def test_incremental_encode(self):
self.assertEquals(
"".join(codecs.iterencode(u"python.org", "idna")),
"python.org"
)
self.assertEquals(
"".join(codecs.iterencode(u"python.org.", "idna")),
"python.org."
)
self.assertEquals(
"".join(codecs.iterencode(u"pyth\xf6n.org.", "idna")),
"xn--pythn-mua.org."
)
self.assertEquals(
"".join(codecs.iterencode(u"pyth\xf6n.org.", "idna")),
"xn--pythn-mua.org."
)
encoder = codecs.getincrementalencoder("idna")()
self.assertEquals(encoder.encode(u"\xe4x"), "")
self.assertEquals(encoder.encode(u"ample.org"), "xn--xample-9ta.")
self.assertEquals(encoder.encode(u"", True), "org")
encoder.reset()
self.assertEquals(encoder.encode(u"\xe4x"), "")
self.assertEquals(encoder.encode(u"ample.org."), "xn--xample-9ta.org.")
self.assertEquals(encoder.encode(u"", True), "")
def test_incremental_encode(self):
self.assertEquals(
"".join(codecs.iterencode(u"python.org", "idna")),
"python.org"
)
self.assertEquals(
"".join(codecs.iterencode(u"python.org.", "idna")),
"python.org."
)
self.assertEquals(
"".join(codecs.iterencode(u"pyth\xf6n.org.", "idna")),
"xn--pythn-mua.org."
)
self.assertEquals(
"".join(codecs.iterencode(u"pyth\xf6n.org.", "idna")),
"xn--pythn-mua.org."
)
encoder = codecs.getincrementalencoder("idna")()
self.assertEquals(encoder.encode(u"\xe4x"), "")
self.assertEquals(encoder.encode(u"ample.org"), "xn--xample-9ta.")
self.assertEquals(encoder.encode(u"", True), "org")
encoder.reset()
self.assertEquals(encoder.encode(u"\xe4x"), "")
self.assertEquals(encoder.encode(u"ample.org."), "xn--xample-9ta.org.")
self.assertEquals(encoder.encode(u"", True), "")
def push_last(self, arg: str, extra_margin: int=0,
human_trunc: bool=False) -> typing.Optional[str]:
last_arg = self.args[-1]
tags, line = self._format()
n = len(line.encode("utf8")) # get length of current line
n += self._margin # margin used for :hostmask
if " " in arg and not " " in last_arg:
n += 1 # +1 for colon on new arg
n += extra_margin # used for things like (more ...)
overflow: typing.Optional[str] = None
if (n+len(arg.encode("utf8"))) > LINE_MAX:
for i, char in enumerate(codecs.iterencode(arg, "utf8")):
n += len(char)
if n > LINE_MAX:
arg, overflow = arg[:i], arg[i:]
if human_trunc and not overflow[0] == " ":
new_arg, sep, new_overflow = arg.rpartition(" ")
if sep:
arg = new_arg
overflow = new_overflow+overflow
break
if arg:
self.args[-1] = last_arg+arg
return overflow
def handler(env, start_response):
inp = env['wsgi.input']
reader = codecs.getreader("utf-8")
state = json.load(reader(inp))
ret = {}
for key,vtype in STATE_KEYS.items():
value = state[key]
assert isinstance(value, vtype)
ret[key] = value
# validate the input, for sure.
n = state["n"]
if n < 2 or n > 201:
ret["n"] = 101
values = valuesiter(ret)
ret["values"] = values
start_response('200 OK', [('Content-Type','application/json')])
je = json.JSONEncoder()
viter = je.iterencode(ret)
return codecs.iterencode(viter,"utf-8")
def handle_charref(self, name):
if name.startswith('x') or name.startswith('X'):
cp = int(name[1:], 16)
else:
cp = int(name)
u = unichr(cp)
self.buf += codecs.iterencode(u, 'utf-8')
def __init__(self, unicode_csvfile, *args, **kwargs):
decoder = codecs.getdecoder('utf-8')
self.decoder = lambda v: decoder(v)[0]
utf8_csvfile = codecs.iterencode(unicode_csvfile, encoding='utf-8')
# bollicks to csv.DictReader being an oldstyle class
csv.DictReader.__init__(self, utf8_csvfile, *args, **kwargs)
self.fieldnames = [self.decoder(f) for f in self.fieldnames]
def handler(env, start_response):
inp = env['wsgi.input']
reader = codecs.getreader("utf-8")
state = json.load(reader(inp))
ret = {}
for key, vtype in STATE_KEYS.items():
value = state[key]
assert isinstance(value, vtype)
ret[key] = value
# validate the input, for sure.
n = state["n"]
if n < 2 or n > 201:
ret["n"] = 101
values = valuesiter(ret)
ret["values"] = values
start_response('200 OK', [('Content-Type', 'application/json')])
je = json.JSONEncoder()
viter = je.iterencode(ret)
return codecs.iterencode(viter, "utf-8")
def run_cgi_script(self, filesystem_path: pathlib.Path,
environ: dict) -> Response:
"""
Execute the given file as a CGI script and return the script's stdout
stream to the client.
