model: Fetch emoji using events #825
Conversation
There was a problem hiding this comment.
@mkp6781 Thanks for starting to put together the ideas from the issues - it's clearer to see what you have here and give feedback 👍
This shows you have the scaffold for this, but note that the realm_emoji data you're adding is in a different format to that previously - that's at least one reason why you have so many test failures.
Note that if you update the fetch_custom_emojis and the associated locations and tests, then this will likely make adapting the tests easier.
I'm not sure why the tests aren't running yet 😕
zulipterminal/model.py
Outdated
| Handle update of emoji | ||
| """ | ||
| realm_emojis = event['realm_emoji'] | ||
| self.active_emoji_data.update(realm_emojis) |
There was a problem hiding this comment.
As I mentioned previously, events are not in the same format as the original data, so a dict update is not feasible here.
zulipterminal/model.py
Outdated
| @@ -145,7 +148,7 @@ def __init__(self, controller: Any) -> None: | |||
| for name, data in unicode_emoji_data.items(): | |||
| data['type'] = 'unicode_emoji' | |||
| typed_unicode_emoji_data = cast(NamedEmojiData, unicode_emoji_data) | |||
| custom_emoji_data = self.fetch_custom_emojis() | |||
| custom_emoji_data = self.initial_data['realm_emoji'] | |||
There was a problem hiding this comment.
This sets custom_emoji_data to a different data format than before; note the processing done in fetch_custom_emojis - we could rename and adjust that to process the realm emoji in the initial data instead.
In principle we could adjust the code to handle original and emoji events to be in the same format, but we still need to integrate those with the format of the active_emoji_data upon update.
|
Opened a pull request from my branch |
Fetching emoji using events ensures that emoji updates are not missed.
Initial state of emoji is updated from initial_data.