fitnesse 中各类fit fixture的python实现

虽然网上都说slim效率很高,无奈找不到支持python的方法,继续用pyfit

1 Column Fixture

特点:行表格展现形式,一条测试用例对应一行数据

Wiki

!define COMMAND_PATTERN {python "%m" %p}

!define TEST_RUNNER {C:\Python27\PyFIT-0.8a2\fit\FitServer.py}

!path E:\selfworkspaces\fitTest

!|ColumnFixtureTest.AddTest |

| a| b| add?|

|1|2|3|

|1|3|3|

Python:

from fit.ColumnFixture import ColumnFixture

class AddTest(ColumnFixture):

_typeDict = {

"a":"Int",

"b":"Int",

"add":"Int",

}

def __init__(self):

ColumnFixture.__init__(self)

self.a=''

self.b=''

def add(self):

c=self.a+self.b

return c

Action Fixture

主要用于测试一系列事件流,可直接调用其他类进行测试

Wiki:

第一行填入ActionFixture

Start:调用的类名

Enter:传入的参数

Press:执行一个没有返回值的方法

Check:检查结果

Wiki:

!define COMMAND_PATTERN {python "%m" %p}

!define TEST_RUNNER {C:\Python27\PyFIT-0.8a2\fit\FitServer.py}

!path E:\selfworkspaces\fitTest

!| ActionFixture |

| start | ActionFixtureTest.AddTest| |

|enter|firstPart|1|

|enter|secondPart|2|

|press|join|

|check|together|3|

python:

from fit.Fixture import Fixture

class AddTest(Fixture):

_typeDict = {}

def __init__(self):

Fixture.__init__(self)

self.__first  = ""    #< Private attributes (Python convention).

self.__second = ""

self.__both   = ""

_typeDict["firstPart"] = "Int"

def firstPart(self, s):

self.__first = s

_typeDict["secondPart"] = "Int"

def secondPart(self, s):

self.__second = s

_typeDict["join"] = "Default"      #< AUTO-DETECT: None = void

def join(self):

self.__both = self.__first+self.__second

_typeDict["together"] = "Int"

def together(self):

return self.__both

3 Row Fixture

非常适用于测试一个列表,和column想比,row验证属性多了missing和surplus,missing表示期望有值,但是实际结果没有返回, surplus表示期望无值,但实际结果返回有值

使用时,如果被测对象有主键(比如id之类的),将主键放在最左侧,便于查看错误报告

Row Fixture 用于检查没有序列的列表,检查有序列的对象,要用ArrayFixture

Wiki:
!|RowFixtureTest.PayRecordTest|

|id|name|

|1|nxy|

|2|nxy|

|3|nxy|

Python:
from fit.RowFixture import RowFixture

class
PayRecordTest(RowFixture):

def getTargetClass(self):

return PayRecord

def query(self):

PayRecord.addId(1,"nxy")

PayRecord.addId(3,"nnn")

PayRecord.addId(4,"nxy")

return PayRecord.idList #< Return copy of
players

class
PayRecord(object):

_typeDict = {

"id":"Int"
,

"name":"String"

}

def __init__(self,id=0,name=""):

self.id=id

self.name=name

idList=[]

@classmethod

def addId(cls,id,name):

cls.idList.append(PayRecord(id,name))

4 Table Fixture

可用于需要自定义表格的解析方式,*设计表格样式

Wiki:
验证总分数和平均分

!| TableFixtureTest |

| name| score|

|math|90|

|chinese|95|

|english|91|

|total|276|

|avg|92|

Python:
from
fitnesse.fixtures.TableFixture import TableFixture

import
types

class
TableFixtureTest(TableFixture):

def doStaticTable(self,
rows):

total=0

for row in range(1, rows-2):

total+=int(self.getText(row,
2))

totalTable = int(self.getText(rows-2, 1))

avg=total/(rows-3)

avgTable=int(self.getText(rows-1, 1))

if total==totalTable:

self.right(self.getCell(rows-2, 1))

else:

self.wrong(self.getCell(rows-2, 1), str(totalTable))

if avg == avgTable:

self.right(self.getCell(rows-1, 1))

else:

self.wrong(self.getCell(rows-1, 1), str(avg))

Comment

protected abstract void doStaticTable(int rows)

