0
0
Fork 0
mirror of https://github.com/matrix-org/dendrite synced 2024-11-16 14:50:52 +01:00
dendrite/syncapi/storage/interface.go

142 lines
10 KiB
Go
Raw Normal View History

Add peer-to-peer support into Dendrite via libp2p and fetch (#880) * Use a fork of pq which supports userCurrent on wasm * Use sqlite3_js driver when running in JS * Add cmd/dendritejs to pull in sqlite3_js driver for wasm only * Update to latest go-sqlite-js version * Replace prometheus with a stub. sigh * Hard-code a config and don't use opentracing * Latest go-sqlite3-js version * Generate a key for now * Listen for fetch traffic rather than HTTP * Latest hacks for js * libp2p support * More libp2p * Fork gjson to allow us to enforce auth checks as before Previously, all events would come down redacted because the hash checks would fail. They would fail because sjson.DeleteBytes didn't remove keys not used for hashing. This didn't work because of a build tag which included a file which no-oped the index returned. See https://github.com/tidwall/gjson/issues/157 When it's resolved, let's go back to mainline. * Use gjson@1.6.0 as it fixes https://github.com/tidwall/gjson/issues/157 * Use latest gomatrixserverlib for sig checks * Fix a bug which could cause exclude_from_sync to not be set Caused when sending events over federation. * Use query variadic to make lookups actually work! * Latest gomatrixserverlib * Add notes on getting p2p up and running Partly so I don't forget myself! * refactor: Move p2p specific stuff to cmd/dendritejs This is important or else the normal build of dendrite will fail because the p2p libraries depend on syscall/js which doesn't work on normal builds. Also, clean up main.go to read a bit better. * Update ho-http-js-libp2p to return errors from RoundTrip * Add an LRU cache around the key DB We actually need this for P2P because otherwise we can *segfault* with things like: "runtime: unexpected return pc for runtime.handleEvent" where the event is a `syscall/js` event, caused by spamming sql.js caused by "Checking event signatures for 14 events of room state" which hammers the key DB repeatedly in quick succession. Using a cache fixes this, though the underlying cause is probably a bug in the version of Go I'm on (1.13.7) * breaking: Add Tracing.Enabled to toggle whether we do opentracing Defaults to false, which is why this is a breaking change. We need this flag because WASM builds cannot do opentracing. * Start adding conditional builds for wasm to handle lib/pq The general idea here is to have the wasm build have a `NewXXXDatabase` that doesn't import any postgres package and hence we never import `lib/pq`, which doesn't work under WASM (undefined `userCurrent`). * Remove lib/pq for wasm for syncapi * Add conditional building to remaining storage APIs * Update build script to set env vars correctly for dendritejs * sqlite bug fixes * Docs * Add a no-op main for dendritejs when not building under wasm * Use the real prometheus, even for WASM Instead, the dendrite-sw.js must mock out `process.pid` and `fs.stat` - which must invoke the callback with an error (e.g `EINVAL`) in order for it to work: ``` global.process = { pid: 1, }; global.fs.stat = function(path, cb) { cb({ code: "EINVAL", }); } ``` * Linting
2020-03-06 11:23:55 +01:00
// Copyright 2020 The Matrix.org Foundation C.I.C.
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
package storage
import (
"context"
"time"
"github.com/matrix-org/dendrite/eduserver/cache"
"github.com/matrix-org/dendrite/internal"
Add peer-to-peer support into Dendrite via libp2p and fetch (#880) * Use a fork of pq which supports userCurrent on wasm * Use sqlite3_js driver when running in JS * Add cmd/dendritejs to pull in sqlite3_js driver for wasm only * Update to latest go-sqlite-js version * Replace prometheus with a stub. sigh * Hard-code a config and don't use opentracing * Latest go-sqlite3-js version * Generate a key for now * Listen for fetch traffic rather than HTTP * Latest hacks for js * libp2p support * More libp2p * Fork gjson to allow us to enforce auth checks as before Previously, all events would come down redacted because the hash checks would fail. They would fail because sjson.DeleteBytes didn't remove keys not used for hashing. This didn't work because of a build tag which included a file which no-oped the index returned. See https://github.com/tidwall/gjson/issues/157 When it's resolved, let's go back to mainline. * Use gjson@1.6.0 as it fixes https://github.com/tidwall/gjson/issues/157 * Use latest gomatrixserverlib for sig checks * Fix a bug which could cause exclude_from_sync to not be set Caused when sending events over federation. * Use query variadic to make lookups actually work! * Latest gomatrixserverlib * Add notes on getting p2p up and running Partly so I don't forget myself! * refactor: Move p2p specific stuff to cmd/dendritejs This is important or else the normal build of dendrite will fail because the p2p libraries depend on syscall/js which doesn't work on normal builds. Also, clean