"""
script_name = str(filesystem_path)
cgi_env = environ.copy()
cgi_env["GATEWAY_INTERFACE"] = "GCI/1.1"
cgi_env["SCRIPT_NAME"] = script_name
# Decode the stream as unicode so we can parse the status line
# Use surrogateescape to preserve any non-UTF8 byte sequences.
out = subprocess.Popen(
[script_name],
stdout=subprocess.PIPE,
env=cgi_env,
bufsize=1,
universal_newlines=True,
errors="surrogateescape",
)
status_line = out.stdout.readline().strip()
status_parts = status_line.split(maxsplit=1)
if len(status_parts) != 2 or not status_parts[0].isdecimal():
return Response(Status.CGI_ERROR, "Unexpected Error")
status, meta = status_parts
# Re-encode the rest of the body as bytes
body = codecs.iterencode(out.stdout,
encoding="utf-8",
errors="surrogateescape")
return Response(int(status), meta, body)
def csvreader(f):
try:
for row in csv.reader(f):
yield row
except UnicodeEncodeError:
for row in csv.reader(codecs.iterencode(f, 'utf-8')):
yield [e.decode('utf-8') for e in row]
def __init__(self, unicode_csvfile, *args, **kwargs):
decoder = codecs.getdecoder('utf-8')
self.decoder = lambda v: decoder(v)[0]
utf8_csvfile = codecs.iterencode(unicode_csvfile, encoding='utf-8')
# bollicks to csv.DictReader being an oldstyle class
csv.DictReader.__init__(self, utf8_csvfile, *args, **kwargs)
self.fieldnames = [self.decoder(f) for f in self.fieldnames]
def serialize(self, out):
"""Write the units back to file."""
# Thanks to iterencode, a possible BOM is written only once
for chunk in iterencode(
(unit.getoutput() for unit in self.units), self.encoding
):
out.write(chunk)
def iter_byte_indices(text, codec):
'''
Iterate over the codepoint offset of each byte (any codec).
'''
for i, b in enumerate(codecs.iterencode(text, codec)):
for _ in b:
yield i
yield len(text)
def _utf8_iter_recoder(stream, encoding):
"""Generator re-encodes input file's lines from a given
encoding to utf-8.
:param stream: file handle.
:param encoding: str of encoding.
"""
return codecs.iterencode(codecs.iterdecode(stream, encoding), "utf-8")
def test_incremental_encode():
from codecs import iterencode
encoded = iterencode(
(c for c in UNICODE.utf8),
'internet'
)
encoded = ''.join(encoded)
assert encoded == UNICODE.utf8.encode('UTF-8')
def __init__(self, f, fromenc=ENCODING, toenc=ENCODING, **kwargs):
""" Reencoder constructor
Args:
f (obj): File-like object
fromenc (str): The input encoding.
toenc (str): The output encoding.
Kwargs:
remove_BOM (bool): Remove Byte Order Marker (default: True)
decode (bool): Decode the text into a string (default: False)
Examples:
>>> encoding = 'utf-16-be'
>>> eff = p.join(DATA_DIR, 'utf16_big.csv')
>>>
>>> with open(eff, 'rb') as f:
... reenc = Reencoder(f, encoding)
... first = reenc.readline(keepends=False)
... first.decode('utf-8') == '\ufeffa,b,c'
... reenc.read().decode('utf-8').split('\\n')[1] == '4,5,ʤ'
True
True
>>> with open(eff, 'rb') as f:
... reenc = Reencoder(f, encoding, decode=True)
... reenc.readline(keepends=False) == '\ufeffa,b,c'
True
>>> with open(eff, 'rU', encoding=encoding) as f:
... reenc = Reencoder(f, remove_BOM=True)
... reenc.readline(keepends=False) == b'a,b,c'
... reenc.readline() == b'1,2,3\\n'
... reenc.readline().decode('utf-8') == '4,5,ʤ'
True
True
True
"""
self.fileno = f.fileno
first_line = next(f)
bytes_mode = isinstance(first_line, BYTE_TYPE)
decode = kwargs.get('decode')
rencode = not decode
if kwargs.get('remove_BOM'):
strip = BOM.encode(fromenc) if bytes_mode else BOM
first_line = first_line.lstrip(strip)
chained = it.chain([first_line], f)
if bytes_mode:
decoded = iterdecode(chained, fromenc)
self.binary = rencode
else:
decoded = chained
self.binary = bytes_mode or rencode
self.stream = iterencode(decoded, toenc) if rencode else decoded
def read_decoded_data(self, f): # This ensures that the raw result bytes we got are, in fact, valid utf-8, # replacing invalid bytes with �. Because python2's unicode support is # wonky, we re-encode the now-valid-utf-8 back into a str object so that # users don't need to deal with `unicode` objects. # The file contents can be large, so be careful to do the conversion in # chunks while streaming the data in, instead of requiring a full copy. n = 1 << 16 chunks = iter(lambda: f.read(n), '') decoded = codecs.iterdecode(chunks, 'utf-8', 'replace') return ''.join(codecs.iterencode(decoded, 'utf-8'))
def __iter__(self):
root = ElementTree.fromstringlist(codecs.iterencode(self.inputreader, 'utf'))
table = root.findall('.//table')[5]
for row in table[2:-1]:
date_str = row.find('.//td[3].nobr').text
tdate = datetime.strptime(date_str, '%d.%m.%Y')
amount_str = row.find('.//td[5].nobr').text
amount_str = amount_str.replace('.', '')
tamount = float(normalize_num(amount_str))
desc = plain_content(row.find('.//td[4]'))
tmessage = normalize_field(desc)
yield TransactionData(tdate, tamount, message=tmessage)
def iter_codepoint_indices(text, codec):
'''
Iterate over the byte offset of each character (any codec).
