floatingghost
floatingghost pushed to develop at AkkomaGang/akkoma 2024-06-12 17:08:08 +00:00
5b75fb2a2f Merge pull request 'pool timeouts/rich media cherry-picks' (#796) from pool-timeouts into develop
4d6fb43cbd No need to spawn() any more
ad52135bf5 Convert rich media backfill to oban task
28d357f52c add diagnostic script
9c5feb81aa fix tests
Compare 25 commits »
floatingghost deleted branch pool-timeouts from AkkomaGang/akkoma 2024-06-12 17:08:08 +00:00
floatingghost merged pull request AkkomaGang/akkoma#796 2024-06-12 17:08:07 +00:00
pool timeouts/rich media cherry-picks
floatingghost commented on pull request AkkomaGang/akkoma#796 2024-06-12 17:07:55 +00:00
pool timeouts/rich media cherry-picks

ship it (tm)

floatingghost commented on issue AkkomaGang/akkoma#797 2024-06-12 15:20:05 +00:00
[question] Does Akkoma handle (created) pseudo-header in HTTP signatures

the created-pseudoheader branch was actually my experiment to do the same thing that doesn't quite work (hence the wip tag)

i'll look at this PR, maybe they've beaten me to the punch and…

floatingghost commented on issue AkkomaGang/akkoma#790 2024-06-12 02:37:54 +00:00
[bug] Rich media parser crashes, resulting in 504 and no posts being shown

point being, if your task stalls for any reason, you have an instant leak on your hands with the current implementation

but eh i've done more than my part in relaying the testing info back so 🍤

floatingghost created pull request AkkomaGang/http_signatures#1 2024-06-12 02:33:03 +00:00
WIP: Allowed supplying "(created)" pseudoheader
floatingghost created branch created-pseudoheader in AkkomaGang/http_signatures 2024-06-12 02:32:31 +00:00
floatingghost pushed to created-pseudoheader at AkkomaGang/http_signatures 2024-06-12 02:32:31 +00:00
a0239bbcce Allowed supplying "(created)" pseudoheader
floatingghost pushed to pool-timeouts at AkkomaGang/akkoma 2024-06-12 01:10:21 +00:00
4d6fb43cbd No need to spawn() any more
floatingghost commented on issue AkkomaGang/akkoma#790 2024-06-12 01:08:35 +00:00
[bug] Rich media parser crashes, resulting in 504 and no posts being shown

I would recommend if you haven't been following my Pleroma refactor of the Rich Media stuff and do the same -- currently the rich media stuff is done *while the posts are being fetched to render…

floatingghost commented on pull request AkkomaGang/akkoma#796 2024-06-12 01:05:21 +00:00
pool timeouts/rich media cherry-picks

nope it does not

especially now that it's just an oban

floatingghost commented on pull request AkkomaGang/akkoma#796 2024-06-12 01:05:06 +00:00
pool timeouts/rich media cherry-picks

yeah seems nowhere

purged

floatingghost commented on pull request AkkomaGang/akkoma#796 2024-06-12 01:03:48 +00:00
pool timeouts/rich media cherry-picks

i tried this, because :url_hash is a :bytea type, it cannot be used as a primary key it seems ecto (and postgres in general) only really wants you to use either fixed-length :binary_id or integer…

floatingghost commented on pull request AkkomaGang/akkoma#796 2024-06-11 17:10:04 +00:00
pool timeouts/rich media cherry-picks

Async tasks we love async tasks don't we folks we love memory management it's the best

floatingghost pushed to pool-timeouts at AkkomaGang/akkoma 2024-06-11 17:07:07 +00:00
ad52135bf5 Convert rich media backfill to oban task
floatingghost pushed to pool-timeouts at AkkomaGang/akkoma 2024-06-10 14:10:57 +00:00
28d357f52c add diagnostic script
floatingghost commented on issue AkkomaGang/akkoma#797 2024-06-09 20:42:09 +00:00
[question] Does Akkoma handle (created) pseudo-header in HTTP signatures

seems to be the case that the upstream lib doesn't support it, and that's been known about for a while (apparently)

see: https://git.pleroma.social/pleroma/elixir-libraries/http_signatures/-/iss

floatingghost pushed to pool-timeouts at AkkomaGang/akkoma 2024-06-09 20:26:41 +00:00
9c5feb81aa fix tests
a360836ce3 fix oembed test
Compare 2 commits »
floatingghost pushed to pool-timeouts at AkkomaGang/akkoma 2024-06-09 17:52:20 +00:00
840c70c4fa remove prints