up main.go to read a bit better. * Update ho-http-js-libp2p to return errors from RoundTrip * Add an LRU cache around the key DB We actually need this for P2P because otherwise we can *segfault* with things like: "runtime: unexpected return pc for runtime.handleEvent" where the event is a `syscall/js` event, caused by spamming sql.js caused by "Checking event signatures for 14 events of room state" which hammers the key DB repeatedly in quick succession. Using a cache fixes this, though the underlying cause is probably a bug in the version of Go I'm on (1.13.7) * breaking: Add Tracing.Enabled to toggle whether we do opentracing Defaults to false, which is why this is a breaking change. We need this flag because WASM builds cannot do opentracing. * Start adding conditional builds for wasm to handle lib/pq The general idea here is to have the wasm build have a `NewXXXDatabase` that doesn't import any postgres package and hence we never import `lib/pq`, which doesn't work under WASM (undefined `userCurrent`). * Remove lib/pq for wasm for syncapi * Add conditional building to remaining storage APIs * Update build script to set env vars correctly for dendritejs * sqlite bug fixes * Docs * Add a no-op main for dendritejs when not building under wasm * Use the real prometheus, even for WASM Instead, the dendrite-sw.js must mock out `process.pid` and `fs.stat` - which must invoke the callback with an error (e.g `EINVAL`) in order for it to work: ``` global.process = { pid: 1, }; global.fs.stat = function(path, cb) { cb({ code: "EINVAL", }); } ``` * Linting
2020-03-06 11:23:55 +01:00
"github.com/matrix-org/dendrite/roomserver/api"
"github.com/matrix-org/dendrite/syncapi/types"
userapi "github.com/matrix-org/dendrite/userapi/api"
Add peer-to-peer support into Dendrite via libp2p and fetch (#880) * Use a fork of pq which supports userCurrent on wasm * Use sqlite3_js driver when running in JS * Add cmd/dendritejs to pull in sqlite3_js driver for wasm only * Update to latest go-sqlite-js version * Replace prometheus with a stub. sigh * Hard-code a config and don't use opentracing * Latest go-sqlite3-js version * Generate a key for now * Listen for fetch traffic rather than HTTP * Latest hacks for js * libp2p support * More libp2p * Fork gjson to allow us to enforce auth checks as before Previously, all events would come down redacted because the hash checks would fail. They would fail because sjson.DeleteBytes didn't remove keys not used for hashing. This didn't work because of a build tag which included a file which no-oped the index returned. See https://github.com/tidwall/gjson/issues/157 When it's resolved, let's go back to mainline. * Use gjson@1.6.0 as it fixes https://github.com/tidwall/gjson/issues/157 * Use latest gomatrixserverlib for sig checks * Fix a bug which could cause exclude_from_sync to not be set Caused when sending events over federation. * Use query variadic to make lookups actually work! * Latest gomatrixserverlib * Add notes on getting p2p up and running Partly so I don't forget myself! * refactor: Move p2p specific stuff to cmd/dendritejs This is important or else the normal build of dendrite will fail because the p2p libraries depend on syscall/js which doesn't work on normal builds. Also, clean up main.go to read a bit better. * Update ho-http-js-libp2p to return errors from RoundTrip * Add an LRU cache around the key DB We actually need this for P2P because otherwise we can *segfault* with things like: "runtime: unexpected return pc for runtime.handleEvent" where the event is a `syscall/js` event, caused by spamming sql.js caused by "Checking event signatures for 14 events of room state" which hammers the key DB repeatedly in quick succession. Using a cache fixes this, though the underlying cause is probably a bug in the version of Go I'm on (1.13.7) * breaking: Add Tracing.Enabled to toggle whether we do opentracing Defaults to false, which is why this is a breaking change. We need this flag because WASM builds cannot do opentracing. * Start adding conditional builds for wasm to handle lib/pq The general idea here is to have the wasm build have a `NewXXXDatabase` that doesn't import any postgres package and hence we never import `lib/pq`, which doesn't work under WASM (undefined `userCurrent`). * Remove lib/pq for wasm for syncapi * Add conditional building to remaining storage APIs * Update build script to set env vars correctly for dendritejs * sqlite bug fixes * Docs * Add a no-op main for dendritejs when not building under wasm * Use the real prometheus, even for WASM Instead, the dendrite-sw.js must mock out `process.pid` and `fs.stat` - which must invoke the callback with an error (e.g `EINVAL`) in order for it to work: ``` global.process = { pid: 1, }; global.fs.stat = function(path, cb) { cb({ code: "EINVAL", }); } ``` * Linting
2020-03-06 11:23:55 +01:00
"github.com/matrix-org/gomatrixserverlib"
)
type Database interface {
internal.PartitionStorer
// AllJoinedUsersInRooms returns a map of room ID to a list of all joined user IDs.
