比较机器人框架中的函数的Json结果

Question

试图在机器人框架中创建一个测试用例。

我们有几个休息api返回我们Json作为结果。 为了调用这种情况，我们在rest2.py中使用了以下代码

def restJSON():
    r = requests.get("http://httpbin.org/get")
#    print "Code" , r.status_code
#    print "Text " , r.text
    return r.json()

我们将json存储在文件输出中。 我们编写了机器人测试用例来评估json比较，如下所示

*** Settings ***
 Library   rest2.py
 Library   OperatingSystem
*** Test Cases ***
 Example that calls a python keyword
  ${result}=   restJSON
  ${json}=   Get file   output
  Should be equal   ${result}   ${json}

但是当我们运行pybot测试用例时，我们得到的错误是两个json都不一样。

--------------------------------
pybot testSuite.txt
--------------------------------
==============================================================================
testSuite
==============================================================================
Example that calls a python keyword                                   | FAIL |
{u'origin': u'10.252.30.94, 69.241.25.16', u'headers': {u'Via': u'1.1 localhost (squid/3.1.14)', u'Accept-Encoding': u'gzip, deflate, compress', u'Accept': u'*/*', u'User-Agent': u'python-requests/2.2.1 CPython/2.7.6 Linux/3.16.0-30-generic', u'Host': u'httpbin.org', u'Cache-Control': u'max-age=259200'}, u'args': {}, u'url': u'http://httpbin.org/get'} != {u'origin': u'10.252.30.94, 69.241.25.16', u'headers': {u'Via': u'1.1 localhost (squid/3.1.14)', u'Accept-Encoding': u'gzip, deflate, compress', u'Accept': u'*/*', u'User-Agent': u'python-requests/2.2.1 CPython/2.7.6 Linux/3.16.0-30-generic', u'Host': u'httpbin.org', u'Cache-Control': u'max-age=259200'}, u'args': {}, u'url': u'http://httpbin.org/get'}
------------------------------------------------------------------------------
testSuite                                                             | FAIL |
1 critical test, 0 passed, 1 failed
1 test total, 0 passed, 1 failed
==============================================================================

json文件是相同的。 但仍然是它的失败测试案例说两者都不一样。 这是比较的正确方法，还是我们有任何其他方法在机器人框架中执行此操作。

Answer 1

为什么不使用现有的库：

示例代码：

*** Settings ***
Library                     RequestsLibrary
Library                     Collections
Library                     XML  use_lxml=True
Force Tags                  REST


*** Variables ***
${SERVICE_ROOT}  http://ip.jsontest.com/
${SERVICE_NAME}  testing

*** Keywords ***
json_property_should_equal
    [Arguments]  ${json}  ${property}  ${value_expected}
    ${value_found} =    Get From Dictionary  ${json}  ${property}
    ${error_message} =  Catenate  SEPARATOR=  Expected value for property "  ${property}  " was "  ${value_expected}  " but found "  ${value_found}  "
    Should Be Equal As Strings  ${value_found}  ${value_expected}  ${error_message}    values=false

*** Test Cases ***
Example REST JSON
  Create session  ${SERVICE_NAME}  ${SERVICE_ROOT}
  ${headers}=  Create Dictionary  Content-Type=application/json  Accept=application/json
  ${result}=  Get Request  ${SERVICE_NAME}  ${SERVICE_ROOT}  headers=${headers}
  Should Be Equal  ${result.status_code}  ${200}
  Log  ${result.content}
  ${json}=    To Json    ${result.content}
  ${pp}=  To Json  ${result.content}  pretty_print=True
  Log  ${pp}
  json_property_should_equal  ${json}  ip  [Your_IP]

记得安装所有需要的库：

pip install robotframework-requests
pip install requests

Answer 2

你做什么回应：

Should Be Equal As Strings    ${result}   ${json}

我发现你实现你想要的有点奇怪的方法，如果我是你，我将使用http库或请求库

https://github.com/peritus/robotframework-httplibrary https://github.com/bulkan/robotframework-requests

