The contact filter APIs were timing out due to the case‑insensitive
filter. There is no index for lower case phone numbers, so it would
perform a table scan, potentially examining 8 million records or more at
a time.
This change should fix the issue.
I am changing the filter to use direct comparison without lower‑case.
**Previous:**
```sql
SELECT contacts.*
FROM contacts
WHERE contacts.account_id = $1
AND (
LOWER(contacts.phone_number) = '<number>'
OR LOWER(contacts.phone_number) = '<other-number>'
)
ORDER BY contacts.created_at DESC NULLS LAST
LIMIT $2
OFFSET $3
```
**Updated:**
```sql
SELECT contacts.*
FROM contacts
WHERE contacts.account_id = $1
AND (
contacts.phone_number = '<number>'
OR contacts.phone_number = '<other-number>'
)
ORDER BY contacts.created_at DESC NULLS LAST
LIMIT $2
OFFSET $3
```
Fixes:
https://linear.app/chatwoot/issue/CW-5582/contact-filter-timing-out
When CAPTAIN_OPEN_AI_ENDPOINT is set to empty string, use .presence to
properly fall back to default OpenAI endpoint instead of creating
malformed URLs.
Fixes#12383
Co-authored-by: Pranav <pranav@chatwoot.com>
Currently, auto-assignment runs only during conversation creation or
update events. If no agents are online when new conversations arrive,
those conversations remain unassigned.
With this change, unassigned conversations will be automatically
assigned once agents become available. The job runs every 15 minutes and
uses a fair distribution threshold of 100 to prevent a large number of
conversations from being assigned to a single available agent. This will
be customizable later.
We now support searching within the actual message content, email
subject lines, and audio transcriptions. This enables a faster, more
accurate search experience going forward. Unlike the standard message
search, which is limited to the last 3 months, this search has no time
restrictions.
The search engine also accounts for small variations in queries. Minor
spelling mistakes, such as searching for slck instead of Slack, will
still return the correct results. It also ignores differences in accents
and diacritics, so searching for Deja vu will match content containing
Déjà vu.
We can also refine searches in the future by criteria such as:
- Searching within a specific inbox
- Filtering by sender or recipient
- Limiting to messages sent by an agent
Fixes https://github.com/chatwoot/chatwoot/issues/11656
Fixes https://github.com/chatwoot/chatwoot/issues/10669
Fixes https://github.com/chatwoot/chatwoot/issues/5910
---
Rake tasks to reindex all the messages.
```sh
bundle exec rake search:all
```
Rake task to reindex messages from one account only
```sh
bundle exec rake search:account ACCOUNT_ID=1
```
This PR adds the ability to modify the embedding model used by Captain
AI.Previously, the embedding model was hardcoded which led to errors when
you used a different API provider which did not support that specific
embedding model.
Co-authored-by: Shivam Mishra <scm.mymail@gmail.com>
This PR migrates the legacy OpenAI integration (where users provide
their own API keys) from using hardcoded `https://api.openai.com`
endpoints to use the configurable `CAPTAIN_OPEN_AI_ENDPOINT` from the
captain configuration. This ensures consistency across all OpenAI
integrations in the platform.
## Changes
- Updated `lib/integrations/openai_base_service.rb` to use captain
endpoint config
- Updated `enterprise/app/models/enterprise/concerns/article.rb` to use
captain endpoint config
- Removed unused `enterprise/lib/chat_gpt.rb` class
- Added tests for endpoint configuration behavior
**Summary**
- Fixed Ruby warning about unescaped closing bracket in character class
within `MENTION_REGEX`
- Properly escaped the `]` character in the regex pattern to follow Ruby
regex syntax standards
**Changes**
- Updated `MENTION_REGEX` in `lib/regex_helper.rb` to escape the closing
bracket in character class `[^]]+` → `[^\\]]+`
Fixes notification display issues when user or team names contain emojis
and special characters. Previously, mention notifications would show
URL-encoded characters instead of properly formatted names with emojis.
**Before:**
Notification: "John Doe: Hey @%F0%9F%91%8D%20customer%20support please
check this"
**After:**
Notification: "John Doe: Hey @👍 customer support please check this"
- Add `actor=app` parameter to Linear OAuth authorization URL for
consistent app-level authorization
https://linear.app/developers/oauth-actor-authorization
- Implement user attribution for Linear issue creation and linking using
`createAsUser` and `displayIconUrl` parameters
- Enhance Linear integration to properly attribute actions to specific
Chatwoot agents
**Note**
- The displayIconUrl parameter is being sent correctly to Linear's
GraphQL API (verified through testing), but there is an issues with icon
is not attaching properly.
- We might need to disconnect the integration connect again.
When users delete the Linear integration from their Chatwoot dashboard,
the access token remains valid in Linear's system. This causes the
integration to still appear as connected in Linear's UI, even though
it's been removed from Chatwoot. Users need to manually disconnect from
Linear's side to fully remove the integration.
https://www.loom.com/share/5c102cbdf02e49bcb7a6fa6d409b531a?sid=0c664250-c867-4fc8-b44d-e1c1165337a7
Fixes
https://linear.app/chatwoot/issue/CW-4507/slack-integration-not-showing-private-channels
## Problem
When the Slack workspace has many channels (requiring multiple API
requests to fetch all of them), our system was only looking for private
channels in the first batch of results. All subsequent batches were
missing the instruction to include private channels, so they only
returned public channels.