Table Fixture is an abstract class that you must derive from. You must override doStaticTable to perform the functions of the fixture. The number of rows in the table is passed in rows.

protected Parse getCell(int row, int column)

Returns the addressed table cell as a Parse.

protected String getText(int row, int column)

Returns the text within the addressed table cell.

protected boolean blank(int row, int column)

Returns true if the addressed table cell is blank.

protected void wrong(int row, int column)

Turns the addressed table cell red.

protected void right(int row, int column)

Turns the addressed table cell green.

protected void wrong(int row, int column, String actual)

Turns the addressed table cell red, and annotates it with the actuall value.

protected void ignore(int row, int column)

Turns the addressed cell gray.

protected int getInt(int row, int column)

Converts the addressed cell to an int, and returns it.

5 SetUpFixture

可以用来做测试数据的准备

Wiki:

!| SetUpFixtureTest |

| player | post code | balance |

| John Smith | SW4 66Z | 10.00 |

| Michael Jordan | NE1 8AT | 12.00 |

Python:

from fitLib.SetUpFixture import SetUpFixture

from domain.Player import Player

class SetUpFixtureTest(SetUpFixture):

_typeDict = {}

def __init__(self):

Player.players = []

_typeDict["playerPostCodeBalance.types"] = [ None, "String", "String", "Float" ]

def playerPostCodeBalance(self, name, postCode, balance):

Player.addPlayer(name, postCode, balance)

Player 类见RowFixture示例

6 CalculateFixture

用来验证一个或多个列的组合值,可以用columnFixture来完成,但代码要比column少

Wiki:

!|CalculateFixtureTest|

|firstPart|secondPart||together|

|Hello|World||Hello, World|

|Houston|We Have a Problem||Houston, We Have a Problem|

Python:

from fitLib.CalculateFixture import CalculateFixture

class CalculateFixtureTest(CalculateFixture):

_typeDict = {}

# JAVA: String togetherFirstPartSecondPart(String firstPart,String secondPart){

_typeDict["togetherFirstPartSecondPart.types"] = ["String", "String", "String"]

def togetherFirstPartSecondPart(self, firstPart, secondPart):

return "%s, %s" % (firstPart, secondPart)

7 DoFixture

用来描述故事型的测试,比ActionFixture更高效

如果DoFixture是测试页的第一个表,则他管理着整个测试页面,允许你将测试表分成多个fixture类型的表,使测试具有可读性

Wiki:

!|DoFixtureTest|

|fill|10|times with|x|

|check|char at|4|x|

|set list|A,B,C,D|

|show|char at|2|

Python:

from fitLib.DoFixture import DoFixture

from typeadapter import buildListTypeAdapterFor

class DoFixtureTest(DoFixture):

_typeDict = {

"letters": "String"

}

def __init__(self):

DoFixture.__init__(self)

self.letters = ""

_typeDict["fillTimesWith.types"] = [None, "Integer", "Char" ]

def fillTimesWith(self, count, c):

self.letters = c * count    #< FILL: Repeat char ``count`` times.

_typeDict["charAtIs.types"] = ["Boolean", "Integer", "Char" ]

def charAtIs(self, position, c):

return self.letters[position] == c

ARRAY_OF_CHAR_TYPE_ADAPTER = buildListTypeAdapterFor("Char")

_typeDict["setList.types"] = [ None, ARRAY_OF_CHAR_TYPE_ADAPTER ]

def setList(self, array):

self.letters = "".join(array)

_typeDict["charAt.types"] = [ "Char", "Integer" ]

def charAt(self, position):

return self.letters[position]

from fit.taBase  import StringAdapter, ListAdapter

from fit.taTable import typeAdapterTable

from fit.FitException import FitException

import types

def buildListTypeAdapterFor(scalarType, typeName=""):

"""Utility method to build a TypeAdapter for a list of scalarType."""

if not typeName:

typeName = "ListOf%s" % scalarType

return ListAdapter(None, "", typeName, { ".scalarType": scalarType })

class CharTypeAdapter(StringAdapter):

"""

    TypeAdapter for Character.

    Characters are normally not of interest in Python (you normally use strings).

    Therefore, Python has no seperate datatype for a character.

