JsonSchema
使用fastjsonschema來校驗數據
# 導入驗證器
import json
import fastjsonschema
# 讀取schema
with open('../schema/oneof-schema.json', encoding='utf8') as f:
my_schema = json.load(f)
# json數據:
with open('../data/test.json', encoding='utf8') as f:
json_data = json.load(f)
# 驗證:
fastjsonschema.validate(my_schema, json_data)
使用jsonschema來校驗數據
import json
# 導入驗證器
from jsonschema import validate, draft7_format_checker, SchemaError, ValidationError
if __name__ == '__main__':
with open('../schema/MySchema.json', encoding='utf8') as f:
my_schema = json.load(f)
# json數據:
with open('../data/cece.json', encoding='utf8') as f:
json_data = json.load(f)
# error_list = check_type(my_schema, json_data)
# print(error_list)
# 驗證:
try:
validate(instance=json_data, schema=my_schema, format_checker=draft7_format_checker)
# Draft7Validator.format_checker
except SchemaError as serr:
print("schema 錯誤 【%s】 \nschema錯誤" % str(serr))
except ValidationError as verr:
print("數據 錯誤 【%s】 \n數據錯誤" % str(verr))
MySchema
JSONSchema缺點
- 錯誤提示英文
- 校驗數據爲一步步校驗,遇到錯誤停止
自定義JSONSchema
schema遵循 http://json-schema.org/,
“$schema”: “http://json-schema.org/draft-07/schema#”,
使用方法 >>>> 點擊這裏
代碼
個人編寫的校驗的代碼,自定義成分較多
目前僅僅擴展了 string類型的數據 format 的選型判斷
CheckDataUti.py
import re
import time
# email 正則表達式
EMAIL_REGEX = "^\w+([-+.]\w+)*@\w+([-.]\w+)*\.\w+([-.]\w+)*$"
# URL 正則表達式
URL_REGEX = "^[a-zA-z]+://[^\s]*$"
# PHONE 正則表達式
PHONE_REGEX = "^([1][3,4,5,6,7,8,9])\d{9}$"
# 身份證 正則表達式
ID_CARD_REGEX = "^((\d{18})|([0-9x]{18})|([0-9X]{18}))$"
# 郵政編碼 正則表達式
ZIP_CODE_REGEX = "^[1-9]\d{5}(?!\d)$"
# IP 地址 正則表達式
IP_REGEX = "^\d+\.\d+\.\d+\.\d+$"
# 正整數
INTEGER_REGEX = "^[1-9]\d*$"
ERR_LIST = []
COMMON_ERR_LIST = []
def log_error(msg, data, schema, is_common=False):
"""
打印錯誤日誌
"""
err_log = "%s,數據:【%s】,校驗規則: %s" % (str(msg), str(data) + " type of " + str(type(data).__name__), str(schema))
if not is_common:
ERR_LIST.append(err_log)
print("=================================================")
print(err_log)
print("=================================================")
else:
COMMON_ERR_LIST.append(err_log)
def check_object(data, schema, is_common):
"""
校驗對象格式
【 properties、required、minProperties、maxProperties、patternProperties、additionalProperties 】
"""
if type(data) != dict:
log_error("當前校驗的json不是一個對象格式", data, schema, is_common)
else:
# 獲取當前校驗數據的所有key
keys = dict.keys(data)
# 處理必需值
if "required" in schema:
required_schema = schema['required']
for schema_key in required_schema:
if schema_key not in keys:
log_error("字段【%s】必填" % schema_key, data, schema, is_common)
# 處理最小key和最大key
if "minProperties" in schema:
min_properties = schema['minProperties']
if len(keys) < min_properties:
log_error("校驗數據的key個數小於【%s】" % str(min_properties), data, schema, is_common)
if "maxProperties" in schema:
max_properties = schema['maxProperties']
if len(keys) > max_properties:
log_error("校驗數據的key個數大於【%s】" % str(max_properties), data, schema, is_common)
# 處理具體的key
if "properties" in schema:
# 處理 properties
properties_schema = schema['properties']
schema_keys = dict.keys(properties_schema)
for data_key in schema_keys:
if data_key in data:
check_data(properties_schema[data_key], data[data_key])
# 處理滿足正則表達式的key
if "patternProperties" in schema:
# 處理 properties
pattern_properties = schema['patternProperties']
schema_keys = dict.keys(pattern_properties)
# 循環所有正則表達式的key
for schema_key in schema_keys:
# 循環當前待校驗的數據key
for data_key in keys:
# 僅僅處理滿足正則表達式的key數據
if re.match(schema_key, data_key):
check_data(pattern_properties[schema_key], data[data_key])
def check_array(data, schema, is_common):
"""
校驗數組格式
【 items、additionalItems、minItems、maxItems、uniqueItems 】
"""
if type(data) != list:
log_error("當前校驗的json不是數組格式", data, schema, is_common)
else:
# minItems、maxItems
# 判斷最小值
if "minItems" in schema:
