contextable.py @@ -15,4 +15,4 @@ def dig_web(): return words if __name__ == "__main__": - print dig_web() \ No newline at end of file + print dig_web() analyser.py @@ -24,6 +24,7 @@ class Cluster: return len(self.contextables) + def create_clusters(contextables, clusters, tags): new_clusters = [ Cluster(cluster.tags.union(set([tag]))) for cluster in clusters for tag in tags if tag not in cluster ] @@ -33,7 +34,10 @@ def create_clusters(contextables, clusters, tags): cluster.add(contextable) return new_clusters -def merge_clusters(contextables, clusters, tags, number=5): +def merge_clusters(contextables, number=5): + contextables = contextables[:3] + tags = order_tags(contextables) + clusters = [ Cluster(set([tag])) for tag in tags ] new_clusters = create_clusters(contextables, clusters, tags) for i in range(number-2): @@ -49,11 +53,10 @@ def order_tags(contextables): tags[tag] += 1 else: tags[tag] = 1 - return sort_dict_by_value(tags, limit=160) + return sort_dict_by_value(tags) -def sort_dict_by_value(d, limit): +def sort_dict_by_value(d): items=d.items() backitems=[ [v[1],v[0]] for v in items] backitems.sort() - return [ item[1] for item in reversed(backitems[-limit:]) ] - + return [ item[1] for item in reversed(backitems) ] clusters.py @@ -4,7 +4,9 @@ import flickrapi from flickr_keys import * import time -class Photo: +from contextable import Contextable + +class Photo(Contextable): def __init__(self,url,tags): self.tags = tags self.url = url @@ -13,8 +15,8 @@ class Photo: def relevant_content(self): return "" - def __str__(self): - return "text: %s tags: %s"%(self.url,self.tags) + def __repr__(self): + return "Flick <%s>"%(self.tags) __repr__ = __str__ flickr_api.py @@ -55,9 +55,10 @@ ctxs = [Contextable("python, brhackday, eventomeeter"), # 4 Contextable("brhackday, senac"), Contextable("brhackday, senac"), + # 2 Contextable("brhackday, senac, python"), Contextable("brhackday, senac, python"), - # 4 + # 2 ] @@ -65,8 +66,7 @@ import unittest class TestCluster(unittest.TestCase): def test_brhackday_senac_python_clustering(self): - tags = order_tags(ctxs) - clusters = merge_clusters(ctxs, [ Cluster(set([tag])) for tag in tags ], tags, 3) + clusters = merge_clusters(ctxs, 3) cluster = Cluster(set(["brhackday", "senac", "python"])) for c in clusters: if c == cluster: @@ -77,8 +77,7 @@ class TestCluster(unittest.TestCase): self.assertEquals(len(c), 8) def test_brhackday_eventomeeter_clustering(self): - tags = order_tags(ctxs) - clusters = merge_clusters(ctxs, [ Cluster(set([tag])) for tag in tags ], tags, 2) + clusters = merge_clusters(ctxs, 2) cluster = Cluster(set(["brhackday", "eventomeeter"])) for c in clusters: if c == cluster: @@ -89,8 +88,7 @@ class TestCluster(unittest.TestCase): self.assertEquals(len(c), 27, c.contextables) def test_php_ruby_clustering(self): - tags = order_tags(ctxs) - clusters = merge_clusters(ctxs, [ Cluster(set([tag])) for tag in tags ], tags, 2) + clusters = merge_clusters(ctxs, 2) cluster = Cluster(set(["php", "ruby"])) for c in clusters: if c == cluster: test_clusters.py @@ -1,11 +1,13 @@ import twitter +from contextable import Contextable + ''' IDEAS - Follow tyniurl no twitter + Follow tinyurl no twitter ''' -class Twit: +class Twit(Contextable): def __init__(self,text): self.text = text self.content_tags = {} @@ -14,7 +16,7 @@ class Twit: return self.text def __str__(self): - return "'%s'"%(self.text) + return "Twit <%s>"%(self.text) __repr__ = __str__ twitter_api.py 29bdb216d52c696403cead7cda043f09d9d5f562 Bruno Gola bgola@pedal.(none) http://github.com/Fabs/eventomeeter/commit/fd757fa1bc45db16f56fe9a47205922f3e599b03 fd757fa1bc45db16f56fe9a47205922f3e599b03 2008-11-08T23:01:52-08:00 2008-11-08T23:01:52-08:00 trying to glue things together d5065fea5abe12dea1a0975783d600dda2851077 Bruno Gola bgola@pedal.(none)