'''
# Note: for encodings with a BOM, the first offset probably shouldn't
# be 0, but 2, 3, or 4, depending on the BOM's length.
# This is ignored due to the lack of expected practical applications.
i = 0
for b in codecs.iterencode(text, codec):
yield i
i += len(b)
yield i
def __iter__(self):
root = ElementTree.fromstringlist(codecs.iterencode(self.inputreader, 'utf'))
table = root.findall('.//table')[5]
for row in table[2:-1]:
date_str = row.find('.//td[3].nobr').text
tdate = datetime.strptime(date_str, '%d.%m.%Y')
amount_str = row.find('.//td[5].nobr').text
amount_str = amount_str.replace('.', '')
tamount = float(normalize_num(amount_str))
desc = plain_content(row.find('.//td[4]'))
tmessage = normalize_field(desc)
yield TransactionData(tdate, tamount, message=tmessage)
def stream(input, encoding=None, errors='strict'): """Safely iterate a template generator, ignoring ``None`` values and optionally stream encoding. Used internally by ``cinje.flatten``, this allows for easy use of a template generator as a WSGI body. """ input = (i for i in input if i) # Omits `None` (empty wrappers) and empty chunks. if encoding: # Automatically, and iteratively, encode the text if requested. input = iterencode(input, encoding, errors=errors) return input
def write_encoded_data(self, f):
# Sometimes users give us invalid utf-8 data. They shouldn't, but it does
# happen every once and a while. Just ignore it, and replace with �.
# We're assuming users only want to write text data out.
# self.data can be large, so be careful to do the conversion in chunks
# while streaming the data out, instead of requiring a full copy.
n = 1 << 16
# This is a generator expression, so this only copies one chunk of
# self.data at any one time.
chunks = (self.data[i:i + n] for i in xrange(0, len(self.data), n))
decoded = codecs.iterdecode(chunks, 'utf-8', 'replace')
for chunk in codecs.iterencode(decoded, 'utf-8'):
f.write(chunk)
def csv_unireader(f, encoding="utf-8"):
if PY3:
f = codecs.open(f, encoding=encoding)
r = csv.reader(f, delimiter='|', quotechar='"')
else:
r = csv.reader(
codecs.iterencode(codecs.iterdecode(open(f), encoding), 'utf-8'),
delimiter=b'|', quotechar=b'"')
for row in r:
if PY3:
yield row
else:
yield [e.decode("utf-8") for e in row]
def error(self,
code: int,
message: str,
environ: Environ,
start_response: StartResponseCallable) -> ResponseStream:
statuses = {
400: 'Bad Request',
405: 'Method Not Allowed',
}
start_response(
f'{code} {statuses[code]}',
[('Content-Type', 'text/plain')]
)
return codecs.iterencode(message, 'utf-8')
def encode_all(f=None, **kwargs):
"""
Encode unicode into bytes (str)
"""
names = kwargs.pop('fieldnames', None)
encoding = kwargs.pop('encoding', None) if f else False
decoded = codecs.iterdecode(f, encoding) if encoding else f
ekwargs = {encode(k): encode(v) for k, v in kwargs.items()}
fmtparams = {k: v for k, v in ekwargs.items() if k in dir(csv.Dialect)}
res = {
'f': codecs.iterencode(decoded, ENCODING) if f else None,
'fieldnames': [encode(x) for x in names] if names else None,
'kwargs': ekwargs,
'fmtparams': fmtparams}
return res
def write_text_resource(foutput, text, encoding='utf-8'):
"""Write a text resource
:param foutput: path or file handle
:type foutput: str, file
:param text: content to write
:type text: str, unicode, iterable
:param encoding: which encoding to use (default: UTF-8)
:type encoding: str
"""
if isinstance(foutput, file):
for chunk in codecs.iterencode(text, encoding=encoding):
foutput.write(chunk)
else:
with codecs.open(foutput, 'w', encoding=encoding) as fhandle:
if isiterable(text):
for line in text:
fhandle.write(u"%s\n" % line)
else:
fhandle.write(text)
def __iter_extended_rows(self):
# For PY2 encode/decode
if six.PY2:
# Reader requires utf-8 encoded stream
bytes = iterencode(self.__chars, 'utf-8')
items = csv.reader(bytes, **self.__options)
for number, item in enumerate(items, start=1):
values = []
for value in item:
value = value.decode('utf-8')
values.append(value)
yield (number, None, list(values))
# For PY3 use chars
else:
items = csv.reader(self.__chars, **self.__options)
for number, item in enumerate(items, start=1):
yield (number, None, list(item))
def __iter_extended_rows(self):
# For PY2 encode/decode
if six.PY2:
# Reader requires utf-8 encoded stream
bytes = iterencode(self.__chars, 'utf-8')
items = csv.reader(bytes, **self.__options)
for number, item in enumerate(items, start=1):
values = []
for value in item:
value = value.decode('utf-8')
values.append(value)
yield (number, None, list(values))
# For PY3 use chars
else:
items = csv.reader(self.__chars, **self.__options)
for number, item in enumerate(items, start=1):
yield (number, None, list(item))
