class IfrsOperatingRevenueAssembler():
def __init__(self):
self.base_xpath = '//html/body'
self.content_screener = ContentScreener()
self.string_utils = StringUtils()
def assemble(self, param):
content, stock_symbol, date = param['content'], param['stock_symbol'], param['date']
self.content_screener.screen(param)
content = self.string_utils.normalize_string(content)
html_object = lxml.html.fromstring(content)
relative_html_object = self.__traverse_to_relative_html_object(html_object)
column_name_list = self.__assemble_column_name_list(relative_html_object)
row_list = self.__assemble_row_list(relative_html_object)
return OperatingRevenueDao(column_name_list, row_list, stock_symbol, date)
def __traverse_to_relative_html_object(self, html_object):
relative_html_object_list = html_object.xpath(self.base_xpath)
assert len(relative_html_object_list) > 0, 'invalid base_xpath'
return relative_html_object_list[0]
def __assemble_column_name_list(self, relative_html_object):
# traverse and sanity check
tr_tags = relative_html_object.xpath('./table[@class="hasBorder"]/tr')
assert len(tr_tags) > 0, 'invalid tr_tags'
# traverse and sanity check
th_texts = tr_tags[0].xpath('./th/text()')
assert len(th_texts) == 2, 'invalid th_texts size, should be 2'
# should be account
account = th_texts[0]
# traverse and sanity check
table_tags = relative_html_object.xpath('./table[@class="noBorder"]')
assert len(table_tags) > 0, 'invalid table_tags'
td_tags = table_tags[2].xpath('./td')
assert len(td_tags) > 0, 'invalid td_tags'
# should be snapdate
snapdate = self.string_utils.from_local_string_to_date(td_tags[1].text)
return [account, snapdate]
def __assemble_row_list(self, relative_html_object):
# skip one row of column name list
tr_tags = relative_html_object.xpath('./table[@class="hasBorder"]/tr')[1:]
return [self.__assemble_row(tr_tag) for tr_tag in tr_tags]
def __assemble_row(self, relative_html_object):
# should be item
th_texts = relative_html_object.xpath('./th/text()')
assert len(th_texts) == 1, 'invalid th_texts size, should be 1'
item = th_texts[0]
# should be number (operating revenue)
td_texts = relative_html_object.xpath('./td/text()')
assert len(th_texts) == 1, 'invalid td_texts size, should be 1'
number_string = td_texts[0]
number = self.string_utils.normalize_number(number_string)
return [item, number]
class OperatingRevenueSummaryAssembler:
def __init__(self):
self.content_screener = ContentScreener()
self.date_utils = DateUtils()
self.string_utils = StringUtils()
def assemble(self, param):
content, date = param["content"], param["date"]
self.content_screener.screen(param)
stmt_date = self.date_utils.get_last_date_of_month(date)
column_name_list, row_list, release_date = self.__assemble_summary(content)
return OperatingRevenueSummaryDao(column_name_list, row_list, stmt_date, release_date)
def __assemble_summary(self, html_object):
try:
return AriesParser().parse(html_object)
except AssertionError:
return self.__assemble_summary_step_1(html_object)
def __assemble_summary_step_1(self, html_object):
try:
return TaurusParser().parse(html_object)
except AssertionError:
return self.__assemble_summary_step_2(html_object)
def __assemble_summary_step_2(self, html_object):
try:
return GeminiParser().parse(html_object)
except AssertionError:
return self.__assemble_summary_step_3(html_object)
def __assemble_summary_step_3(self, html_object):
try:
return CancerParser().parse(html_object)
except AssertionError:
return LeoParser().parse(html_object)
def __init__(self):
self.base_xpath = '//html/body'
self.content_screener = ContentScreener()
self.string_utils = StringUtils()
def __init__(self):
self.base_xpath = '//html/body[@id="content_d"]/center/table[@class="result_table hasBorder"]'
self.content_screener = ContentScreener()
self.string_utils = StringUtils()
class XbrlBalanceSheetAssembler():
def __init__(self):
self.base_xpath = '//html/body[@id="content_d"]/center/table[@class="result_table hasBorder"]'
self.content_screener = ContentScreener()
self.string_utils = StringUtils()
def assemble(self, param):
content, stock_symbol, date = param['content'], param['stock_symbol'], param['date']
self.content_screener.screen(param)
content = self.string_utils.normalize_string(content)
html_object = lxml.html.fromstring(content)
relative_html_object = self.__traverse_to_relative_html_object(html_object)
column_name_list = self.__assemble_column_name_list(relative_html_object)
