Improve Markdown Parsing to Reduce Unintended Block Quotes

[dirkrombauts]

The situation

It is possible in Gherkin to provide free-text descriptions for the Feature and Scenario parts in a .feature file. Pickles parses them as MarkDown and generates formatted HTML code.

In MarkDown, indented text is rendered as a code block. However, SpecFlow provides an indented description block ("As a math idiot ...") by default, and many people like to arrange all their description text indented.

That leads to a lot of code blocks in the output of Pickles.

So for example, the description in this feature

Scenario: Addition of Two Numbers
    As a math idiot
    In order to avoid mistakes
    I want to be told the sum of two numbers

will result in this code block in Pickles

As a math idiot
In order to avoid mistakes
I want to be told the sum of two numbers

A Possible Solution

Before the markdown processor transforms the description element, we remove the leading spaces from each line in the description element.

So

    As a math idiot
    In order to avoid mistakes
    I want to be told the sum of two numbers

Will be trimmed to

As a math idiot
In order to avoid mistakes
I want to be told the sum of two numbers

However, real code blocks should be preserved.
So

    As a math idiot
    In order to avoid mistakes
    I want to be told the sum of two numbers

        This is a real code block

Will be trimmed to

As a math idiot
In order to avoid mistakes
I want to be told the sum of two numbers

    This is a real code block

If there are blank lines in the description, that should not prevent us from trimming:

    As a math idiot
    In order to avoid mistakes
    I want to be told the sum of two numbers

    Another line after a blank line

Will be trimmed to

As a math idiot
In order to avoid mistakes
I want to be told the sum of two numbers

Another line after a blank line

Important

There should be unit tests to prove the functionality in detail.

[dirkrombauts]

Released in version 2.10.0.