Add peer-to-peer support into Dendrite via libp2p and fetch (#880) * Use a fork of pq which supports userCurrent on wasm * Use sqlite3_js driver when running in JS * Add cmd/dendritejs to pull in sqlite3_js driver for wasm only * Update to latest go-sqlite-js version * Replace prometheus with a stub. sigh * Hard-code a config and don't use opentracing * Latest go-sqlite3-js version * Generate a key for now * Listen for fetch traffic rather than HTTP * Latest hacks for js * libp2p support * More libp2p * Fork gjson to allow us to enforce auth checks as before Previously, all events would come down redacted because the hash checks would fail. They would fail because sjson.DeleteBytes didn't remove keys not used for hashing. This didn't work because of a build tag which included a file which no-oped the index returned. See https://github.com/tidwall/gjson/issues/157 When it's resolved, let's go back to mainline. * Use gjson@1.6.0 as it fixes https://github.com/tidwall/gjson/issues/157 * Use latest gomatrixserverlib for sig checks * Fix a bug which could cause exclude_from_sync to not be set Caused when sending events over federation. * Use query variadic to make lookups actually work! * Latest gomatrixserverlib * Add notes on getting p2p up and running Partly so I don't forget myself! * refactor: Move p2p specific stuff to cmd/dendritejs This is important or else the normal build of dendrite will fail because the p2p libraries depend on syscall/js which doesn't work on normal builds. Also, clean up main.go to read a bit better. * Update ho-http-js-libp2p to return errors from RoundTrip * Add an LRU cache around the key DB We actually need this for P2P because otherwise we can *segfault* with things like: "runtime: unexpected return pc for runtime.handleEvent" where the event is a `syscall/js` event, caused by spamming sql.js caused by "Checking event signatures for 14 events of room state" which hammers the key DB repeatedly in quick succession. Using a cache fixes this, though the underlying cause is probably a bug in the version of Go I'm on (1.13.7) * breaking: Add Tracing.Enabled to toggle whether we do opentracing Defaults to false, which is why this is a breaking change. We need this flag because WASM builds cannot do opentracing. * Start adding conditional builds for wasm to handle lib/pq The general idea here is to have the wasm build have a `NewXXXDatabase` that doesn't import any postgres package and hence we never import `lib/pq`, which doesn't work under WASM (undefined `userCurrent`). * Remove lib/pq for wasm for syncapi * Add conditional building to remaining storage APIs * Update build script to set env vars correctly for dendritejs * sqlite bug fixes * Docs * Add a no-op main for dendritejs when not building under wasm * Use the real prometheus, even for WASM Instead, the dendrite-sw.js must mock out `process.pid` and `fs.stat` - which must invoke the callback with an error (e.g `EINVAL`) in order for it to work: ``` global.process = { pid: 1, }; global.fs.stat = function(path, cb) { cb({ code: "EINVAL", }); } ``` * Linting
2020-03-06 11:23:55 +01:00
AllJoinedUsersInRooms(ctx context.Context) (map[string][]string, error)
// Events lookups a list of event by their event ID.
// Returns a list of events matching the requested IDs found in the database.
// If an event is not found in the database then it will be omitted from the list.
// Returns an error if there was a problem talking with the database.
// Does not include any transaction IDs in the returned events.
Events(ctx context.Context, eventIDs []string) ([]gomatrixserverlib.HeaderedEvent, error)
// WriteEvent into the database. It is not safe to call this function from multiple goroutines, as it would create races
// when generating the sync stream position for this event. Returns the sync stream position for the inserted event.
// Returns an error if there was a problem inserting this event.
WriteEvent(ctx context.Context, ev *gomatrixserverlib.HeaderedEvent, addStateEvents []gomatrixserverlib.HeaderedEvent,
addStateEventIDs []string, removeStateEventIDs []string, transactionID *api.TransactionID, excludeFromSync bool) (types.StreamPosition, error)
// GetStateEvent returns the Matrix state event of a given type for a given room with a given state key
// If no event could be found, returns nil
// If there was an issue during the retrieval, returns an error
GetStateEvent(ctx context.Context, roomID, evType, stateKey string) (*gomatrixserverlib.HeaderedEvent, error)
// GetStateEventsForRoom fetches the state events for a given room.
// Returns an empty slice if no state events could be found for this room.
// Returns an error if there was an issue with the retrieval.
GetStateEventsForRoom(ctx context.Context, roomID string, stateFilterPart *gomatrixserverlib.StateFilter) (stateEvents []gomatrixserverlib.HeaderedEvent, err error)
// SyncPosition returns the latest positions for syncing.
SyncPosition(ctx context.Context) (types.StreamingToken, error)
// IncrementalSync returns all the data needed in order to create an incremental
// sync response for the given user. Events returned will include any client
// transaction IDs associated with the given device. These transaction IDs come
// from when the device sent the event via an API that included a transaction
Send-to-device support (#1072) * Groundwork for send-to-device messaging * Update sample config * Add unstable routing for now * Send to device consumer in sync API * Start the send-to-device consumer * fix indentation in dendrite-config.yaml * Create send-to-device database tables, other tweaks * Add some logic for send-to-device messages, add them into sync stream * Handle incoming send-to-device messages, count them with EDU stream pos * Undo changes to test * pq.Array * Fix sync * Logging * Fix a couple of transaction things, fix client API * Add send-to-device test, hopefully fix bugs * Comments * Refactor a bit * Fix schema * Fix queries * Debug logging * Fix storing and retrieving of send-to-device messages * Try to avoid database locks * Update sync position * Use latest sync position * Jiggle about sync a bit * Fix tests * Break out the retrieval from the update/delete behaviour * Comments * nolint on getResponseWithPDUsForCompleteSync * Try to line up sync tokens again * Implement wildcard * Add all send-to-device tests to whitelist, what could possibly go wrong? * Only care about wildcard when targeted locally * Deduplicate transactions * Handle tokens properly, return immediately if waiting send-to-device messages * Fix sync * Update sytest-whitelist * Fix copyright notice (need to do more of this) * Comments, copyrights * Return errors from Do, fix dendritejs * Review comments * Comments * Constructor for TransactionWriter * defletions * Update gomatrixserverlib, sytest-blacklist
2020-06-01 18:50:19 +02:00
// ID. A response object must be provided for IncrementaSync to populate - it
// will not create one.