Answer 3

Get File读取Get File的内容并返回一个字符串; 同时python函数返回一个dict对象。 所以看起来你正在将字典与字符串进行比较 - 而且它们根本无法返回相等。

如果将文件输出转换为dict，则检查很可能会通过：

# your other code
${json_file}=   Get file   output
${json_file}=   Evaluate   json.loads("""${json_file}""")    json     # use the json module to transform str->dict

Should Be Equal    ${result}    ${json_file}

# even better - more verbose logging if values differ, or there are missing keys:
Dictionaries Should Be Equal    ${result}    ${json_file}

Answer 4

工作实例

把它放在.robot文件中

*** Settings ***
Library    jsonlib.py

*** Variables ***
${response_json}    {"key1": 33, "key2": "string", "key3": [1,2,3,"sring"], "name": "hans"}
${expected_json}    {"key1": 77, "key2": "string", "key3": [1,2,3,"sring"], "name": "OTTO", "age": 33}

*** Test Cases ***
response meets expectation
    &{json1}=    Evaluate    json.loads('''${response_json}''')    json
    &{json2}=    Evaluate    json.loads('''${expected_json}''')    json
    Log Dictionary    ${json1}
    Log Dictionary    ${json2}
    # ${type}=    evaluate    type(${json1}).__name__
    # ${type}=    evaluate    type(&{json2})
    response_meets_expectation  ${json1}  ${json2}

compare json payloads
    &{json1}=    Evaluate    json.loads('''${response_json}''')    json
    &{json2}=    Evaluate    json.loads('''${expected_json}''')    json
    compare_json_payloads  ${json1}  ${json2}

json payloads should match
    &{json1}=    Evaluate    json.loads('''${response_json}''')    json
    &{json2}=    Evaluate    json.loads('''${expected_json}''')    json
    json_paylodas_should_match  ${json1}  ${json2}

把它放在jsonlib.py中与.robot文件相同的文件夹中

# -*- coding: utf-8 -*-

import json
import jsondiff
from deepdiff import DeepDiff
# from deepdiff import DeepSearch, grep
from pprint import pprint
from robot.api import logger


class JsonCompareError(Exception):
    pass

def response_meets_expectation(response, expectation):
    union = jsondiff.patch(response, expectation)
    difference = jsondiff.diff(response, union)
    if difference:
            raise JsonCompareError("Diff found: {}".format(difference))
    else:
        return True

def compare_json_payloads(response_json, expected_json):
    logger.debug("Compare JSON payloads")
    diff = DeepDiff(response_json, expected_json, ignore_order=True, verbose_level=2)
    logger.debug("Values changed: {}".format(diff['values_changed']))
    return diff.to_dict()

def json_paylodas_should_match(response_json, expected_json):
    diff = DeepDiff(response_json, expected_json, verbose_level=2)
    if diff != {}:
        raise JsonCompareError("Payloads do not match! Differences: {}".format(diff))
    else:
        return True

Answer 5

使用https://github.com/Accruent/robotframework-zoomba它将是这样的：

Library         Zoomba.APILibrary

*** Test Cases ***
Example Test
   ${headers}=  Create Dictionary  Content-Type=application/json  Accept=application/json
   ${actual_response}=   Call Get Request   ${headers}    http://httpbin.org/get
   Validate Response Contains Expected Response     ${actual_response}    ${expected_response}

这将通过密钥来验证结果，任何失败的东西都会如下所示：

Key(s) Did Not Match:
------------------
Key: pear
Expected: fish
Actual: bird
------------------
Full List Breakdown:
Expected: [{'apple': 'cat', 'banana': 'dog', 'pear': 'fish'}, {'apple': 'cat', 'banana': 'mice', 'pear': 'bird'}, {'apple': 'dog', 'banana': 'mice', 'pear': 'cat'}]
Actual: [{'apple': 'cat', 'banana': 'dog', 'pear': 'bird'}]

Please see differing value(s)"

文档： https ： //accruent.github.io/robotframework-zoomba/docs/APILibraryDocumentation.html