Commit message (Collapse) | Author | Age | Files | Lines | ||
---|---|---|---|---|---|---|
... | ||||||
* | | core.modulemanager: Fix resource location compat with LuaRocks 2.x | Kim Alvefur | 2021-01-16 | 1 | -1/+1 | |
| | | | | | | | | | | The path doesn't include lua version, at least least on Debian, which still has luarocks 2.x | |||||
* | | MUC: Allow modules to add to the room registration form, and handle the result | Matthew Wild | 2021-01-15 | 1 | -0/+23 | |
| | | ||||||
* | | doc/doap.xml: Fix attribute name | Matthew Wild | 2021-01-15 | 1 | -1/+1 | |
| | | ||||||
* | | mod_mam: Workaround scansion test flakyness | Kim Alvefur | 2021-01-13 | 1 | -12/+2 | |
| | | | | | | | | The empty ID fields are sometimes accepted, sometimes not. | |||||
* | | mod_storage_sql: Fix compatibilty with Lua 5.1 | Kim Alvefur | 2021-01-13 | 1 | -1/+2 | |
| | | | | | | | | | | | | string.rep() doesn't have the 3rd argument before Lua 5.2 Revert once we drop support for Lua 5.1 | |||||
* | | mod_muc_mam: Copy extended MAM support from mod_mam | Kim Alvefur | 2021-01-12 | 1 | -4/+21 | |
| | | | | | | | | | | | | mod_mam and mod_muc_mam are just different enough that having them be the same module gets complicated, but close enough to have overlapping code. | |||||
* | | mod_mam: Advertise extended MAM when archive storage is capable | Kim Alvefur | 2021-01-12 | 2 | -3/+2 | |
| | | ||||||
* | | mod_mam: Allow querying by set of IDs | Kim Alvefur | 2021-01-12 | 1 | -2/+8 | |
| | | ||||||
* | | mod_storage_sql: Support query for set of IDs | Kim Alvefur | 2021-01-12 | 1 | -0/+10 | |
| | | | | | | | | Not compatible with Lua 5.1 | |||||
* | | mod_storage_internal: Support query for set of IDs | Kim Alvefur | 2021-01-12 | 1 | -0/+8 | |
| | | | | | | | | Based on implementation in mod_storage_memory | |||||
* | | mod_storage_memory: Support query for set of IDs | Kim Alvefur | 2021-01-12 | 1 | -0/+8 | |
| | | ||||||
* | | storage tests: Add test for querying a set of IDs | Kim Alvefur | 2021-01-12 | 1 | -0/+24 | |
| | | ||||||
* | | mod_storage_internal: Add support for full ID range query | Kim Alvefur | 2021-01-12 | 1 | -2/+5 | |
| | | | | | | | | Based on the mod_storage_memory implementation | |||||
* | | mod_storage_memory: Support full ID range queries | Kim Alvefur | 2021-01-12 | 1 | -2/+5 | |
| | | ||||||
* | | storage tests: Test querys with both before and after IDs | Kim Alvefur | 2021-01-12 | 1 | -0/+21 | |
| | | | | | | | | Needed for extended MAM, supported by SQL storage already. | |||||
* | | mod_mam: Add support for before-id and after-id fields | Kim Alvefur | 2021-01-09 | 1 | -4/+9 | |
| | | ||||||
* | | mod_storage_sql: Advertise support for id range query | Kim Alvefur | 2021-01-09 | 1 | -0/+1 | |
| | | | | | | | | That is, a query for items between two IDs (not inclusive). | |||||
* | | mod_mam: Implement extended MAM metadata query | Kim Alvefur | 2020-11-30 | 2 | -1/+49 | |
| | | ||||||
* | | mod_mam: Add support for page flipping | Kim Alvefur | 2020-11-30 | 2 | -3/+52 | |
| | | | | | | | | | | This was easy. Needs tests tho, 100% chance of the logic being backwards. | |||||
* | | mod_mam: Advertise extended MAM 0.7.x behind a feature flag | Kim Alvefur | 2020-11-30 | 3 | -2/+92 | |
| | | | | | | | | | | | | | | In order to ease testing until the extended feautres are all implemented. Also TODOs for all the sub-features. | |||||
* | | Merge 0.11->trunk | Kim Alvefur | 2021-01-12 | 1 | -0/+5 | |
|\| | ||||||
| * | net.adns: Prevent empty packets from being sent on "connect" (fix #1619) | Kim Alvefur | 2021-01-12 | 1 | -0/+5 | |
| | | | | | | | | Thanks Ge0rG for testing | |||||
| * | net.server_epoll: Fix off-by-one in 2c559953ad41 | Kim Alvefur | 2021-01-10 | 1 | -1/+1 | |
| | | | | | | | | Thanks tmolitor | |||||
* | | Merge 0.11->trunk | Kim Alvefur | 2021-01-08 | 3 | -5/+34 | |
|\| | ||||||
| * | util.timer: Ensure timers can't run more than once per tick (fixes #1620) | Kim Alvefur | 2021-01-08 | 1 | -1/+14 | |
| | | | | | | | | See longer explanation in 2c559953ad41 | |||||
| * | net.server_epoll: Ensure timers can't run more than once per tick | Kim Alvefur | 2021-01-08 | 1 | -2/+18 | |
| | | | | | | | | | | | | | | | | This makes sure that a timer that returns 0 (or less) does not prevent runtimers() from completing, as well as making sure a timer added with zero timeout from within a timer does not run until the next tick. Thanks tmolitor | |||||
| * | util.stanza: Move misplaced argument to correct place | Kim Alvefur | 2020-12-24 | 1 | -2/+2 | |
| | | | | | | | | | | valid_utf8() takes only one argument, so the false was probably meant to be valid_xml_cdata(text, attribute=false) | |||||
