在網上查閱了好幾個EXCEL轉Json的代碼，有的是僅支持一個層級的Json（這個就不支持多層級的json），有的太過復雜看的不是很懂，所以就自己寫了一個目前符合自己使用的代碼。
我先講下實現的方式。
如果遇到一個如下的json格式，我該怎么存到excel方便讀取出來呢？
{
"name": "haha",
"value": 12,
"floor_area": 43.5,
"categories": [
{
"id": 1,
"extra_property": "xixi",
"renovation_type": [
1,
2
],
"subcategories": [
{
"subcategory_id": 1,
"subcategory_value": 2
}
]
}
]
}
這是個多維json，存在excel中不是很好存放，讀取出來也得花點心思，畢竟你不知道后期又會有什么格式的json出現。為了應對千奇百怪的json格式，我想到如下方式
上圖中特別標注的，可以划分為三個部分，以下對這三個部分進行講解。
區域一（功能校驗區域）：
user：登陸的用戶或操作的用戶
real_code：預計接口執行操作后需要返回的code值，如200、401等
real_message：預計接口執行操作后需要返回的message，如“請求成功‘、”無操作權限“等
區域二（參數類型區域）：
傳遞參數的類型無外乎這幾種：int、str、float、boolean。同樣我們需要告訴代碼這個參數是什么類型的參數，以免傳遞數據的時候發生參數類型錯誤。在代碼中我做了處理，可以識別常用的四種類型（不區分大小寫）。
區域三（參數區域）：
如果是數組類型的數據，用"[]"來標記，不同的數據存儲在不同的單元格中，如：
看到圖的時候你會問：為什么sheet里面會有參數的名稱？
我們可以看出“categories”是個哈希，底下還存在不同的參數。為了知道哪些參數是在“categories”下的，我們可以用一張sheet去存儲“categories”參，這樣我們可以通過代碼方便去找到。
有可能“categories”存在多組數據，所以我們需要用"[]"來告知代碼。要讀取那些數據，我們可以存儲行號，不同的行號可以用頓號分隔
附上代碼：
# -*- coding: utf-8 -*-
# !/usr/bin/python3
import os
import re
import xlrd
import json
import unicodedata
_author_ = 'garcia'
class DataCenter:
def __init__(self):
# Default File Path:
self.data_dir = os.getenv('AUTO_DATACENTER', '存放excel的文件夾地址')
self.filename = ''
self.path = ''
self.sheetNames = ''
@staticmethod
def is_number(val):
"""判斷val是否是str"""
try:
return True
except ValueError:
pass
try:
unicodedata.numeric(val)
return True
except (TypeError, ValueError):
pass
def keep_integer_type_from_excel(self, value):
# Keep integer number as integer type. When reading from excel it has been changed to float type.
if self.is_number(value) and type(value) != str and value % 1 == 0:
return int(value)
else:
return value
def change_field_type(self, table, col, val):
# 字段類型轉換
field_type = table.cell(0, col).value
val = self.keep_integer_type_from_excel(val)
try:
if val == '' or val is None:
pass
elif field_type.lower() == 'int':
return int(val)
elif field_type.lower() == 'float':
return float(val)
elif field_type.lower() == 'boolean':
return int(bool(val))
elif field_type.lower() == 'str' or field_type == '' or field_type is None:
return str(val)
else:
return '字段類型錯誤！'
except Exception as e:
print(e)
@staticmethod
def unic(item):
# Resolved Chinese mess code.
try:
item = json.dumps(item, ensure_ascii=False, encoding='UTF-8')
except UnicodeDecodeError:
try:
item = json.dumps(item, ensure_ascii=False, encoding='UTF-8')
except:
pass
except:
pass
# Convert json data string back
item = json.loads(item, encoding='UTF-8')
return item
@staticmethod
def get_sheet_names(wb):
"""
Returns the names of all the worksheets in the current workbook.
