def test_errors_property(self):
import _io
with _io.open(self.tmpfile, "w") as f:
assert f.errors == "strict"
with _io.open(self.tmpfile, "w", errors="replace") as f:
assert f.errors == "replace"
def test_attributes(tempfile):
import warnings
with _io.open(tempfile, "wb", buffering=0) as f:
assert f.mode == "wb"
with warnings.catch_warnings(record=True) as l:
warnings.simplefilter("always")
with _io.open(tempfile, "U") as f:
assert f.name == tempfile
assert f.buffer.name == tempfile
assert f.buffer.raw.name == tempfile
assert f.mode == "U"
assert f.buffer.mode == "rb"
assert f.buffer.raw.mode == "rb"
assert isinstance(l[0].message, DeprecationWarning)
with _io.open(tempfile, "w+") as f:
assert f.mode == "w+"
assert f.buffer.mode == "rb+"
assert f.buffer.raw.mode == "rb+"
with _io.open(f.fileno(), "wb", closefd=False) as g:
assert g.mode == "wb"
assert g.raw.mode == "wb"
assert g.name == f.fileno()
assert g.raw.name == f.fileno()
def test_errors_property(self):
import _io
with _io.open(self.tmpfile, "w") as f:
assert f.errors == "strict"
with _io.open(self.tmpfile, "w", errors="replace") as f:
assert f.errors == "replace"
def open_as_bytes_stream(filename):
''' If filesize < TWO_GB, read whole file as BytesIO object '''
filesize = os.path.getsize(filename)
if filesize < TWO_GB:
with open(filename, 'rb') as f:
return BytesIO(f.read(filesize))
else:
return open(filename, 'rb', buffering=bestIOBufferSize)
def test_open(tempfile):
f = _io.open(tempfile, "rb")
assert f.name.endswith('tempfile')
assert f.mode == 'rb'
f.close()
with _io.open(tempfile, "rt") as f:
assert f.mode == "rt"
def loadjson(name, objHook=None) -> dict:
""" Load json from file and return dict """
try:
with _io.open(name, encoding='utf-8', errors='replace') as f:
return json.loads(f.read(), encoding='utf-8', object_hook=objHook)
except Exception as e:
if 'BOM' in str(e):
with _io.open(name, encoding='utf-8-sig', errors='replace') as f:
return json.loads(f.read(), encoding='utf-8-sig', object_hook=objHook)
def loadjson(name, objHook=None) -> dict:
''' Load json from file return dict '''
try:
with _io.open(name, encoding='utf-8', errors='replace') as f:
return json.loads(f.read(), encoding='utf-8', object_hook=objHook)
except Exception as e:
if 'BOM' in str(e):
with _io.open(name, encoding='utf-8-sig', errors='replace') as f:
return json.loads(f.read(), encoding='utf-8-sig', object_hook=objHook)
def read(self, fn, mode='rb', encoding='utf-8-sig', errors='replace'):
bufsize = bestIOBufferSize
parser_ = self.parse_line
if 'b' in mode:
for line in open(fn, mode, buffering=bufsize):
parser_(line.rstrip())
else:
for line in open(fn, mode, buffering=bufsize, encoding=encoding, errors=errors):
parser_(line.rstrip())
def test_array_write(tempfile):
a = array.array('i', range(10))
n = len(a.tostring())
with _io.open(tempfile, "wb", 0) as f:
res = f.write(a)
assert res == n
with _io.open(tempfile, "wb") as f:
res = f.write(a)
assert res == n
def test_truncate(tempfile):
with _io.open(tempfile, "w+") as f:
f.write("abc")
with _io.open(tempfile, "w+") as f:
f.truncate()
with _io.open(tempfile, "r+") as f:
res = f.read()
assert res == ""
def createFile(self, path, name):
"""
Metoda kreira fajl sa zadatim imenom u zadatom direktorijumu
Args:
path: putanje do direktorijuma u kom se kreira fajl
name: ime fajla koji ce biti kreiran
"""
open(os.path.join(path, name), mode='w').close()
def test_array_write(self):
import _io, array
a = array.array(b'i', range(10))