def __emit_items(self):
# For PY2 encode/decode
if six.PY2:
# Reader requires utf-8 encoded stream
bytes = iterencode(self.__chars, 'utf-8')
items = csv.reader(bytes, **self.__options)
for item in items:
values = []
for value in item:
value = value.decode('utf-8')
values.append(value)
yield (None, tuple(values))
# For PY3 use chars
else:
items = csv.reader(self.__chars, **self.__options)
for item in items:
yield (None, tuple(item))
def write_text_resource(foutput, text, encoding='utf-8'):
"""Write a text resource
:param foutput: path or file handle
:type foutput: str, file
:param text: content to write
:type text: str, unicode, iterable
:param encoding: which encoding to use (default: UTF-8)
:type encoding: str
"""
if isinstance(foutput, file):
for chunk in codecs.iterencode(text, encoding=encoding):
foutput.write(chunk)
else:
with codecs.open(foutput, 'w', encoding=encoding) as fhandle:
if isiterable(text):
for line in text:
fhandle.write(u"%s\n" % line)
else:
fhandle.write(text)
def read_docs(filename, column):
ret_arr = []
#print_err("reading %s"%(filename))
dupes = 0
with open(filename, 'r') as infile:
coder = codecs.iterencode(codecs.iterdecode(infile, "utf-8"), "utf-8")
csvfile = csv.reader(coder, delimiter=',', quotechar='"')
#next(csvfile)
dup_checker = {}
for row in csvfile:
test = row[column].lower()
if test not in dup_checker:
dup_checker[test] = 1
ret_arr.append(row[column])
else:
dupes += 1
print_err("total dupes: %d"%dupes)
return ret_arr
def testParseResponse(self):
# Parse a simple response to a subscribe:
self.conn.write_socket(self.subscribe_cmd)
parsed_resp = self.conn.parse_response()
self.assertEqual([u'subscribe', u'tmp.20', 1L], parsed_resp)
# Parse a more complicated incoming msg:
# '*3\r\n' +\
# '$7\r\n' +\
# 'PUBLISH\r\n' +\
# '$4\r\n' +\
# 'test\r\n' +\
# '$11\r\n{"Hello World"}'
# Make the OnDemandPublisher thread send us this
# message:
OneShotTester.answer_server.sendMessage(OneShotTester.test_msg,
OneShotTester.from_channel)
parsed_resp = self.conn.parse_response()
# Get the message that was sent back to us:
sent_bus_msg = OneShotTester.answer_server.outMsg
out_id = sent_bus_msg.id
rx_encoded = []
for el in codecs.iterencode(parsed_resp, 'UTF-8'):
rx_encoded.append(el)
expected = [
'message', OneShotTester.from_channel,
'{"content": "Hello world", "id": "%s", ' % out_id
]
# Time of sent msg will be different each time; cut it out of the
# rx-ed string:
rxed_msg_body = rx_encoded[2]
rxed_body_chop_pos = rxed_msg_body.index('"time": ')
chopped_rxed_body = rxed_msg_body[:rxed_body_chop_pos]
# Replaced msg body part of received by the
# truncated version that doesn't include the time:
rx_encoded[2] = chopped_rxed_body
self.assertEqual(expected, rx_encoded)
def __iter_extended_rows(self):
# For PY2 encode/decode
if six.PY2:
# Reader requires utf-8 encoded stream
bytes = iterencode(self.__chars, 'utf-8')
sample, dialect = self.__prepare_dialect(bytes)
items = csv.reader(chain(sample, bytes), dialect=dialect)
for number, item in enumerate(items, start=1):
values = []
for value in item:
value = value.decode('utf-8')
values.append(value)
yield (number, None, list(values))
# For PY3 use chars
else:
sample, dialect = self.__prepare_dialect(self.__chars)
items = csv.reader(chain(sample, self.__chars), dialect=dialect)
for number, item in enumerate(items, start=1):
yield (number, None, list(item))
def encode_all(f=None, **kwargs):
"""
Encode unicode into bytes (str)
"""
names = kwargs.pop('fieldnames', None)
encoding = kwargs.pop('encoding', None) if f else False
if PY2:
decoded = codecs.iterdecode(f, encoding) if encoding else f
ekwargs = {encode(k): encode(v) for k, v in kwargs.items()}
else:
decoded, ekwargs = f, kwargs
res = {
'f': codecs.iterencode(decoded, ENCODING) if f and PY2 else decoded,
'fieldnames': [encode(x) for x in names] if names and PY2 else names,
'drkwargs': use_keys_from(ekwargs, READER_KEYS),
'dwkwargs': use_keys_from(ekwargs, WRITER_KEYS),
'fmtparams': use_keys_from(ekwargs, FMTKEYS)}
return res
def encode_all(f=None, **kwargs):
"""
Encode unicode into bytes (str)
"""
names = kwargs.pop('fieldnames', None)
encoding = kwargs.pop('encoding', None) if f else False
if PY2:
decoded = codecs.iterdecode(f, encoding) if encoding else f
ekwargs = {encode(k): encode(v) for k, v in kwargs.items()}
else:
decoded, ekwargs = f, kwargs
res = {
'f': codecs.iterencode(decoded, ENCODING) if f and PY2 else decoded,
'fieldnames': [encode(x) for x in names] if names and PY2 else names,
'drkwargs': use_keys_from(ekwargs, READER_KEYS),
'dwkwargs': use_keys_from(ekwargs, WRITER_KEYS),
'fmtparams': use_keys_from(ekwargs, FMTKEYS)}
return res
def __iter_extended_rows(self):
