def main():
args = docopt(__doc__)
c = get_config()
# create new Graph object
g = rdflib.Graph()
print "Loading files: ",
sys.stdout.flush()
# parse data files (could take a while: 4,000,000 triples will take ~10 minutes )
for fn in files:
print fn,
sys.stdout.flush()
with gzip.open(os.path.join(c.ttl_dir, fn)) as f:
result = g.parse(f, format='turtle')
print "\nQuerying..."
# compile query
qres = g.query("""
PREFIX dc: <http://purl.org/dc/elements/1.1/>
PREFIX dcterms: <http://purl.org/dc/terms/>
PREFIX rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#>
PREFIX rdfs: <http://www.w3.org/2000/01/rdf-schema#>
PREFIX lp: <http://purl.org/linkedpolitics/>
PREFIX lpv: <http://purl.org/linkedpolitics/vocabulary/>
PREFIX xml: <http://www.w3.org/XML/1998/namespace>
PREFIX xsd: <http://www.w3.org/2001/XMLSchema#>
PREFIX foaf: <http://xmlns.com/foaf/0.1/>
SELECT ?date ?speaker ?firstname ?lastname ?country ?text
WHERE {
?sessionday dcterms:hasPart ?agendaitem.
?sessionday dc:date ?date.
?agendaitem dcterms:hasPart ?speech.
?speech lpv:speaker ?speaker.
?speaker lpv:countryOfRepresentation ?countryobj.
?countryobj lpv:acronym ?country.
?speaker foaf:givenName ?firstname.
?speaker foaf:familyName ?lastname.
?speech lpv:text ?text.
}
""")
# The query is actually executed now, it takes a while (~3 min)
print "Found %d records" % len(qres)
# Write out csv file
with gzip.open(os.path.join(c.textdb_dir, 'English.csv.gz'), 'wb') as csvfile:
csv_headers = ['Date', 'SpeakerURI', 'Firstname', 'Lastname', 'Country', 'Speech']
speechwriter = csv.writer(csvfile, delimiter=',', quotechar='"', quoting=csv.QUOTE_ALL)
speechwriter.writerow(csv_headers)
for row in progress.bar(qres, label='Writing CSV ', expected_size=len(qres), every=1000):
csv_line = [x.encode('utf8').strip() for x in row]
speechwriter.writerow(csv_line)
print 'Done'
def main():
args = docopt(__doc__)
feature_name = args['<feature_name>']
assert feature_name == 'words'
assert args['<experimentset_name>'] in EXPERIMENT_SETS, '<experimentset_name> must be one of %s' % str(EXPERIMENT_SETS.keys())
c = get_config()
experiment_set = EXPERIMENT_SETS[args['<experimentset_name>']](feature_name=feature_name)
print "Computing foreground group sums using %d cores..." % c.num_cores
pool = Pool(c.num_cores, init_worker)
fg_groups = experiment_set.list_foreground_groups()
cache = {}
try:
for group_name, sum_vector in progress.bar(pool.imap_unordered(ComputeForegroundGroupSumCallable(experiment_set), fg_groups), label="Progress ", expected_size=len(fg_groups)):
cache[group_name] = sum_vector
except KeyboardInterrupt:
print "Terminating pool.."
pool.terminate()
pool.join()
print "Computing background sums..."
bg_groups = experiment_set.list_background_groups()
for g in bg_groups:
sum_vector = experiment_set.compute_background_group_sum(g, cache)
cache[g] = sum_vector
print "Saving sums to ZODB..."
zodb_root = open_zodb(read_only=False)
if getattr(zodb_root, 'group_sums', None) is None:
zodb_root.group_sums = BTrees.OOBTree.OOBTree()
transaction.commit()
if feature_name not in zodb_root.group_sums:
zodb_root.group_sums[feature_name] = BTrees.OOBTree.OOBTree()
transaction.commit()
for k, v in cache.iteritems():
zodb_root.group_sums[feature_name][k] = v
transaction.commit()
print "Creating output db tables..."
create_db(c.resultsdb_url)
session_out = open_db(c.resultsdb_url)
print "Computing overrepresentation using %d cores..." % c.num_cores
exps = experiment_set.list_experiments()
cls = experiment_set.result_table_class()
try:
for fg, bg, results in progress.bar(pool.imap_unordered(ComputeOverrepresentedWordsCallable(experiment_set), exps), label="Progress ", expected_size=len(exps)):
for w, odds, pval in results:
c = cls(foreground_group_name=fg, background_group_name=bg, word=w, odds=odds, pval=pval)
session_out.add(c)
except KeyboardInterrupt:
print "Terminating pool.."
pool.terminate()
pool.join()
print "Committing..."
session_out.commit()
print "Done"
def open_db(db_url = None):
"Returns an initialized Session object. If db_url is not specified, uses get_config().db_url"
if db_url is None:
from talkofeuropedb.config import get_config
db_url = get_config().db_url
e = create_engine(db_url)
Session = sessionmaker(e)
return Session()
def main():
args = docopt(__doc__)
c = get_config()
print "Downloading files into %s" % c.ttl_dir
for g in graph_list:
target_file = os.path.join(c.ttl_dir, g.split('/')[-1] + '.ttl.gz')
source_url = "http://linkedpolitics.ops.few.vu.nl/api/export_graph?graph=%s&mimetype=text%%2Fplain&format=turtle" % urllib.quote(g, '')
print "Downloading %s..." % g
download_gzipped(source_url, target_file)
print "Done"
def open_zodb(config=None, read_only=False):
"Opens a Zope database and returns a root object. If config not specified get_config() is used."