n = len(a.tostring())
with _io.open(self.tmpfile, "wb", 0) as f:
res = f.write(a)
assert res == n
with _io.open(self.tmpfile, "wb") as f:
res = f.write(a)
assert res == n
def test_opener(tempfile):
import os
with _io.open(tempfile, "w") as f:
f.write("egg\n")
fd = os.open(tempfile, os.O_RDONLY)
def opener(path, flags):
return fd
with _io.open("non-existent", "r", opener=opener) as f:
assert f.read() == "egg\n"
def test_array_write(self):
import _io, array
a = array.array('i', range(10))
n = len(a.tostring())
with _io.open(self.tmpfile, "wb", 0) as f:
res = f.write(a)
assert res == n
with _io.open(self.tmpfile, "wb") as f:
res = f.write(a)
assert res == n
def test_truncate(self):
import _io
with _io.open(self.tmpfile, "w+") as f:
f.write("abc")
with _io.open(self.tmpfile, "w+") as f:
f.truncate()
with _io.open(self.tmpfile, "r+") as f:
res = f.read()
assert res == ""
def test_truncate(self):
import _io
with _io.open(self.tmpfile, "w+") as f:
f.write(u"abc")
with _io.open(self.tmpfile, "w+") as f:
f.truncate()
with _io.open(self.tmpfile, "r+") as f:
res = f.read()
assert res == ""
def test_seek_append_bom(self):
# Same test, but first seek to the start and then to the end
import _io, os
filename = self.tmpfile + '_x3'
for charset in ('utf-8-sig', 'utf-16', 'utf-32'):
with _io.open(filename, 'w', encoding=charset) as f:
f.write('aaa')
with _io.open(filename, 'a', encoding=charset) as f:
f.seek(0)
f.seek(0, os.SEEK_END)
f.write('xxx')
with _io.open(filename, 'rb') as f:
assert f.read() == 'aaaxxx'.encode(charset)
def gen_cfg(condition):
token = condition.split('-')
f_condition = open('condition.yaml')
data = ordered_yaml_load('condition.yaml')
data_enc = data[condition]['encoder']
data_dec = data[condition]['decoder']
f_condition.close()
f_ratepoint = open('ratepoint.yaml')
rps = yaml.load(f_ratepoint, Loader=yaml.Loader)
f_ratepoint.close()
f_sequence = open('sequences.yaml')
seqs = yaml.load(f_sequence, Loader=yaml.Loader)
f_sequence.close()
for seq_name in seqs:
seq = seqs[seq_name]
rps_geom = rps[token[1]][seq['category']]
rps_attr = rps[token[2]]
for r in rps_attr:
dir = '{}/{}/{}'.format(condition, seq_name, r)
mkdir_p(dir)
# encoder cfg
data_r = data_enc
if len(rps_geom) == 1:
rate_geom = rps_geom['']
else:
rate_geom = rps_geom[r]
rate_attr = rps_attr[r]
for key in rate_geom:
if key in data_r:
data_r[key] = rate_geom[key]
for key in rate_attr:
if key in data_r:
data_r[key] = rate_attr[key]
for key in seq:
if key in data_r:
data_r[key] = seq[key]
with open('{}/encoder.cfg'.format(dir), 'w') as f:
ordered_yaml_dump(data_r, f, default_flow_style=False)
# decoder cfg
data_r = data_dec
with open('{}/decoder.cfg'.format(dir), 'w') as f:
ordered_yaml_dump(data_r, f, default_flow_style=False)
def file2matrix(filename):
fr = open(filename)
numberOfLines = len(fr.readlines()) #get the number of lines in the file
returnMat = zeros((numberOfLines, 3)) #prepare matrix to return
classLabelVector = [] #prepare labels return
fr = open(filename)
index = 0
for line in fr.readlines():
line = line.strip()
listFromLine = line.split('\t')
returnMat[index, :] = listFromLine[0:3]
classLabelVector.append(int(listFromLine[-1]))
index += 1
return returnMat, classLabelVector
def file2matrix(filename):
fr = open(filename)
numberOfLines = len(fr.readlines()) #get the number of lines in the file
returnMat = zeros((numberOfLines,3)) #prepare matrix to return
classLabelVector = [] #prepare labels return
fr = open(filename)