* | | mod_csi_simple,mod_carbons,mod_mam: Update comment about XEP-0353 | Kim Alvefur | 2021-01-08 | 3 | -3/+3 | |
| | | | | | | | | It went back to Deferred | |||||
* | | doap: XEP-0004 Multiple Items in Form Results is not supported | Kim Alvefur | 2020-12-29 | 1 | -1/+1 | |
| | | ||||||
* | | doap: LibExpat was used from the start and thus XEP-0044 supported | Kim Alvefur | 2020-12-29 | 1 | -0/+2 | |
| | | ||||||
* | | doap: util.dataforms first appeared in 0.4.0 | Kim Alvefur | 2020-12-29 | 1 | -0/+1 | |
| | | ||||||
* | | doap: Add removed status | Kim Alvefur | 2020-12-29 | 1 | -0/+8 | |
| | | ||||||
* | | doap: Specify exact Prosody version | Kim Alvefur | 2020-12-29 | 1 | -62/+62 | |
| | | | | | | | | Rarely is anything added in minor versions but it's nice to be exact. | |||||
* | | mod_scansion_record: Flush after writes | Matthew Wild | 2021-01-08 | 1 | -0/+3 | |
| | | ||||||
* | | net.unbound: Delay loading until server has started or first query | Kim Alvefur | 2021-01-05 | 1 | -1/+4 | |
| | | | | | | | | | | Shouldn't need a DNS resolver until later anyways. Might even be sensible to only initialize if a query is actually attempted. | |||||
* | | net.unbound: Move libunbound initialization into a function | Kim Alvefur | 2021-01-05 | 1 | -4/+7 | |
| | | | | | | | | Prepare for lazy-loading it. | |||||
* | | scansion: Allow specifying a logfile for prosody via environment | Kim Alvefur | 2021-01-05 | 1 | -1/+1 | |
| | | | | | | | | This should allow Buildbot to pick up those logs along with scansions | |||||
* | | net.unbound: Allow tracing individual queries with a logger per query | Kim Alvefur | 2021-01-05 | 1 | -6/+9 | |
| | | ||||||
* | | net.unbound: Don't pass error as logger formatting string | Kim Alvefur | 2021-01-05 | 1 | -1/+1 | |
| | | | | | | | | | | This could cause weirdness if the error contains formatting options, but should be reasonably safe with util.format | |||||
* | | net.unbound: Log net.server interactions | Kim Alvefur | 2021-01-05 | 1 | -0/+2 | |
| | | | | | | | | Noticed the potential need for this thanks to Ge0rG | |||||
* | | prosodyctl: Remove non-existant command from hidden commands | Kim Alvefur | 2021-01-05 | 1 | -1/+1 | |
| | | | | | | | | | | Apparently there existed a plugin installer for a brief time in 2010, between c5d3905c98a6 and 8f78e8164032 | |||||
* | | MUC: Reject probes from non-occupants | JC Brand | 2020-10-02 | 2 | -13/+85 | |
| | | | | | | | | Also test for self-probes | |||||
* | | MUC: Add support for presence probes (fixes #1535) | JC Brand | 2020-04-19 | 3 | -24/+202 | |
| | | | | | | | | | | | | | | | | | | | | | | The following patch allows Prosody to respond to `probe` presences and send out the probed occupant's current presence. This is based on line 17.3 in XEP-0045: A MUC service MAY handle presence probes sent to the room JID <room@service> or an occupant JID <room@service/nick> (e.g, these might be sent by an occupant's home server to determine if the room is still online or to synchronize presence information if the user or the user's server has gone offline temporarily or has started sharing presence again, as for instance when Stanza Interception and Filtering Technology (XEP-0273) is used). | |||||
* | | doap: Use logo property from schema.org | Kim Alvefur | 2021-01-02 | 1 | -2/+2 | |
| | | | | | | | | As requested by Link Mauve | |||||
* | | doap: Apply XML canonicalization | Kim Alvefur | 2021-01-02 | 1 | -1/+1 | |
| | | | | | | | | xmllint --c14n doc/doap.xml | xmllint --format - | sponge doc/doap.xml | |||||
* | | doap: Link to the actual logo (thanks Link Mauve) | Kim Alvefur | 2021-01-01 | 1 | -1/+1 | |
| | | ||||||
* | | doap: Link to logo | Kim Alvefur | 2021-01-01 | 1 | -0/+1 | |
| | | | | | | | | | | | | | | | | Link Mauve added logo to the xmpp doap namespace, waiting for https://github.com/ewilderj/doap/pull/68 to be merged Also, Happy New Year! | |||||
* | | mod_component: Distinguish missing stream @to from unknown local component ↵ | Kim Alvefur | 2020-12-18 | 1 | -0/+4 | |
| | | | | | | | | | | | | | | | | | | | | | | (thanks Daniel_W) Spectrum2 did not include a stream @to attribute due to a configuration error. "nil does not match ..." is not very clear in pointing this out and is somewhat confusing, tho it did help pinpoint what went wrong. It's also not great leak to programming language internals like "nil" out into user-facing strings like this. | |||||
* | | net.server_epoll: Increase log level for error in callback | Kim Alvefur | 2020-12-16 | 1 | -1/+1 | |
| | | | | | | | | It's an error, it should be logged at error level. | |||||
* | | Merge 0.11->trunk | Kim Alvefur | 2020-12-16 | 1 | -1/+1 | |
|\| |