def __init__(self, test_docs_dir="./test"):
self.test_docs_dir = test_docs_dir
self.files = []
self.index = NearDuplicatesIndex()
# Calculate near-duplicates index
# Try to connect
try:
conn=psycopg2.connect("dbname='djangology' user='******' password=''")
except:
print "I am unable to connect to the database."
cur = conn.cursor()
try:
cur.execute("""SELECT * from dj_document""")
except:
print "I can't SELECT from dj_document"
rows = cur.fetchall()
#print "\nRows: \n"
for row in rows:
#print " ", row[1]
self.index.append(row[1], row[0])
self.files.append(row[0])
cur.close()
def __init__(self, test_docs_dir="./test"):
self.test_docs_dir = test_docs_dir
self.files = [
d for d in os.listdir(test_docs_dir)
if os.path.isfile(os.path.join(test_docs_dir, d)) and d[0] != "."
]
self.index = NearDuplicatesIndex()
# Calculate near-duplicates index
for file in self.files:
filename = self.filename(file)
with open(filename) as f:
doc = f.read().strip().strip(
",.!|&-_()[]<>{}/\"'").strip().split(" ")
self.index.append(doc, filename)
class Detector:
def __init__(self, test_docs_dir="./test"):
self.test_docs_dir = test_docs_dir
self.files = [
d for d in os.listdir(test_docs_dir)
if os.path.isfile(os.path.join(test_docs_dir, d)) and d[0] != "."
]
self.index = NearDuplicatesIndex()
# Calculate near-duplicates index
for file in self.files:
filename = self.filename(file)
with open(filename) as f:
doc = f.read().strip().strip(
",.!|&-_()[]<>{}/\"'").strip().split(" ")
self.index.append(doc, filename)
# Public: returns the full relative path from the base dir of the project
# to the filename input
#
# filename - the filename relative to the test directory
#
# Returns full filename (including test directory)
def filename(self, filename):
return "%s/%s" % (self.test_docs_dir, filename)
# Public: checks for near-duplicates in the set of files based on jaccard
# coefficient threshold of 0.5
#
# Returns a string containing formatted names and coefficients of
# documents whose jaccard coefficient is greater than 0.5
def check_for_duplicates(self):
matches = []
for indx1, f1 in enumerate(self.files):
file1 = self.filename(f1)
for indx2, f2 in enumerate(self.files[indx1 + 1:]):
file2 = self.filename(f2)
jaccard = self.index.get_jaccard(file1, file2)
if jaccard > 0.5:
matches.append(
"%s and %s are near-duplicates, with Jaccard value of %0.3f."
% (f1, f2, jaccard))
return "\n".join(matches)
def __init__(self, test_docs_dir="./test"):
self.test_docs_dir = test_docs_dir
self.files = [d for d in os.listdir(test_docs_dir) if os.path.isfile(os.path.join(test_docs_dir, d)) and d[0] != "." ]
self.index = NearDuplicatesIndex()
# Calculate near-duplicates index
for file in self.files:
filename = self.filename(file)
with open(filename) as f:
doc = f.read().strip().strip(",.!|&-_()[]<>{}/\"'").strip().split(" ")
self.index.append(doc, filename)
class Detector:
def __init__(self, test_docs_dir="./test"):
self.test_docs_dir = test_docs_dir
self.files = [d for d in os.listdir(test_docs_dir) if os.path.isfile(os.path.join(test_docs_dir, d)) and d[0] != "." ]
self.index = NearDuplicatesIndex()
# Calculate near-duplicates index
for file in self.files:
filename = self.filename(file)
with open(filename) as f:
doc = f.read().strip().strip(",.!|&-_()[]<>{}/\"'").strip().split(" ")
self.index.append(doc, filename)
# Public: returns the full relative path from the base dir of the project
# to the filename input
#
# filename - the filename relative to the test directory
#
# Returns full filename (including test directory)
def filename(self, filename):
return "%s/%s" % (self.test_docs_dir, filename)
