DSL Kata

Python solution uses a different approach because the language lacks a block syntax and multi-line lambdas. Each of these objects is built inline using readable method names.

This implementation overloads the operators __str__ for operations on the text and range fields.

print(
    Form(
        CharField(name="user", size="140", label="User Name"),
        id="adidas-form",
    )
)

print("-"*5)

print(
    Form(
        CharField(name="user", size="140", label="User Name"),
        PasswordField(label="Password"),
        id="adidas-form",
    )
)

<form id='adidas-form'>
<label>User Name<label/>
<input name='user' size='140' label='User Name' type='text'/>
<form/>
-----
<form id='adidas-form'>
<label>User Name<label/>
<input name='user' size='140' label='User Name' type='text'/>

<label>Password<label/>
<input label='Password' type='password'/>
<form/>

Run

$ python main.py

Strengths

• Single step for creating the request
• Easy to dynamically create criteria
• Operator overloading improves readability

Weaknesses

• Repetition of the class name
• Literal list used as argument

TL;DR

In general, there are 3 types of DSL implementations: fluent interface, readable inline method arguments and blocks. Fluent interfaces is the most effective approach for the static languages. For Python, readable inline method arguments seemed to be the most idiomatic.