Sometimes the message scheduled & message sent notifications are
received out of order.
Now we use a single channel for both so order is maintained.
This was causing the tests to be flaky and it might have happened in
production as well.
We need to discard installations with no metadata, as some might be sent
before, so we need to check in the loop that the one we care about is
received.
This commit fixes one source of flakyness in the tests, which was an
actual bug.
If 1 device is registering with a push notification server, if there's
another device with the same public key, both would mark themselves as
registered, while maybe only one has been actually registered.
To fix this, we keep track of the request ids we send (in memory for
now), and only mark it as registered if the request was originating on
this device.
* Protobufs and adapters
* Added basic anon metric service and config init
* Added fibonacci interval incrementer
* Added basic Client.Start func and integrated interval incrementer
* Added new processed field to app metrics table
* Added id column to app metrics table
* Added migration clean up
* Added appmetrics GetUnprocessed and SetToProcessedByIDs and tests
There was a wierd bug where metrics in the db that did not explicitly insert a value would be NULL, so could not be found by . In addition I've added a new primary id field to the app_metrics table so that updates could be done against very specific metric rows.
* Updated adaptors and db to handle proto_id
I need a way to distinguish individual metric items from each other so that I can ignore the ones that have been seen before.
* Moved incrementer into dedicated file
* Resolve incrementer test fail
* Finalised the main loop functionality
* Implemented delete loop framework
* Updated adaptors file name
* Added delete loop delay and quit, and tweak on RawMessage gen
* Completed delete loop logic
* Added DBLock to prevent deletion during mainLoop
* Added postgres DB connection, integrated into anonmetrics.Server
* Removed proto_id from SQL migration and model
* Integrated postgres with Server and updated adaptors
* Function name update
* Added sample config files for client and server
* Fixes and testing for low level e2e
* make generate
* Fix lint
* Fix for receiving an anonMetricBatch not in server mode
* Postgres test fixes
* Tidy up, make vendor and make generate
* delinting
* Fixing database tests
* Attempted fix of does: cannot open `does' (No such file or directory)
not: cannot open `not' (No such file or directory)
exist: cannot open `exist' (No such file or directory) error on sql resource loas
* Moved all anon metric postgres migration logic and sources into a the protocol/anonmetrics package or sub packages. I don't know if this will fix the does: cannot open `does' (No such file or directory)
not: cannot open `not' (No such file or directory)
exist: cannot open `exist' (No such file or directory) error that happens in Jenkins but this could work
* Lint for the lint god
* Why doesn't the linter list all its problems at once?
* test tweaks
* Fix for wakuV2 change
* DB reset change
* Fix for postgres db migrations fails
* More robust implementation of postgres test setup and teardown
* Added block for anon metrics functionality
* Version Bump to 0.84.0
* Added test to check anon metrics broadcast is deactivated
* Protobufs and adapters
* Added basic anon metric service and config init
* Added new processed field to app metrics table
* Added id column to app metrics table
* Added migration clean up
* Added appmetrics GetUnprocessed and SetToProcessedByIDs and tests
There was a wierd bug where metrics in the db that did not explicitly insert a value would be NULL, so could not be found by . In addition I've added a new primary id field to the app_metrics table so that updates could be done against very specific metric rows.
* Updated adaptors and db to handle proto_id
I need a way to distinguish individual metric items from each other so that I can ignore the ones that have been seen before.
* Added postgres DB connection, integrated into anonmetrics.Server
* Removed proto_id from SQL migration and model
* Integrated postgres with Server and updated adaptors
* Added sample config files for client and server
* Fix lint
* Fix for receiving an anonMetricBatch not in server mode
* Postgres test fixes
* Tidy up, make vendor and make generate
* Moved all anon metric postgres migration logic and sources into a the protocol/anonmetrics package or sub packages. I don't know if this will fix the does: cannot open `does' (No such file or directory)
not: cannot open `not' (No such file or directory)
exist: cannot open `exist' (No such file or directory) error that happens in Jenkins but this could work
AllMessagesFromChatsAndCommunitiesWhichMatchTerm method added which returns all messages which
match the search term, if they belong to either any chat from the chatIds array or any channel of
any community from communityIds array.
This api point is necessary for desktop issue 2934.