min_items = schema['minItems']
if len(data) < min_items:
log_error("當前校驗的數據數組長度小於【%s】" % str(min_items), data, schema, is_common)
# 判斷最大值
if "maxItems" in schema:
max_properties = schema['maxItems']
if len(data) > max_properties:
log_error("當前校驗的數據數組長度大於【%s】" % str(max_properties), data, schema, is_common)
# uniqueItems true 數組元素不能重複
if "uniqueItems" in schema:
unique_items_schema = schema['uniqueItems']
if unique_items_schema:
# 數組元素不能重複
try:
if len(set(data)) != len(data):
log_error("當前校驗的數據數組元素不能重複", data, schema, is_common)
except TypeError:
# 存在數組內部元素是dict格式
pass
# 判斷每一個items
if "items" in schema:
items_schema = schema["items"]
# 判斷items_schema 是數組還是對象
if type(items_schema) is list:
# 如果是數組 每一個item都是一個jsonSchema 索引對應的數組內索引的格式
index = 0
for item_sc in items_schema:
check_data(item_sc, data[index])
index += 1
# additionalItems該關鍵字只有在items是數組的時候纔會有效
# additionalItems 除了上述規定之外的數據必需符合指定的規則
if "additionalItems" in schema:
additional_items_schema = schema['additionalItems']
for i in range(index, len(data)):
check_data(additional_items_schema, data[i])
# items如果是對象 當前schema規範了數組內所有元素的格式
elif type(items_schema) is dict:
for item_data in data:
check_data(items_schema, item_data)
def check_number(data, schema, is_common):
"""
校驗數字類型
"""
if type(data) not in (int, float):
log_error("當前校驗的json不是一個數字格式", data, schema, is_common)
else:
# 判斷最大值 maximum 如果exclusiveMaximum該關鍵字是True 包含本身
if "maximum" in schema:
maximum_schema = schema['maximum']
if 'exclusiveMaximum' in schema and schema['exclusiveMaximum']:
if data >= maximum_schema:
log_error("當前校驗的數據大於等於【%s】" % maximum_schema, data, schema, is_common)
else:
if data > maximum_schema:
log_error("當前校驗的數據大於【%s】" % maximum_schema, data, schema, is_common)
# minimum、exclusiveMinimum
if "minimum" in schema:
minimum_schema = schema['minimum']
if 'exclusiveMinimum' in schema and schema['exclusiveMinimum']:
if data <= minimum_schema:
log_error("當前校驗的數據小於等於【%s】" % minimum_schema, data, schema, is_common)
else:
if data < minimum_schema:
log_error("當前校驗的數據小於【%s】" % minimum_schema, data, schema, is_common)
# multipleOf 整除
if "multipleOf" in schema:
multiple_of_schema = schema['multipleOf']
if not data % multiple_of_schema == 0:
log_error("當前校驗的數據不能被%s整除" % multiple_of_schema, data, schema, is_common)
def check_str(data, schema, is_common):
"""
校驗字符串類型
涉及的關鍵字 【maxLength、minLength、pattern、format】
"""
if type(data) != str:
log_error("當前校驗的數據不是一個字符串格式", data, schema, is_common)
else:
# maxLength
if "maxLength" in schema:
max_length_schema = schema['maxLength']
if len(data) > max_length_schema:
log_error("當前校驗的數據長度大於%d" % max_length_schema, data, schema, is_common)
# minLength
if "minLength" in schema:
min_length_schema = schema['minLength']
if len(data) < min_length_schema:
log_error("當前校驗的數據長度小於%d" % min_length_schema, data, schema, is_common)
# pattern
if "pattern" in schema:
pattern_schema = schema['pattern']
if not re.match(pattern_schema, data):
log_error("當前校驗的數據不符合正則表達式規則【%s】" % pattern_schema, data, schema, is_common)
# format
if 'format' in schema:
format_schema = schema['format']
if format_schema == 'email' and not re.match(EMAIL_REGEX, data):
log_error("當前校驗的數據不是正確的郵箱格式", data, schema, is_common)
elif format_schema == 'phone' and not re.match(PHONE_REGEX, data):
log_error("當前校驗的數據不是正確的手機號碼格式", data, schema, is_common)
elif format_schema == 'hostname' and not re.match(IP_REGEX, data):
log_error("當前校驗的數據不是正確的IP地址格式", data, schema, is_common)
elif format_schema == 'idCard' and not re.match(ID_CARD_REGEX, data):
log_error("當前校驗的數據不是正確的身份證格式", data, schema, is_common)
elif format_schema == 'date':
format_patten = '%Y-%m-%d'
if 'format_patten' in schema:
format_patten = schema['format_patten']
try:
time.strptime(data, format_patten)
except ValueError:
