Status: ACTIVE

Under active development and maintenance.

Ideas behind this project

• Input is HTML code, output is the same HTML code with changes in typography (entities, spaces, quotes, etc.).

• You can’t parse HTML with regex.

• The best existing HTML parser and tokenizer for Python is lxml.

• There are more languages than English in the world. Each of them has different typographic rules.

Installation

Easy:

$ pip install tipi

Quickstart

Usage of tipi is very straightforward:

>>> from tipi import tipi
>>> html = '<p>"Zavolej mi na číslo <strong class="tel">765-876-888</strong>," řekla, a zmizela...</p>'
>>> html = tipi(html, lang='cs')
>>> html
'<p>\u201eZavolej mi na \u010d\xed\xadslo <strong class="tel">765\u2013876\u2013888</strong>,\u201c \u0159ekla, a\xa0zmizela\u2026</p>'
>>> print html
<p>„Zavolej mi na čí­slo <strong class="tel">765–876–888</strong>,“ řekla, a zmizela…</p>

Remember that tipi is designed to work with HTML. In case you need to perform replacements on plaintext, escape it first:

>>> fron tipi import tipi
>>> tipi('b -> c')  # this works only by coincidence!
u'b → c'
>>> tipi('a <- b -> c')
u'a  c'
>>> import cgi
>>> html = cgi.escape(u'a <- b -> c')
>>> html
u'a &lt;- b -&gt; c'
>>> tipi(html)
u'a ← b → c'

Features

• Support for multiple languages.

• Language-sensitive replacements for single quotes and double quotes.

• Ellipsis, dashes, nonbreakable spaces, …

• Arrows (–> turned into → ), dimensions (12 × 30).

• Symbols (trademark, registered, copyright, EUR, …)

Alternatives

• Typogrify - English only, adds markup for styling, on top of smartypants

• cstypo - Czech only, not working well with HTML

Plans

• Inspiration from Typogrify?

• Get some inspiration from Dero’s and Typomil’s typography guides.

• Get some inspiration from Liteera.cz) (source).

• Maybe also some inspiration from here.

License: MIT

© 2013-2014 Jan Javorek <mail@honzajavorek.cz>

This work is licensed under MIT license.