Switch to download counts from Cloud Front

Previous numbers were from S3 direct logs. Cloud front requests from S3
on cache miss only. The user agent was also cloud front instead of the
actual requester. Cloud front logs have the actual user agent and we
use it to filter out bot requests.

So, the old numbers (relatively) under counted popular boards and over
counted boards that only bots cared about. These numbers should be truer
to reality.
This commit is contained in:
Scott Shawcroft 2025-05-12 15:33:13 -07:00
parent 380e08dba3
commit 17555d0082
No known key found for this signature in database

File diff suppressed because it is too large Load diff