IncrementalSync(ctx context.Context, res *types.Response, device userapi.Device, fromPos, toPos types.StreamingToken, numRecentEventsPerRoom int, wantFullState bool) (*types.Response, error)
Send-to-device support (#1072) * Groundwork for send-to-device messaging * Update sample config * Add unstable routing for now * Send to device consumer in sync API * Start the send-to-device consumer * fix indentation in dendrite-config.yaml * Create send-to-device database tables, other tweaks * Add some logic for send-to-device messages, add them into sync stream * Handle incoming send-to-device messages, count them with EDU stream pos * Undo changes to test * pq.Array * Fix sync * Logging * Fix a couple of transaction things, fix client API * Add send-to-device test, hopefully fix bugs * Comments * Refactor a bit * Fix schema * Fix queries * Debug logging * Fix storing and retrieving of send-to-device messages * Try to avoid database locks * Update sync position * Use latest sync position * Jiggle about sync a bit * Fix tests * Break out the retrieval from the update/delete behaviour * Comments * nolint on getResponseWithPDUsForCompleteSync * Try to line up sync tokens again * Implement wildcard * Add all send-to-device tests to whitelist, what could possibly go wrong? * Only care about wildcard when targeted locally * Deduplicate transactions * Handle tokens properly, return immediately if waiting send-to-device messages * Fix sync * Update sytest-whitelist * Fix copyright notice (need to do more of this) * Comments, copyrights * Return errors from Do, fix dendritejs * Review comments * Comments * Constructor for TransactionWriter * defletions * Update gomatrixserverlib, sytest-blacklist
2020-06-01 18:50:19 +02:00
// CompleteSync returns a complete /sync API response for the given user. A response object
// must be provided for CompleteSync to populate - it will not create one.
CompleteSync(ctx context.Context, res *types.Response, device userapi.Device, numRecentEventsPerRoom int) (*types.Response, error)
// GetAccountDataInRange returns all account data for a given user inserted or
// updated between two given positions
// Returns a map following the format data[roomID] = []dataTypes
// If no data is retrieved, returns an empty map
// If there was an issue with the retrieval, returns an error
GetAccountDataInRange(ctx context.Context, userID string, r types.Range, accountDataFilterPart *gomatrixserverlib.EventFilter) (map[string][]string, error)
// UpsertAccountData keeps track of new or updated account data, by saving the type
// of the new/updated data, and the user ID and room ID the data is related to (empty)
// room ID means the data isn't specific to any room)
// If no data with the given type, user ID and room ID exists in the database,
// creates a new row, else update the existing one
// Returns an error if there was an issue with the upsert
Add peer-to-peer support into Dendrite via libp2p and fetch (#880) * Use a fork of pq which supports userCurrent on wasm * Use sqlite3_js driver when running in JS * Add cmd/dendritejs to pull in sqlite3_js driver for wasm only * Update to latest go-sqlite-js version * Replace prometheus with a stub. sigh * Hard-code a config and don't use opentracing * Latest go-sqlite3-js version * Generate a key for now * Listen for fetch traffic rather than HTTP * Latest hacks for js * libp2p support * More libp2p * Fork gjson to allow us to enforce auth checks as before Previously, all events would come down redacted because the hash checks would fail. They would fail because sjson.DeleteBytes didn't remove keys not used for hashing. This didn't work because of a build tag which included a file which no-oped the index returned. See https://github.com/tidwall/gjson/issues/157 When it's resolved, let's go back to mainline. * Use gjson@1.6.0 as it fixes https://github.com/tidwall/gjson/issues/157 * Use latest gomatrixserverlib for sig checks * Fix a bug which could cause exclude_from_sync to not be set Caused when sending events over federation. * Use query variadic to make lookups actually work! * Latest gomatrixserverlib * Add notes on getting p2p up and running Partly so I don't forget myself! * refactor: Move p2p specific stuff to cmd/dendritejs This is important or else the normal build of dendrite will fail because the p2p libraries depend on syscall/js which doesn't work on normal builds. Also, clean up main.go to read a bit better. * Update ho-http-js-libp2p to return errors from RoundTrip * Add an LRU cache around the key DB We actually need this for P2P because otherwise we can *segfault* with things like: "runtime: unexpected return pc for runtime.handleEvent" where the event is a `syscall/js` event, caused by spamming sql.js caused by "Checking event signatures for 14 events of room state" which hammers the key DB repeatedly in quick succession. Using a cache fixes this, though the underlying cause is probably a bug in the version of Go I'm on (1.13.7) * breaking: Add Tracing.Enabled to toggle whether we do opentracing Defaults to false, which is why this is a breaking change. We need this flag because WASM builds cannot do opentracing. * Start adding conditional builds for wasm to handle lib/pq The general idea here is to have the wasm build have a `NewXXXDatabase` that doesn't import any postgres package and hence we never import `lib/pq`, which doesn't work under WASM (undefined `userCurrent`). * Remove lib/pq for wasm for syncapi * Add conditional building to remaining storage APIs * Update build script to set env vars correctly for dendritejs * sqlite bug fixes * Docs * Add a no-op main for dendritejs when not building under wasm * Use the real prometheus, even for WASM Instead, the dendrite-sw.js must mock out `process.pid` and `fs.stat` - which must invoke the callback with an error (e.g `EINVAL`) in order for it to work: ``` global.process = { pid: 1, }; global.fs.stat = function(path, cb) { cb({ code: "EINVAL", }); } ``` * Linting
2020-03-06 11:23:55 +01:00
UpsertAccountData(ctx context.Context, userID, roomID, dataType string) (types.StreamPosition, error)
// AddInviteEvent stores a new invite event for a user.
