You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Adding an emoji to the annotation text breaks the annotations view and does not show any annotations.
Reproduce: Create an annotation with an emoji. E.g. "This kills your annotations 馃挘馃挘馃挘"
What happens?
The annotations are serialized and inserted in to the matomo_option table. But the serialized string is truncated at the first emoji. This is because utf8_general_ci is limited in the character range. The values can't be retrieved from the database.
Manual fix for users:
Find the entry < site id>_annotations in the matomo_option and fix the serialized php data set. Replace < site id> with the id of your site.
Fix:
My preferred solution would be to use utf8mb4_general_ci instead - this should support all characters.
The text was updated successfully, but these errors were encountered:
A non breaking change would be to check the strings for invalid characters and report an error instead of corrupting the database with an invalid entry and trashing all existing annotations.
Not everybody is able to change the database content and fix the broken php data.
This is not just a small inconvenience that the entries are not displayed - they are simply lost.
Adding an emoji to the annotation text breaks the annotations view and does not show any annotations.
Reproduce: Create an annotation with an emoji. E.g. "This kills your annotations 馃挘馃挘馃挘"
What happens?
The annotations are serialized and inserted in to the
matomo_option
table. But the serialized string is truncated at the first emoji. This is because utf8_general_ci is limited in the character range. The values can't be retrieved from the database.Manual fix for users:
Find the entry
< site id>_annotations
in thematomo_option
and fix the serialized php data set. Replace < site id> with the id of your site.Fix:
My preferred solution would be to use utf8mb4_general_ci instead - this should support all characters.
The text was updated successfully, but these errors were encountered: