Skip to content

Latest commit

 

History

History
74 lines (53 loc) · 1.7 KB

README.MD

File metadata and controls

74 lines (53 loc) · 1.7 KB
Raw

Gherkin Objects

The goal of Gherkin Objects is: To be able to programmatically read, create, and modify gherkin

Gherkin Objects

  • Feature
  • Scenario
  • Step
  • Example Table
  • Tag
  • Comment
  • (Potentially others)

There should be one class for each distinct piece of gherkin syntax.

These classes represent an abstract notion of the object, throwing away a lot of information about how it is presented. No line numbers, no whitespace, just the abstract tree.

Constructing Gherkin Objects

Each gherkin object should be able to be constructed in at least two ways:

  1. Using the class constructor
  2. Parsed from raw text

For example:

Scenario.from_text("""
    Scenario: scenario
        Given step
""")

should be exactly equivalent to:

Scenario(name=scenario, steps=[
    Step(keyword='Given', text='step')
])

The from_text variant should handle any valid, parsable gherkin. For example:

                   Scenario: scenario
Given foo

                # Some random comment

Then bar

should parse into exactly the same object as something more reasonably formatted, like

Scenario: scenario
    Given foo
    # Some random comment
    Then bar

Writing Gherkin Objects

Since gherkin objects do not contain formatting information, it is impossible to write the data without specifying how to format it. This is done with a formatter configuration. For example:

step = Step.from_text("Given foo")
print(step.to_text(FormatterConfig(indent_step=4)))

A reasonable default formatter config is provided, but a custom formatter can be specified. A config file can be specified using either a command line argument or with the environment variable GHERKIN_FORMATTER_CONFIG