// If the invite was successfully stored this returns the stream ID it was stored at.
// Returns an error if there was a problem communicating with the database.
AddInviteEvent(ctx context.Context, inviteEvent gomatrixserverlib.HeaderedEvent) (types.StreamPosition, error)
// RetireInviteEvent removes an old invite event from the database. Returns the new position of the retired invite.
// Returns an error if there was a problem communicating with the database.
RetireInviteEvent(ctx context.Context, inviteEventID string) (types.StreamPosition, error)
// SetTypingTimeoutCallback sets a callback function that is called right after
// a user is removed from the typing user list due to timeout.
Add peer-to-peer support into Dendrite via libp2p and fetch (#880) * Use a fork of pq which supports userCurrent on wasm * Use sqlite3_js driver when running in JS * Add cmd/dendritejs to pull in sqlite3_js driver for wasm only * Update to latest go-sqlite-js version * Replace prometheus with a stub. sigh * Hard-code a config and don't use opentracing * Latest go-sqlite3-js version * Generate a key for now * Listen for fetch traffic rather than HTTP * Latest hacks for js * libp2p support * More libp2p * Fork gjson to allow us to enforce auth checks as before Previously, all events would come down redacted because the hash checks would fail. They would fail because sjson.DeleteBytes didn't remove keys not used for hashing. This didn't work because of a build tag which included a file which no-oped the index returned. See https://github.com/tidwall/gjson/issues/157 When it's resolved, let's go back to mainline. * Use gjson@1.6.0 as it fixes https://github.com/tidwall/gjson/issues/157 * Use latest gomatrixserverlib for sig checks * Fix a bug which could cause exclude_from_sync to not be set Caused when sending events over federation. * Use query variadic to make lookups actually work! * Latest gomatrixserverlib * Add notes on getting p2p up and running Partly so I don't forget myself! * refactor: Move p2p specific stuff to cmd/dendritejs This is important or else the normal build of dendrite will fail because the p2p libraries depend on syscall/js which doesn't work on normal builds. Also, clean up main.go to read a bit better. * Update ho-http-js-libp2p to return errors from RoundTrip * Add an LRU cache around the key DB We actually need this for P2P because otherwise we can *segfault* with things like: "runtime: unexpected return pc for runtime.handleEvent" where the event is a `syscall/js` event, caused by spamming sql.js caused by "Checking event signatures for 14 events of room state" which hammers the key DB repeatedly in quick succession. Using a cache fixes this, though the underlying cause is probably a bug in the version of Go I'm on (1.13.7) * breaking: Add Tracing.Enabled to toggle whether we do opentracing Defaults to false, which is why this is a breaking change. We need this flag because WASM builds cannot do opentracing. * Start adding conditional builds for wasm to handle lib/pq The general idea here is to have the wasm build have a `NewXXXDatabase` that doesn't import any postgres package and hence we never import `lib/pq`, which doesn't work under WASM (undefined `userCurrent`). * Remove lib/pq for wasm for syncapi * Add conditional building to remaining storage APIs * Update build script to set env vars correctly for dendritejs * sqlite bug fixes * Docs * Add a no-op main for dendritejs when not building under wasm * Use the real prometheus, even for WASM Instead, the dendrite-sw.js must mock out `process.pid` and `fs.stat` - which must invoke the callback with an error (e.g `EINVAL`) in order for it to work: ``` global.process = { pid: 1, }; global.fs.stat = function(path, cb) { cb({ code: "EINVAL", }); } ``` * Linting
2020-03-06 11:23:55 +01:00
SetTypingTimeoutCallback(fn cache.TimeoutCallbackFn)
// AddTypingUser adds a typing user to the typing cache.
// Returns the newly calculated sync position for typing notifications.
