@sequential
Feature: Parse data types in step parameters (type transformation)
As a test writer
I want write steps with parameter that should have floating-point data types
So that I can just use the step parameter with the correct type
behave uses the parse module (inverse of Python string.format).
Therefore, the following ``parse types`` for floats are already supported:
===== =========================================== ============= ================================
Type Characters Matched (regex class) Output Type Example(s)
===== =========================================== ============= ================================
% Percentage (converted to value/100.0) float 51.2%
f Fixed-point numbers float 1.23 -1.45
e Floating-point numbers with exponent float 1.1e-10 -12.3E+5
g General number format (either d, f or e) float 123 1.23 -1.45E+12
===== =========================================== ============= ================================
SEE ALSO:
* http://pypi.python.org/pypi/parse
* string.format: http://docs.python.org/library/string.html#format-string-syntax
@setup
Scenario: Feature Setup
Given a new working directory
And a file named "features/steps/float_param_steps.py" with:
"""
from behave import then, step
class NotMatched(object): pass
@step('a float param with "{value:g}"')
def step_float_param_with(context, value):
assert type(value) is float
context.value = value
@step('a float param with "{value}"')
def step_float_param_otherwise(context, value):
context.value = NotMatched
@step('a generic float param with "{value:g}"')
def step_generic_float_param_with(context, value):
step_float_param_with(context, value)
@step('a generic float param with "{value}"')
def step_generic_float_param_otherwise(context, value):
step_float_param_otherwise(context, value)
@step('a float with exponent param with "{value:e}"')
def step_float_with_exponent_param_with(context, value):
step_float_param_with(context, value)
@step('a float with exponent param with "{value}"')
def step_float_with_exponent_param_otherwise(context, value):
step_float_param_otherwise(context, value)
@step('a percentage param with "{value:%}"')
def step_percentage_param_with(context, value):
step_float_param_with(context, value)
@step('a percentage param with "{value}"')
def step_percentage_param_otherwise(context, value):
step_float_param_otherwise(context, value)
@step('a fixed-point number param with "{value:f}"')
def step_fixed_point_number_param_with(context, value):
step_float_param_with(context, value)
@step('a fixed-point number param with "{value}"')
def step_fixed_point_number_param_otherwise(context, value):
step_float_param_otherwise(context, value)
@then('the value should be {outcome} as float number')
def step_value_should_be_matched_as_float_number(context, outcome):
expected_type = float
if outcome == "matched":
assert type(context.value) is expected_type, \
"Unexpected type: %s" % type(context.value)
else:
assert context.value is NotMatched
@then('the value should be {outcome} as float number with "{expected:g}"')
def step_value_should_be_matched_as_float_number_with_expected(context, outcome, expected):
step_value_should_be_matched_as_float_number(context, outcome)
assert context.value == expected, \
"FAILED: value(%s) == expected(%s)" % (context.value, expected)
"""
Scenario: Float parameter values with type "%" (percentage)
Given a file named "features/example.float_param.with_percent.feature" with:
"""
Feature: Float parameter values with type "%" (percentage)
Scenario Outline: Good cases
Given a percentage param with "<Value>"
Then the value should be matched as float number with "<Expected Value>"
Examples:
| Value | Expected Value | Case |
| 0% | 0 | Zero |
| 20% | 0.2 | Positive number |
| 120% | 1.2 | Larger than 100% |
| 10.5% | 0.105 | Float number |
| -10% | -0.1 | Negative number |
| +10% | 0.1 | With plus sign |
Scenario Outline: Bad cases (not matched)
Given a percentage param with "<Value>"
