過去数か月分のデータをDFに追加していく

APIが混雑している際に再取得するdef fetch_url(url):を追加

import pandas as pd
import plotly.offline as pyo
import plotly.graph_objs as go
import urllib.request
from time import sleep
from datetime import datetime, date, timedelta
from dateutil.relativedelta import relativedelta
import ssl
ssl._create_default_https_context = ssl._create_unverified_context

#パラメーターの作成
#trade data type C Commodities S Services 
datatype = "type=C"
#china=156,UK=826,US=842,jp=392or all
reporting_area = "&r=842"
#Annual=A,Monthly=M
frequency = "&freq=M"
#YYYY or YYYYMM or now or recent　
time_period = "&ps=201201"
# Harmonized System (HS)
classification = "&px=H4"
#china=156,UK=826,US=842,or all
partner_area = "&p=all"
# 1 (imports) and 2 (exports),
import_or_export = "&rg=1"
#HS CODE
classification_code = "&cc=610910"
#csv or json
fmt = "&fmt=csv"
#取得する相手国の数
amount_country = 6
#取得期間（年or月)
term=72
# term = 72
#一番古いデータの年月（更新あり）
oldest = "201201"

url = ("https://comtrade.un.org/api/get?"+datatype + reporting_area+frequency+time_period+classification+partner_area+
      import_or_export+classification_code+fmt)


#csvをdfに入れる

res = urllib.request.urlopen(url)
df = pd.read_csv(res)

#最新年を取得
newest = (df["Period"].max())


#最新年のデータのみを取得
df_n = df[df["Period"] == newest]

#valueが高い順に並べ替え
df_s = df_n.sort_values('Trade Value (US$)', ascending=False)

#上位10件を取得してリスト化
partner = []
for var in range(1, amount_country):
    partner.append(df_s.iat[var, 12])

    



    
#ここに過去の日付のcsvを追加するコードを入れる
def get_oldtime(tdatetime):
    tdatetime = datetime.strptime(str(oldest),"%Y%m")
    tdatetime =(tdatetime + relativedelta(months=1))  
    return tdatetime.strftime('%Y%m')



#一か月ごとのパラメータを生成
def add_old(old):
    addurl = ("https://comtrade.un.org/api/get?"+datatype + reporting_area+frequency+old+classification+partner_area+
      import_or_export+classification_code+fmt)
    sleep(2)
    return addurl




#APIが混雑している際に再取得する
def fetch_url(url):
    max_times=10
    sleep_sec=5
    retry_count = 0
    while True:
        f = urllib.request.urlopen(url)
        
        try:
            retry_count += 1
            if f.getcode() == 200 or retry_count >= max_times:
                # 200が返ってくるか、最大試行回数に到達した場合ループを抜ける
                return f
                break
            time.sleep(sleep_sec)
        except ValueError:
            print("Oops!  That was no valid number.  Try again...")   



for i in range(1,term):
    oldest = get_oldtime(oldest)
    time_period = "&ps=" + str(oldest)
    added_url = add_old(time_period)
    added_res = fetch_url(added_url)
    print(added_res.getcode())
    added_df = pd.read_csv(added_res)
    df = pd.concat([df,added_df])
    print(oldest)



data = []
for i in partner:
    
    df2 = df[df["Partner"] == i]
    trace = go.Scatter(x=df2["Period Desc."],
                       y=df2["Trade Value (US$)"],
                       mode = 'lines',
                       name = i)
    data.append(trace)

    
layout = go.Layout(
    title="アメリカが輸入するTシャツ等HS6109.10該当貨物"
)

fig = go.Figure(data=data, layout=layout)

pyo.plot(fig,filename="ustshirt.html")

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

import pandas as pd

import plotly.offline as pyo

import plotly.graph_objs as go

import urllib.request

from time import sleep

from datetime import datetime, date, timedelta

from dateutil.relativedelta import relativedelta

import ssl

ssl._create_default_https_context = ssl._create_unverified_context

#パラメーターの作成

#trade data type C Commodities S Services

datatype = "type=C"

#china=156,UK=826,US=842,jp=392or all

reporting_area = "&r=842"

#Annual=A,Monthly=M

frequency = "&freq=M"

#YYYY or YYYYMM or now or recent　

time_period = "&ps=201201"

# Harmonized System (HS)

classification = "&px=H4"

#china=156,UK=826,US=842,or all

partner_area = "&p=all"

# 1 (imports) and 2 (exports),

import_or_export = "&rg=1"

#HS CODE

classification_code = "&cc=610910"

#csv or json

fmt = "&fmt=csv"

#取得する相手国の数

amount_country = 6

#取得期間（年or月)

term=72

# term = 72

#一番古いデータの年月（更新あり）

oldest = "201201"

url = ("https://comtrade.un.org/api/get?"+datatype + reporting_area+frequency+time_period+classification+partner_area+

import_or_export+classification_code+fmt)

#csvをdfに入れる

res = urllib.request.urlopen(url)

df = pd.read_csv(res)

#最新年を取得

newest = (df["Period"].max())

#最新年のデータのみを取得

df_n = df[df["Period"] == newest]

#valueが高い順に並べ替え

df_s = df_n.sort_values('Trade Value (US$)', ascending=False)

#上位10件を取得してリスト化

partner = []

for var in range(1, amount_country):

partner.append(df_s.iat[var, 12])

#ここに過去の日付のcsvを追加するコードを入れる

def get_oldtime(tdatetime):

tdatetime = datetime.strptime(str(oldest),"%Y%m")

tdatetime =(tdatetime + relativedelta(months=1))

return tdatetime.strftime('%Y%m')

#一か月ごとのパラメータを生成

def add_old(old):

addurl = ("https://comtrade.un.org/api/get?"+datatype + reporting_area+frequency+old+classification+partner_area+

import_or_export+classification_code+fmt)

sleep(2)

return addurl

#APIが混雑している際に再取得する

def fetch_url(url):

max_times=10

sleep_sec=5

retry_count = 0

while True:

f = urllib.request.urlopen(url)

try:

retry_count += 1

if f.getcode() == 200 or retry_count >= max_times:

# 200が返ってくるか、最大試行回数に到達した場合ループを抜ける

return f

break

time.sleep(sleep_sec)

except ValueError:

print("Oops! That was no valid number. Try again...")

for i in range(1,term):

oldest = get_oldtime(oldest)

time_period = "&ps=" + str(oldest)

added_url = add_old(time_period)

added_res = fetch_url(added_url)

print(added_res.getcode())

added_df = pd.read_csv(added_res)

df = pd.concat([df,added_df])

print(oldest)

data = []

for i in partner:

df2 = df[df["Partner"] == i]

trace = go.Scatter(x=df2["Period Desc."],

y=df2["Trade Value (US$)"],

mode = 'lines',

name = i)

data.append(trace)

layout = go.Layout(

title="アメリカが輸入するTシャツ等HS6109.10該当貨物"

)

fig = go.Figure(data=data, layout=layout)

pyo.plot(fig,filename="ustshirt.html")

[`yahoo` not found]

Reader Interactions

コメントを残す コメントをキャンセル

コメントを残すコメントをキャンセル