Add peer-to-peer support into Dendrite via libp2p and fetch (#880) * Use a fork of pq which supports userCurrent on wasm * Use sqlite3_js driver when running in JS * Add cmd/dendritejs to pull in sqlite3_js driver for wasm only * Update to latest go-sqlite-js version * Replace prometheus with a stub. sigh * Hard-code a config and don't use opentracing * Latest go-sqlite3-js version * Generate a key for now * Listen for fetch traffic rather than HTTP * Latest hacks for js * libp2p support * More libp2p * Fork gjson to allow us to enforce auth checks as before Previously, all events would come down redacted because the hash checks would fail. They would fail because sjson.DeleteBytes didn't remove keys not used for hashing. This didn't work because of a build tag which included a file which no-oped the index returned. See https://github.com/tidwall/gjson/issues/157 When it's resolved, let's go back to mainline. * Use gjson@1.6.0 as it fixes https://github.com/tidwall/gjson/issues/157 * Use latest gomatrixserverlib for sig checks * Fix a bug which could cause exclude_from_sync to not be set Caused when sending events over federation. * Use query variadic to make lookups actually work! * Latest gomatrixserverlib * Add notes on getting p2p up and running Partly so I don't forget myself! * refactor: Move p2p specific stuff to cmd/dendritejs This is important or else the normal build of dendrite will fail because the p2p libraries depend on syscall/js which doesn't work on normal builds. Also, clean up main.go to read a bit better. * Update ho-http-js-libp2p to return errors from RoundTrip * Add an LRU cache around the key DB We actually need this for P2P because otherwise we can *segfault* with things like: "runtime: unexpected return pc for runtime.handleEvent" where the event is a `syscall/js` event, caused by spamming sql.js caused by "Checking event signatures for 14 events of room state" which hammers the key DB repeatedly in quick succession. Using a cache fixes this, though the underlying cause is probably a bug in the version of Go I'm on (1.13.7) * breaking: Add Tracing.Enabled to toggle whether we do opentracing Defaults to false, which is why this is a breaking change. We need this flag because WASM builds cannot do opentracing. * Start adding conditional builds for wasm to handle lib/pq The general idea here is to have the wasm build have a `NewXXXDatabase` that doesn't import any postgres package and hence we never import `lib/pq`, which doesn't work under WASM (undefined `userCurrent`). * Remove lib/pq for wasm for syncapi * Add conditional building to remaining storage APIs * Update build script to set env vars correctly for dendritejs * sqlite bug fixes * Docs * Add a no-op main for dendritejs when not building under wasm * Use the real prometheus, even for WASM Instead, the dendrite-sw.js must mock out `process.pid` and `fs.stat` - which must invoke the callback with an error (e.g `EINVAL`) in order for it to work: ``` global.process = { pid: 1, }; global.fs.stat = function(path, cb) { cb({ code: "EINVAL", }); } ``` * Linting
2020-03-06 11:23:55 +01:00
AddTypingUser(userID, roomID string, expireTime *time.Time) types.StreamPosition
// RemoveTypingUser removes a typing user from the typing cache.
// Returns the newly calculated sync position for typing notifications.
Add peer-to-peer support into Dendrite via libp2p and fetch (#880) * Use a fork of pq which supports userCurrent on wasm * Use sqlite3_js driver when running in JS * Add cmd/dendritejs to pull in sqlite3_js driver for wasm only * Update to latest go-sqlite-js version * Replace prometheus with a stub. sigh * Hard-code a config and don't use opentracing * Latest go-sqlite3-js version * Generate a key for now * Listen for fetch traffic rather than HTTP * Latest hacks for js * libp2p support * More libp2p * Fork gjson to allow us to enforce auth checks as before Previously, all events would come down redacted because the hash checks would fail. They would fail because sjson.DeleteBytes didn't remove keys not used for hashing. This didn't work because of a build tag which included a file which no-oped the index returned. See https://github.com/tidwall/gjson/issues/157 When it's resolved, let's go back to mainline. * Use gjson@1.6.0 as it fixes https://github.com/tidwall/gjson/issues/157 * Use latest gomatrixserverlib for sig checks * Fix a bug which could cause exclude_from_sync to not be set Caused when sending events over federation. * Use query variadic to make lookups actually work! * Latest gomatrixserverlib * Add notes on getting p2p up and running Partly so I don't forget myself! * refactor: Move p2p specific stuff to cmd/dendritejs This is important or else the normal build of dendrite will fail because the p2p libraries depend on syscall/js which doesn't work on normal builds. Also, clean up main.go to read a bit better. * Update ho-http-js-libp2p to return errors from RoundTrip * Add an LRU cache around the key DB We actually need this for P2P because otherwise we can *segfault* with things like: "runtime: unexpected return pc for runtime.handleEvent" where the event is a `syscall/js` event, caused by spamming sql.js caused by "Checking event signatures for 14 events of room state" which hammers the key DB repeatedly in quick succession. Using a cache fixes this, though the underlying cause is probably a bug in the version of Go I'm on (1.13.7) * breaking: Add Tracing.Enabled to toggle whether we do opentracing Defaults to false, which is why this is a breaking change. We need this flag because WASM builds cannot do opentracing. * Start adding conditional builds for wasm to handle lib/pq The general idea here is to have the wasm build have a `NewXXXDatabase` that doesn't import any postgres package and hence we never import `lib/pq`, which doesn't work under WASM (undefined `userCurrent`). * Remove lib/pq for wasm for syncapi * Add conditional building to remaining storage APIs * Update build script to set env vars correctly for dendritejs * sqlite bug fixes * Docs * Add a no-op main for dendritejs when not building under wasm * Use the real prometheus, even for WASM Instead, the dendrite-sw.js must mock out `process.pid` and `fs.stat` - which must invoke the callback with an error (e.g `EINVAL`) in order for it to work: ``` global.process = { pid: 1, }; global.fs.stat = function(path, cb) { cb({ code: "EINVAL", }); } ``` * Linting
2020-03-06 11:23:55 +01:00
RemoveTypingUser(userID, roomID string) types.StreamPosition
// GetEventsInStreamingRange retrieves all of the events on a given ordering using the given extremities and limit.
GetEventsInStreamingRange(ctx context.Context, from, to *types.StreamingToken, roomID string, limit int, backwardOrdering bool) (events []types.StreamEvent, err error)
// GetEventsInTopologicalRange retrieves all of the events on a given ordering using the given extremities and limit.
GetEventsInTopologicalRange(ctx context.Context, from, to *types.TopologyToken, roomID string, limit int, backwardOrdering bool) (events []types.StreamEvent, err error)
// EventPositionInTopology returns the depth and stream position of the given event.
