State of the Mission, 2017-05-03: Lots of Coding

I know I've been very quiet lately, but there's a good reason for it.

Here's what I've been doing on Verbalyser:

And this is what I've been focused on for the last few days.  I'm not at liberty to share my code, but I think it's pretty clear how I've gone about it.

 

Refactoring and Unicode

Ruby's implementation of Unicode for Lithuanian, an admittedly complicated language, is not perfect but I have modified a useful script on Gist for identifying the unicode value of a given character.

Now that I can identify the unicode value of tricky diacritics and accents, I can refine my code.

Details of the script here.

Pirmas darbo pokalbis baigę

via Alexander Garber http://ift.tt/2qdd0DX

Lithuanian gives you good practice with regex

I'm parsing Lithuanian verbs, and in tidying up the raw HTML, I'm compelled to come up with little beauties like this:

<a href="\/[a-z]*\/[a-z]*\W*[a-z]*\W*[a-z]*">

...in order to to clear away the HTML formatting.

The problem lies in the fact that Lithuanian diacritics (č, ž, ė, ę and so on) are not recognised as letters [a-z] but rather as non-word characters \W.

How I Am Studying Lithuanian

via Alexander Garber http://ift.tt/2n0dqQQ

Geras Savaitgalis

via Alexander Garber http://ift.tt/2mWEBJm

Daug dirbiau šiandien

via Alexander Garber http://ift.tt/2ndCJe6

Labas!

via Alexander Garber http://ift.tt/2lm4xip