def MEASURE_CODE(code, mlist):
if str(code) in measure_file['code2']:
return measure_file['code2'][code]
elif code == '' or str(code) == 'nan':
return ''
else:
logging.info(mlist.keys())
ERROR('Measures未知代碼: ' + code)
def START_DATE(freq):
if find_unknown == False:
if freq == 'A':
return dealing_start_year
elif freq == 'Q':
return str(dealing_start_year) + '-Q1'
elif freq == 'M':
return str(dealing_start_year) + '-01'
else:
ERROR('頻率錯誤: ' + freq)
else:
return None
def SUBJECT_CODE(code, slist):
if code in subject_file['code2']:
return subject_file['code2'][code]
else:
logging.info(slist.keys())
ERROR('Subjects未知代碼: ' + code)
def COUNTRY_NAME(location):
if location in country['Country_Name']:
return country['Country_Name'][location]
else:
ERROR('找不到國家: ' + location)
def COUNTRY_CODE(location):
if location in country['Country_Code']:
return country['Country_Code'][location]
else:
ERROR('國家代碼錯誤: ' + location)
def MEI_DATA(i, name, MEI_t, code_num, table_num, KEY_DATA, DATA_BASE,
db_table_t, DB_name, snl, freqlist, frequency):
freqlen = len(freqlist)
NonValue = ['nan', '']
if code_num >= 200:
db_table = DB_TABLE + frequency + '_' + str(table_num).rjust(4, '0')
DATA_BASE[db_table] = db_table_t
DB_name.append(db_table)
table_num += 1
code_num = 1
db_table_t = pd.DataFrame(index=freqlist, columns=[])
value = MEI_t[MEI_t.columns[i]]
db_table = DB_TABLE + frequency + '_' + str(table_num).rjust(4, '0')
db_code = DB_CODE + str(code_num).rjust(3, '0')
#db_table_t[db_code] = ['' for tmp in range(freqlen)]
db_table_t = pd.concat([
db_table_t,
pd.DataFrame(
['' for tmp in range(freqlen)], index=freqlist, columns=[db_code])
],
axis=1)
start_found = False
last_found = False
found = False
for k in range(len(value)):
try:
freq_index = int(value.index[k])
except:
freq_index = str(value.index[k])
if freq_index in db_table_t.index and (
(find_unknown == False
and int(str(freq_index)[:4]) >= dealing_start_year)
or find_unknown == True):
if str(value.iloc[k]).strip() in NonValue:
db_table_t[db_code][freq_index] = ''
else:
try:
db_table_t[db_code][freq_index] = float(value.iloc[k])
except ValueError:
ERROR('Nontype Value detected: ' + str(value.iloc[k]))
found = True
if start_found == False:
if frequency == 'A':
start = int(freq_index)
else:
start = str(freq_index)
start_found = True
else:
continue
if start_found == False:
if found == True:
ERROR('start not found: ' + str(name))
try:
last = db_table_t[db_code].loc[~db_table_t[db_code].isin(NonValue
)].index[-1]
except IndexError:
if found == True:
ERROR('last not found: ' + str(name))
if found == False:
start = 'Nan'
last = 'Nan'
Subject = subjects_list[MEI_t.columns[i][1]]
Measure = measures_list[str(MEI_t.columns[i][2])]
PowerCode = MEI_t.columns[i][4]
if str(MEI_t.columns[i][3]).find('Unnamed') >= 0 or str(
MEI_t.columns[i][3]).find('nan') == 0:
if dataset == 'MEI_BTS_COS':
Unit = 'Percentage'
else:
Unit = ''
else:
Unit = MEI_t.columns[i][3]
if Measure == '':
desc_e = str(Subject) + ', ' + str(PowerCode) + ' of ' + str(Unit)
else:
desc_e = str(Subject) + ', ' + str(Measure) + ', ' + str(
PowerCode) + ' of ' + str(Unit)
if str(Subject).find('>') > 0:
sub = str(Subject).find('>') - 1
form_e = str(Subject)[:sub]
elif dataset == 'MEI_CLI':
form_found = False
for form in form_e_dict1:
if MEI_t.columns[i][1] in form_e_dict1[form]:
form_e = str(form)
form_found = True
break
if form_found == False:
form_e = 'Others'
elif dataset == 'MEI_BTS_COS':
form_found = False
for form in form_e_dict2:
if MEI_t.columns[i][1] in form_e_dict2[form]:
form_e = str(form)
form_found = True
break
if form_found == False:
form_e = 'Others'
else:
form_e = 'Others'
if str(MEI_t.columns[i][2]) == 'nan':
desc_c = str(MEI_t.columns[i][1])
else:
desc_c = str(MEI_t.columns[i][1]) + str(MEI_t.columns[i][2])
unit = str(PowerCode) + ' of ' + str(Unit)
name_ord = MEI_t.columns[i][0]
book = COUNTRY_NAME(MEI_t.columns[i][0])
desc_e = desc_e + ' - ' + book
if str(MEI_t.columns[i][5]).isnumeric():
form_c = int(MEI_t.columns[i][5])
if desc_e.find('=') < 0:
desc_e = desc_e.replace('of Index',
'of Index(' + str(form_c) + ')')
elif str(MEI_t.columns[i][5]).find('Unnamed') >= 0 or str(
MEI_t.columns[i][5]).find('nan') == 0 or str(
MEI_t.columns[i][5]).strip() == '':
form_c = ''
else:
form_c = MEI_t.columns[i][5]
if desc_e.find('=') < 0:
desc_e = desc_e.replace('of Index',
'of Index(' + str(form_c) + ')')
#flags = MEI_t['Flags'][i]
key_tmp = [
databank, name, db_table, db_code, desc_e, desc_c, frequency, start,
last, unit, name_ord, snl, book, form_e, form_c
]
KEY_DATA.append(key_tmp)
snl += 1
code_num += 1
return code_num, table_num, DATA_BASE, db_table_t, DB_name, snl
break
sys.stdout.write("\n\n")
if merging:
logging.info('Process: File Merging\n')
elif updating:
logging.info('Process: File Updating\n')
logging.info('Reading main key: ' + NAME + 'key' + main_suf +
'.xlsx, Time: ' + str(int(time.time() - tStart)) + ' s' +
'\n')
main_file = readExcelFile(out_path + NAME + 'key' + main_suf + '.xlsx',
header_=0,
index_col_=0,
sheet_name_=NAME + 'key',
acceptNoFile=False)
if main_file.empty:
ERROR('Empty updated_file')
logging.info('Reading main database: ' + NAME + 'database' + main_suf +
'.xlsx, Time: ' + str(int(time.time() - tStart)) + ' s' +
'\n')
main_database = readExcelFile(out_path + NAME + 'database' + main_suf +
'.xlsx',
header_=0,
index_col_=0,
acceptNoFile=False)
if merge_file_loaded:
merge_file = df_key
merge_database = DATA_BASE_dict
else:
logging.info('Reading original key: ' + NAME + 'key' + merge_suf +
'.xlsx, Time: ' + str(int(time.time() - tStart)) + ' s' +
'\n')