index = 0
for line in fr.readlines():
line = line.strip()
listFromLine = line.split('\t')
returnMat[index,:] = listFromLine[0:3]
classLabelVector.append(int(listFromLine[-1]))
index += 1
return returnMat,classLabelVector
def test_newlines_attr(tempfile):
with _io.open(tempfile, "r") as f:
assert f.newlines is None
with _io.open(tempfile, "wb") as f:
f.write(b"hello\nworld\n")
with _io.open(tempfile, "r") as f:
res = f.readline()
assert res == "hello\n"
res = f.readline()
assert res == "world\n"
assert f.newlines == "\n"
assert type(f.newlines) is str
def test_append_bom(tempfile):
# The BOM is not written again when appending to a non-empty file
for charset in ["utf-8-sig", "utf-16", "utf-32"]:
with _io.open(tempfile, "w", encoding=charset) as f:
f.write("aaa")
pos = f.tell()
with _io.open(tempfile, "rb") as f:
res = f.read()
assert res == "aaa".encode(charset)
with _io.open(tempfile, "a", encoding=charset) as f:
f.write("xxx")
with _io.open(tempfile, "rb") as f:
res = f.read()
assert res == "aaaxxx".encode(charset)
def test_seek_bom(self):
# The BOM is not written again when seeking manually
import _io
filename = self.tmpfile + '_x3'
for charset in ('utf-8-sig', 'utf-16', 'utf-32'):
with _io.open(filename, 'w', encoding=charset) as f:
f.write('aaa')
pos = f.tell()
with _io.open(filename, 'r+', encoding=charset) as f:
f.seek(pos)
f.write('zzz')
f.seek(0)
f.write('bbb')
with _io.open(filename, 'rb') as f:
assert f.read() == 'bbbzzz'.encode(charset)
def create_load_data_directory (triplesFileName, rootDir, outputDir):
with open(triplesFileName, 'r') as tf:
for line in tf:
opHolder,opinion,topic = line.split(',')
opHolder = sub(' ', '_', opHolder.strip())
opinion = opinion.strip()
topic = sub(' ', '_', topic.strip())
try:
ohfname = rootDir+'/opinion_holders/'+opHolder
tfname = rootDir+'/topics/'+topic
outfname = outputDir+'/'+opinion+'/'+opHolder+'__'+topic
with codecs.open(ohfname, 'r',errors='ignore') as ohFile, codecs.open(tfname,'r', errors='ignore') as tFile:
content = ''
for line1 in ohFile:
content = content+line1
for line2 in tFile:
content = content+line2
with codecs.open(outfname, 'w', errors='ignore') as outFile:
print ('Writing...'+ opHolder+ ', '+ topic)
outFile.write(content)
outFile.close()
except Exception:
print ("Something wrong with either "+ opHolder+ ' or '+
topic + '...skipping...')
pass
def __init__(self, fileName=None):
'''
Constructor
'''
self.vertex = {}
self.edge = {}
self.setOfEdges = set()
if fileName != None:
file = open(fileName, 'r')
# iterates through the lines of the file
for line in file:
splittedLine = line.split(None)
if len(splittedLine) >= 3:
# Vertex
if splittedLine[0] == 'V':
self.vertex[splittedLine[1]] = [
int(splittedLine[2]), True
]
self.edge[splittedLine[1]] = {}
# Edge
elif splittedLine[0] == 'E':
self.edge[splittedLine[1]][splittedLine[2]] = [
int(splittedLine[3]),
int(splittedLine[3])
]
self.edge[splittedLine[2]][splittedLine[1]] = [
int(splittedLine[3]),
int(splittedLine[3])
]
self.setOfEdges.add((splittedLine[1], splittedLine[2]))
def dumpjson(obj:dict, name=None) -> str:
''' Dump json(dict) to file '''
str = json.dumps(obj, indent=4, ensure_ascii=False)
if name:
with _io.open(name, 'w') as f:
f.write(str)
return str
def UpdateStopWords(self, stopWordsFileName):
# Open the stop words file name
fin = open(stopWordsFileName, 'r', encoding='utf-8')
# Read the file line by line
for line in fin:
# The word is the first entry in the line
rawWord = line.split()[0]
