-
Notifications
You must be signed in to change notification settings - Fork 0
/
adh_prep_algeria.py
203 lines (170 loc) · 6.16 KB
/
adh_prep_algeria.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
# -*- coding: utf-8 -*-
"""
Created on Mon Sep 19 12:52:22 2022
@author: heiko
"""
import tabula
import pandas as pd
from datetime import datetime, timedelta
import glob, shutil
import re, os
from translate import Translator
# functions and definitions
def get_last_date_of_month(year, month):
"""Return the last date of the month.
Args:
year (int): Year, i.e. 2022
month (int): Month, i.e. 1 for January
Returns:
date (datetime): Last date of the current month
"""
if month == 12:
last_date = datetime(year, month, 31)
else:
last_date = datetime(year, month + 1, 1) + timedelta(days=-1)
return last_date.strftime("%Y-%m-%d")
months = dict({'Jan':1,
'Fev':2,
'Mar':3,
'Avr':4,
'Mai':5,
'Juin':6,
'Juil':7,
'Aout':8,
'Sep':9,
'Oct':10,
'Nov':11,
'Dec':12
})
def mapp_values(df,template):
template = template.loc[:,['Indicator.Name','Indicator.Code']]
# note: unique to Algeria
values = ['All',
'Food',
'Tobacco',
'Clothing',
#'Communication',
'Education',
'Housing',
'Furni',
'Health',
'Miscellaneous',
'Recreation',
'Restaurants',
'Transport',
'Insurance']
for i in range(len(values)):
val = template[template['Indicator.Name'].str.contains(values[i],case=False)==True]
try:
df['Indicator.Name'][df['Indicator.Name'].str.contains(values[i],case=False)==True] = val['Indicator.Name'].values
except:
print('ERROR with: {}'.format(values[i]))
df = pd.merge(template,df,how='left',on = 'Indicator.Name')
df = df.round(2)
return df
#%%
def translate_divisions(df):
divs = df['Indicator.Name'].to_list()
translator= Translator(from_lang="french",to_lang="english")
divs2 = []
for i in range(0,len(divs)):
try:
divs2.append(translator.translate(divs[i]))
except:
try:
divs2.append(translator.translate(divs[i].split(' ')[2]))
except:
print('tanslation failed: {}'.format(divs[i]))
divs2.append(divs[i])
df = df.rename(columns={'Indicator.Name': 'original_language'})
df['Indicator.Name'] = divs2
#translation = translator.translate("Guten Morgen")
#divs2 = [translator.translate(x) for x in divs]
return df
def execute(data_path, country):
#codes = pd.read_csv('./data/codeList.csv')
# get template
if '_' in country:
c = country.split('_')
c = [i.capitalize() for i in c]
country2 = ' '.join(c)
else:
country2 = country.capitalize()
tables = tabula.read_pdf("{}.pdf".format(data_path), pages=(2), stream=True)
df = tables[0]
df = df.drop([0,1])
df = df.iloc[:,[0,4]]
month = [val for key, val in months.items() if key.lower() in data_path][0]
year = re.search(r'.*([1-3][0-9]{3})',data_path).group(1) # [1-3] = num between 1-3, [0-9]{3} = num 0-9 repeat 3 times
year = int(year)
last = get_last_date_of_month(year, month)
df.columns = ['Indicator.Name',last]
df = translate_divisions(df)
df['Indicator.Name'][df['Indicator.Name'].str.contains('ASSEMBLY',case=False)==True] = "All"
df['Indicator.Name'][df['Indicator.Name'].str.contains('Expenses',case=False)==True] = "Housing"
df['Indicator.Name'][df['Indicator.Name'].str.contains('Personal',case=False)==True] = "Health"
df['Indicator.Name'][df['Indicator.Name'].str.contains('Shoes',case=False)==True] = "Clothing"
# save this in csv folder
csv_folder = './data/%s/csv/'% country
# create csv_folder folder
if not os.path.exists(csv_folder):
os.makedirs(csv_folder)
df.to_csv('{}{}.csv'.format(csv_folder,data_path.split('raw/')[1]),index=False)
#%% check if there are new files
country = 'algeria'
base_data_path ='./data/%s/raw/'% country
files_list = glob.glob('%s*.pdf'% base_data_path)
for i in range(len(files_list)):
files_list[i] = files_list[i].replace("\\","/")
#%%
#check for data log
data_log = glob.glob('%sdata_log.txt'% base_data_path)
if len(data_log)==0:
f = open('%sdata_log.txt'% base_data_path,'w')
for i in range(len(files_list)):
data_path = files_list[i].split('.pdf')[0]
execute(data_path, country)
f.write(files_list[i])
f.write('\n')
f.close()
else:
logs = pd.read_csv('%sdata_log.txt'% base_data_path,header=None)
logs.columns=['done']
logs = logs.done.to_list()
files = pd.DataFrame()
files['files'] = files_list
file = files[~files.files.isin(logs)]
if len(file) != 0:
print('Preparing %s data...'% country)
f = open('%sdata_log.txt'% base_data_path,'w')
for i in range(len(file)):
data_path = file.files.to_list()[i].split('.pdf')[0]
print(data_path)
try:
execute(data_path, country)
f.write(file.files.to_list()[i])
f.write('\n')
except:
print('failed %s'% data_path)
f.close()
else:
print('No new %s country data'% country)
#%%
def template(country):
#codes = pd.read_csv('./data/codeList.csv')
# get template
if '_' in country:
c = country.split('_')
c = [i.capitalize() for i in c]
country2 = ' '.join(c)
else:
country2 = country.capitalize()
file = './outputs/ckan/bk/template.csv'
df_template = pd.read_csv(file)
df_template = df_template[df_template['Country']==country2]
#df_template = df_template.iloc[:,[0,1,2,3,4,-2,-1]]
# save this in csv folder
csv_folder = './data/%s/csv/'% country
df_template.to_csv('{}{}_template.csv'.format(csv_folder,country),index=False)
#template(country)
#data_path = 'C:/Users/heiko/Documents/Work/OCL/ADH/Inflation/adh_inflation_database_v2/data/algeria/raw/ipc_septembre2022'