def get_keyword_cloud(self, minimum_count=None, limit=50,
minimum_pct=0, maximum_pct=100,
start_time=None, end_time=None, timespan=None):
"""Returns a list of tuples of keywords and their size relative to the
others, as a percentage with set bounds. Sorted alphabetically.
"""
cloud = []
parser = HTMLParser()
if timespan is not None:
start_time = int(time.time()) - int(timespan)
end_time = int(time.time())
else:
if start_time is None and end_time is None:
start_time = int(time.time()) - 3600
end_time = int(time.time())
elif start_time is None:
start_time = int(end_time) - 3600
elif end_time is None:
end_time = int(time.time())
keywords = self._storage.get_keywords(minimum_count=minimum_count,
start_time=start_time, end_time=end_time)
limitval = 0
vals = keywords.values()
vals.sort(reverse=True)
if limit < len(keywords):
vals = vals[:limit]
limitval = vals[-1]
if len(vals) == 1:
totalcount = vals[0]
truncate = 0
else:
totalcount = sum(vals) - (limitval * len(vals))
truncate = limitval
pct_range = maximum_pct - minimum_pct
keys = keywords.keys()
keys.sort()
for keyword in keys:
if limitval and keywords[keyword] < limitval:
continue
pct = ((keywords[keyword] - truncate) / float(totalcount)
* pct_range + minimum_pct)
keyword = parser.escape(keyword)
cloud.append((keyword, round(pct)))
return sorted(cloud)
def list_searches(self, keyword=None, limit=None):
"""List all the search phrases which contain the given keyword, or all
phrases if no keyword given.
"""
phrases = []
sources = map(operator.itemgetter('source'), self._hits)
sources = {}.fromkeys(sources).keys() # make unique
htmlparser = HTMLParser()
for source in sources:
if source.startswith('searches'):
qpos = source.find(': ')
if qpos > 0:
phrase = source[qpos+2:]
if keyword is None or phrase.find(keyword) != -1:
phrases.append(htmlparser.escape(phrase))
if limit is not None:
return phrases[:limit]
return phrases
def list_searches(self, keyword=None, limit=None):
"""List all the search phrases which contain the given keyword, or all
phrases if no keyword given.
"""
phrases = []
urlparser = URLParser(self._conf)
htmlparser = HTMLParser()
if keyword is None:
referrers = self.list_referrers(limit=limit)
else:
referrers = self.list_referrers(limit=limit, refsearch=keyword)
for ref in referrers:
ref = list(urlparse.urlsplit(ref))
if ref is not None:
querydata = urlparser.searchquery(ref)
if querydata is not None:
phrase = querydata[1]
phrases.append(htmlparser.escape(phrase))
return phrases