| Commit message (Collapse) | Author | Age | Files | Lines |
... | |
| |
|
|
|
|
| |
could be a strong selling point.
|
|
|
|
| |
month which means we can find dead channels - because they won't be pinging the directory server once a month.
|
|
|
|
| |
uplink back to them without any ambiguity.
|
| |
|
| |
|
|
|
|
|
|
|
|
|
| |
possibly recurse and blow up the matrix. Hard to say. Do you feel lucky? Well do ya' ... punk? Rule #1 - don't mess with anything unless it's blowing up the matrix. If it doesn't blow up the matrix, but doesn't work, just let it go and let's figure out what it is doing and what it isn't doing.
The flow is as follows:
Once a day go out to all the directory servers besides yourself and grab a list of updates. This happens in the poller. If we've never seen them before add them to the updates table. The poller also looks to see if we're a directory server and have updates that haven't yet been processed. It calls onedirsync.php to process each one. If we contact the channel to update and don't find anything (we're just doing a basic zot_finger), set a ud_last timestamp. If this is set we will only try once a day for seven days. Then we stop trying to update.
This will probably cause a spike the first time through because you haven't seen any updates before, but we spread out the load over your delivery interval.
|
| |
|
|
|
|
| |
the parent xchan is safe or not. Otherwise we'll have tags that lead to nowhere because the directory entry is hidden but the tag isn't. A successful porn site in the matrix could also swamp the directory with x-rated tags, even if the site was playing nice and did everything right to self-censor. Accomplishing this with joins would be horrendously inefficient, though it will take a bit of code re-org to get this flag where it needs to be when it's time to set keywords.
|
|
|
|
| |
to update table to store the target address since it's possible the mirroring directory won't yet have an xchan or hubloc they can link the ud_hash to and therefore mayn't know how to contact them.
|
|
|
|
| |
(directory sync) updates have been processed in some way.
|
| |
|
| |
|
| |
|
|
|
|
| |
updated their site access policy which old sites do not have)
|
|
|
|
| |
as a block editor, and probably a layout editor as well. Eventually, these should all probably just be switches onto a single editor instance. Decided to put the layout_mid into the item table directory rather than re-use resource_id, so that we can still have pages attached to different resources like photos and events and stuff. The block editor is far from finished, at this point I've only cloned it and changed the name and type of item it looks for.
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
|
|
|
| |
(lots of) work. Oh and the intro table is no longer used and won't be - so it's gone.
|
|
|
|
| |
identity. This will be set to the url of the Red site that they are connected with. Only the nomadic identity clone at that site can communicate directly with them. Other instances will need to relay through that site - and if it goes down, that connection is stuffed.
|
| |
|
| |
|
|
|
|
| |
directory - such as to inform people what this channel is about before they connect. This will work best with a "channel detail" popup in the directory to provide more info than is available in the micro-profiles (when hovered, or clicked or something). And of course, the first thing is to enable this data to be stored in the directory.
|
| |
|
| |
|
| |
|
| |
|
|
|
|
| |
our nomadic channel clones don't.
|
| |
|
|
|
|
| |
part is still missing
|
| |
|
|
|
|
| |
but potentially can be exchanged through directory mirrors to automatically create "open site" lists.
|
| |
|
|
|
|
| |
we have permission to comment on something , and we'll need to send it out with all communications. The current check is not only flawed but also a huge performance hit. Also provide the ability for an item to disable commenting completely - such as for a webpage or wherever you want to prevent comments on one item, without requiring a change to your entire permission scheme. All of this is only partially implemented at the moment but we need the structures in place on several sites in order to finish it without breaking everything.
|
| |
|
|
|
|
| |
with multiple choice
|
| |
|
| |
|
| |
|
| |
|
|
|
|
| |
longer remotely resemble uri's and are actually message_id's. This change is potentially destabilising because it touches a lot of code and structure. But it has to get done and there's no better time than the present.
|
| |
|
| |
|
| |
|
|
|
|
| |
postings, and experiment with compressing pcss on the fly
|
|
|
|
| |
all" suggestions)
|