# For PY2 encode/decode
if six.PY2:
# Reader requires utf-8 encoded stream
bytes = iterencode(self.__chars, 'utf-8')
sample, dialect = self.__prepare_dialect(bytes)
items = csv.reader(chain(sample, bytes), dialect=dialect)
for row_number, item in enumerate(items, start=1):
values = []
for value in item:
value = value.decode('utf-8')
values.append(value)
yield (row_number, None, list(values))
# For PY3 use chars
else:
sample, dialect = self.__prepare_dialect(self.__chars)
items = csv.reader(chain(sample, self.__chars), dialect=dialect)
for row_number, item in enumerate(items, start=1):
yield (row_number, None, list(item))
def testParseResponse(self):
# Parse a simple response to a subscribe:
self.conn.write_socket(self.subscribe_cmd)
parsed_resp = self.conn.parse_response()
self.assertEqual([u'subscribe', u'tmp.20', 1L], parsed_resp)
# Parse a more complicated incoming msg:
# '*3\r\n' +\
# '$7\r\n' +\
# 'PUBLISH\r\n' +\
# '$4\r\n' +\
# 'test\r\n' +\
# '$11\r\n{"Hello World"}'
# Make the OnDemandPublisher thread send us this
# message:
OneShotTester.answer_server.sendMessage(OneShotTester.test_msg, OneShotTester.from_channel)
parsed_resp = self.conn.parse_response()
# Get the message that was sent back to us:
sent_bus_msg = OneShotTester.answer_server.outMsg
out_id = sent_bus_msg.id
rx_encoded = []
for el in codecs.iterencode(parsed_resp, 'UTF-8'):
rx_encoded.append(el)
expected = ['message',
OneShotTester.from_channel,
'{"content": "Hello world", "id": "%s", ' % out_id
]
# Time of sent msg will be different each time; cut it out of the
# rx-ed string:
rxed_msg_body = rx_encoded[2]
rxed_body_chop_pos = rxed_msg_body.index('"time": ')
chopped_rxed_body = rxed_msg_body[:rxed_body_chop_pos]
# Replaced msg body part of received by the
# truncated version that doesn't include the time:
rx_encoded[2] = chopped_rxed_body
self.assertEqual(expected, rx_encoded)
def testIncrementalEncoder(self):
# Tests derived from Python standard library test/test_codecs.py
incremental_tests = (
(u"python.org", b"python.org"),
(u"python.org.", b"python.org."),
(u"pyth\xf6n.org", b"xn--pythn-mua.org"),
(u"pyth\xf6n.org.", b"xn--pythn-mua.org."),
)
for decoded, encoded in incremental_tests:
self.assertEqual(b"".join(codecs.iterencode(decoded, "idna")),
encoded)
encoder = codecs.getincrementalencoder("idna")()
self.assertEqual(encoder.encode(u"\xe4x"), b"")
self.assertEqual(encoder.encode(u"ample.org"), b"xn--xample-9ta.")
self.assertEqual(encoder.encode(u"", True), b"org")
encoder.reset()
self.assertEqual(encoder.encode(u"\xe4x"), b"")
self.assertEqual(encoder.encode(u"ample.org."), b"xn--xample-9ta.org.")
self.assertEqual(encoder.encode(u"", True), b"")
def testIncrementalEncoder(self):
# Tests derived from Python standard library test/test_codecs.py
incremental_tests = (
("python.org", b"python.org"),
("python.org.", b"python.org."),
("pyth\xf6n.org", b"xn--pythn-mua.org"),
("pyth\xf6n.org.", b"xn--pythn-mua.org."),
)
for decoded, encoded in incremental_tests:
self.assertEqual(b"".join(codecs.iterencode(decoded, "idna")),
encoded)
encoder = codecs.getincrementalencoder("idna")()
self.assertEqual(encoder.encode("\xe4x"), b"")
self.assertEqual(encoder.encode("ample.org"), b"xn--xample-9ta.")
self.assertEqual(encoder.encode("", True), b"org")
encoder.reset()
self.assertEqual(encoder.encode("\xe4x"), b"")
self.assertEqual(encoder.encode("ample.org."), b"xn--xample-9ta.org.")
self.assertEqual(encoder.encode("", True), b"")
def csv_unireader(f, encoding="utf-8"):
"""
abre um arquivo utf-8 com a lib csv
"""
for row in csv.reader(codecs.iterencode(codecs.iterdecode(f, encoding), "utf-8"), delimiter="\t"):
yield [e.decode("utf-8") for e in row]
file_list = sorted(glob.glob(args.inputs))
file_count = len(file_list)
url_dict = {}
#
# step through each file
#
for filename_index in range(file_count):
filename = file_list[filename_index]
print "opening \"%s\" (%d of %d)" % (
filename, filename_index+1, file_count)
with io.open(filename, "r", encoding="utf-8") as f:
input_csv = csv.DictReader(codecs.iterencode(f, "utf-8"))
for row in input_csv:
url = row['url']
if url not in url_dict:
url_dict[url] = row
with io.open(args.output, "w", encoding="utf-8") as outfile:
outfile.write(u"id,url,text\n")
for row in url_dict.values():
tweet_id = row["id"]
url = row["url"]
text = row["text"]
if text is not None:
text = text.replace("\"", "\"\"")
def on_get(self, request, resp, **kwargs):
"""
Falcon resource method, for handling HTTP request GET method
Falcon request provides: parameters embedded in URL via a keyword args
dict, as well as convenience class variables falcon.HTTP_*
FIXME: remove moduel pylint:disable= & refactor this overlong code block!
