| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
| |
get a status post - leave it manual.
|
|
|
|
| |
filtered away.
|
|\ |
|
| |
| |
| |
| |
| | |
initialize some variables before using them.
commented out some unused variables.
|
|/
|
|
| |
it into a bona-fide on/off toggle. This takes up one more vertical line, but we have it to spare, and the single button (while concise visually) provided no feedback of the current state, or indicate whether a second click toggled it. Some addons use this setting so addons will need to be pulled when this is pulled in order to remain in sync.
|
|
|
|
|
|
|
|
|
|
| |
Check if post_photos permission is available otherwise hide the photo
upload button.
Or should we show the button and just disable it with a title, about
missing permission to upload photos?
In comments we use the same icon, but just add [img] bb-code. Should we
provide this if no permission to upload photos. Do we need different
icons for [img] and photo upload buttons?
|
|
|
|
| |
firehose. Don't let it.
|
| |
|
| |
|
|
|
|
| |
just from the navbar, and fix some saved-search weirdness related to the delete-term icon
|
| |
|
| |
|
|
|
|
| |
going forward (was 99). This way it can be adjusted down later without requiring you to change all your existing connections upward (since your existing connections are likely all sitting at 99 at the moment). The default setting is also configurable with a pconfig system.new_abook_closeness
|
|
|
|
| |
blocked because it isn't really a reddress and won't work if you copy it and try and use it somewhere. We should really convert the symbol back to '@' on copy and allow it to be copied, but this isn't as easy as it sounds and is left as an exercise for the community. If we just allow it to be copied we'll get a lot of bugs that making friends doesn't work. It does, but that isn't a legitimate reddress and even if we made allowances for it, Diaspora and Friendica and other webfinger based services wouldn't and would just say it can't be found or it's an illegal address. So if we block copy we'll just get bugs that it can't be copied. Eventually somebody will see this checkin and take it on themselves to figure out how to fix the address when copied to clipboard and then allow it to be copied. And there will be joy.
|
|
|
|
| |
in archive.org - which has the complete unescaped target url, scheme and all, as part of its own.
|
| |
|
|
|
|
| |
changes
|
|\
| |
| | |
RedDAV: fix renaming files.
|
| |
| |
| |
| | |
How can no one have spot this bug yet?!? Is no one using WebDAV?
Took me 2 days to figure out it was not my RedDAV rewrite, but that this should have never worked. :-!
|
| |
| |
| |
| | |
main editor window). Eventually the link specific icons (image, link, audio, video) will all be handled by inserting a link using the link icon (here and in the main editor window) and the separate buttons will go away. We also need to tie this to oembed so that there's one place to insert a link - no matter what kind of link it is, and we'll try to find the best way to render that as content. Similarly upload will upload content, no matter what kind of content; and we'll likewise try and figure out the best way to present whatever content was uploaded (img, file, audio, video, whatever).
|
| |
| |
| |
| | |
with no album specified will go now into an album based on YYYY-MM.
|
| | |
|
| | |
|
| | |
|
|/
|
|
| |
return the bbcode for the appropriate content type. This results in two web fetches of the url in question, but the first one is just a HEAD. This way we won't try and download and parse an entire video. TODO: img's are checked to see if they should be zid-ified, but audio/video currently are not.
|
| |
|
| |
|
|\ |
|
| |
| |
| |
| |
| |
| |
| | |
Fixed some wrong variable names.
Initialized some variables before using them.
Removed some checks for STATUSNET_PRIVACY_COMPATIBILITY in
include/security.php as it does not seem to be defined anywhere.
|
| |
| |
| |
| | |
successfully connect with a site. This way we can mark hubs dead on non-directory servers; but we have to wait before implementing this until we have a history of site_update being updated.
|
|/
|
|
| |
we're currently talking to.
|
| |
|
|
|
|
| |
$hash wasn't defined in this case
|
| |
|
|
|
|
|
|
| |
to start
marking them as dead sites.
|
| |
|
|
|
|
| |
last successful connection or indication that the hub was marked offline, but these are potentially expensive queries.
|
|
|
|
| |
function to mark a hubloc dead.
|
|\ |
|
| | |
|
| |
| |
| |
| | |
This reverts commit b57010e3da6c0521121a622c27013c8af7678132.
|
| | |
|
|/ |
|
| |
|
|
|
|
| |
hackish but I don't see any other way out.
|
|
|
|
| |
The issue remains what to do about comments which are already in the DB and have ITEM_ORIGIN incorrectly set. We can't exactly reset them because the "original" context has been lost. (Sorry but couldn't resist an insider pun that none of you will be able to follow anyway). Read the comments.
|
|
|
|
| |
There are some fields in the actual stored item that may change when we start the second delivery chain so a few plugins may not deal with these items correctly. Since we only source public items these effects will be slight.
|
| |
|
| |
|
|
|
|
| |
if the realm is so configured.
|