row_list = self.__assemble_row_list(relative_html_object)
return BalanceSheetDao(column_name_list, row_list, stock_symbol, date)
def __traverse_to_relative_html_object(self, html_object):
relative_html_object_list = html_object.xpath(self.base_xpath)
assert len(relative_html_object_list) == 1, 'invalid base_xpath'
return relative_html_object_list[0]
def __assemble_column_name_list(self, relative_html_object):
# traverse and sanity check
tr_tags = relative_html_object.xpath('./tr[@class="tblHead"]')
assert len(tr_tags) == 2, 'invalid tr_tags'
# traverse and sanity check
statement_th_texts = tr_tags[1].xpath('./th/text()')
assert len(statement_th_texts) == 1, 'invalid statement_th_texts'
assert unicode(statement_th_texts[0]) == u'資產負債表', 'invalid statement_th_texts[0]'
column_name_list = []
# should be account type
column_th_texts = tr_tags[0].xpath('./th/text()')
account_type = column_th_texts[0] # of unicode type
column_name_list.append(account_type)
# should be snapshot dates
for local_string in column_th_texts[1:]:
snapshot_date = self.string_utils.from_local_string_to_date(local_string) # of datetime.date type
column_name_list.append(snapshot_date)
return column_name_list
def __assemble_row_list(self, relative_html_object):
# skip one row of statement name and one row of column name list
tr_tags = relative_html_object.xpath('./tr')[2:]
return [self.__assemble_row(tr_tag) for tr_tag in tr_tags]
def __assemble_row(self, relative_html_object):
row = []
td_texts = relative_html_object.xpath('./td/text()')
# should be account type
account_type = td_texts[0].strip()
row.append(account_type)
# should be number
for number_string in td_texts[1:]:
number = self.string_utils.normalize_number(number_string)
row.append(number)
return row
def __init__(self):
self.base_xpath = '//html/body/div[@id="SysJustIFRAMEDIV"]/table/tr/td/table/tr/td/table'
self.content_screener = ContentScreener()
self.string_utils = StringUtils()
class DividendPolicyAssembler():
def __init__(self):
self.base_xpath = '//html/body/div[@id="SysJustIFRAMEDIV"]/table/tr/td/table/tr/td/table'
self.content_screener = ContentScreener()
self.string_utils = StringUtils()
def assemble(self, param):
content, stock_symbol = param['content'], param['stock_symbol']
self.content_screener.screen(param)
html_object = self.__get_html_object(content)
relative_html_object = self.__traverse_to_relative_html_object(html_object)
column_name_list = self.__assemble_column_name_list(relative_html_object)
row_list = self.__assemble_row_list(relative_html_object)
return DividendPolicyDao(column_name_list, row_list, stock_symbol)
def __get_html_object(self, content):
content = self.string_utils.normalize_string(content)
content = content.replace(u'<br>', u'')
return lxml.html.fromstring(content)
def __traverse_to_relative_html_object(self, html_object):
relative_html_object_list = html_object.xpath(self.base_xpath)
assert len(relative_html_object_list) > 0, 'invalid base_xpath (table_tags)'
return relative_html_object_list[0]
def __assemble_column_name_list(self, relative_html_object):
# traverse and sanity check
tr_tags = relative_html_object.xpath('./tr')
assert len(tr_tags) > 2, 'invalid tr_tags'
# skip the first row of header
td_texts = tr_tags[2].xpath('./td/text()')
assert len(td_texts) == 7, 'invalid td_texts size, should be 7'
return [text.strip() for text in td_texts]
def __assemble_row_list(self, relative_html_object):
# skip the first row of header
# skip the second row of empty lines
# skip the third row of column name list
tr_tags = relative_html_object.xpath('./tr')
assert len(tr_tags) > 2, 'invalid tr_tags'
return [self.__assemble_row(tr_tag) for tr_tag in tr_tags[3:]]
def __assemble_row(self, relative_html_object):
# should be item
td_texts = relative_html_object.xpath('./td/text()')
assert len(td_texts) == 7, 'invalid td_texts size, should be 7'
row = []
# should be stmt_date
stmt_date = self.string_utils.from_local_string_to_date(td_texts[0])
row.append(stmt_date)
# should be number
for number_string in td_texts[1:-1]:
number = self.string_utils.normalize_number(number_string)
row.append(number)
# should be number in percentage
for number_string in td_texts[-1:]:
number = self.string_utils.normalize_number(number_string) * 0.01
row.append(number)
return row
def __init__(self):
self.content_screener = ContentScreener()
self.date_utils = DateUtils()
self.string_utils = StringUtils()