# Stem the word if enabled
if (self.enableStemming == "true"):
word = self.stemmer.stem(rawWord)
else:
word = rawWord
# Insert the word only if it doesn't exist before in the model
if not word in self.stopWords:
# Put the word frequency as 1 since it's the first incident
self.stopWords[word] = 1
else:
# Increment the frequency
self.stopWords[word] += 1
# Close the file
fin.close()
def dumpjson(obj:dict, name=None) -> str:
""" Dump json(dict) to file """
jstr = json.dumps(obj, indent=4, ensure_ascii=False)
if name:
with _io.open(name, 'w') as f:
f.write(jstr)
return jstr
def __init__(self, fileName=None):
'''
Constructor
'''
self.vertex = {};
self.edge = {};
self.setOfEdges = set();
if fileName != None:
file = open(fileName, 'r');
# iterates through the lines of the file
for line in file:
splittedLine = line.split(None);
if len(splittedLine) >= 3:
# Vertex
if splittedLine[0] == 'V':
self.vertex[splittedLine[1]] = [int(splittedLine[2]), True];
self.edge[splittedLine[1]] = {};
# Edge
elif splittedLine[0] == 'E':
self.edge[splittedLine[1]][splittedLine[2]] = [int(splittedLine[3]), int(splittedLine[3])];
self.edge[splittedLine[2]][splittedLine[1]] = [int(splittedLine[3]), int(splittedLine[3])];
self.setOfEdges.add((splittedLine[1], splittedLine[2]));
def test_append_bom(self):
import _io
# The BOM is not written again when appending to a non-empty file
for charset in ["utf-8-sig", "utf-16", "utf-32"]:
with _io.open(self.tmpfile, "w", encoding=charset) as f:
f.write("aaa")
pos = f.tell()
with _io.open(self.tmpfile, "rb") as f:
res = f.read()
assert res == "aaa".encode(charset)
with _io.open(self.tmpfile, "a", encoding=charset) as f:
f.write("xxx")
with _io.open(self.tmpfile, "rb") as f:
res = f.read()
assert res == "aaaxxx".encode(charset)
def test_newlines_attr(self):
import _io
with _io.open(self.tmpfile, "r") as f:
assert f.newlines is None
with _io.open(self.tmpfile, "wb") as f:
f.write("hello\nworld\n")
with _io.open(self.tmpfile, "r") as f:
res = f.readline()
assert res == "hello\n"
res = f.readline()
assert res == "world\n"
assert f.newlines == "\n"
assert type(f.newlines) is unicode
def __init__(self, filename):
self.filename = filename
with _io.open(filename, 'rb') as wav:
self.__raw_header = wav.read(72)
self.__struct_header = struct.unpack('4sL4s4sLHHLLHH4sL', self.__raw_header)
self.chunk_id, \
self.chunk_size, \
self.format, \
self.sub_chunk_id_1, \
self.sub_chunk_size_1, \
self.format, \
self.num_channel, \
self.sample_rate_bit, \
self.byte_rate, \
self.block_align, \
self.bit_per_sample, \
self.sub_chunk_id_2, \
self.sub_chunk_size_2 = self.__struct_header
self.is_compressed = False if self.format == 1 else True
self.byte_per_sample = int(self.bit_per_sample / 8)
self.data = wav.read()
self.len = len(self.data)
self.__struct_mask = {1:'B',2:'H',4:'I',8:'Q'}[self.byte_per_sample]
self.__struct_mask = self.__struct_mask * self.num_channel
self.__sample_bytes = self.byte_per_sample * self.num_channel
def create_stdio(fd, writing, name, encoding, errors, unbuffered):
import _io
# stdin is always opened in buffered mode, first because it
# shouldn't make a difference in common use cases, second because
# TextIOWrapper depends on the presence of a read1() method which
# only exists on buffered streams.
buffering = 0 if unbuffered and writing else -1
mode = 'w' if writing else 'r'
try:
buf = _io.open(fd, mode + 'b', buffering, closefd=False)
except OSError as e:
if e.errno != errno.EBADF:
