the client
for s in media_link_page['links']:
cleaned_source_info = csv.dict2row(basic_media_props, s['source_info'])
cleaned_ref_info = csv.dict2row(basic_media_props, s['ref_info'])
row_string = ','.join(cleaned_source_info) + ',' + ','.join(cleaned_ref_info) + '\n'
yield row_string
# set up to grab the next page
if 'next' in media_link_page['link_ids']:
link_id = media_link_page['link_ids']['next']
else:
more_media = False
def _topic_story_link_list_by_page_as_csv_row(user_key, topics_id, props, **kwargs):
local_mc = user_admin_mediacloud_client(user_key)
spec_props = [
'source_stories_id', 'source_publish_date', 'source_title', 'source_url', 'source_language', 'source_ap_syndicated',
'source_inlink_count', 'source_outlink_count', 'ref_stories_id', 'ref_publish_date', 'ref_title', 'ref_url', 'ref_language',
'ref_ap_syndicated', 'ref_inlink_count', 'ref_outlink_count'
# 'media_pub_country', 'media_pub_state', 'media_language', 'media_about_country', 'media_media_type'
]
yield u','.join(spec_props) + u'\n' # first send the column names
link_id = 0
more_pages = True
while more_pages:
story_link_page = topic_story_link_list_by_page(user_key, topics_id, link_ids=link_id, **kwargs)
story_src_ids = [str(s['source_stories_id']) for s in story_link_page['links']]
story_ref_ids = [str(s['ref_stories_id']) for s in story_link_page['links']]
story_src_ids = story_src_ids + story_ref_ids
stories_info_list = local_mc.topicStoryList(topics_id, stories_id=story_src_ids)
for s in story_link_page['links']:
for s_info in stories_info_list['stories']:
if s['source_stories_id'] == s_info['stories_id']:
s['source_info'] = s_info
if s['ref_stories_id'] == s_info['stories_id']:
s['ref_info'] = s_info
if 'next' in story_link_page['link_ids']:
link_id = story_link_page['link_ids']['next']
else:
more_pages = False
for s in story_link_page['links']:
cleaned_source_info = csv.dict2row(props, s['source_info'])
cleaned_ref_info = csv.dict2row(props, s['ref_info'])
row_string = u','.join(cleaned_source_info) + ',' + u','.join(cleaned_ref_info) + u'\n'
yield row_string
def _stream_media_by_page(user_mc_key, topics_id, props, **kwargs):
yield ','.join(props) + '\n' # first send the column names
more_media = True
while more_media:
page = apicache.topic_media_list_page(user_mc_key, topics_id, **kwargs)
media = page['media']
for m in media:
row = csv.dict2row(props, m)
row_string = ','.join(row) + '\n'
yield row_string
if 'next' in page['link_ids']:
kwargs['link_id'] = page['link_ids']['next']
more_media = True
else:
more_media = False
def _topic_story_list_by_page_as_csv_row(user_key, topics_id, props, **kwargs):
yield u','.join(props) + u'\n' # first send the column names
link_id = 0
more_pages = True
while more_pages:
page = _topic_story_page_with_media(user_key, topics_id, link_id,
**kwargs)
if 'next' in page['link_ids']:
link_id = page['link_ids']['next']
else:
more_pages = False
for s in page['stories']:
cleaned_row = csv.dict2row(props, s)
row_string = u','.join(cleaned_row) + u'\n'
yield row_string
def _topic_story_link_list_by_page_as_csv_row(user_key, topics_id, props, **kwargs):
local_mc = user_admin_mediacloud_client(user_key)
spec_props = [
'source_stories_id', 'source_publish_date', 'source_title', 'source_url', 'source_language',
'source_ap_syndicated', 'source_inlink_count', 'source_outlink_count', 'ref_stories_id', 'ref_publish_date',
'ref_title', 'ref_url', 'ref_language', 'ref_ap_syndicated', 'ref_inlink_count', 'ref_outlink_count'
# 'media_pub_country', 'media_pub_state', 'media_language', 'media_about_country', 'media_media_type'
]
yield ','.join(spec_props) + '\n' # first send the column names
link_id = 0
more_pages = True
while more_pages:
story_link_page = apicache.topic_story_link_list_by_page(user_key, topics_id, link_id=link_id, **kwargs)
story_src_ids = [str(s['source_stories_id']) for s in story_link_page['links']]
story_ref_ids = [str(s['ref_stories_id']) for s in story_link_page['links']]
story_src_ids = story_src_ids + story_ref_ids
stories_info_list = local_mc.topicStoryList(topics_id, stories_id=story_src_ids)
for s in story_link_page['links']:
for s_info in stories_info_list['stories']:
if s['source_stories_id'] == s_info['stories_id']:
