Free Behavioral Targeting and Segmentation - BTBuckets

Hi guys!

We are from Russia, so we need to use cyrillic letters in 'came from' field.

We tried just to type 'тест' in that input and after we clicked on 'Update the bucket profile' they were transformed to the Unicode sequence (like that - '%u0442%u0435%u0441%u0442'). But it didn't work - the bucket was empty.

Then we tried to use percent-encoding (URL encoding). We wrote '%D1%82%D0%B5%D1%81%D1%82' and this was transformed to 'тест'. It didn't work either, though document.referrer contains '%D1%82%D0%B5%D1%81%D1%82'.

So, how can we use cyrillic letters or can't we?

Reply to This

Replies to This Discussion

Hello Max,

We've never had that issue before, so after further investigation we made some modifications on the configuration tool to accept cyrillic letters.

Just go back and edit the behavior and input 'тест' on the came from field. This time it will be correctly saved to the proper format and your bucket should work when the referrer contains the '%D1%82%D0%B5%D1%81%D1%82'.

Thanks for your feedback, it was helpful to us to improve the tool.

If you have more questions, let us know.
BTBuckets Team

Reply to This

Thank you for the quick response!

OK, now string 'тест' doesn't transform to Unicode and stay the same. But it's not working.

I made a little test case http://webprofiters.ru/0.html
We can just add any keywords after '?'.

There are two links, first page should alert document.referrer and 'test_2' if we came from http://webprofiters.ru/0.html?тест
Second one should alert document.referrer and 'test_3' if we came from http://webprofiters.ru/0.html?test

Second page's working fine, but first one's not.

I guess, may be we can try to not transform Unicode strings like '%D1%82%D0%B5%D1%81%D1%82' if it's possible?

Reply to This

Hello Max,

We're investigating this issue and will update this post when we have more information.

Thanks for your feedback.
BTBuckets Team

Reply to This

Hello Max,

This was a very specific case, as the browser automatically converted the cyrillic letters to UTF-8 according it's configurations of character encoding, even if you don't do it. So we updated the application to cover cyrillic letters both escaped or unmodified and the URL with your test case is now working just as expected.

Thanks for your feedback.

If you have more questions please let us know
BTBuckets Team

Reply to This

Thank you! Everything is working fine now! =)

Reply to This

RSS

Follow us on Twitter

© 2010   Created by BTBuckets.

Badges  |  Report an Issue  |  Terms of Service