Fixes: #2934
* Added community sync protobuf
* Updated community sync send logic
* Integrated syncCommunity handling
* Added synced_at field and tidied up some other logic
* persistence testing
* Added testing and join functionality
* Fixed issue with empty scan params
* Finshed persistence tests for new db funcs
* Midway debug of description not persisting after sync
* Resolved final issues and tidied up
* Polish
* delint
* Fix error not handled on SetPrivateKey
* fix infinite loop, again
* Added muted option and test fix
* Added Muted to syncing functions, not just in persistence
* Fix bug introduced with Muted property
* Added a couple of notes for future devs
* Added most of the sync RequestToJoin functionality
Tests need to be completed and tests are giving some errors
* Finished tests for getJoinedAndPending
* Added note
* Resolving lint
* Fix of protobuf gen bug
* Fixes to community sync tests
* Fixes to test
* Continued fix of e2e
* Final fix to e2e testing
* Updated migration position
* resolve missing import
* Apparently the linter spellchecks
* Fix bug from #2276 merge
* Bug fix for leaving quirkiness
* Addressed superfluous MessengerResponse field
* Addressed feedback
* VERSION bump
* feature(@status-go/chat): implement search on sqlcipher (status-go side)
Searching messages by some term for a specific channel is added on the side of status-go.
Fixes: #2912
* Linting
Co-authored-by: Andrea Maria Piana <andrea.maria.piana@gmail.com>
Changes the community invitation text notifications from
“Upgrade to see a community invitation”
to
**InviteUsersToCommunity**
```golang
fmt.Sprintf("You have been invited to the community %s", community.Name())
```
**ShareCommunity**
```golang
fmt.Sprintf("Community %s has been shared with you", community.Name())
```
* Adding wakunode module
* Adding wakuv2 fleet files
* Add waku fleets to update-fleet-config script
* Adding config items for waku v2
* Conditionally start waku v2 node depending on config
* Adapting common code to use go-waku
* Setting log level to info
* update dependencies
* update fleet config to use WakuNodes instead of BootNodes
* send and receive messages
* use hash returned when publishing a message
* add waku store protocol
* trigger signal after receiving store messages
* exclude linting rule SA1019 to check deprecated packages
This commit adds a setting for mailserver, that allows the client to
specify the syncing period.
Also fixes a minor issue with deletion of public chats.
* feat: Add edit communities
Allow Communities to be edited, including display name, description, color, membership, and permissions.
* Added EditCommunity request type
* Fix lint errors
* Allow editing community without changing image
Previously, retaining an existing community image was not possible because the existing community image path had to be provided in the `editCommunity` RPC call to retain the image. However, once the image is processed by status-go, it is encoded as a base64 string and therefore it is not possible to get the original file path back from this string.
This commit allows for the original to be retained by passing an empty string for the image field in the RPC call.
* Don't change permissions. Fixed clock updating
Co-authored-by: Volodymyr Kozieiev <vkjr.sp@gmail.com>
This property is useful for clients to know when a channel or chat
was joined so they can use that to calculate the order of channels
and chats shown in applications.
Changes include a new joined property on the Chat struct,
as well as adjustments in the persistence layer to retreive and
update chat data in the database.
In addition there's a migration script that alters the existing
chat table to introduce a new column for the joined field.
It also updates all existing rows in the database to set `joined`
to `0`.
* add PinMessage and PinnedMessage
* fix gruop pin messages
* add SkipGroupMessageWrap to pin messages
* update pinMessage ID generation to be symmetric
* Added anon metrics send opt in setting
* resolved rebase conflict, renamed migration to use unixtimestamp
Theres always conflicts with migrations using sequential numbers, less so with unix timestamp
* Add extra event to capture other type of navigations, allow empty screen name, rename cofx to get rid of clj ns, update tests
* Some view ids are greater than 16 characters. Made it 32 to be safe.
* Tab navigation events occur outside nav, add a new validator for them
* Remove navigate to cofx event, capture screens on will focus, get rid of enum and make valid screens a string less than 32 characters
* Run make generate
* Fix test
* Bump version to 0.75.1
This commit introduces the following changes:
- `local-notifications` require as body an interface complying with
`json.Marshaler`
- removed unmarshaling of `Notifications` as not used (we only Marshal
notifications)
- `protocol/messenger.go` creates directly a `Notification` instead of
having an intermediate format
- add community notifications on request to join
- move parsing of text in status-go for notifications
* Migrations in place, how to run them?
