units = {
    "15ml":["1큰술","1T","1Ts","3t"],
    "5ml":["1작은술","1t","1ts"],    
    "200ml":["1컵","1Cup","1C"],
    "250ml":["16T", "1C"],
    "180ml":["1종이컵"],
    "28.3g":["1oz"],
    "453g":["1파운드","lb"],
    "3780ml":["1갤런","gallon"],
    "2g":["1꼬집",'약간'],
    "4g":["조금"],
    "6g":["조금"],
    "10g":["적당량"],
    "50g":["1줌"], # 나물
    "100g":["1큰줌"], # 나물
    "13마리":["1줌"], # 멸치
    "26마리":["1큰줌"], # 멸치
    "100g":["1주먹"], # 여자 어른의 주먹크기
    "3cm":["1토막"],
    "1알":["1톨","1쪽"], # 마늘, 생강 등
    "600g":["1근"], # 고기
    "400g":["1근"], # 채소
    "200g":["1봉지"],
}

if len(tokens_key) == 1:
    tokens_key = tokens_key[0]
tokens_key

# if len(tokens_values)%2 == 1:
#     tokens_values = ['레시피'] + tokens_values

for _ in tokens_values:
    

import re

df_safe.loc[993]["재료정보"]

df_safe["재료정보"][528]

for _ in df_safe_recipe:
    

[_ for _ in df_safe_recipe  if len(_) > 3]

pd.DataFrame(df_safe_recipe).to_csv("recipe_temp.csv", index=None, sep="|", header=None)

df_safe_recipe[410]

# 재료 목록을 우선 모아서 정리하기

df_safe["재료정보"]

df_org = df_orf[df_orf["구분"]!='국/탕류']

pd.concat([df_taged, df_org], axis=0).to_csv("momukji_taged_temp.csv", encoding="cp949")

# 식약처 레시피 호출
import pandas as pd
df_menu_mfds = pd.read_excel(file_name, sheet_name=sht_names[0])
df_menu_mfds = df_menu_mfds.loc[:,['메뉴명','조리방법','요리종류','해쉬태그','재료정보']]
df_menu_mfds.재료정보 = [_.split('\n')  for _ in df_menu_mfds.재료정보.fillna('')]
df_menu_mfds.head(2)

df_menu_10000.recipes[100]

x = ast.literal_eval(df_menu_10000.recipes[0])
x

json.loads(df_menu_10000.recipes[0])
type(df_menu_10000.recipes[0])

# 10,000 개의 레시피에서 세부정보 수집 및 정리
df_menu_10000.recipes = [json.loads(_) for _ in df_menu_10000.recipes]

# Excel 파일 내부 Sheet 목록의 출력
file_name = r'data/momukji_lab_full.xlsx'
xls       = xlrd.open_workbook(file_name, on_demand=True)
sht_names = [_ for _ in xls.sheet_names()]
df_menu_zen = pd.read_excel(file_name, sheet_name=sht_names[-2])
df_menu_zen.tail(3)

# df_menu_zen.iloc[:, 2:]

# Excel 파일 내부 Sheet 목록의 출력
file_name = r'data/momukji_lab_Tag작업.xlsx'
import xlrd
xls = xlrd.open_workbook(file_name, on_demand=True)
sht_names = [_ for _ in xls.sheet_names()]
select_sheet = sht_names[1]
select_sheet, "/".join(sht_names)

df_menu_mfds.iloc[159,:].재료정보

df_menu_mfds.iloc[159,:]

Counter([len(_)  for no, _ in enumerate(df_menu_mfds.재료정보)])

[_  for no, _ in enumerate(df_menu_mfds.재료정보)  if len(_) == 3][:2]

df_menu_mfds.재료정보[1].split('\n')

