mazko / jstaggregator Goto Github PK

https://github.com/mazko/jstaggregator/blob/fb6f5dbaaf76cec9dab195bc450b272a7c3fe36a/crossbrowser/test_materials/fixmenewlines.html

Expected: Ювентус 3

Got: 2 + Н.СансонеЮвентус

Even http://stackoverflow.com/questions/28899298/extract-the-text-out-of-html-string-using-javascript

function extractContent(html) {
    return (new DOMParser).parseFromString(html, "text/html") . 
        documentElement . textContent;

}

Produce same issue.

Find better html->text impl

Relative path to the data directory (to make it re-usable for add-on HTML content) do not currently work. Instead you have to use following syntax:

resource://ID/data/icon-name.png

https://developer.mozilla.org/en-US/Add-ons/SDK/Tools/package_json#icon

Wrong Danish language identification:

./tika-app-1.2.jar -l http://da.wikipedia.org/wiki/Danmark
no

To fix this we need as more Norwegian source text as possible, but ~10 mb of plain Norwegian text would be enough i think. Ngrams creation process descried here http://nongreedy.ru/javascript/2013/01/18/universal-language-identifier-for-every-modern-browser.

Opera Violent monkey fails loading Taggregator.js

1. Make Taggregator.js to pass JSLint (currently fails)

2. Snowball.js >> Taggregator.js

3. Check Violent again