Commit message (Collapse) | Author | Age | Files | Lines | |
---|---|---|---|---|---|
* | Remove unnecessary text string | Badlop | 2022-10-12 | 1 | -2/+1 |
| | |||||
* | Handle case that module exports mod_doc, but nothing in it | Badlop | 2022-10-12 | 1 | -1/+4 |
| | |||||
* | Annotate new options | Badlop | 2022-10-12 | 1 | -0/+2 |
| | |||||
* | Fix cache invalidation in shared roster | Paweł Chmielowski | 2022-09-28 | 1 | -0/+1 |
| | | | | | | Doing srg_get_info for not existing group, then srg_create, will make srg_add_user fail because it will get info that group doesn't exist from cache. | ||||
* | Export mod_muc_admin:get_room_pid/2 | Paweł Chmielowski | 2022-09-22 | 1 | -1/+2 |
| | |||||
* | The archive_msg export fun requires MUC Service for room archives | Badlop | 2022-09-16 | 2 | -3/+9 |
| | |||||
* | Allow MUC service admins to bypass max_user_conferences limitation | Badlop | 2022-09-16 | 1 | -3/+3 |
| | |||||
* | Remove unused macro definitions detected by rebar3_hank | Badlop | 2022-09-16 | 3 | -10/+0 |
| | | | | | | | | | | | | | | | | | | | | | | | ?OFFLINE_TABLE_LOCK_THRESHOLD is unused: definition and usage added in 4103f30812ed99c984a2fdb16aabca139c9255d7 definition copied, and usage moved to mod_offline_mnesia, in 901d2e0aed83d195a4d1cf2929114b07dcac0dd8 definition and usage removed in mod_offline_mnesia in d88e4d495ffc2ae950f77e440aa7c5d06c864309 this definition looks useless ?VERSION_MULTICAST is unused: added in 0cf8d1fa6e931367b63b133396478593146a8d07 it was never used ?DISCO_QUERY_TIMEOUT is unused: added in 0cf8d1fa6e931367b63b133396478593146a8d07 it was never used ?FEATURE/1 is unused: added in 0cf8d1fa6e931367b63b133396478593146a8d07 usage removed in bc33a3873dbb0828714eef69094213575fc979b6 ?DEFAULT_MAX_USERS_PRESENCE is unused: added in f817762cc48c13476a531bc40059cd17aea4cf0b usage removed in ba2b650464bd3aae2b6b0f3a3177476360cb6d08 ?CT_XML is unused: added in 46568fb9596d8203c635477b30c91d2928847dea removed in 3dc55c6d47e3093a6147ce275c7269a7d08ffc45 | ||||
* | Adapt to change of mix_annotate type to boolean in roster_query | Linus Jahn | 2022-09-16 | 1 | -2/+1 |
| | |||||
* | Adapt to renaming of 'participant-id' from mix_roster_channel record | Linus Jahn | 2022-09-16 | 1 | -2/+2 |
| | |||||
* | Don't persist 'none' role (thanks to Blake Miller)(#3330) | Badlop | 2022-09-13 | 1 | -1/+10 |
| | |||||
* | Store role, and use it when joining a moderated room (#3330) | Badlop | 2022-09-13 | 2 | -4/+22 |
| | |||||
* | mod_admin_extra: Handle empty roster group names | Holger Weiss | 2022-09-09 | 1 | -1/+1 |
| | | | | | | Don't let 'add_rosteritem' create a roster group with an empty name. Thanks to Licaon_Kter for reporting the issue. | ||||
* | Simplify recent commit (thanks to Simon Lipp)(#3894) | Badlop | 2022-09-07 | 1 | -4/+2 |
| | |||||
* | Fix unix socket support for ejabberd_http (thanks to Simon Lipp)(#3894) | Badlop | 2022-09-06 | 1 | -0/+2 |
| | |||||
* | Add function for getting room diagnostics | Paweł Chmielowski | 2022-09-01 | 1 | -5/+27 |
| | |||||
* | Catch all errors from jose_jwt:verify and log debugging details (#3890) | Badlop | 2022-09-01 | 1 | -1/+4 |
| | |||||
* | The #roster_item.ask allowed values are subscribe|undefined | Badlop | 2022-09-01 | 1 | -2/+1 |
| | |||||
* | mod_host_meta: Complain at start if listener is not encrypted | Badlop | 2022-09-01 | 1 | -1/+20 |
| | |||||
* | mod_pubsub: Don't crash on command failures | Holger Weiss | 2022-08-31 | 1 | -2/+2 |
| | | | | Handle 'delete_old_items' and 'delete_expired_items' errors gracefully. | ||||
* | Handle invalid values in X-Forwarded-For header more gracefuly | Paweł Chmielowski | 2022-08-29 | 1 | -6/+8 |
| | |||||
* | Update roster_get hook in mod_shared_roster_ldap to use #roster_item{} | Paweł Chmielowski | 2022-08-24 | 1 | -23/+14 |
| | |||||
* | Fix mucsub unsubscribe notification payload to have muc_unsubcribe in it | Paweł Chmielowski | 2022-08-24 | 1 | -2/+2 |
| | | | | ...instead of muc_subscribe | ||||
* | Fix filter_nodes | Jonah Brüchert | 2022-08-17 | 1 | -4/+2 |
| | | | | | | The previous implementation always returned an empty list while testing it. However I don't really understand why that happened. The list comprehension based one works, although it looks equivalent to me. | ||||
* | Parse sub elements of the mix join remote result | Jonah Brüchert | 2022-08-17 | 1 | -3/+15 |
| | |||||
