lektor-fixedlang 0.1

Why?

Say you have a multilingual site where one of the languages you support is Turkish. In Turkish, the uppercase form for i is İ, and the lowercase form for I is ı. If you use an English word containing these letters in Turkish text, a text-transform CSS rule will cause the word to be shown incorrectly. For example, the word “Wikipedia” transformed to uppercase will become “WİKİPEDİA”.

To prevent this, you have to mark such pieces of text using a lang attribute, as in <span lang="en">Wikipedia</span>. This can of course be done manually when writing the content, but that’s going to be inconvenient for content writers and it can also be easily overlooked.

Although this “dotless-i / dotted-I” problem was the initial motivation behind this plugin, it can be useful in other cases as well.

Installation

To use the plugin, add it to your project:

lektor plugin add lektor-fixedlang

The plugin can be configured using the configs/fixedlang.ini file. Every section is the name of the tag that will wrap the pattern. Every key in the section is a regular expression pattern and the value is the language for the pattern.

Example:

[span]
\bWikipedia\b = en

Note: At the moment, regular expression matches are case insensitive.