* Remove down migrations and touch database.go
* Database and Database Test package in place, added functions to get and store app metrics
* make generate output
* Minor bug fix on app metrics insert and select
* Add a validation layer to restrict what can be saved in the database
* Make validation more terse, throw error if schema doesn't exist, expose appmetrics service
* service updates
* Compute all errors before sending them out
* Trying to bring a closjure to appmetrics go
* Expose appmetrics via an api, skip fancy
* Address value as Jason Dawt Rawmasage to ease parsing
* Introduce a buffered chan with magic cap of 8 to minimize writes to DB. Tests for service and API. Also expose GetAppMetrics function.
* Lint issues
* Remove autoincrement, undo waku.json changes, fix error being shadowed, return nil where nil ought to be returned, get rid of buffered channel
* Bump migration number
* Fix API factory usage
* Add comment re:json.RawMessage instead of strings
* Get rid of test vars, throw save error inside the loop
* Update version
Co-authored-by: Samuel Hawksby-Robinson <samuel@samyoul.com>
This commit expands the confirmation mechanism to allow private group
chat messages to be confirmed:
Changes:
- Added a separate table for message confirmations as group chat
messages have same messageID but multiple datasyncID
- Removed DataSyncID from raw message (I haven't removed the column name
as it can't be done in sqlite without copying over the table)
* Revert "Revert "Expand Local Notifications to support multiple Notification types (#2100)""
This reverts commit 5887337b88.
* Revert "Revert "fix protocol.MessageNotificationBody marshalling""
This reverts commit cf0a16dff1.
* Bump version to 0.70.0
* Added localnotifications for Transaction messages
* Fixed bug where Message.SigPubKey was presumed to be set
* Added lookup for contact existing in Messenger.allContacts
Additionally added functionality to add a contact to the messenger store if it isn't present
* Get chat directly from Messenger.allChats store
Co-authored-by: Andrea Maria Piana <andrea.maria.piana@gmail.com>
It looks like profile chats are created as public chats, and the code
will drop messages for it.
This commit fixes the issue by checking for the "@" prefix in profile
chat names, until we fix the issue by migrating those chats.
We were not locking before accessing the contacts map and it would panic
in some cases.
I have changed the code to pull contacts from db so we move away from
having locks.
There was a bug on status-react where it would save filters that were
not listened to.
This commit adds a task to clean up those filters as they might result
in long syncing times.
This commit also returns topics/ranges/mailserves from messenger in
order to make the initialization of the app simpler and start moving
logic to status-go.
It also removes whisper from vendor.
One of the issues we noticed is that the partitioned topic
in push notification is heavy in traffic, as any user using a particular
mailserver will use that partitioned topic to register for PNs.
This commit moves from the partitioned topic to the personal topic of
the PN server, so it does not clash with other users that might happen
to have the same partitioned topic as the mailserver, resulting in long
sync times.
Another issue that will need to be addressed separately is that once you
send a message to a topic, because of the way how waku/whisper works,
you will have to register to that topic, meaning that you will receive
that data. Currently waku does not support unsubscribing from a topic
without logging in and out, so that needs also to be addressed.
When sending a message in a private group chat we send the whole history
for redundancy and allow out-of-order processing.
This can be very expensive in some chats, resulting in long delay when
sending a message and calculating the POW.
This commit improves the performance by only forwarding the events
necessary for the user to be able to construct a group chat and process
correctly the message.
This commit fixes a couple of issues:
1) Emojis were sent to any member of the group chat, regardless of
whether they joined
2) We don't want to wrap emojis, as there's no need to do so, only
messages are to be wrapped
This commit re-introduces a feature that we lost during the migration to
status-go.
Messages are cached for a couple of days if processed correctly by
status-go, to avoid performance issues.
* Initial work on expanding Local Notifications
Adding functionality to support multiple notification types in Notification.Body. Currently have a bug that I think is caused by a the jsonMarshal func not working as intented, need to resolve this next before proceeding
* Fixed json.Marshaller issue and implemented json.Unmarshaller
* Tweak errors, go convention is errors don't begin with capital letters
* Added notificationMessageBody with un/marshalling
Also removed the Body interface
* Added check for bodyType mismatch
* Implement building and sending new message notifications
* Refactor to remove cycle imports
* Resolved linting issue ... Hopefully
* Resolving an implicit memory aliasing in a for loop
* version bump
* Added Notification.Category consts
Since we fixed re-sending of messages, it has been noticed a performance
degradation in private group chats.