EventPositionInTopology(ctx context.Context, eventID string) (types.TopologyToken, error)
// BackwardExtremitiesForRoom returns a map of backwards extremity event ID to a list of its prev_events.
BackwardExtremitiesForRoom(ctx context.Context, roomID string) (backwardExtremities map[string][]string, err error)
// MaxTopologicalPosition returns the highest topological position for a given room.
MaxTopologicalPosition(ctx context.Context, roomID string) (types.TopologyToken, error)
// StreamEventsToEvents converts streamEvent to Event. If device is non-nil and
// matches the streamevent.transactionID device then the transaction ID gets
// added to the unsigned section of the output event.
StreamEventsToEvents(device *userapi.Device, in []types.StreamEvent) []gomatrixserverlib.HeaderedEvent
// SyncStreamPosition returns the latest position in the sync stream. Returns 0 if there are no events yet.
Add peer-to-peer support into Dendrite via libp2p and fetch (#880) * Use a fork of pq which supports userCurrent on wasm * Use sqlite3_js driver when running in JS * Add cmd/dendritejs to pull in sqlite3_js driver for wasm only * Update to latest go-sqlite-js version * Replace prometheus with a stub. sigh * Hard-code a config and don't use opentracing * Latest go-sqlite3-js version * Generate a key for now * Listen for fetch traffic rather than HTTP * Latest hacks for js * libp2p support * More libp2p * Fork gjson to allow us to enforce auth checks as before Previously, all events would come down redacted because the hash checks would fail. They would fail because sjson.DeleteBytes didn't remove keys not used for hashing. This didn't work because of a build tag which included a file which no-oped the index returned. See https://github.com/tidwall/gjson/issues/157 When it's resolved, let's go back to mainline. * Use gjson@1.6.0 as it fixes https://github.com/tidwall/gjson/issues/157 * Use latest gomatrixserverlib for sig checks * Fix a bug which could cause exclude_from_sync to not be set Caused when sending events over federation. * Use query variadic to make lookups actually work! * Latest gomatrixserverlib * Add notes on getting p2p up and running Partly so I don't forget myself! * refactor: Move p2p specific stuff to cmd/dendritejs This is important or else the normal build of dendrite will fail because the p2p libraries depend on syscall/js which doesn't work on normal builds. Also, clean up main.go to read a bit better. * Update ho-http-js-libp2p to return errors from RoundTrip * Add an LRU cache around the key DB We actually need this for P2P because otherwise we can *segfault* with things like: "runtime: unexpected return pc for runtime.handleEvent" where the event is a `syscall/js` event, caused by spamming sql.js caused by "Checking event signatures for 14 events of room state" which hammers the key DB repeatedly in quick succession. Using a cache fixes this, though the underlying cause is probably a bug in the version of Go I'm on (1.13.7) * breaking: Add Tracing.Enabled to toggle whether we do opentracing Defaults to false, which is why this is a breaking change. We need this flag because WASM builds cannot do opentracing. * Start adding conditional builds for wasm to handle lib/pq The general idea here is to have the wasm build have a `NewXXXDatabase` that doesn't import any postgres package and hence we never import `lib/pq`, which doesn't work under WASM (undefined `userCurrent`). * Remove lib/pq for wasm for syncapi * Add conditional building to remaining storage APIs * Update build script to set env vars correctly for dendritejs * sqlite bug fixes * Docs * Add a no-op main for dendritejs when not building under wasm * Use the real prometheus, even for WASM Instead, the dendrite-sw.js must mock out `process.pid` and `fs.stat` - which must invoke the callback with an error (e.g `EINVAL`) in order for it to work: ``` global.process = { pid: 1, }; global.fs.stat = function(path, cb) { cb({ code: "EINVAL", }); } ``` * Linting
2020-03-06 11:23:55 +01:00
SyncStreamPosition(ctx context.Context) (types.StreamPosition, error)
Send-to-device support (#1072) * Groundwork for send-to-device messaging * Update sample config * Add unstable routing for now * Send to device consumer in sync API * Start the send-to-device consumer * fix indentation in dendrite-config.yaml * Create send-to-device database tables, other tweaks * Add some logic for send-to-device messages, add them into sync stream * Handle incoming send-to-device messages, count them with EDU stream pos * Undo changes to test * pq.Array * Fix sync * Logging * Fix a couple of transaction things, fix client API * Add send-to-device test, hopefully fix bugs * Comments * Refactor a bit * Fix schema * Fix queries * Debug logging * Fix storing and retrieving of send-to-device messages * Try to avoid database locks * Update sync position * Use latest sync position * Jiggle about sync a bit * Fix tests * Break out the retrieval from the update/delete behaviour * Comments * nolint on getResponseWithPDUsForCompleteSync * Try to line up sync tokens again * Implement wildcard * Add all send-to-device tests to whitelist, what could possibly go wrong? * Only care about wildcard when targeted locally * Deduplicate transactions * Handle tokens properly, return immediately if waiting send-to-device messages * Fix sync * Update sytest-whitelist * Fix copyright notice (need to do more of this) * Comments, copyrights * Return errors from Do, fix dendritejs * Review comments * Comments * Constructor for TransactionWriter * defletions * Update gomatrixserverlib, sytest-blacklist
2020-06-01 18:50:19 +02:00
// AddSendToDevice increases the EDU position in the cache and returns the stream position.
