| 10 May 2023 |
 Sumner Evans | In reply to @zhaofeng:zhaofeng.li
I left the room as the second wave started, do we have a list of the redacted media IDs to quarantine (again)? Travis got mad at me for editing my database. You should probably use the xargs approach. | 20:30:10 |
 Zhaofeng Li | In reply to @sumner:nevarro.space
Travis got mad at me for editing my database. You should probably use the xargs approach. Yeah, I tried the xargs command after I did the database edit, seemed to be noop but at least it confirmed that the database edit worked | 20:31:26 |
 @grahamc:nixos.org | Zhaofeng Li: https://gist.github.com/grahamc/b1d72cc8b79cce1dc18ac71cd013b7c9 | 20:31:45 |
 K900 | Is this updated? | 20:31:57 |
| @grahamc:nixos.org | yea | 20:32:35 |
| Sumner Evans | In reply to @zhaofeng:zhaofeng.li
Yeah, I tried the xargs command after I did the database edit, seemed to be noop but at least it confirmed that the database edit worked That is what I figured. I assume that he's mainly concerned about people who don't know how to be this admins running things on their database | 20:32:36 |
| K900 | Cool | 20:32:44 |
| * K900 fires up the redact-a-thon | 20:32:47 |
| * K900 * fires up the redact-a-tron | 20:33:06 |
| Zhaofeng Li | Oh ok, it always returns an empty JSON object, thought it meant it did nothing:
Trying to fetch homeserver name via localhost...
{}
Trying to fetch homeserver name via localhost...
{}
Trying to fetch homeserver name via localhost...
{}
Trying to fetch homeserver name via localhost...
{}
Trying to fetch homeserver name via localhost...
{}
| 20:33:34 |
| K900 | I think it still does something | 20:33:51 |
| K900 | But I'm bad at this | 20:36:05 |
| K900 | So you shouldn't listen to me | 20:36:08 |
| Zhaofeng Li | So now how do I find the paths to delete? Last time I just nuked the remote_content for the two homeservers (quarantined everything from the homeservers) | 20:37:08 |
| Sumner Evans | In reply to @k900:0upti.me
I think it still does something Yes, I look to the documentation and it says that it just returns an empty chase on object if it is successful | 20:38:12 |
| Sumner Evans | In reply to @zhaofeng:zhaofeng.li
So now how do I find the paths to delete? Last time I just nuked the remote_content for the two homeservers (quarantined everything from the homeservers) You have to join over to event_json and extract from there. | 20:38:36 |
| @grahamc:nixos.org | gonna go pet my dog and be outside for a while, hopefully this no-image thing deals with any abuse in the meantime | 20:39:19 |
| K900 | And the third person got banned on matrix.org too | 20:39:32 |
| @grahamc:nixos.org | T&S jobs must be inhumane | 20:39:35 |
 Alyssa Ross | I'll be around for the next little while | 20:40:45 |
| Alyssa Ross | thanks for everything you've done Graham | 20:41:09 |
| Sumner Evans | * Yes, I look to the documentation and it says that it just returns an empty JSON object if it is successful | 20:42:33 |
| Sumner Evans | In reply to @qyliss:fairydust.space
I wonder if we can/should mass contact the homeserver operators We may want to contact the admins of the homeservers where the content came from. | 20:43:50 |
| Zhaofeng Li | Wait a sec, I'm dumb, those are event IDs not media IDs | 20:46:23 |
| Zhaofeng Li | So I think the xargs synadm media quarantine -i didn't actually do anything | 20:46:57 |
| Zhaofeng Li | So I do need to go from event_json, extract the media ID from the mxc://, quarantine them in remote_media_cache and get their filesystem_ids | 20:48:53 |
| Zhaofeng Li | ... and transform those filesystem_ids to paths that I can actually delete | 20:49:07 |
| Alyssa Ross | if we think this is a good idea, if anybody wants to take this on like doing this I would be much obliged, as I think I'm going to be close to my limit just forcing myself to stay online watching to see if further moderation action is needed | 20:49:09 |
| K900 | FWIW those all seem to be public servers | 20:49:49 |
| K900 | https://matrix.im/ is some German thing | 20:50:00 |
