Question

I have a python file filled with functions like h1 and img and strong for use styling text. Each of these functions are defined as follows:

def _wrapTag(tag, text, **attributes):
    out = _createTag(tag, **attributes)
    out += text
    out += "</" + tag + ">"
    return out

def _createTag(tag, **attributes):
    out = "<" + tag
    if attributes:
        for attr, value in attributes:
            out += " " + attr + "=\"" + value + "\""
    out += ">"
    return out

def h2(text, **attributes):
    return _wrapTag("h2", text, **attributes)

In an ideal world, to create a div with the class modal, I would call div(content, class="modal") however class is a restricted keyword. Is there any way to bypass this without adding a special case to _createTag?

Answer 1

The PEP 8 standard way of handling that is to add a trailing underscore:

• single_trailing_underscore_: used by convention to avoid conflicts with Python keyword, e.g.

Tkinter.Toplevel(master, class_='ClassName')

That’s a common workaround and won’t surprise anyone. You could implement that in your code like:

def _createTag(tag, **attributes):
    out = "<" + tag
    if attributes:
        for attr, value in attributes.items():
            out += " " + attr.rstrip('_') + "=\"" + value + "\""
    out += ">"
    return out

so that it automatically removes the extra underscore from any and all attributes. Then you could call:

>>> h2('contents', class_='myh2tag', id_='contenttag')
'<h2 class="myh2tag" id="contenttag">contents</h2>'

and work around two separate Python namespace conflicts without handling either of them as a special case.