* | mod_mix: Return user jid on join | Jonah Brüchert | 2022-08-17 | 1 | -0/+6 |
| | |||||
* | mod_shared_roster: Fix wrong hook type #roster{} (now #roster_item{}) | Linus Jahn | 2022-08-12 | 1 | -6/+5 |
| | | | | | | The hook type was changed from #roster{} to #roster_item{} by me in the recent MIX changes. Unfortunately I've overseen this one (and unit tests + dialyzer passed). | ||||
* | mod_mix, mod_mix_pam: Add new MIX namespaces to disco features | Linus Jahn | 2022-08-12 | 2 | -5/+6 |
| | |||||
* | mod_mix, mod_mix_pam: Add handling of IQs with newer MIX namespaces | Jonah Brüchert | 2022-08-12 | 2 | -19/+27 |
| | | | | Supported are mix:core:{0,1} and mix:pam:{0,2} | ||||
* | mod_roster: Change hook type from #roster{} to #roster_item{} | Linus Jahn | 2022-08-11 | 6 | -101/+78 |
| | | | | | | | | The problem with #roster{} is that every new record entry is also stored in the mnesia roster table. Adding the mix_participant_id there makes no sense because the normal roster items are no MIX channels. Using \#roster_item{} for the hook and #roster{} for storing the normal items seems to be a better idea. | ||||
* | mod_roster: Respect MIX <annotate/> setting | Linus Jahn | 2022-08-11 | 1 | -4/+65 |
| | | | | | MIX extensions are not sent to clients if they haven't been enabled by the client with <annotate/> in the roster get request. | ||||
* | mod_mix_pam: Do roster pushes on join/leave | Linus Jahn | 2022-08-11 | 1 | -1/+21 |
| | | | | | Before the channels were added/removed from the database but the clients were not notified of it. | ||||
* | mod_mix_pam: Provide MIX channels as roster entries via hook | Linus Jahn | 2022-08-11 | 2 | -1/+31 |
| | |||||
* | ejabberd_listener: Let module take over socket | Holger Weiss | 2022-08-10 | 1 | -0/+3 |
| | | | | | For some use cases, it may be necessary to allow a module to take over socket ownership during initialization. | ||||
* | ext_mod: Support managing remote nodes in the cluster | Badlop | 2022-08-05 | 1 | -5/+8 |
| | |||||
* | ext_mod: Handle correctly when COMMIT.json not found | Badlop | 2022-08-05 | 1 | -7/+15 |
| | |||||
* | Show allow_query_users room option in disco info (#3830) | Badlop | 2022-08-04 | 1 | -0/+1 |
| | |||||
* | Add log_burst_limit_* options (#3865) | Mark Zealey | 2022-08-01 | 4 | -1/+34 |
| | | | | | | | | | | * Add log_burst_limit_* options On our ejabberd deployment we were sometimes seeing more than 500 msgs/sec of legitimate traffic, however this was getting silently dropped. Provide config options to enable this limit to be configured from the config file. * Pass new logging vars in via ejabberdctl | ||||
* | Skip connection exit message when we triggered reconnection | Paweł Chmielowski | 2022-07-28 | 1 | -0/+2 |
| | |||||
* | Handle errors reported from begin/commit inside transaction | Paweł Chmielowski | 2022-07-28 | 1 | -35/+45 |
| | |||||
* | Make connection close errors bubble up from inside sql transaction | Paweł Chmielowski | 2022-07-28 | 1 | -22/+50 |
| | |||||
* | Add syntax_tools to applications, required when using ejabberd_sql_pt (#3869) | Badlop | 2022-07-27 | 1 | -1/+1 |
| | |||||
* | Don't bother with COMMIT.json user-friendly feature in automated user case | Badlop | 2022-07-27 | 1 | -1/+9 |
| | |||||
* | Handle not found COMMIT.json, for example in GH Actions | Badlop | 2022-07-27 | 1 | -3/+9 |
| | |||||
* | Add WebAdmin page for managing external modules | Badlop | 2022-07-26 | 1 | -2/+415 |
| | |||||
* | Catch mysql connection being close earlier | Paweł Chmielowski | 2022-07-19 | 1 | -1/+4 |
| | |||||
* | Make first sql reconnect wait shorter time | Paweł Chmielowski | 2022-07-19 | 1 | -4/+9 |
| | |||||
* | React to sql driver process exit earlier | Paweł Chmielowski | 2022-07-15 | 1 | -8/+13 |
| | | | | | | | | | If there are queued request when connection closes we may try to process those requests (by trying to send them to already terminated db process, and waiting until we hit timeout) before we see that driver is not longer alive. This change adds check for driver exit before we process next queued sql request, and immediately switch to connection state if we have one. | ||||
* | Don't set affiliation to 'none' if it's already 'none' in ↵ | Alexey Shchepin | 2022-07-01 | 1 | -16/+20 |
| | | | | mod_muc_room:process_item_change/3 | ||||
* | mod_mix_pam: Display joined channels on webadmin page | Linus Jahn | 2022-06-28 | 1 | -1/+66 |
| |