🇩🇪 Kura / くら / クラ / 区裸 Woomy :disconnecting: on Nostr: Alright, HaiMmrf is technically and theoretically done. I will do some local testing ...
Alright, HaiMmrf is technically and theoretically done.
I will do some local testing tomorrow, and then ship it to 
haizs.
When everything runs, I will create a patch or something for people to apply.
Also I will look into:
- Converting Akkomas MRF and Pleromas MRF to HaiMmrf Check and Action system
- Sharkey Patch Request
- Scriptable HaiMmrf Checks and Actions[1]
- Excel-like formula System for Checks and Actions. Will likely need a slight internal rewrite to make it work like I imagine it to.
- Im- and Exportsystem, so that you can share your settings with others (or transfer them to a new instance). This will not import settings from Pleroma or Akkoma, they use a complete different MRF.
1: The scriptable rules must be configured to be available.
The system runs through nodejs's vm module, which allows me to completly seperate the environment for the filter part from HaiSchluessel. This allows the exact control over what gets injected to the script, and what is taken out. This should prevent rogue scripts to do malicious actions. To make things more "secure", you need to explicitly enable the (server side) scripting explicitly in your config.yml. I do not want to be responsible for broken instances. You must conciously activate them.
I will do some local testing tomorrow, and then ship it to
When everything runs, I will create a patch or something for people to apply.
Also I will look into:
- Converting Akkomas MRF and Pleromas MRF to HaiMmrf Check and Action system
- Sharkey Patch Request
- Scriptable HaiMmrf Checks and Actions[1]
- Excel-like formula System for Checks and Actions. Will likely need a slight internal rewrite to make it work like I imagine it to.
- Im- and Exportsystem, so that you can share your settings with others (or transfer them to a new instance). This will not import settings from Pleroma or Akkoma, they use a complete different MRF.
1: The scriptable rules must be configured to be available.
The system runs through nodejs's vm module, which allows me to completly seperate the environment for the filter part from HaiSchluessel. This allows the exact control over what gets injected to the script, and what is taken out. This should prevent rogue scripts to do malicious actions. To make things more "secure", you need to explicitly enable the (server side) scripting explicitly in your config.yml. I do not want to be responsible for broken instances. You must conciously activate them.