# Excel 파일 내부 Sheet 목록의 출력
file_name = r'data/momukji_lab_Tag작업.xlsx'
import xlrd
xls = xlrd.open_workbook(file_name, on_demand=True)
sht_names = [_ for _ in xls.sheet_names()]
select_sheet = sht_names[1]
select_sheet, "/".join(sht_names)

import pandas as pd
df_org = pd.read_excel(file_name, sheet_name=select_sheet)
df = df_org[df_org.구분=='국/탕류']
# df = df.reset_index(drop=True)
df.shape, df.tail(3)

result = []
for _ in df.메뉴명:
    if _.find("찌개") != -1:
        result.append("찌개")
    else: result.append("국")
df.대분류 = result
df.구분  = '국/찌개'
# df.to_csv("muyong_tags.csv", index=None)
df.tail(3)

df_mid_term = sorted(set(df.중분류))
",".join(df_mid_term)

# 0 번 컬럼의 내용 보완
menu_name = df_mid_term[0]
df.원메뉴[df.중분류 == menu_name] = menu_name
df[df.중분류 == menu_name]

# 1 계란국 편집
menu_name = df_mid_term[1]
df.원메뉴[df.중분류 == menu_name] = menu_name
df.loc[98].원메뉴 = "부추국"
df[df.중분류 == menu_name]

# 2 근대국 편집
menu_name = df_mid_term[2]
df.원메뉴[df.중분류 == menu_name] = menu_name
# df.loc[98].원메뉴 = "부추국"
df[df.중분류 == menu_name]

df.loc[26].대분류

# 3 김치국/찌개
menu_name = df_mid_term[3]
for _ in list(df[df.중분류 == menu_name].index):
    if df.loc[_].대분류 == '국':
        df.loc[_].원메뉴 = '김칫국'
    else:
        df.loc[_].원메뉴 = '김치찌개'
df[df.중분류 == menu_name]

df.원메뉴[df.중분류 == menu_name]# = menu_name

# df.loc[98].원메뉴 = "부추국"
df.to_csv("muyong_tags.csv", index=None)
df[df.중분류 == menu_name]

df[df.중분류 == df_mid_term[0]].원메뉴 = '감자탕'
# = df_mid_term[0]
# df.loc[list(df[df.중분류 == df_mid_term[0]].index), :]

df[df.중분류 == df_mid_term[0]]

df = df.loc[:, ["메뉴명","구분","중분류","원메뉴","분류명"]]

import pandas as pd
df = pd.read_csv("muyong_tags.csv")
df.tail(3)

list(set(df.원메뉴))

df[df.원메뉴.isna()].index

len(["찌개" for _ in df.메뉴명   if _.find("찌개") != -1])

df.

import pandas as pd
food_menu_org = pd.read_csv("data/food_recipe.csv")
food_menu_org.shape, food_menu_org.head(1)

from momukji import Nlp
nlp = Nlp('data/nouns_tokens.txt')
menu_valid = [["_".join(nlp.food_nouns(_)), _]  for _ in food_menu_org.메뉴명]
food_menu_org.insert(2, 'menu_token',[_[0]  for _ in menu_valid])
print(food_menu_org.head(2))

food_menu_org.to_csv("data/food_recipe_addToken.csv", index=None)

# 저장된 내용 확인하기
import pandas as pd
df_menu = pd.read_csv("data/food_recipe_10000.csv", sep=';', header=None)
df_menu.columns = ['url','menu','tags','recipe']
menu_valid = [["_".join(nlp.food_nouns(_)), _]  for _ in df_menu.menu]
df_menu.head(1), menu_valid[:5]

import re
df_menu.menu = [re.sub("\d+회", "", _).strip()   for _ in df_menu.menu]
df_menu.menu = [re.sub("\d+분", "", _).strip()   for _ in df_menu.menu]
",".join(list(df_menu.menu)[:20])

menu_valid = [["_".join(nlp.food_nouns(_)), _]  for _ in df_menu.menu.fillna("")]
menu_valid[:5]

"///".join([_[0]+":"+_[1]  for _ in menu_valid])