This is due to too aggressive re-sending of messages.
This commit disables resending of private group chat messages for now
(same as 1.9, so no changes), but keeps it enabled for 1-to-1s.
In some instances the retry mechanism would get into a busy loop.
That's due to the fact that we would fetch some non-retriable
notifications but not act on them.
This commit fixes the issue by filtering them from the database query
and making sure that we at least wait 1 second.
In some instances the communities migration would be skipped but not
marked as `dirty`.
This commit addresses the issue by:
- Making sure that if dirty is set the migration is not skipped but
replayed
- If the version is on the communities migration and dirty is false, we
check for the presence of the communities table. If not present we
replay the communities migration.
- Make community_id field in user_messages nullable
It also removes all the `down` migration, as we can't use them
effectively, as explained in the README.md added.
This commit changes the behavior so that when the image is updated it
will be published on the contact code topic.
If that does not happen because we are offline, it will be scheduled for
the next time we are online.
* feat: allow getPreviewLinkData for all links
Only YouTube links were supported in `getPreviewLinkData` previously.
Now, any whitelisted links can have their data retreived using this function, returning the true content type (by examining the header bytes of the content).
feat: add tenor.com and giphy.com to whitelisted urls list
* Fix json format
* bump VERSION
* Lint urls.go
Co-authored-by: Andrea Maria Piana <andrea.maria.piana@gmail.com>
This commit adds an endpoint to batch the sending of messages.
This is useful to simplify client logic when sending a batch of messages
and ensuring the correct order in the message stream.
It currently implements only what's needed, and naively return an error
if any of the messages fail.
if an error is returned on the Send function, datasync will keep
retrying a message at each epoch. If the message cannot be sent (for
example is too large), then no messages will be sent until logout.
This commit fixes a bug on the mvds library where the nextEpoch would be
incorrectly summed to the retry time, resulting in messages not being
retried, or retried much less frequently the longer the app was running.
It also updates the retry timing to backoff exponentially at multiple of
30 seconds.
In some cases no error is returned but the name is not verified.
Before this ENS names would get verified again every 30s.
This commit changes the logic so that if they fail to verify without
error they will still exponentially backoff.
When sending messages in quick succession, it might be that multiple
messages are batched together in datasync, resulting in a single large
payload.
This commit changes the behavior so that we can pass a max-message-size
and we split the message in batches before sending.
A more elegant way would be to split at the transport layer (i.e
waku/whisper), but that would be incompatible with older client.
We can still do that eventually to support larger messages.
We were checking for the wrong error kind when pulling messages from the
database, which resulted in the code not retrying to pull the message,
giving flaky tests / race condition (that's present in production as
well)
The code had an issue where it would not register a chat if just joined
as re-register push notifications was called before the chat had been
added.
This commit fixes the behavior by making sure that the chat just joined
is included.
This commit does two things:
1) Add an index on seen & update only the not-seen messages in the query
2) Hide long messages in the database, as that's likely spam
Currently messenger has no notion of being online.
This might cause a problem as we retry to register with a push
notification server even if not connected to any peer, which will
inevitably fail.
This commit adds a method `handleConnectionChange` that will be called
every time the connection change state.
When sending a contact update we automatically added the contact,
but that resulted in the contact not being synced correctly as
`saveContact` will not trigger the side effects.
For now I have removed this behavior. Ideally we should have a single
call that handles the side effects, but for that ENS names should be
stored in messenger, so we can propagate it.
LastMessage in chat was encoded in bytes so that we don't have to
encoded/decode everytime we save to db or pass the client.
An issue with emoji surfaced a problem with this approach.
Chat.LastClockValue represent the last clock value of any type of
message exchanged in a chat (emoji,group membership updates, contact
updates).
So when receving a new message, we should update LastMessage if the
clock of the LastMessage is lower than the received message, and we
should not only check LastClockValue, otherwise the message might be
discarded although it is the most recent.
This commit fixes the issue by keeping LastMessage as an object and
comparing LastMessage.Clock instead of LastClockValue
If a message was inserted before the migration the field
audio_duration_ms would be set to NULL, and would not be serialized into
go correctly, as uint is non-nullable.
this commit fixes the issue by calling COALESCE on the value.