    In addition, a TypeAdapter for Character is also not provided by PyFIT.

    """

def parse(self, text):

text = text.strip()

if len(text) != 1:

raise FitException, ("ParseError CharValue", text)

return text[0]

def toString(self, c):

if isinstance(c, (types.IntType, types.LongType)):

return chr(c)

elif isinstance(c, types.StringTypes):

assert len(c) >= 1

return c[0]

else:

raise FitException, ("toString.ValueError CharValue", c)

typeAdapterTable["Char"] = CharTypeAdapter

8 ArrayFixture

用于严格有序的序列测试

Wiki

!include -seamless SetUpFixture

!|ArrayFixtureTest|

|name|post code|credit limit|

|John Smith|SW4 66Z|10|

|Michael Jordan|NE1 8AT|12|

其中setupFixture见1.5,player类见之前例子

Python:

from fitLib.ArrayFixture import ArrayFixture

from domain.Player import Player

class ArrayFixtureTest(ArrayFixture):

def __init__(self):

ArrayFixture.__init__(self)

self.paramCollection = Player.players

self.setActualCollection()

def getTargetClass(self):

return Player   #< TYPE-HINT: For ValueObject class.

9 Combination Fixture

CombinationFixture用于描述有两个参数的计算规则,如下面两个例子,除法和乘法,其中theFirst指第一列各值,theSecond指第一行各值

Wiki:

!|CombinationFixtureTest.DivisionTest|

|  |1 |2|3|

|6 |6 |3|2|

|12|12|6|4|

!|CombinationFixtureTest.MultiplicationTest|

|  |1 |2|3|

|6 |6 |12|18|

|12|12|24|3|

Python:

from fitLib.CombinationFixture import CombinationFixture

class DivisionTest(CombinationFixture):

_typeDict = {}

# PY3K: combine(theFirst : int, theSecond : int) : int

_typeDict["combine.types"] = [ "Int", "Int", "Int" ]

def combine(self, theFirst, theSecond):

return theFirst / theSecond

class MultiplicationTest(CombinationFixture):

_typeDict = {}

# PY3K: combine(theFirst : int, theSecond : int) : int

_typeDict["combine.types"] = [ "Int", "Int", "Int" ]

def combine(self, theFirst, theSecond):

return theFirst * theSecond

10 ConstraintFixture

与calculatefixture不同 的是,constraintfixture每一次计算期望值都为true

Wiki中都为输入值,代码需有一个返回布尔类型的方法

Wiki
!|ConstraintFixtureTest|

|firstPart|secondPart|

|1|2|

|2|3|

Python:

from fitLib.ConstraintFixture import ConstraintFixture

class ConstraintFixtureTest(ConstraintFixture):

_typeDict = {}

# PY3K: firstPartSecondPart(firstPart : int, secondPart : int) : bool

_typeDict["firstPartSecondPart.types"] = [ "Boolean", "Int", "Int" ]

def firstPartSecondPart(self, firstPart, secondPart):

return firstPart < secondPart

11 Fixture Arguments

表的第一行表示类名,在类名后面添加单元格传递参数

Arguments 可以参数化字符串,并且重复使用,比如可以向RowFixture传递参数,他的参数不是在构造函数里初始化,而是通过加载类初始化

Wiki:
!|ArgumentsTest|Hello You Houston We Have a Problem|

|word|

|Hello|

|World|

|Houston|

|We|

|Have|

|a|

|Problem|

|Problem|

Python:

from
fitLib.SetFixture import
SetFixture

import
types

class
Word(object):

"""Simple ValueObject class
to store a word as string."""

_typeDict = { "word": "String" }

def __init__(self,
word):

assert isinstance(word, types.StringTypes)

self.word = word

class
ArgumentsTest(SetFixture):

def getTargetClass(self):

return Word #< CLASS-HINT: For _typeDict lookup.

def doTable(self,
table):

wordSet = set()

for s in self.args:#*argslist

for word in s.split(" "):

wordSet.add( Word(word) )

# -- WEIRD: setActualCollection() takes no arg -> Preassign
first.

self.paramCollection = wordSet

self.setActualCollection()

SetFixture.doTable(self, table)

上一篇:HTML5之选择上传图片文件


下一篇:【Scala-ML】使用Scala构建机器学习工作流