Then the value should be <Outcome> as float number
Examples:
| Value | Outcome | Reason |
| 123 | not-matched | Percentage sign is missing |
| 1.23 | not-matched | Float number without percentage sign |
| Alice | not-matched | Not a number |
"""
When I run "behave -f plain features/example.float_param.with_percent.feature"
Then it should pass with:
"""
9 scenarios passed, 0 failed, 0 skipped
"""
Scenario: Fixed-point parameter values with type "f"
Given a file named "features/example.float_param.with_type_f.feature" with:
"""
Feature: Float parameter values with type "f" (fixed-point number)
Scenario Outline: Good cases
Given a fixed-point number param with "<Value>"
Then the value should be matched as float number with "<Expected Value>"
Examples:
| Value | Expected Value | Case |
| 0.23 | 0.23 | |
| 1.23 | 1.23 | |
| 123.45 | 123.45 | |
| +1.23 | 1.23 | With plus sign |
| -1.23 | -1.23 | Negative float |
Scenario Outline: Bad cases (not matched)
Given a fixed-point number param with "<Value>"
Then the value should be <Outcome> as float number
Examples:
| Value | Outcome | Reason |
| 123 | not-matched | Integer number |
| 1.23E-7 | not-matched | Float number with exponent |
| Alice | not-matched | Not a number |
"""
When I run "behave -f plain features/example.float_param.with_type_f.feature"
Then it should pass with:
"""
8 scenarios passed, 0 failed, 0 skipped
"""
Scenario: Float with exponent parameter values with type "e"
Given a file named "features/example.float_param.with_type_e.feature" with:
"""
Feature: Float parameter values with type "e" (float with exponents)
Scenario Outline: Good cases
Given a float with exponent param with "<Value>"
Then the value should be matched as float number with "<Expected Value>"
Examples:
| Value | Expected Value | Case |
| 1.0E-10 | 1E-10 | With mantisse, negative exponent |
| 1.23E+5 | 123E3 | Exponent with plus sign |
| 123.0E+3 | 1.23E5 | Exponent with plus sign |
| -1.23E5 | -123E3 | Negative number with mantisse and exponent |
| INF | +INF | Infinity (INF) |
| +INF | INF | |
| -INF | -INF | |
| +inf | INF | Lower-case special names |
Scenario Outline: Bad cases (not matched)
Given a float with exponent param with "<Value>"
Then the value should be <Outcome> as float number
Examples:
| Value | Outcome | Reason |
| 1E10 | not-matched | Without mantissa |
| 1.E10 | not-matched | Short mantissa |
| 123 | not-matched | Integer number |
| Alice | not-matched | Not a number |
"""
When I run "behave -f plain features/example.float_param.with_type_e.feature"
Then it should pass with:
"""
12 scenarios passed, 0 failed, 0 skipped
"""
Scenario: Generic float parameter values with type "g"
Given a file named "features/example.float_param.with_type_g.feature" with:
"""
Feature: Float parameter values with type "g" (generic float)
Scenario Outline: Good cases
Given a generic float param with "<Value>"
Then the value should be matched as float number with "<Expected Value>"
Examples:
| Value | Expected Value | Case |
| 1 | 1.0 | Integer number format |
| 1E10 | 1.0E10 | Float with exponent and shortened mantissa |
| 1.23E5 | 1.23E5 | Float with exponent and mantissa |
| 1.23e5 | 1.23E5 | Float with exponent and mantissa (lower-case) |
| 1.0E-10 | 1E-10 | With mantisse, negative exponent |
| 1.23E+5 | 123E3 | Exponent with plus sign |
| 123.0E+3 | 1.23E5 | Exponent with plus sign |
| -1.23E5 | -123E3 | Negative number with mantisse and exponent |
Scenario Outline: Bad cases (not matched)
Given a generic float param with "<Value>"
Then the value should be <Outcome> as float number
Examples:
| Value | Outcome | Reason |
| 0b101 | not-matched | Binary number |
| 0o17 | not-matched | Octal number |
| 0x1A | not-matched | Hexadecimal number |
| 1.E10 | not-matched | Short mantissa |
| Alice | not-matched | Not a number |
"""
When I run "behave -f plain features/example.float_param.with_type_g.feature"
Then it should pass with:
"""
13 scenarios passed, 0 failed, 0 skipped
"""