def cut(self, sentence, punctuation=True):
sentence_length = len(sentence)
cursor = 0
while cursor < sentence_length:
if self.is_chinese_char(sentence[cursor]):
DEBUG('begin pos 中文')
chunks = self.__get_chunks(sentence,
cursor) # Matching Algorithm
if len(chunks) > 0:
words, length = self.__ambiguity_resolution(
chunks) # Ambiguity Resolution Rules
for term in list(filter(None, words)):
yield term
# Set the next begin pos
cursor += length
else:
# for a single char in Chinese
yield sentence[cursor]
cursor += 1
elif self.V.is_punctuation(sentence[cursor]) and punctuation:
DEBUG("begin pos 字符")
yield sentence[cursor]
cursor += 1
else:
DEBUG('begin pos 非中文单词(英文单词, etc.)')
word, cursor = self.__match_none_chinese_words(
sentence, cursor)
yield word
DEBUG("cursor %s" % cursor)
DEBUG("char %s" % sentence[cursor - 1])
def __match_none_chinese_words(sentence, begin_pos):
'''
切割出非中文词
'''
DEBUG('__match_none_chinese_words %s' % sentence[begin_pos])
cursor = begin_pos
# Skip pre-word whitespaces and punctuations
# 跳过中英文标点和空格
while cursor < len(sentence):
ch = sentence[cursor]
if Tokenizer.is_ascii_char(ch) or Tokenizer.is_chinese_char(ch):
break
cursor += 1
# 得到英文单词的起始位置
start = cursor
# 找出英文单词的结束位置
while cursor < len(sentence):
ch = sentence[cursor]
if not Tokenizer.is_ascii_char(ch):
break
cursor += 1
end = cursor
#Skip chinese word whitespaces and punctuations
# 跳过中英文标点和空格
while cursor < len(sentence):
ch = sentence[cursor]
if Tokenizer.is_ascii_char(ch) or Tokenizer.is_chinese_char(ch):
break
cursor += 1
# 返回英文单词和游标地址
return sentence[start:end], cursor
def __ambiguity_resolution(self, chunks):
'''
根据当前游标位置进行切词
'''
DEBUG('__ambiguity_resolution 开始消岐')
for x in chunks:
[DEBUG(y.text) for y in x.words]
DEBUG('-' * 20)
if len(chunks) > 1: # Rule 1: 根据 total_word_length 进行消岐
DEBUG("# Rule 1: 根据 total_word_length 进行消岐")
score = max([x.total_word_length for x in chunks])
chunks = list(filter(None, \
[ x if x.total_word_length == score \
else None for x in chunks]))
# for x in chunks: [DEBUG(y.text) for y in x.words]; DEBUG('-'*20)
if len(chunks) > 1: # Rule 2: 根据 average_word_length 进行消岐
DEBUG("# Rule 2: 根据 average_word_length 进行消岐")
score = max([x.average_word_length for x in chunks])
chunks = list(filter(None, \
[ x if x.average_word_length == score \
else None for x in chunks]))
if len(chunks) > 1: # Rule 3: 根据 standard_deviation 进行消岐
score = max([x.standard_deviation for x in chunks])
chunks = list(filter(None, \
[ x if x.standard_deviation == score \
else None for x in chunks]))
if len(chunks) > 1: # Rule 4: 根据 word_frequency 进行消岐
score = max([x.word_frequency for x in chunks])
chunks = list(filter(None, \
[ x if x.word_frequency == score \
else None for x in chunks]))
if len(chunks) != 1:
'''
分词失败
'''
DEBUG("分词失败")
Tokenizer.__print_chunks(chunks)
return '', 1
words = chunks[0].words
return [w.text for w in words], reduce(lambda x, y: x + y.length,
words, 0)
smpv.setDebug(debug)
if smpv.connected is not True:
print('An issue with connection to the SolarmanPV API')
sys.exit(1)
power_details = smpv.getPower(datetime.date.today().strftime("%Y-%m-%d"), True)
if power_details is not None:
power = power_details['power']
# need to convert date and time value from UTC to local time before uploading to pvoutput.org
power_datetime_local = utc_to_local(
datetime.datetime.strptime(power_details['time'],
"%Y-%m-%dT%H:%M:%SZ"))
power_date = power_datetime_local.strftime("%Y%m%d")
power_time = power_datetime_local.strftime("%H:%M")
DEBUG('power == ' + str(power) + ' and power_date == ' + power_date)
try:
# Create connection to pvoutput.org
pvout = PVoutput_Connection(pvo_key, pvo_system_id)
# update pvoutput - but only if there is a value power value (i.e. > 0)
# add_output() is for end of day output - therfore the peak power, pear time, etc
# pvout.add_output(power_date, generated=power)
if power > 0:
pvout.add_status(power_date, power_time, power_exp=power)
else:
DEBUG('no need to update - power %dW' % power)
except:
print('An error with PVoutput ', sys.exc_info()[0])
def getPower(self, date_to_retrieve=None, most_recent_value=None):
if date_to_retrieve is None:
date_to_retrieve = datetime.date.today().strftime('%Y-%m-%d')
if self.debug:
DEBUG('today == ' + date_to_retrieve)
DEBUG('Getting plant power (for a day)')
# Get the power data for a specified date or today
url = solarman_pv_api_base + '/plant/power'
params = {
'plant_id': self.__plant_id,
'date': date_to_retrieve,
'timezone_id': 'Europe/Amsterdam'
}
try:
response = requests.get(url,
verify=self.__requests_verify,
headers=self.__auth_headers,
params=params)
except requests.exceptions.ConnectionError as e:
print '%s: connection failed - %s' % (self.__class__.__name__, e)
return None
response.encoding = 'utf-8'
if self.debug:
print response
print response.url
print response.encoding
print response.text
#print response.json()
# validate response
if 'data' not in response.json() and 'powers' not in response.json():
# should return None, maybe an exception
print '%s: data or powers not in response: %s' % (
self.__class__.__name__, response.text)
return None
if most_recent_value is True:
# Sort the json() (just to be sure), take the last value
power_data = response.json()['data']['powers']
most_recent_power_data = None
if isinstance(power_data, list):
power_data.sort(key=self.__extractTime, reverse=True)
# temporary exception handler to nut out but first thing in the morning
try:
most_recent_power_data = power_data[0]
except IndexError:
# Some error in response from the API, i.e. an empty list
most_recent_power_data = None
except:
# Effectively an unhandled error - retaining this debug whilst in beta testing mode
print '%s: Exception: getPower(): An error with power_data %s' % (
self.__class__.__name__, sys.exc_info()[0])
print str(power_data)
print power_data
else:
# temporary debug - whilst in beta testing mode
print str(power_data)
return most_recent_power_data
else:
return response.json()
def __print_words(words, tag='__print_words'):
for x in words:
DEBUG('%s: %s length: %d' % (tag, x.text, x.length))
def __print_chunks(chunks, tag='__print_chunks'):
for x in chunks:
for y in x.words:
DEBUG('%s: %s' % (tag, ' '.join([w.text for w in y])))
def __init__(self, dict_path=os.path.join(curdir, 'dict.txt')):
DEBUG('Vocabulary loaded.')
self.V = Vocabulary(dict_path=dict_path)