AddSendToDevice() types.StreamPosition
// SendToDeviceUpdatesForSync returns a list of send-to-device updates. It returns three lists:
// - "events": a list of send-to-device events that should be included in the sync
// - "changes": a list of send-to-device events that should be updated in the database by
// CleanSendToDeviceUpdates
// - "deletions": a list of send-to-device events which have been confirmed as sent and
// can be deleted altogether by CleanSendToDeviceUpdates
// The token supplied should be the current requested sync token, e.g. from the "since"
// parameter.
SendToDeviceUpdatesForSync(ctx context.Context, userID, deviceID string, token types.StreamingToken) (events []types.SendToDeviceEvent, changes []types.SendToDeviceNID, deletions []types.SendToDeviceNID, err error)
// StoreNewSendForDeviceMessage stores a new send-to-device event for a user's device.
StoreNewSendForDeviceMessage(ctx context.Context, streamPos types.StreamPosition, userID, deviceID string, event gomatrixserverlib.SendToDeviceEvent) (types.StreamPosition, error)
// CleanSendToDeviceUpdates will update or remove any send-to-device updates based on the
// result to a previous call to SendDeviceUpdatesForSync. This is separate as it allows
// SendToDeviceUpdatesForSync to be called multiple times if needed (e.g. before and after
// starting to wait for an incremental sync with timeout).
// The token supplied should be the current requested sync token, e.g. from the "since"
// parameter.
CleanSendToDeviceUpdates(ctx context.Context, toUpdate, toDelete []types.SendToDeviceNID, token types.StreamingToken) (err error)
// SendToDeviceUpdatesWaiting returns true if there are send-to-device updates waiting to be sent.
SendToDeviceUpdatesWaiting(ctx context.Context, userID, deviceID string) (bool, error)
// GetFilter looks up the filter associated with a given local user and filter ID.
// Returns a filter structure. Otherwise returns an error if no such filter exists
// or if there was an error talking to the database.
GetFilter(ctx context.Context, localpart string, filterID string) (*gomatrixserverlib.Filter, error)
// PutFilter puts the passed filter into the database.
// Returns the filterID as a string. Otherwise returns an error if something
// goes wrong.
PutFilter(ctx context.Context, localpart string, filter *gomatrixserverlib.Filter) (string, error)
// RedactEvent wipes an event in the database and sets the unsigned.redacted_because key to the redaction event
RedactEvent(ctx context.Context, redactedEventID string, redactedBecause *gomatrixserverlib.HeaderedEvent) error
Add peer-to-peer support into Dendrite via libp2p and fetch (#880) * Use a fork of pq which supports userCurrent on wasm * Use sqlite3_js driver when running in JS * Add cmd/dendritejs to pull in sqlite3_js driver for wasm only * Update to latest go-sqlite-js version * Replace prometheus with a stub. sigh * Hard-code a config and don't use opentracing * Latest go-sqlite3-js version * Generate a key for now * Listen for fetch traffic rather than HTTP * Latest hacks for js * libp2p support * More libp2p * Fork gjson to allow us to enforce auth checks as before Previously, all events would come down redacted because the hash checks would fail. They would fail because sjson.DeleteBytes didn't remove keys not used for hashing. This didn't work because of a build tag which included a file which no-oped the index returned. See https://github.com/tidwall/gjson/issues/157 When it's resolved, let's go back to mainline. * Use gjson@1.6.0 as it fixes https://github.com/tidwall/gjson/issues/157 * Use latest gomatrixserverlib for sig checks * Fix a bug which could cause exclude_from_sync to not be set Caused when sending events over federation. * Use query variadic to make lookups actually work! * Latest gomatrixserverlib * Add notes on getting p2p up and running Partly so I don't forget myself! * refactor: Move p2p specific stuff to cmd/dendritejs This is important or else the normal build of dendrite will fail because the p2p libraries depend on syscall/js which doesn't work on normal builds. Also, clean up main.go to read a bit better. * Update ho-http-js-libp2p to return errors from RoundTrip * Add an LRU cache around the key DB We actually need this for P2P because otherwise we can *segfault* with things like: "runtime: unexpected return pc for runtime.handleEvent" where the event is a `syscall/js` event, caused by spamming sql.js caused by "Checking event signatures for 14 events of room state" which hammers the key DB repeatedly in quick succession. Using a cache fixes this, though the underlying cause is probably a bug in the version of Go I'm on (1.13.7) * breaking: Add Tracing.Enabled to toggle whether we do opentracing Defaults to false, which is why this is a breaking change. We need this flag because WASM builds cannot do opentracing. * Start adding conditional builds for wasm to handle lib/pq The general idea here is to have the wasm build have a `NewXXXDatabase` that doesn't import any postgres package and hence we never import `lib/pq`, which doesn't work under WASM (undefined `userCurrent`). * Remove lib/pq for wasm for syncapi * Add conditional building to remaining storage APIs * Update build script to set env vars correctly for dendritejs * sqlite bug fixes * Docs * Add a no-op main for dendritejs when not building under wasm * Use the real prometheus, even for WASM Instead, the dendrite-sw.js must mock out `process.pid` and `fs.stat` - which must invoke the callback with an error (e.g `EINVAL`) in order for it to work: ``` global.process = { pid: 1, }; global.fs.stat = function(path, cb) { cb({ code: "EINVAL", }); } ``` * Linting
2020-03-06 11:23:55 +01:00
}