"""
# obtain logged in API user ID (if available)
api_session_user = auth.get_user_id(request)
# select data
with warehouse.get_source_model_session() as dwsupport_model:
sources = source.SourceUtil.get_list_of_data_sources(
request.url
,auth.get_user_id(request)
,dwsupport_model)
str_dataset_id = get_requested_dataset_id( sources, request, resp, kwargs)
list_variables_requested_source = variables.get_list_of_variables( str_dataset_id)
# convert 'datasets' into a list of variables
list_requested_datasets = parameters.get_requested_datasets( request)
list_variables_from_datasets = []
for str_id in list_requested_datasets:
if str_dataset_id == str_id:
list_variables_from_datasets = list_variables_requested_source
break
if str_dataset_id == 'warehouse': #FIXME: refactor this into a source.warehouse function
#obtain the 'warehouse' field aliases for each dataset
list_source_variables = variables.get_list_of_variables( str_id)
for var in list_source_variables:
warehouse_utils = api.resources.source.warehouse.warehouse
str_alias = warehouse_utils.prefix_field_name( var, str_id)
list_variables_from_datasets.append( str_alias)
else: #error; not a warehouse request & Dataset does not match requested ID
raise falcon.HTTPNotFound(description= "Unrecognized dataset: "
+ str_id)
list_requested_variables = parameters.get_requested_variables( request)
# add default variables
if len(list_requested_variables) < 1:
requested_default_query = parameters.get_list_requested_parameter(
defaults.PARAMETER_NAME, request)
try:
default_variables = defaults.get_default_variables(
requested_default_query
,str_dataset_id
,dwsupport_model)
except defaults.UndefinedDefaultQuery as error:
msg = ("Value {} is not defined for dataset: '{}'"
.format(error, str_dataset_id))
raise falcon.HTTPInvalidParam(msg, defaults.PARAMETER_NAME)
except defaults.AmbiguousDefaultQuery as error:
msg = "More than one value was specified: {}".format(error)
raise falcon.HTTPInvalidParam(msg, defaults.PARAMETER_NAME)
except defaults.AmbiguousQueryHierarchy as error:
raise falcon.HTTPBadRequest( #TODO: add functional test coverage
title="Missing Parameter"
,description=(
"Selection defaults not clear for"
" data source: '{}'."
" Selection must specify one or more 'variables='"
" selection parameters (or a 'defaults=' parameter"
" value from the following list: {})"
).format(str_dataset_id, error)
)
list_requested_variables.extend(default_variables)
# add variables derived from 'datasets' param
list_requested_variables.extend( list_variables_from_datasets)
list_requested_filters = parameters.get_requested_filters( request)
# process pivot columns parameter
try:
pivot_column_variables = parameters.get_requested_pivot_columns(
request
,str_dataset_id
,dwsupport_model['tables'])
except parameters.PivotVariableError as err:
raise falcon.HTTPInvalidParam(
msg=str(err)
,param_name=parameters.ReservedParameterNames.pivot_columns
) from err
# process 'Empty_cells' parameter
try:
empty_cell_dimensions = parameters.get_requested_empty_cells(
request
,str_dataset_id
,dwsupport_model['tables']
,dwsupport_model['associations']
)
except (parameters.EmptyCellsSourceError
,parameters.EmptyCellsDimensionError) as err:
raise falcon.HTTPInvalidParam(
msg=str(err)
,param_name=parameters.ReservedParameterNames.empty_cells
) from err
# retrieve data
start_time = datetime.now(pytz.timezone('US/Pacific'))
try:
result_generator = data.get_data(str_dataset_id
,list_requested_variables
,list_requested_filters
,pivot_column_variables
,empty_cell_dimensions
,user_id=api_session_user)
except sqlalchemy.exc.DatabaseError as err:
raise falcon.HTTPInternalServerError(
title='500'
,description="Please try again"
) from err
except data.NoSourceException as err:
raise falcon.HTTPNotFound(description=("Source '{}' dataset not found:"
" {}").format(str_dataset_id,err)) from err
except parameters.FilterVariableError as err:
#TODO: the bad HTTP parameter not always 'filters',sometimes a user-defined param (implicit-filter)
#TODO: perhaps parameters should raise two different Exceptions?
raise falcon.HTTPInvalidParam(str(err), 'filters') from err
except data.NotAuthorizedException as error:
raise falcon.HTTPUnauthorized(
title='401'
,description=("Selection from sensitive data source '{}'"
" not authorized").format(str_dataset_id)
) from error
str_format_type = get_requested_format_type( kwargs)
resp.content_type = FormatUtil.get_http_content_type(str_format_type)
for data_source in sources:
if data_source['id'] == str_dataset_id:
formatter = FormatUtil(str_format_type, data_source, request, start_time)
result_stream = formatter.format(result_generator)
break
chunked_stream = streaming.biggerchunks_stream(result_stream, 4)#2(13.6),3(13),4(
if str_format_type == 'xlsx':
byte_stream = chunked_stream #already bytes
else:
encoding = 'utf-8'
if resp.content_type == 'text/csv':
encoding = 'utf-8-sig'