Why make this change?
We are adding support of audio recorded files, similarly to how we did
with images
What has changed?
- Added protobuf definition, only AAC supported
- Added migrations to store files
- Fixed an issue with nil pointer when transaction would fail to be
created, causing the application to crash
The index for message was fairly inefficient as it was only using the
cursor, as it was referring to the old `chat_id` field.
This meant that newer messages would be fetched much faster then older
messages.
The index has been changed so that now it includes `local_chat_id`
(which is currently used for filtering), and not using `hide`.
The reason being is that `hide` is a low cardinality index, so there's
no performance benefit to have it in, also it's mostly ignored by the
query planner.
This commit also adds the missing migrations, we generated the file, but
the source was missing, probably I forgot to add them in a rebase. They
have been generated from the migration file, using `RestoreAsset`.
'm trying to add more documentation to parts of the repo that I go into.
I feel this basic history will make it easier to understand why the protocol package is as big as it is compared to the other packages in the repo, and help in understanding its existence.
We used *Raw method in message_processor as before we had non-Raw method
during the transition from status-react to status-go. This naming is not
meaningful anymore, so I have changed it.
Images are too large to be sent over waku/whisper with the
current PoW (0.002).
This commit lowers the PoW for sending messages to a lower value,
depending on the size.
This means that older clients will not be able to receive messages that
are greater than 50KB (none of the messages we currently send is anywhere close to
that, most of them are less than 1KB).
I would have preferred to set the PoW explicitly for images to be lower,
but it's not trivial as we use `datasync` to send messages and its
interface takes a payload that is to be dispatched and a function to
dispatch, at initialization stage.
This make it difficult (impossible?) to set a different PoW for a
particular message, without changing function signature in datasync
(which is agnostic to the transport used).
So a less cumbersome approach is to just fingerprint on size.
This commit adds support for images in protobuf messages.
The client can specify a path which will be used to load the image
and set the corresponding fields.
This makes the assumption that the RCP server runs on the same machine
as the client and they have access to the same files. This holds
currently for both status-react and status-console-client, we could
revisit and adds an upload if that changes in the future.
Why make the changes?
Mainly performance, those fields are almost always present in the
database but they are re-calculated on load by the client as it does not
have necessarily access to it.
What has changed?
- Remove `_legacy` persistence namespaces as it's a vestige of the
initial move frmo status-react to status-go
- Pulling chats is now a join with contacts to add contact & alias
* fix: close resultsets so we don't leak them
* Refactor browsers/database
To implement PR suggestions and improve code quality.
* Refactor services/permissions/database
To implement PR suggestions and improve code quality.
Co-authored-by: Samuel Hawksby-Robinson <samuel@samyoul.com>
If the user deletes/leaves a group chat, the chat is set as not active.
This means that if we are re-invited to the chat it won't be shown to
the user.
This commit changes this behavior so that if we are re-invited to the
chat it is set as active again.
When receiving a message with save a contact in the database in order to
avoid re-calculating image/profile.
This contact is then passed to the client, which can negatively impact
performance.
This commit changes the behavior so that only those contacts that have
some custom fields (have been explicitly added by the user, have been
blocked by the user, have sent a contact request or have a verified ens
name) are passed to the client.
Currently replies to messages are handled in status-react.
This causes some issues with the fact that sometimes replies might come
out of order, they might be offloaded to the database etc.
This commit changes the behavior so that status-go always returns the
replies, and in case a reply comes out of order (first the reply, later
the message being replied to), it will include in the messages the
updated message.
It also adds some fields (RTL,Replace,LineCount) to the database which
were not previously saved, resulting in some potential bugs.
The method that we use to pull replies is currently a bit naive, we just
pull all the message again from the database, but has the advantage of
being simple. It will go through performance testing to make sure
performnace are acceptable, if so I think it's reasonable to avoid some
complexity.
Richard Ramos
Roman Volosovskyi
Andrea Maria Piana
Parvesh Monu
andrey
Brian Sztamfater
Samuel Hawksby-Robinson
Michele Balistreri
Anthony Laibe
saledjenic
Jakub Sokołowski
Ismael
Jonathan Rainville
frank
Eric Mastro
Volodymyr Kozieiev
Andrea Franz
Brian Sztamfater
Pascal Precht
Shivek Khurana
flexsurfer
André Medeiros
Adam Babik