log_error("當前校驗的數據不是正確的日期格式格式【%s】" % format_patten, data, schema, is_common)
def check_common(schema, data):
"""
校驗通用的
涉及到關鍵字:
【 enum、const、allOf、anyOf、oneOf、not、 if……then…… 】
"""
if "enum" in schema:
enum_schema = schema['enum']
if data not in enum_schema:
log_error("當前校驗的數據值不存在【%s】中" % str(enum_schema), data, schema)
if "const" in schema:
const_schema = schema['const']
if data != const_schema:
log_error("當前校驗數據值不等於【%s】" % str(const_schema), data, schema)
if "allOf" in schema:
all_of_schema = schema['allOf']
for item_schema in all_of_schema:
check_data(item_schema, data)
if "anyOf" in schema:
any_of_schema = schema['anyOf']
begin_len = len(COMMON_ERR_LIST)
for item_schema in any_of_schema:
check_data(item_schema, data, True)
end_len = len(COMMON_ERR_LIST)
if end_len - begin_len == len(any_of_schema):
log_error("當前校驗的數據不符合當前anyof中的任一規則", data, schema)
if "oneOf" in schema:
one_of_schema = schema['oneOf']
begin_len = len(COMMON_ERR_LIST)
for item_schema in one_of_schema:
check_data(item_schema, data, True)
end_len = len(COMMON_ERR_LIST)
if end_len - begin_len != len(one_of_schema) - 1:
log_error("待校驗JSON元素不能通過oneOf的校驗", data, schema)
if "not" in schema:
not_schema = schema['not']
begin_len = len(COMMON_ERR_LIST)
check_data(not_schema, data, True)
end_len = len(COMMON_ERR_LIST)
if end_len == begin_len:
log_error("待校驗JSON元素不能通過not規則的校驗", data, schema)
# if……then……
if 'if' in schema:
if_schmea = schema['if']
begin_len = len(COMMON_ERR_LIST)
check_data(if_schmea, data, True)
end_len = len(COMMON_ERR_LIST)
if end_len == begin_len:
if "then" in schema:
then_schema = schema['then']
check_data(then_schema, data, False)
else:
if "else" in schema:
else_schema = schema['else']
check_data(else_schema, data, False)
def get_data_type(data):
"""
獲取type
"""
if type(data) == dict:
return 'object'
if type(data) == list:
return 'array'
if type(data) in (int, float):
return 'number'
if type(data) == str:
return 'string'
if type(data) == bool:
return 'boolean'
def check_data(schema, data, is_common=False):
# 優先處理 通用的
check_common(schema, data)
# 沒有type的情況
# type 默認爲string
type_name = schema['type'] if "type" in schema else get_data_type(data)
if type_name == 'object':
check_object(data, schema, is_common)
elif type_name == 'array':
check_array(data, schema, is_common)
elif type_name in ['integer', 'number']:
check_number(data, schema, is_common)
elif type_name == 'string':
check_str(data, schema, is_common)
# type是布爾類型
elif type_name == 'boolean':
if type(data) != bool:
log_error("當前校驗的數據不是一個boolean格式", data, schema, is_common)
JsonSchmea 的數據示例
{
"$schema": "http://json-schema.org/draft-07/schema#",
"type": "object",
"minProperties": 1,
"maxProperties": 200,
"properties": {
"name": {
"type": "string",
"enum": [
"shaofei",
"upuptop",
"pyfysf"
]
},
"email": {
"type": "string",
"format": "email",
"const": "[email protected]"
},
"idCard": {
"type": "string",
"format": "idCard",
"pattern": "\\d+"
},
"phone": {
"type": "string",
"format": "phone"
},
"hostname": {
"type": "string",
"format": "hostname"
},
"createTime": {
"format": "date",
"format_patten": "%Y%m%d"
},
"is": {
"type": "boolean"
},
"age": {
"type": "integer",
"maximum": 20,
"minimum": 1,
"multipleOf": 2
},
"like": {
"type": "array"
}
},
"allOf": [
{
"type": "string"
}
],
"patternProperties": {
"^\\S+123$": {
"type": "integer"
}
},
"required": [
"email"
]
}
使用方式
import json
from CheckDataUti import check_data
if __name__ == '__main__':
with open('../schema/MySchema.json', encoding='utf8') as f:
my_schema = json.load(f)
# json數據:
with open('../data/cece.json', encoding='utf8') as f:
json_data = json.load(f)
check_data(my_schema, json_data)
# print(ERR_LIST)
參考:
schema遵循 http://json-schema.org/,
“$schema”: “http://json-schema.org/draft-07/schema#”,
使用方法 >>>> 點擊這裏