byte_stream = codecs.iterencode(chunked_stream, encoding)
resp.stream = byte_stream#content
re_strip_mention = re.compile(u'(via\s)?@\w+', flags=re.IGNORECASE)
re_strip_cc = re.compile(u'-CC', flags=re.IGNORECASE)
re_strip_url = re.compile(u'(?i)\b((?:https?:?//|www\d{0,3}[.]|[a-z0-9.\-]+[.][a-z]{2,4}/)(?:[^\s()<>]+|\(([^\s()<>]+|(\([^\s()<>]+\)))*\))+(?:\(([^\s()<>]+|(\([^\s()<>]+\)))*\)|[^\s`!()\[\]{};:\'".,<>?«»“”‘’]))', flags=re.IGNORECASE)
re_strip_tco_url = re.compile(u'https?:?//[\w\-\.]*(/[\w\-\%]*)*( via)?', flags=re.IGNORECASE)
re_strip_punct = re.compile(u'[\.,;\:\-\!\[\]\"\'\?\(\)\|@\/]', flags=re.IGNORECASE)
re_strip_newline = re.compile(u'[\r\n\t]+')
re_strip_ending_http = re.compile(u'(\s)ht?t?t?p?:?/?/?([\w\-\.]*)?\s?$', flags=re.IGNORECASE)
word_split_re = re.compile(u'\@?\w+|[\$?-]*[\d\.]+|[a-zA-Z0-9\-\'\/]+|[!\?\.\,\/\@\:\-\_&\$\s]*', flags=re.IGNORECASE)
for csvinput in args.inputfiles:
with open(csvinput, 'r') as infile:
coder = codecs.iterencode(codecs.iterdecode(infile, "utf-8"), "utf-8")
csvreader = csv.reader(coder, delimiter=',', quotechar='"')
# skip header
next(csvreader)
for row in csvreader:
text = row[2]
# strip retweet
text = re_strip_rt.sub('', text)
text = re_strip_mention.sub('', text)
text = re_strip_cc.sub('', text)
text = re_strip_url.sub('', text)
text = re_strip_tco_url.sub('', text)
def handle(self, *args, **options):
for file_name in args:
csv_file = codecs.open(file_name, 'rb', 'cp1252')
encoded_file = codecs.iterencode(csv_file, 'utf-8')
self.consume_csv(encoded_file)
def csv_unireader(f, encoding="utf-8"):
for row in csv.reader(codecs.iterencode(codecs.iterdecode(f, encoding), "utf-8")):
yield [e.decode("utf-8") for e in row]
def _csv_unireader(f, encoding='utf-8'):
return csv.DictReader(codecs.iterencode(codecs.iterdecode(f, encoding),
'utf-8'),
delimiter=',')
sniffer = csv.Sniffer()
for f in files:
encoding = 'UTF-8'
if f is not sys.stdin and quiet < 2:
print ('Processing %s...' % f.name),
sys.stdout.flush()
data = f.read()
prefix = data[:1024 ** 2]
if f is not sys.stdin and chardet:
encoding = chardet.detect(prefix)['encoding']
dialect = sniffer.sniff(prefix, delimiters=',\t')
f = StringIO(data)
if f is not sys.stdin and quiet < 2:
print '(as %s)' % encoding
if encoding != 'UTF-8':
f = codecs.iterencode(
codecs.iterdecode(f, encoding), 'utf-8')
if args.dry_run:
# just the headers
data = [f.next()]
else:
data = list(f)
reader = csv.DictReader(data, dialect=dialect)
if args.dry_run:
d = GetSupportingDefaultDict()
importer(d)
fieldnames = set(reader.fieldnames)
d = set(d)
used = fieldnames & d
print 'Used keys (%i):' % len(used)
for k in sorted(used):
print '*', k
def serialize(self, out):
"""Write the units back to file."""
# Thanks to iterencode, a possible BOM is written only once
for chunk in iterencode((unit.getoutput() for unit in self.units), self.encoding):
out.write(chunk)
#coding=utf-8
'''
Created on 2012-5-4
@author: zhaojp
'''
from codecs import iterencode;
from codecs import iterdecode;
echars = iterencode(['你', '好', '啊'], 'utf-8');
dchars = iterdecode(echars, 'utf-8');
for c in dchars:
print(c)
def encode(input, output):
# TODO tests
for chunk in codecs.iterencode(chunks_text(input), 'utf-8'):
output.write(chunk)
def csv_unireader(self, f, encoding="utf-8"):
import codecs
import csv
for row in csv.reader(codecs.iterencode(codecs.iterdecode(f, encoding), "utf-8")):
yield [e.decode("utf-8") for e in row]
def test_basics(self):
s = u"abc123" # all codecs should be able to encode these
for encoding in all_unicode_encodings:
name = codecs.lookup(encoding).name
if encoding.endswith("_codec"):
name += "_codec"
elif encoding == "latin_1":
name = "latin_1"
self.assertEqual(encoding.replace("_", "-"), name.replace("_", "-"))
(bytes, size) = codecs.getencoder(encoding)(s)
if encoding != "unicode_internal":
self.assertEqual(size, len(s), "%r != %r (encoding=%r)" % (size, len(s), encoding))
(chars, size) = codecs.getdecoder(encoding)(bytes)
self.assertEqual(chars, s, "%r != %r (encoding=%r)" % (chars, s, encoding))
if encoding not in broken_unicode_with_streams:
# check stream reader/writer
q = Queue()
writer = codecs.getwriter(encoding)(q)
encodedresult = ""
for c in s:
writer.write(c)
encodedresult += q.read()
q = Queue()
reader = codecs.getreader(encoding)(q)
decodedresult = u""
for c in encodedresult:
q.write(c)
decodedresult += reader.read()
self.assertEqual(decodedresult, s, "%r != %r (encoding=%r)" % (decodedresult, s, encoding))
# check incremental decoder/encoder (fetched via the Python
# and C API) and iterencode()/iterdecode()
try:
encoder = codecs.getincrementalencoder(encoding)()
cencoder = _testcapi.codec_incrementalencoder(encoding)
except LookupError: # no IncrementalEncoder
pass
else:
# check incremental decoder/encoder
encodedresult = ""
for c in s:
encodedresult += encoder.encode(c)
encodedresult += encoder.encode(u"", True)
decoder = codecs.getincrementaldecoder(encoding)()
decodedresult = u""