raise
return None
raw = buf.raw if buffering else buf
raw.name = name
# We normally use newline='\n' below, which turns off any translation.
# However, on Windows (independently of -u), then we must enable
# the Universal Newline mode (set by newline = None): on input, \r\n
# is translated into \n; on output, \n is translated into \r\n.
# We must never enable the Universal Newline mode on POSIX: CPython
# never interprets '\r\n' in stdin as meaning just '\n', unlike what
# it does if you explicitly open a file in text mode.
newline = None if sys.platform == 'win32' else '\n'
stream = _io.TextIOWrapper(buf,
encoding,
errors,
newline=newline,
line_buffering=unbuffered or raw.isatty())
stream.mode = mode
return stream
def read_csv(name: str,
delim=',',
withhead=False,
strip=True,
convert=None,
encoding='utf-8',
errors='strict') -> ([], []):
""" Read csv file, return head and body as list """
if convert and strip:
make = lambda s: convert(s.strip(b'" ').decode(encoding, errors))
elif convert:
make = lambda s: convert(s.decode(encoding, errors))
elif strip:
make = lambda s: s.strip(b'" ').decode(encoding, errors)
else:
make = lambda s: s.decode(encoding, errors)
head, body = [], []
delim = delim.encode()
if isinstance(name, str):
name = _io.open(name, 'rb')
if withhead:
line = next(name).rstrip()
head = [i.decode() for i in line.rstrip(b',').split(delim)]
for line in name:
body.append([make(i) for i in line.rstrip(b'"\r\n, ').split(delim)])
return head, body
def _loadFile(self):
try:
f = open(self._fileName, "r")
s = f.readline()
while len(s)>1:
tok = s.split(",")
"""
function to turn string into entity
"""
if len(tok) == 3:
tok[2] = tok[2].split("\n")
book = Book(int(tok[0]), tok[1], tok[2][0])
else:
tok[3] = tok[3].split("\n")
book = Book(int(tok[0]), tok[1], tok[2], tok[3][0])
BookRepository.add(self, book)
s = f.readline()
except Exception as e:
raise RepoError("Error reading input file: " + str(e))
finally:
f.close()
def create_data_file():
global x_max, x_min, x_step, n_max, n_min, n_step, nb_opr, current_opr, outfilename, iSWorking, label_stat_prc
out = open(outfilename, "w")
#out.write("x_value\tn_value\tfast_way_time\tnormal_way_time\n")
x = x_min
n = n_min
while (x <= x_max):
while (n <= n_max):
t1 = power_fast_calculted_time(x, n)
t2 = power(x, n)
line = str(x) + "\t" + str(n) + "\t" + str(t1) + "\t" + str(
t2) + "\n"
out.write(line)
n = n + n_step
current_opr = current_opr + 1
if (iSWorking == False):
label_stat_prc.config(bg="red")
messagebox.showerror("Work", "STOP")
return
out.write("\n")
x = x + x_step
n = n_min
out.close()
iSWorking = False
messagebox.showinfo("Work", "Job Done [%s], thank you" % current_opr)
def test_newlines_attr(self):
import _io
with _io.open(self.tmpfile, "r") as f:
assert f.newlines is None
with _io.open(self.tmpfile, "wb") as f:
f.write(b"hello\nworld\n")
with _io.open(self.tmpfile, "r") as f:
res = f.readline()
assert res == "hello\n"
res = f.readline()
assert res == "world\n"
assert f.newlines == "\n"
assert type(f.newlines) is str
def test_fileread_line_bin(self):
f = _io.open('test_io.py', 'rb')
f.readline() # skip 1st line
b = f.readline()
self.assertEqual(b,
b'#\xe3\x81\x82\xe3\x81\x84\xe3\x81\x86\xe3\x81\x88\xe3\x81\x8a\n')
f.close()
def test_issue1902(tempfile):
with _io.open(tempfile, 'w+b', 4096) as f:
f.write(b'\xff' * 13569)
f.flush()
f.seek(0, 0)
f.read(1)
f.seek(-1, 1)
f.write(b'')
def SaveModel(self):
