def _public_safe_topic_story_count(topics_id, q):
if access_public_topic(topics_id):
total = topic_story_count(TOOL_API_KEY, topics_id, q=add_to_user_query(None))
matching = topic_story_count(TOOL_API_KEY, topics_id, q=add_to_user_query(q)) # force a count with just the query
elif is_user_logged_in():
total = topic_story_count(user_mediacloud_key(), topics_id, q=add_to_user_query(None))
matching = topic_story_count(user_mediacloud_key(), topics_id, q=add_to_user_query(q)) # force a count with just the query
else:
return jsonify({'status': 'Error', 'message': 'Invalid attempt'})
return jsonify({'counts': {'count': matching['count'], 'total': total['count']}})
def _public_safe_topic_story_count(topics_id, q):
if access_public_topic(topics_id):
total = apicache.topic_story_count(TOOL_API_KEY, topics_id, q=apicache.add_to_user_query(None))
# force a count with just the query
matching = apicache.topic_story_count(TOOL_API_KEY, topics_id, q=apicache.add_to_user_query(q))
elif is_user_logged_in():
total = apicache.topic_story_count(user_mediacloud_key(), topics_id, q=apicache.add_to_user_query(None))
# force a count with just the query
matching = apicache.topic_story_count(user_mediacloud_key(), topics_id, q=apicache.add_to_user_query(q))
else:
return jsonify({'status': 'Error', 'message': 'Invalid attempt'})
return jsonify({'counts': {'count': matching['count'], 'total': total['count']}})
def _public_safe_topic_story_count(topics_id, q):
total = apicache.topic_story_count(user_mediacloud_key(),
topics_id,
q=apicache.add_to_user_query(None))
# force a count with just the query
matching = apicache.topic_story_count(user_mediacloud_key(),
topics_id,
q=apicache.add_to_user_query(q))
return jsonify(
{'counts': {
'count': matching['count'],
'total': total['count']
}})
def story_words_csv(topics_id, stories_id):
query = add_to_user_query('stories_id:'+stories_id)
ngram_size = request.args['ngram_size'] if 'ngram_size' in request.args else 1 # default to word count
word_counts = topic_ngram_counts(user_mediacloud_key(), topics_id, ngram_size, q=query,
num_words=WORD_COUNT_DOWNLOAD_NUM_WORDS)
return csv.stream_response(word_counts, WORD_COUNT_DOWNLOAD_COLUMNS,
'topic-{}-story-{}-sampled-ngrams-{}-word'.format(topics_id, stories_id, ngram_size))
def _matching_ratio(topics_id, query_clause):
total = apicache.topic_story_count(user_mediacloud_key(), topics_id)
sub_query_clause = None
if query_clause:
sub_query_clause = apicache.add_to_user_query(query_clause)
matching = apicache.topic_story_count(user_mediacloud_key(), topics_id, q=sub_query_clause)
return {'count': matching['count'], 'total': total['count']}
def media_words_csv(topics_id, media_id):
query = apicache.add_to_user_query('media_id:'+media_id)
ngram_size = request.args['ngram_size'] if 'ngram_size' in request.args else 1 # default to word count
word_counts = apicache.topic_ngram_counts(user_mediacloud_key(), topics_id, ngram_size=ngram_size, q=query,
num_words=WORD_COUNT_DOWNLOAD_NUM_WORDS,
sample_size=WORD_COUNT_DOWNLOAD_SAMPLE_SIZE)
return csv.stream_response(word_counts, apicache.WORD_COUNT_DOWNLOAD_COLUMNS,
'topic-{}-media-{}-sampled-ngrams-{}-word'.format(topics_id, media_id, ngram_size))
def topic_words_csv(topics_id):
query = apicache.add_to_user_query(None)
sample_size = request.args['sample_size'] if 'sample_size' in request.args else WORD_COUNT_SAMPLE_SIZE
ngram_size = request.args['ngram_size'] if 'ngram_size' in request.args else 1 # default to word count
word_counts = apicache.topic_ngram_counts(user_mediacloud_key(), topics_id, ngram_size=ngram_size, q=query,
num_words=WORD_COUNT_DOWNLOAD_NUM_WORDS, sample_size=sample_size)
return csv.stream_response(word_counts, apicache.WORD_COUNT_DOWNLOAD_COLUMNS,
'topic-{}-sampled-ngrams-{}-word'.format(topics_id, ngram_size))
def topic_word_usage_sample(topics_id, word):
# gotta respect the manual query if there is one
q = apicache.add_to_user_query(word)
# need to use tool API key here because non-admin users can't pull sentences
results = apicache.topic_sentence_sample(TOOL_API_KEY, topics_id, sample_size=1000, q=q)
# only pull the 5 words before and after so we aren't leaking full content to users
fragments = [_sentence_fragment_around(word, s['sentence']) for s in results if s['sentence'] is not None]
fragments = [f for f in fragments if f is not None]
return jsonify({'fragments': fragments})
def topic_word_usage_sample(topics_id, word):
# gotta respect the manual query if there is one
q = apicache.add_to_user_query(word)
# need to use tool API key here because non-admin users can't pull sentences
results = apicache.topic_sentence_sample(TOOL_API_KEY, topics_id, sample_size=1000, q=q)
# TODO: only pull the 5 words before and after so we
fragments = [_sentence_fragment_around(word, s['sentence']) for s in results if s['sentence'] is not None]
fragments = [f for f in fragments if f is not None]
return jsonify({'fragments': fragments})
def topic_word_associated_words_csv(topics_id, word):
query = apicache.add_to_user_query(word)
ngram_size = request.args[
'ngram_size'] if 'ngram_size' in request.args else 1 # default to word count
word_counts = apicache.topic_ngram_counts(user_mediacloud_key(),
topics_id,
ngram_size=ngram_size,
q=query)
return csv.stream_response(
word_counts, apicache.WORD_COUNT_DOWNLOAD_COLUMNS,
'topic-{}-{}-sampled-ngrams-{}-word'.format(topics_id, word,
ngram_size))
def media_words(topics_id, media_id):
query = apicache.add_to_user_query('media_id:' + media_id)
word_list = apicache.topic_word_counts(user_mediacloud_key(),
topics_id,
q=query)[:100]
return jsonify(word_list)
def media_words(topics_id, media_id):
query = apicache.add_to_user_query('media_id:'+media_id)
word_list = apicache.topic_word_counts(user_mediacloud_key(), topics_id, q=query)[:100]
return jsonify(word_list)
def topic_word_associated_words(topics_id, word):
query = apicache.add_to_user_query(word)
response = apicache.topic_word_counts(user_mediacloud_key(), topics_id, q=query)[:100]
return jsonify(response)
def topic_word_associated_words(topics_id, word):
query = apicache.add_to_user_query(word)
response = apicache.topic_word_counts(user_mediacloud_key(), topics_id, q=query)[:100]
return jsonify(response)
def media_words_csv(topics_id, media_id):
query = add_to_user_query('media_id:'+media_id)
word_list = topic_word_counts(user_mediacloud_key(), topics_id, q=query)
props = ['term', 'stem', 'count']
return csv.stream_response(word_list, props, 'media-'+str(media_id)+'-words')
