aboutsummaryrefslogtreecommitdiffstats
path: root/include/items.php
Commit message (Collapse)AuthorAgeFilesLines
* in hubzilla we're going to convert doco to items. define it here so that an ↵redmatrix2015-07-031-2/+5
| | | | accidental import/export doesn't chuck a wobbly.
* provide backward compatibility with hubzilla extended item exportredmatrix2015-07-011-0/+55
|
* turn the connection filter into a feature.redmatrix2015-06-291-0/+5
|
* no this isn't it. well ok, it's part of it, but not the important part. it's ↵redmatrix2015-06-281-3/+62
| | | | still pretty cool.
* ensure we always sign a trimmed item body.redmatrix2015-06-221-3/+3
|
* Make item_store_update only affect the data we requestedAlexandre Hannud Abdo2015-06-201-9/+10
|
* hubzilla compatibility functionredmatrix2015-06-111-0/+14
|
* asjld workredmatrix2015-06-021-3/+13
|
* bare bones ASLDredmatrix2015-05-311-4/+44
|
* some basic background work for activitystreams-json-ld-2.0 conversionsredmatrix2015-05-311-0/+37
|
* there was a reason we included that fileredmatrix2015-05-281-3/+3
|
* Merge pull request #332 from solstag/zotfeedcleanupKlaus2015-05-241-3/+1
|\ | | | | Import security only when necessary, use correct variable in log
| * Import security only when necessary, use correct variable in logAlexandre Hannud Abdo2015-05-221-3/+1
| | | | | | | | Just some cleanup, seemed like the intent of previous changes was so
* | federation problem from diaspora comments relayed through red channels due ↵redmatrix2015-05-231-1/+1
|/ | | | to typo
* cloning abook entries was broken in several ways if the target xchan wasn't ↵redmatrix2015-05-201-0/+8
| | | | available on the clone site. This should make it less broken.
* don't allow item_id to update without a valid post idredmatrix2015-05-201-0/+5
|
* PRIVACY: possible privacy leakage under a defined set of circumstancesredmatrix2015-05-201-5/+8
|
* product was a bad string to use because we may require it for e-commerce, ↵redmatrix2015-05-131-1/+1
| | | | and RED_PLATFORM is equally problematic for other reasons.
* zotfeed issuesfriendica2015-04-241-3/+3
|
* block unfederated posts from zot_feedfriendica2015-04-231-0/+6
|
* issue #176, sender copy of item not obscured if using private mentionfriendica2015-04-201-1/+3
|
* match parensfriendica2015-04-141-1/+1
|
* post_activity_item incorrectly setting the obj_type when a valid type was ↵friendica2015-04-141-1/+1
| | | | specified
* Some documentation and whitespaces at line endings.Klaus Weidenbach2015-04-101-159/+232
|
* update features, allow sys account page deletionfriendica2015-04-081-0/+5
|
* Correcting reported Doxygen syntax warnings.Klaus Weidenbach2015-03-291-457/+405
| | | | | | | Fixed wrong Doxygen syntax and add some of the available FIXME to Doxygen documentation. Updated Doxygen configuration to add also all capital letter tags. Adding some more Doxygen documentation.
* remove bogus grouping from fix for #938friendica2015-03-281-2/+2
|
* issue #938friendica2015-03-281-2/+8
|
* also allow service classes to set post expiration. These take priority over ↵friendica2015-03-241-3/+3
| | | | the site default.
* more work on expiration, make system limits useful - even in retrospectfriendica2015-03-241-1/+5
|
* provide a site-wide expiration policy without loopholes - still need work to ↵friendica2015-03-241-1/+1
| | | | reduce the impact of expiration queries.
* sql errors on postgres with yestderday's optimisationfriendica2015-03-241-2/+2
|
* improve performance of the zotfeed queriesfriendica2015-03-231-10/+9
|
* some tabs title statuseditor restructuringMario Vavti2015-03-211-1/+1
|
* The random bad signatures are because something somewhere is trimming the ↵friendica2015-03-171-2/+2
| | | | body text. It could be any one of hundreds of functions that touch the message body. We really want to trim the body text, so I'm putting back all the trim statements - in mod/item and item_store and item_store_update. The last fix for random bad sigs noted that one of the trims wasn't there, so the others were removed. The correct fix is for all the trims to be there. We will probably have a few (quite a few) bad sigs during the transition back to trimmed text but this should nail it for anybody on recent code and with new content.
* The never ending saga of parent = 0 bugs on Dreamhost.friendica2015-03-011-0/+10
|
* this may actually fix the deliver loop when deleting existing items. It's ↵friendica2015-02-251-0/+15
| | | | hackish but I don't see any other way out.
* Found the cause of the delete looping, and I can prevent it going forward. ↵friendica2015-02-251-1/+10
| | | | 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.
* call post_local hooks prior to storing a post that is going to be sourced. ↵friendica2015-02-251-0/+2
| | | | 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.
* Fixes tag delivery regex for when multiple tags are presentAlexandre Hannud Abdo2015-02-131-2/+2
| | | | | Previous regex would cause matches to span several tags when present, thus never recognizing the mention.
* provide relief to sites that are severely impacted by the slow ITEM_UNSEEN ↵friendica2015-02-121-5/+4
| | | | searches. This does not incorporate any other flag optimisations as that will require a major DB update and possibly involve significant downtime. This is just to bite off a little chunk now and provide some much needed relief.
* Dot not trim body in order to avoid issues with signature checksAlexandre Hannud Abdo2015-02-121-2/+2
|
* issue #843friendica2015-02-111-0/+3
|
* more efficient processing of response activities, plus a proof of concept ↵friendica2015-02-111-2/+9
| | | | fix for the problem of receiving a delete notification before we actually get the item which we're supposed to delete.
* item voting toolsfriendica2015-02-101-2/+2
|
* issue #889 private tags for forums no longer recognised after tag refactor a ↵friendica2015-02-021-0/+1
| | | | few weeks back. This solution is not optimal as it also leaves you in the privacy ACL in some cases and not in others but it restores the previous functionality
* remote_user => remote_channelfriendica2015-01-281-1/+1
|
* local_user => local_channelfriendica2015-01-281-5/+5
|
* limit the number of forums that can be tagged in a single post - default is ↵friendica2015-01-281-8/+46
| | | | 2. The reason is simple - count how many posts would be in transit simultaneously if this was unlimited and somebody tagged 40-50 forums. In practice when used legitimately - we've rarely seen more than two, in fact I don't recall seeing more than two ever. Typically it is one and occasionally two. Changing the default is tricky - a client system cannot do it, but the site hosting a forum can choose to. Since not all sites that host forums will choose to do so, the ordering of the mentions would then be important.
* improved route mismatch detection. We will be less strict about the absolute ↵friendica2015-01-271-33/+20
| | | | route matching and only look at the last hop before it got to us - which is ultimately all we should care about (since that sender controls the thread permissions). Route mismatches seem to occur somewhat frequently from yamkote (for unknown reasons), and the logging has been improved a bit so it should provide some slightly more useful debugging info in case it still happens going forward. Oh, also we'll set the parent on comments when we store the initial post (item_store()) and only go back and set the parent for top-level posts. This should reduce the number of comments with missing parents on shared hosts, but may increase the number of missing threads. Probably worthwhile to do a query occasionally for parent = 0 and see how we're doing and how many have shared host related delivery issues.