# You must close and open to append to the binary file
# Open the serialization file
serializationFile = open(self.languageModelSerializationFileName, 'wb')
# Save the model
pickle.dump(self.languageModel, serializationFile)
# Open the serialization file
serializationFile.close()
def test_chunk_size(self):
import _io
with _io.open(self.tmpfile) as f:
assert f._CHUNK_SIZE >= 1
f._CHUNK_SIZE = 4096
assert f._CHUNK_SIZE == 4096
raises(ValueError, setattr, f, "_CHUNK_SIZE", 0)
def test_chunk_size(self):
import _io
with _io.open(self.tmpfile) as f:
assert f._CHUNK_SIZE >= 1
f._CHUNK_SIZE = 4096
assert f._CHUNK_SIZE == 4096
raises(ValueError, setattr, f, "_CHUNK_SIZE", 0)
def write_module(module, modulename, write_path):
global _f
with io.open(write_path + '/' + modulename + '.txt', 'w',
encoding="ascii") as f:
_f = f
p(handle_module(module, modulename))
_f = None
def findjsonkeyval(self, file_input_path, json_key, data_key_num):
int_data_key_num = int(data_key_num)
json_data= open(file_input_path,'r')
json_items= ijson.items(json_data,json_key)
#column=list(json_items)
list_convert=list(json_items)
output = list_convert[int_data_key_num]
return output
def readFileContent(self,file_name,read_op="r+"):
with open(file_name,read_op) as f:
lines=f.readlines()
return lines
def __init__(self):
self.pageNo=1
self.houseItems=[]
self.houseItem=[]
self.xiaoquItems=[]
self.keepdoing=True
self.fp=open('houseInfo.txt','w',encoding="utf-8")
self.conn=pymysql.connect(host='localhost',user='******',password='',db='app_',charset='utf8mb4',cursorclass=pymysql.cursors.DictCursor)
def split_by_lines(self, nline:int):
''' Split by every `nline` lines '''
nfile = self.__howmany(self._nline, nline)
newnames = self.splited_names(nfile)
inames = iter(newnames)
fout = open(next(inames), 'wb')
n = 1
with open_as_bytes_stream(self._fname) as stream:
for line in stream:
if not (n % nline):
fout.close()
fout = open(next(inames), 'wb')
n += 1
fout.write(line)
fout.close()
return newnames
def test_issue1902(self):
import _io
with _io.open(self.tmpfile, 'w+b', 4096) as f:
f.write(b'\xff' * 13569)
f.flush()
f.seek(0, 0)
f.read(1)
f.seek(-1, 1)
f.write(b'')
def test_issue1902(self):
import _io
with _io.open(self.tmpfile, "w+b", 4096) as f:
f.write(b"\xff" * 13569)
f.flush()
f.seek(0, 0)
f.read(1)
f.seek(-1, 1)
f.write(b"")
def test_instance(self):
f = _io.open('test_io.py')
self.assertTrue(isinstance(f, _io.TextIOWrapper))
f.close()
f = _io.open('/tmp/foo', 'bw')
self.assertTrue(isinstance(f, _io.BufferedWriter))
f.close()
f = _io.open('/tmp/foo', 'br')
self.assertTrue(isinstance(f, _io.BufferedReader))
f.close()
f = _io.open('/tmp/foo', 'w')
self.assertTrue(isinstance(f, _io.TextIOWrapper))
f.close()
f = _io.open('/tmp/foo', 'r')
self.assertTrue(isinstance(f, _io.TextIOWrapper))
f.close()
def spamTest():
docList = []
classList = []
fullText = []
for i in range(1,26):
wordList = textParse(open('C:\\db\\email\\spam\\'+str(i)+'.txt').read())
docList.append(wordList)
fullText.append(wordList)
classList.append(1)
wordList = textParse(open('C:\\db\\email\\ham\\'+str(i)+'.txt').read())
docList.append(wordList)
fullText.append(wordList)
classList.append(0)
vocabList = createVocabList(docList)
trainingSet = list(range(50))
testSet = []
for i in range(10):
randIndex = int(random.uniform(0,len(trainingSet)))
testSet.append(trainingSet[randIndex])
del(trainingSet[randIndex])
trainMat=[]; trainClasses = []
for docIndex in trainingSet:#train the classifier (get probs) trainNB0