if config is None:
from talkofeuropedb.config import get_config
config = get_config()
storage = ZODB.FileStorage.FileStorage(os.path.join(config.zodb_dir, 'zodb.fs'), read_only=read_only)
db = ZODB.DB(storage)
connection = db.open()
root = connection.root
return root
def main():
args = docopt(__doc__)
c = get_config()
session = open_db()
speeches = session.query(Speech).all()
total_speeches = len(speeches) # For progress bar purposes
print "Computing using %d cores..." % c.num_cores
pool = Pool(c.num_cores, init_worker)
try:
for id, lang in progress.bar(pool.imap_unordered(detect_language, speeches), label='Progress ', expected_size=total_speeches, every=1000):
session.query(Speech).get(id).lang = lang
except KeyboardInterrupt:
print "Terminating pool.."
pool.terminate()
pool.join()
print "Committing..."
session.commit()
num_english_texts = session.query(Speech).filter(Speech.lang == 'en').count()
print "Done. English texts: %d" % num_english_texts
def main():
args = docopt(__doc__)
c = get_config()
e = create_engine(c.db_url)
Base.metadata.drop_all(e)
Base.metadata.create_all(e)
Session = sessionmaker(e)
s = Session()
with gzip.open(os.path.join(c.textdb_dir, 'English.csv.gz'), 'rb') as csv_file:
reader = csv.reader(csv_file, delimiter=',', quotechar='"', quoting=csv.QUOTE_ALL)
reader.next() # Skip header
for row in progress.mill(reader, label='Writing to DB ', expected_size=254253, every=1000):
sp = Speech(date=datetime.strptime(row[0], '%Y-%m-%d'),
speaker_uri=unicode(row[1], 'utf-8'),
first_name=unicode(row[2], 'utf-8'),
last_name=unicode(row[3], 'utf-8'),
country=row[4],
speech=unicode(row[5], 'utf-8'))
s.add(sp)
print "Committing..."
s.commit()
print "Done"
def main():
args = docopt(__doc__)
extractor_name = args['<feature_name>']
extractor = getattr(talkofeuropewords.extract, extractor_name, None)
if extractor is None:
print "Unknown extractor name"
sys.exit(1)
c = get_config()
s = open_db()
print "Preparing ZODB"
zodb_root = open_zodb(read_only=False)
if getattr(zodb_root, 'features', None) is None:
zodb_root.features = BTrees.OOBTree.OOBTree()
transaction.commit()
if extractor_name not in zodb_root.features:
zodb_root.features[extractor_name] = BTrees.OOBTree.OOBTree()
transaction.commit()
runner = TaskRunner(extractor)
print "Querying database..."
speeches = s.query(Speech).filter(Speech.lang == 'en').all()
total_speeches = len(speeches)
print "Computing using %d cores..." % c.num_cores
pool = Pool(c.num_cores, init_worker)
try:
for i, (id, result) in enumerate(progress.bar(pool.imap_unordered(runner, speeches), label='Progress ', expected_size=total_speeches, every=1000), 1):
zodb_root.features[extractor_name][id] = result
if i % 1000 == 0:
transaction.commit()
transaction.commit()
except KeyboardInterrupt:
print "Terminating pool.."
pool.terminate()
pool.join()
print "Done"
def main():
args = docopt(__doc__)
c = get_config()
session = open_db()
print "Finding 5 most active countries"
countries = session.query(Speech.country, func.count(Speech.id)).filter(Speech.lang == 'en').group_by(Speech.country).order_by(desc(func.count(Speech.id))).limit(5).all()
print countries
country_codes = [c[0] for c in countries]
print "Collecting words used by each country using 5 cores"
pool = Pool(5, init_worker)
try:
word_sets = pool.map(country_words, country_codes)
except KeyboardInterrupt:
print "Terminating pool.."
pool.terminate()
pool.join()
print "Collected word sets with sizes: ", map(len, word_sets)
print "Computing intersection..."
word_set = reduce(lambda x, y: x & y, word_sets)
print "Result size: ", len(word_set)
print "Subtracting stopwords..."
nltk.download('stopwords')
langs = ['english', 'dutch', 'french', 'italian', 'portuguese', 'swedish', 'german', 'spanish']
all_stopwords = reduce(lambda x, y: x | y, [set(nltk.corpus.stopwords.words(lng)) for lng in langs])
all_stopwords = set(map(unidecode, all_stopwords))
word_set = word_set - all_stopwords
print "Resulting word set size: ", len(word_set)
print "Saving..."
zodb_root = open_zodb()
zodb_root.all_words = word_set
transaction.commit()
print "Done"
def __init__(self, pval_cutoff=0.01, feature_name='words'):
self.pval_cutoff = pval_cutoff
self.feature_name = feature_name
self.config = get_config()
