| 5 Sep 2024 |
 K900 | Ugh hold on I need coffee | 07:41:14 |
| K900 | I can't fit this three level cache key structure into my head right now | 07:41:28 |
 outfoxxed | In reply to @k900:0upti.me
We could backport not necessarily
I've ended up looking through a few versions of this code for other reasons and it changes a lot | 07:41:57 |
| outfoxxed | plus then we'd have a different qmlc format again | 07:42:18 |
| outfoxxed | and that would break anything not using it | 07:42:27 |
| outfoxxed | im not sure how much of a concern that is | 07:42:39 |
| K900 | Well we'd have cache disabled for all the older versions | 07:42:49 |
| K900 | So that's fine | 07:42:51 |
| K900 | We can just start at version whatever-it-currently-is+1 | 07:43:01 |
| outfoxxed | In reply to @k900:0upti.me
I can't fit this three level cache key structure into my head right now The actual key is just the url of the qml file | 07:43:11 |
| outfoxxed | the rest is just validation | 07:43:16 |
| K900 | Not quite | 07:43:55 |
| K900 | It's also the QApplication name | 07:43:59 |
| outfoxxed | thats just to pick the cache path iirc | 07:44:16 |
| K900 | Well yes | 07:44:21 |
| K900 | But effectively that makes it part of the key | 07:44:26 |
| K900 | In the global space | 07:44:28 |
| outfoxxed | the application can also change that | 07:44:32 |
| K900 | Yes | 07:44:35 |
| outfoxxed | we shouldnt touch it | 07:44:35 |
| K900 | I'm not saying we should | 07:44:39 |
| K900 | But it matters for cleanup purposes | 07:45:37 |
| outfoxxed | the only relevant part is the qmlc url | 07:45:36 |
| K900 | Because e.g. we will never clean up other applications' caches | 07:45:45 |
| outfoxxed | i don't think we should be patching in extra cleanup | 07:45:50 |
| outfoxxed | * i don't think we should be patching in extra cleanup that isn't upstream | 07:46:01 |
| outfoxxed | but we should maintain the same level | 07:46:06 |
| outfoxxed | trying to do extra cleanup will put us into some position where we're deleting caches of something that expected the normal behavior | 07:47:10 |
| outfoxxed | and I don't see a reasonable way to do it, because theres no definite point where all qml resources will be loaded in the application's lifetime | 07:47:44 |
| K900 | There's a weird problem | 07:47:50 |
