Utf8 not displayed properly on the Index page

Hello There,

I have a simple model with a list of string tags. Some of those tags are emojis :cheese::chocolate_bar:
I use the algoliasearch python lib with a simple:


Where batch is a list of recipes dict.

{'id': 177, 'created': datetime.datetime(2017, 4, 12, 22, 17, 4, 47000, tzinfo=<UTC>), 'modified': datetime.datetime(2017, 4, 14, 10, 29, 8, 514000, tzinfo=<UTC>), 'title': 'Spaghetti balls aux Pringles', 'video_url': 'https://www.facebook.com/ChefClub.tv/videos/1186965131328098/', 'tags': ['🧀', 'apero', 'boulette de pâte', 'chips', 'edam', 'farine', 'fingerfood', 'fromage', 'huile', 'oeuf', 'pâtes', 'pringles', 'recette apéritive', 'recette revisitée', 'salé', 'sauce tomate', 'spaghetti']}

Algolia shows � on the browsing index page.
Yet the search does work and match the emoji :tada:

Maybe its simply a frontend issue on Algolia side?
I wanted to double check.

Keep up the good work :sunny:

1 Like

Yes, unfortunately we are using UCS-2 internally, which doesn’t work with characters that are outside of the “binary plane”. We plan to switch to UTF-8 or UTF-16, but that’s a lot of work and hasn’t happened yet.

1 Like

So you do confirm that the search is still working but that is only a display issue related to UCS-2?

Yes, search on emoji name is a feature, but the display is “broken”, with some (not all) emoji.

You might have luck when not using the highlighted value.

1 Like