"""
sheet_names = wb.sheet_names()
return sheet_names
@staticmethod
def __convert_to_list(val):
"""轉換字符串為list"""
value_list = re.split(',|，|、', val)
for i in range(len(value_list)):
value_list[i] = int(value_list[i])
return value_list
def get_table(self, sheet_name):
if self.path is None:
# Default Data Directory
file = os.path.join(self.data_dir, self.filename)
else:
file = os.path.join(self.path, self.filename)
try:
excel_date = xlrd.open_workbook(file)
# 得到excel的全部sheet標簽值
self.sheetNames = self.get_sheet_names(excel_date)
my_sheet_index = self.sheetNames.index(sheet_name)
table = excel_date.sheet_by_index(my_sheet_index)
except Exception as e:
print(e)
return table
@staticmethod
def get_row_and_col(table):
"""獲取列數、行數"""
total_row = table.nrows
total_col = table.ncols
return total_row, total_col
@staticmethod
def get_param(table, start_col, total_col):
param_list = [] # 獲取參數
for col in range(start_col, total_col):
param = table.cell(1, col).value # 獲取字段名
if param is None or param == '':
param_list.append(param_list[-1])
else:
param_list.append(param)
return param_list
def get_child_param(self, param, row, includeEmptyCells):
if param in self.sheetNames:
table = self.get_table(param)
child_total_row, child_total_col = self.get_row_and_col(table)
child_param = self.get_param(table, 1, child_total_col)
data_dic = {}
count = 0
for col in range(1, child_total_col):
# Solve issue that get integer data from Excel file would be auto-changed to float type.
val = self.change_field_type(table, col, table.cell(row, col).value)
param = child_param[count]
count += 1
if '[]' in param:
if val == '' or val is None:
pass
else:
param = param[:param.index('[')]
data_dic[param] = [] if param not in data_dic.keys() else data_dic[param]
if param in self.sheetNames:
val_list = self.__convert_to_list(val)
for i in range(len(val_list)):
data_dic[param].append(
self.get_child_param(param, val_list[i] - 1, includeEmptyCells))
else:
data_dic[param].append(val)
else:
if param in self.sheetNames:
if val is not None and val != '':
val_list = self.__convert_to_list(val)
for i in range(len(val_list)):
data_dic[param] = self.get_child_param(param, val_list[i] - 1, includeEmptyCells)
elif (val == '' or val is None) and includeEmptyCells == 'False':
pass
else:
data_dic[param] = val
return data_dic
def param_to_json(self, filename, sheet_name, includeEmptyCells, path=None):
"""
獲取指定sheet中接口參數
:param filename: 文件名
:param sheet_name: 讀取excel的sheet名稱
:param path:文件路徑
:return:
"""
try:
self.filename = filename
self.path = path
table = self.get_table(sheet_name)
total_row, total_col = self.get_row_and_col(table)
function_point_list = []
check_list = []
user_list = []
all_data_list = []
param_list = self.get_param(table, 4, total_col)
for row in range(2, total_row):
data_dic = {}
get_check_list = []
count = 0
for col in range(4, total_col):
# Solve issue that get integer data from Excel file would be auto-changed to float type.
val = self.change_field_type(table, col, table.cell(row, col).value)
param = param_list[count]
count += 1
if '[]'in param:
if val == '' or val is None:
pass
else:
param = param[:param.index('[')]
data_dic[param] = [] if param not in data_dic.keys() else data_dic[param]
if param in self.sheetNames:
val_list = self.__convert_to_list(val)
for i in range(len(val_list)):
data_dic[param].append(
self.get_child_param(param, val_list[i] - 1, includeEmptyCells))
else:
data_dic[param].append(val)
else:
if param in self.sheetNames:
if val is not None and val != '':
val_list = self.__convert_to_list(val)
for i in range(len(val_list)):
data_dic[param] = self.get_child_param(param, val_list[i] - 1, includeEmptyCells)
else:
pass
elif (val == '' or val is None) and includeEmptyCells == 'False':
pass
else:
data_dic[param] = val
print(data_dic)
get_check_list.append(self.keep_integer_type_from_excel(table.cell(row, 2).value))
get_check_list.append(self.keep_integer_type_from_excel(table.cell(row, 3).value))
check_list.append(get_check_list)
all_data_list.append(data_dic)
user_list.append(table.cell(row, 1).value)
function_point_list.append(table.cell(row, 0).value)
except Exception as e:
print(e)
# return all_data_list, function_point_list
return user_list, all_data_list, function_point_list, check_list
if __name__ == '__main__':
dc = DataCenter()
userlist, allList, FunctionPoint, checklist = dc.param_to_json('存放數據的excel名稱', 'sheet名', 'False')
print(userlist, allList, FunctionPoint, checklist)
說到這，我們來講講她的缺點：
1、如果存在多張表的嵌套，代碼執行的時間比較長