trainMat.append(bagOfWords2VecMN(vocabList, docList[docIndex]))
trainClasses.append(classList[docIndex])
p0V,p1V,pSpam = trainNB0(array(trainMat),array(trainClasses))
errorCount = 0
for docIndex in testSet: #classify the remaining items
wordVector = setOfWord2Vec(vocabList, docList[docIndex])
if classifyNB(array(wordVector),p0V,p1V,pSpam) != classList[docIndex]:
errorCount += 1
print ("classification error",docList[docIndex])
print ('the error rate is: ',float(errorCount)/len(testSet))
errorCount = 0
for docIndex in testSet: #classify the remaining items
wordVector = bagOfWords2VecMN(vocabList, docList[docIndex])
if classifyNB(array(wordVector),p0V,p1V,pSpam) != classList[docIndex]:
errorCount += 1
print ("classification error",docList[docIndex])
print ('the error rate is: ',float(errorCount)/len(testSet))
def test_seek_and_tell(self):
import _io
with _io.open(self.tmpfile, "wb") as f:
f.write(b"abcd")
with _io.open(self.tmpfile) as f:
decoded = f.read()
# seek positions
for i in range(len(decoded) + 1):
# read lenghts
for j in [1, 5, len(decoded) - i]:
with _io.open(self.tmpfile) as f:
res = f.read(i)
assert res == decoded[:i]
cookie = f.tell()
res = f.read(j)
assert res == decoded[i:i + j]
f.seek(cookie)
res = f.read()
assert res == decoded[i:]
def write(self, overwrite=False):
"""Write file to disk"""
#dirty, dirty hack
cwd = os.listdir()[0][0:3] if self.filename[0:3] not in ['0:/','1:/'] else ''
if cwd+self.filename in os.listdir() and not overwrite: raise IOError('File exists')
file = _io.open(cwd+self.filename,'wb')
header = b''
for t in zip(self.__header_slist, self.__get_header_values()):
header += struct.pack(t[0], t[1])
file.write(header)
file.write(self.data)
file.close()
return len(header)+len(self.data)
def test_issue1902_3(self):
import _io
buffer_size = 4096
with _io.open(self.tmpfile, 'w+b', buffer_size) as f:
f.write(b'\xff' * buffer_size * 3)
f.flush()
f.seek(0, 0)
f.read(1)
f.seek(-1, 1)
f.write(b'\xff')
f.seek(1, 0)
f.read(buffer_size * 2)
assert f.tell() == 1 + buffer_size * 2
def write_csv(name:str, delim:str, body:list, head:list, body_format=None, head_format=None) -> int:
''' Write data to csv file, note body should be a bivariate table '''
if not head_format:
head_format = (delim).join(['%s'] * len(head)) + '\n'
elif not head_format.endswith('\n'):
head_format += '\n'
if not body_format:
try:
body_format = delim.join(['%s'] * len(body[0])) + '\n'
except TypeError:
try:
body_format = delim.join(['%s'] * numpy.shape(body)[1]) + '\n'
except IndexError:
body_format = delim.join(['%s'] * numpy.shape(body)[0]) + '\n'
elif not body_format.endswith('\n'):
body_format += '\n'
with _io.open(name, 'wb') as fout:
fout.write(BOM_UTF8)
with _io.open(name, 'a', encoding='utf-8') as fout:
nlines = 0
if head:
nlines += 1
fout.write(head_format % tuple(head))
if len(numpy.shape(body)) == 1:
for i in body:
fout.write('%s\n' % i)
nlines += 1
else:
for row in body:
fout.write(body_format % tuple(row))
nlines += 1
return nlines
def test_attributes(self):
import _io
with _io.open(self.tmpfile, "wb", buffering=0) as f:
assert f.mode == "wb"
with _io.open(self.tmpfile, "U") as f:
assert f.name == self.tmpfile
assert f.buffer.name == self.tmpfile
assert f.buffer.raw.name == self.tmpfile
assert f.mode == "U"
assert f.buffer.mode == "rb"
assert f.buffer.raw.mode == "rb"
with _io.open(self.tmpfile, "w+") as f:
assert f.mode == "w+"
assert f.buffer.mode == "rb+"
assert f.buffer.raw.mode == "rb+"
