Generate HTML conveniently and efficiently with Python generators

1 month ago 9

fast_html is a fast, minimalist HTML generator.

It is an alternative to templating engines, like Jinja, for use with, e.g., htmx.

Pros:

use familiar python syntax
use efficient concatenation techniques
optional automatic indentation

Unlike other HTML generators (e.g. Dominate) that use python objects to represent HTML snippets, fast_html represents HTML snippets using string generators that can be rendered extremely fast using join. (see here)

Like other HTML generators, one needs to remember:

the name of some tags and attributes is changed (e.g., class_ instead of class, due to Python parser)
there may be conflicts of function names with your code base

pip install fast_html or copy the (single) source file in your project.

Don't forget to add a star on GitHub <https://github.com/pcarbonn/fast_html>_ ! Thanks.

>>> from fast_html import *

A tag is created by calling a function of the corresponding name, and rendered using render:

>>> print(render(p("text"))) <p>text</p>

Tag attributes are specified using named arguments:

>>> print(render(br(id=1))) <br id="1"> >>> print(render(br(id=None))) <br> >>> print(render(ul(li("text", selected=True)))) <ul><li selected>text</li></ul> >>> print(render(ul(li("text", selected=False)))) <ul><li>text</li></ul>

The python parser introduces some constraints:

The following tags require a trailing underscore: del_, input_, map_, object_.
The following tag attributes require a trailing underscore: class_, for_.

In fact, the trailing underscore in attribute names is always removed by fast_html, and other underscores are replaced by - (except for the _ attribute). For example, the htmx attribute hx-get is set using hx_get="url".

>>> print(render(object_("text", class_="s12", hx_get="url"))) <object class="s12" hx-get="url">text</object> >>> print(render(button("Click me", hx_post="/clicked", hx_swap="outerHTML"))) <button hx-post="/clicked" hx-swap="outerHTML">Click me</button>

The innerHTML can be a list:

>>> print(render(div(["text", ... span("item 1"), ... span("item 2") ... ]))) <div>text<span>item 1</span><span>item 2</span></div>

The innerHTML can also be a list of lists:

>>> print(render(div(["text", ... [span(f"item {i}") for i in [1,2]] ... ]))) <div>text<span>item 1</span><span>item 2</span></div> >>> print(render([br(), br()])) <br><br>

You can call generators too:

>>> def row(number, name): ... yield "<tr>" ... yield td(number) ... yield td(name) ... yield "</tr>" >>> def table(): ... yield "<table>" ... yield from row("1", "A") ... yield from row("2", "B") ... yield "</table>" >>> print(render(table())) <table><tr><td>1</td><td>A</td></tr><tr><td>2</td><td>B</td></tr></table>

The innerHTML can also be specified using the i parameter, after the other attributes, to match the order of rendering:

>>> print(render(ul(class_="s12", i=[ ... li("item 1"), ... li("item 2")] ... ))) <ul class="s12"><li>item 1</li><li>item 2</li></ul>

You can create your own tag using the tag function:

>>> def my_tag(inner=None, **kwargs): ... yield from tag("my_tag", inner, **kwargs) >>> print(render(my_tag("text"))) <my_tag>text</my_tag>

By default, the inner string of a tag is not escaped: characters &, < and > in it are not converted to HTML-safe sequences.

>>> print(render(p("<bold>text</bold>"))) <p><bold>text</bold></p>

Of course, you can escape strings before calling fast_html:

>>> from html import escape >>> print(render(p(escape("<bold>text</bold>")))) <p><bold>text</bold></p>

If your policy is to escape every inner string, you can activate escaping by setting the variable escape to True (or by calling escape_it(True)).

>>> escape_it(True) >>> print(render(p("<bold>text</bold>"))) <p><bold>text</bold></p>

When debugging your code, you can set global variable indent to True (or call indent_it(True)) to obtain HTML with tag indentation, e.g.,

>>> indent_it(True) >>> print(render(div(class_="s12", i=["text\n", span("item 1"), span("item 2")]))) <div class="s12"> text <span> item 1 </span> <span> item 2 </span> </div> <BLANKLINE>

You can also convert an HTML string to a function-based code representation:

>>> print(html_to_code('<div class="example"><p>Some text</p></div>')) [div([p(['Some text'], )], class_="example")]

Read Entire Article