中文 README

Guide：https://ngta.readthedocs.io

Note:

• from version 1.0, only support python>=3.5.2.
• from version 2.0, don't compatible with unittest.
• from version 4.0, only support python>=3.6.5.

Introduce:

NGTA's full name is Next Generation Test Automation, it designed with following keywords:

• Business independent.
• Cross platform.
• Powerful parametrize for testcase.
• Add testbench concept.
• Add events mechanism.
• Local mode:
  • powerful yml configuration with schema.
  • HTML report can be extended and customized.
  • Support running testcases with multi-process when testbench is not exclusive.
• Agent mode:
  • Consume testcase from RabbitMQ with multi-processes.
  • Provide Http Restful interface.
  • Support Distributed.
  • Integrate with cavia platform.
    • backend: magna
    • frontend: porcellus

Installation

1. Install ngta：

   pip install ngta

2. Generate work dir：

   python -m ngta init --dest-dir=${WORKDIR} --include-sample

Note: ${WORKDIR} indicate the target dir.

Why we need work dir:

1. Test code also need version control, please use GIT.
2. Good working structure is helpful for managing testcases.

The generated work dir:

${WORKDIR}                          root dir
|-- bin                             
    |-- agent.exe                   exe file for installing test agent as service in Windows.
    |-- agent.service               systemd service file for linux. 
    |-- agent.xml                   the config file for installing as service in Windows.
|-- cases                           tesetcase dir
    |-- sample1                     project or product dir1(GIT submodule1)
    |-- sample2                     project or product dir2(GIT submodule2)
|-- conf                            agent config files 
|-- lib                             internal libraries.
|-- logs                            userd for saving log, result and report.
|-- .ngta                           anchor file, also used to store configuration.

Node: For integrating with platform, create each project or product as git repository, and put it as git submodule into ${WORKDIR}/cases.

Sample：run test by yml config

1. Open command-line, run following commands:

   cd ${WORKDIR}
   python -m ngta run --config=cases\sample\base\config.yml

2. After running finished, we can found log and report in ${WORKDIR}\logs.
3. In ${WORKDIR}\cases\sample\base\config.xml, it config how to load test in ${WORKDIR}\cases\sample\base\test_equal.py and inject the parameters.

Sample：run test by path

1. Open command-line, run following commands:

   cd ${WORKDIR}
   python -m ngta run --locate=sample.base.test_equal

2. After runing finished, we can found log and report in ${WORKDIR}\logs.

Note：

• --locate can be a package, module, class, method, py file or dir.
• When specify --locate, it also support other arguments --tag, --include, --failfast, --repeat-number and etc.

Sample：run .py file

1. Open command-line, run following commands:

   cd ${WORKDIR}\cases\sample\base
   python test_equal.py

2. After runing finished, we can found report.html in ${WORKDIR}\cases\sample\base.

Different with unittest.TestCase:

Review ${WORKDIR}\cases\sample\base\test_equal.py:

# coding: utf-8
from ngta import TestCase, tag, test


class EqualTestCase(TestCase):
    @tag("regression")
    @test(u"Test {value1} equal with {value2}")
    def test_int(self, value1=1, value2=2):
        msg = "%s should equal with %s." % (value1, value2)
        self.assert_that(value1, msg).is_equal_to(value2)

• @tag specify the tag regression
• @test used to specify this is a testcase. Test method or function name start with test are also supported.
• assert_that reference: assertpy

Report Sample:

• Screenshot:
• HTML: report.html

More features: Guide