s['source_info'] = s_info
if s['ref_stories_id'] == s_info['stories_id']:
s['ref_info'] = s_info
if 'next' in story_link_page['link_ids']:
link_id = story_link_page['link_ids']['next']
else:
more_pages = False
for s in story_link_page['links']:
cleaned_source_info = csv.dict2row(props, s['source_info'])
cleaned_ref_info = csv.dict2row(props, s['ref_info'])
row_string = ','.join(cleaned_source_info) + ',' + ','.join(cleaned_ref_info) + '\n'
yield row_string
def _topic_story_list_by_page_as_csv_row(user_key, topics_id, props, **kwargs):
yield u','.join(props) + u'\n' # first send the column names
link_id = 0
more_pages = True
while more_pages:
page = _topic_story_page_with_media(user_key, topics_id, link_id, **kwargs)
if 'next' in page['link_ids']:
link_id = page['link_ids']['next']
else:
more_pages = False
for s in page['stories']:
# first foci down to just the readable names
s['subtopics'] = [u"{}: {}".format(f['focal_set_name'], f['name']) for f in s['foci']]
cleaned_row = csv.dict2row(props, s)
row_string = u','.join(cleaned_row) + u'\n'
yield row_string
def _topic_story_list_by_page_as_csv_row(user_key, topics_id, props, **kwargs):
yield ','.join(props) + '\n' # first send the column names
link_id = 0
more_pages = True
while more_pages:
page = _topic_story_page_with_media(user_key, topics_id, link_id, **kwargs)
if 'next' in page['link_ids']:
link_id = page['link_ids']['next']
else:
more_pages = False
for s in page['stories']:
# first foci down to just the readable names
s['subtopics'] = ["{}: {}".format(f['focal_set_name'], f['name']) for f in s['foci']]
cleaned_row = csv.dict2row(props, s)
row_string = ','.join(cleaned_row) + '\n'
yield row_string
def _topic_story_list_by_page_as_csv_row(user_key, topics_id, props, **kwargs):
yield ','.join(props) + '\n' # first send the column names
include_all_url_shares = kwargs[
'include_all_url_shares'] if 'include_all_url_shares' in kwargs else False
story_count = 0
link_id = 0
more_pages = True
yet_to_hit_story_limit = True
has_story_limit = ('story_limit' in kwargs) and (kwargs['story_limit']
is not None)
# page through the story list results, until we run out or we hit the user's desired limit
while more_pages and ((not has_story_limit) or
(has_story_limit and yet_to_hit_story_limit)):
page = _topic_story_page_with_media(user_key, topics_id, link_id,
**kwargs)
if 'next' in page['link_ids']:
link_id = page['link_ids']['next']
else:
more_pages = False
for s in page['stories']:
if include_all_url_shares:
topic_seed_queries = kwargs['topic_seed_queries']
# add in each header col
for item in s['url_sharing_counts']:
seed_query = [
tsq for tsq in topic_seed_queries
if tsq['topic_seed_queries_id'] ==
item['topic_seed_queries_id']
][0]
prefix = platform_csv_column_header_prefix(seed_query)
s[prefix + "post_count"] = item['post_count']
s[prefix + "channel_count"] = item['channel_count']
s[prefix + "author_count"] = item['author_count']
# first foci down to just the readable names
s['subtopics'] = [
"{}: {}".format(f['focal_set_name'], f['name'])
for f in s['foci']
]
cleaned_row = csv.dict2row(props, s)
row_string = ','.join(cleaned_row) + '\n'
yield row_string
story_count += len(page['stories'])
yet_to_hit_story_limit = has_story_limit and (story_count < int(
kwargs['story_limit']))
def _topic_story_list_by_page_as_csv_row(user_key, topics_id, props, **kwargs):
yield ','.join(props) + '\n' # first send the column names
story_count = 0
link_id = 0
more_pages = True
yet_to_hit_story_limit = True
has_story_limit = ('story_limit' in kwargs) and (kwargs['story_limit'] is not None)
while more_pages and ((not has_story_limit) or (has_story_limit and yet_to_hit_story_limit)):
page = _topic_story_page_with_media(user_key, topics_id, link_id, **kwargs)
if 'next' in page['link_ids']:
link_id = page['link_ids']['next']
else:
more_pages = False
for s in page['stories']:
# first foci down to just the readable names
s['subtopics'] = ["{}: {}".format(f['focal_set_name'], f['name']) for f in s['foci']]
cleaned_row = csv.dict2row(props, s)
row_string = ','.join(cleaned_row) + '\n'
yield row_string
story_count += len(page['stories'])
yet_to_hit_story_limit = has_story_limit and (story_count < int(kwargs['story_limit']))