with _io.open(f.fileno(), "wb", closefd=False) as g:
assert g.mode == "wb"
assert g.raw.mode == "wb"
assert g.name == f.fileno()
assert g.raw.name == f.fileno()
def split_by_size(self, nbyte:int):
''' Split by every `nbyte` bytes '''
nfile = self.__howmany(self._nbyte, nbyte)
newnames = self.splited_names(nfile)
inames = iter(newnames)
with open_as_bytes_stream(self._fname) as stream:
buf = stream.read(nbyte)
while buf:
fout = open(next(inames), 'wb')
fout.write(buf)
fout.close()
buf = stream.read(nbyte)
return newnames
def test_io_after_close(self):
import _io
for kwargs in [
{"mode": "w"},
{"mode": "wb"},
{"mode": "w", "buffering": 1},
{"mode": "w", "buffering": 2},
{"mode": "wb", "buffering": 0},
{"mode": "r"},
{"mode": "rb"},
{"mode": "r", "buffering": 1},
{"mode": "r", "buffering": 2},
{"mode": "rb", "buffering": 0},
{"mode": "w+"},
{"mode": "w+b"},
{"mode": "w+", "buffering": 1},
{"mode": "w+", "buffering": 2},
{"mode": "w+b", "buffering": 0},
]:
print kwargs
if "b" not in kwargs["mode"]:
kwargs["encoding"] = "ascii"
f = _io.open(self.tmpfile, **kwargs)
f.close()
raises(ValueError, f.flush)
raises(ValueError, f.fileno)
raises(ValueError, f.isatty)
raises(ValueError, f.__iter__)
if hasattr(f, "peek"):
raises(ValueError, f.peek, 1)
raises(ValueError, f.read)
if hasattr(f, "read1"):
raises(ValueError, f.read1, 1024)
if hasattr(f, "readall"):
raises(ValueError, f.readall)
if hasattr(f, "readinto"):
raises(ValueError, f.readinto, bytearray(1024))
raises(ValueError, f.readline)
raises(ValueError, f.readlines)
raises(ValueError, f.seek, 0)
raises(ValueError, f.tell)
raises(ValueError, f.truncate)
raises(ValueError, f.write, b"" if "b" in kwargs["mode"] else u"")
raises(ValueError, f.writelines, [])
raises(ValueError, next, f)
def test_pickling(self):
import _io
import pickle
# Pickling file objects is forbidden
for kwargs in [
{"mode": "w"},
{"mode": "wb"},
{"mode": "wb", "buffering": 0},
{"mode": "r"},
{"mode": "rb"},
{"mode": "rb", "buffering": 0},
{"mode": "w+"},
{"mode": "w+b"},
{"mode": "w+b", "buffering": 0},
]:
for protocol in range(pickle.HIGHEST_PROTOCOL + 1):
with _io.open(self.tmpfile, **kwargs) as f:
raises(TypeError, pickle.dumps, f, protocol)
def _get_data(archive, toc_entry):
datapath, compress, data_size, file_size, file_offset, time, date, crc = toc_entry
if data_size < 0:
raise ZipImportError('negative data size')
with _io.open(archive, 'rb') as fp:
# Check to make sure the local file header is correct
try:
fp.seek(file_offset)
except OSError:
raise ZipImportError(f"can't read Zip file: {archive!r}", path=archive)
buffer = fp.read(30)
if len(buffer) != 30:
raise EOFError('EOF read where not expected')
if buffer[:4] != b'PK\x03\x04':
# Bad: Local File Header
raise ZipImportError(f'bad local file header: {archive!r}', path=archive)
name_size = _unpack_uint16(buffer[26:28])
extra_size = _unpack_uint16(buffer[28:30])
header_size = 30 + name_size + extra_size
file_offset += header_size # Start of file data
try:
fp.seek(file_offset)
except OSError:
raise ZipImportError(f"can't read Zip file: {archive!r}", path=archive)
raw_data = fp.read(data_size)
if len(raw_data) != data_size:
raise OSError("zipimport: can't read data")
if compress == 0:
# data is not compressed
return raw_data
# Decompress with zlib
try:
decompress = _get_decompress_func()
except Exception:
raise ZipImportError("can't decompress data; zlib not available")
return decompress(raw_data, -15)