# Public: checks for near-duplicates in the set of files based on jaccard
# coefficient threshold of 0.5
#
# Returns a string containing formatted names and coefficients of
# documents whose jaccard coefficient is greater than 0.5
def check_for_duplicates(self):
matches = []
for indx1, f1 in enumerate(self.files):
file1 = self.filename(f1)
for indx2, f2 in enumerate(self.files[indx1+1:]):
file2 = self.filename(f2)
jaccard = self.index.get_jaccard(file1, file2)
if jaccard > 0.5:
matches.append("%s and %s are near-duplicates, with Jaccard value of %0.3f." % (f1, f2, jaccard))
return "\n".join(matches)
def setUp(self):
self.docs = []
self.docs.append(['this','is','a','document'])
self.docs.append(['this','is','b','document'])
self.index = NearDuplicatesIndex()
class TestNearDuplicatesIndex(unittest.TestCase):
def setUp(self):
self.docs = []
self.docs.append(['this','is','a','document'])
self.docs.append(['this','is','b','document'])
self.index = NearDuplicatesIndex()
def test_should_allow_to_append_documents(self):
self.index.append(self.docs[0], 'doc1')
self.index.append(self.docs[1], 'doc2')
self.assertEqual(len(self.index), 2)
def test_should_raise_an_error_when_docname_is_duplicated(self):
self.index.append(self.docs[0], 'doc1')
with self.assertRaises(Exception):
self.index.append(self.docs[1], 'doc1')
def test_should_calculate_jaccard_coefficient(self):
self.index.append(self.docs[0], 'doc1')
self.index.append(self.docs[0], 'doc2')
self.assertEqual(self.index.get_jaccard('doc1', 'doc2'), 1.0)
def test_should_raise_an_error_if_document_does_not_exist(self):
with self.assertRaises(Exception):
self.index.get_jaccard('doc1', 'doc3')
def test_should_append_a_document_if_its_not_duplicated(self):
self.index.append(self.docs[0], 'doc1')
self.index.appendif(self.docs[1], 'doc2', 1.0)
self.assertEqual(len(self.index), 2)
def test_should_not_append_a_document_if_its_duplicated(self):
self.index.append(self.docs[0], 'doc1')
self.index.appendif(self.docs[1], 'doc2', -1.0)
self.assertEqual(len(self.index), 1)
class Detector:
def __init__(self, test_docs_dir="./test"):
self.test_docs_dir = test_docs_dir
self.files = []
self.index = NearDuplicatesIndex()
# Calculate near-duplicates index
# Try to connect
try:
conn=psycopg2.connect("dbname='djangology' user='******' password=''")
except:
print "I am unable to connect to the database."
cur = conn.cursor()
try:
cur.execute("""SELECT * from dj_document""")
except:
print "I can't SELECT from dj_document"
rows = cur.fetchall()
#print "\nRows: \n"
for row in rows:
#print " ", row[1]
self.index.append(row[1], row[0])
self.files.append(row[0])
cur.close()
#for file in self.files:
# filename = self.filename(file)
# with open(filename) as f:
# doc = f.read().strip().strip(",.!|&-_()[]<>{}/\"'").strip().split(" ")
#self.index.append(doc, filename)
# Public: returns the full relative path from the base dir of the project
# to the filename input
#
# filename - the filename relative to the test directory
#
# Returns full filename (including test directory)
def filename(self, filename):
return "%s/%s" % (self.test_docs_dir, filename)
# Public: checks for near-duplicates in the set of files based on jaccard
# coefficient threshold of 0.5
#
# Returns a string containing formatted names and coefficients of
# documents whose jaccard coefficient is greater than 0.5
def check_for_duplicates(self):
matches = []
for indx1, f1 in enumerate(self.files):
for indx2, f2 in enumerate(self.files[indx1+1:]):
jaccard = self.index.get_jaccard(f1, f2)
if jaccard > 0.5:
matches.append("%s and %s are near-duplicates, with Jaccard value of %0.3f." % (f1, f2, jaccard))
return "\n".join(matches)