for c in encodedresult:
decodedresult += decoder.decode(c)
decodedresult += decoder.decode("", True)
self.assertEqual(decodedresult, s, "%r != %r (encoding=%r)" % (decodedresult, s, encoding))
# check C API
encodedresult = ""
for c in s:
encodedresult += cencoder.encode(c)
encodedresult += cencoder.encode(u"", True)
cdecoder = _testcapi.codec_incrementaldecoder(encoding)
decodedresult = u""
for c in encodedresult:
decodedresult += cdecoder.decode(c)
decodedresult += cdecoder.decode("", True)
self.assertEqual(decodedresult, s, "%r != %r (encoding=%r)" % (decodedresult, s, encoding))
# check iterencode()/iterdecode()
result = u"".join(codecs.iterdecode(codecs.iterencode(s, encoding), encoding))
self.assertEqual(result, s, "%r != %r (encoding=%r)" % (result, s, encoding))
# check iterencode()/iterdecode() with empty string
result = u"".join(codecs.iterdecode(codecs.iterencode(u"", encoding), encoding))
self.assertEqual(result, u"")
def csv_unireader(f, encoding="utf-8", **fmtparams):
data = csv.reader(
codecs.iterencode(codecs.iterdecode(f, encoding), "utf-8"),
**fmtparams)
for row in data:
yield [e.decode("utf-8") for e in row]
def csv_unireader(f, encoding="utf-8"):
for row in csv.reader(codecs.iterencode(codecs.iterdecode(f, encoding), "utf-8"), delimiter=';', quotechar='"'):
yield [e.decode("utf-8") for e in row]
def csv_rows(file_obj, dialect=csv.excel, **kwargs):
csvfile = iterencode(file_obj, 'utf-8') if PY2 else file_obj
csvreader = csv.reader(csvfile, dialect=dialect, **kwargs)
csvreader = (list(iterdecode(i, 'utf-8')) for i in csvreader) if PY2 else csvreader
for row in csvreader:
yield row
def csv_unireader(f, encoding="utf-8", delimiter=',', quotechar='"'):
for row in csv.reader(codecs.iterencode(codecs.iterdecode(f, encoding),
"utf-8"),
delimiter=delimiter,
quotechar=quotechar):
yield [e.decode("utf-8") for e in row]
def csv_unireader(f, encoding="utf-8"):
for row in csv.reader(
codecs.iterencode(codecs.iterdecode(f, encoding), "utf-8")):
yield [e.decode("utf-8") for e in row]
def writelines(self, lines):
# function scope import, but this is a bug workaround for pandas.
from codecs import iterencode
encoded_lines = iterencode(lines, self.encoding)
self.bytes_filelike.writelines(encoded_lines)
def test_basics(self):
s = u"abc123" # all codecs should be able to encode these
for encoding in all_unicode_encodings:
name = codecs.lookup(encoding).name
if encoding.endswith("_codec"):
name += "_codec"
elif encoding == "latin_1":
name = "latin_1"
self.assertEqual(encoding.replace("_", "-"),
name.replace("_", "-"))
(bytes, size) = codecs.getencoder(encoding)(s)
if encoding != "unicode_internal":
self.assertEqual(
size, len(s),
"%r != %r (encoding=%r)" % (size, len(s), encoding))
(chars, size) = codecs.getdecoder(encoding)(bytes)
self.assertEqual(chars, s,
"%r != %r (encoding=%r)" % (chars, s, encoding))
if encoding not in broken_unicode_with_streams:
# check stream reader/writer
q = Queue()
writer = codecs.getwriter(encoding)(q)
encodedresult = ""
for c in s:
writer.write(c)
encodedresult += q.read()
q = Queue()
reader = codecs.getreader(encoding)(q)
decodedresult = u""
for c in encodedresult:
q.write(c)
decodedresult += reader.read()
self.assertEqual(
decodedresult, s,
"%r != %r (encoding=%r)" % (decodedresult, s, encoding))
# check incremental decoder/encoder (fetched via the Python
# and C API) and iterencode()/iterdecode()
try:
encoder = codecs.getincrementalencoder(encoding)()
cencoder = _testcapi.codec_incrementalencoder(encoding)
except LookupError: # no IncrementalEncoder
pass
else:
# check incremental decoder/encoder
encodedresult = ""
for c in s:
encodedresult += encoder.encode(c)
encodedresult += encoder.encode(u"", True)
decoder = codecs.getincrementaldecoder(encoding)()
decodedresult = u""
for c in encodedresult:
decodedresult += decoder.decode(c)
decodedresult += decoder.decode("", True)
self.assertEqual(
decodedresult, s, "%r != %r (encoding=%r)" %
(decodedresult, s, encoding))
# check C API
encodedresult = ""
for c in s:
encodedresult += cencoder.encode(c)
encodedresult += cencoder.encode(u"", True)
cdecoder = _testcapi.codec_incrementaldecoder(encoding)
decodedresult = u""
for c in encodedresult:
decodedresult += cdecoder.decode(c)
decodedresult += cdecoder.decode("", True)
self.assertEqual(
decodedresult, s, "%r != %r (encoding=%r)" %
(decodedresult, s, encoding))
# check iterencode()/iterdecode()
result = u"".join(
codecs.iterdecode(codecs.iterencode(s, encoding),
encoding))
self.assertEqual(
result, s,
"%r != %r (encoding=%r)" % (result, s, encoding))
# check iterencode()/iterdecode() with empty string
result = u"".join(
codecs.iterdecode(codecs.iterencode(u"", encoding),
encoding))
self.assertEqual(result, u"")
def handle(self, *args, **options):
for file_name in args:
csv_file = codecs.open(file_name, 'rb', 'cp1252')
encoded_file = codecs.iterencode(csv_file, 'utf-8')
self.consume_csv(encoded_file)