## Root Cause
- Initial API call correctly specified `types:
'public_channel,private_channel'`
- Pagination loop only passed `cursor` parameter, omitting `types` and
`exclude_archived`
- Subsequent pages defaulted to public channels only
## Changes
- Fixed parameter formatting in `types` (removed space:
`'public_channel, private_channel'` →
`'public_channel,private_channel'`)
- Added missing `types` and `exclude_archived` parameters to paginated
`conversations_list` calls
---------
Co-authored-by: Sojan Jose <sojan@pepalo.com>
## Summary
- support region option when configuring Dialogflow integration
- connect to region endpoint when set
- use session identification based on the region
Fixes: https://github.com/chatwoot/chatwoot/issues/4129
Introduce custom roles in account seed data to support testing of
realistic role-based access scenarios. Includes six distinct roles with
varied permission sets and assigns them to multiple users for
comprehensive test coverage.
Description
In Rails 7.1+, accessing `schema_format` via `ActiveRecord::Base` is no
longer supported, causing a method error.
The correct approach is to use
`ActiveRecord.schema_format`, which aligns
with the public API.
This issue occurs locally because the environment uses newer versions of
Ruby and Rails than others.
Fixes: https://github.com/chatwoot/chatwoot/issues/11594
Co-authored-by: Sojan Jose <sojan@pepalo.com>
- Add API support for creating a thread
- Add API support for creating a message
- Remove uuid from thread (no longer required, we will use existing
websocket connection to send messages)
- Update message_type to a column (user, assistant, assistant_thinking)
* feat: add MESSAGES_UNREAD event and implement unread action in ConversationsController
* feat: rename MESSAGES_UNREAD event to CONVERSATION_UNREAD and update unread action in ConversationsController
* feat: implement send_unread_conversation method and update conversation_unread event handling
* feat: add test for dispatching conversation.unread event in ConversationsController
* feat: add conversation_unread event handling in ChannelListener
* feat: add send_unread_conversation method and corresponding tests in Whatsapp channel
* feat: add logging for nil last_message in send_unread_conversation method
* feat: add send_unread_conversation to error handling methods
* feat: implement send_unread_conversation method with logging for nil last_message
* fix: add nil check and logging for last_message in send_unread_conversation method
* fix: update conversation_unread method to call unread_conversation instead of send_unread_conversation
* refactor: rename send_unread_conversation to unread_conversation and update method implementation
* refactor: rename send_unread_conversation to send_unread_messages and update implementation to handle multiple messages
* refactor: rename send_read_messages to read_messages and send_unread_messages to unread_message for clarity
* refactor: rename send_read_messages to read_messages and send_unread_messages to unread_message for clarity
* refactor: rename send_read_messages to read_messages for consistency
# Pull Request Template
## Description
Updates the public swagger spec to be OpenAPI 3.0.4 compliant. Regarding
#7893, I'm investigating the use of
[oas_rails](https://github.com/a-chacon/oas_rails) to auto-generate the
documentation along with correct expected payloads.
Mostly fixes#10531, related to #7893
## Type of change
- [ ] Bug fix (non-breaking change which fixes an issue)
- [x] New feature (non-breaking change which adds functionality)
- [ ] Breaking change (fix or feature that would cause existing
functionality not to work as expected)
- [ ] This change requires a documentation update
## How Has This Been Tested?
Copied spec into swagger editor, reviewed outputs:
Note that this shows two errors with the `DELETE` endpoints as
technically these should not have a `requestBody` - in which case we
should be making use of another HTTP verb or another endpoint.
## Checklist:
- [x] My code follows the style guidelines of this project
- [x] I have performed a self-review of my code
- [x] I have commented on my code, particularly in hard-to-understand
areas
- [x] I have made corresponding changes to the documentation
- [x] My changes generate no new warnings
- [x] I have added tests that prove my fix is effective or that my
feature works
- [x] New and existing unit tests pass locally with my changes
- [x] Any dependent changes have been merged and published in downstream
modules
---------
Co-authored-by: Daniel Jimenez <devildan.knex@gmail.com>
Co-authored-by: Daniel Jimenez <daniel.jimenez@spark64.com>
Show captain messages under the name of the assistant which generated
the message.
- Add support for `Captain::Assistant` sender type
- Add push_event_data for captain_assistants
- Add activity message handler for captain_assistants
- Update UI to show captain messages under the name of the assistant
- Fix the issue where openAI errors when image is sent
- Add support for custom name of the assistant
---------
Co-authored-by: Muhsin Keloth <muhsinkeramam@gmail.com>
Co-authored-by: Sivin Varghese <64252451+iamsivin@users.noreply.github.com>
* feat: implement send_read_messages method for WhatsApp channel
* feat: implement messages_read event handling and dispatch for conversations
* feat: enhance messages_read handling to include last_seen_at and conversation context
* feat: update last_seen handling to reference agent_last_seen_at for messages read event
* chore: fix rebase
* feat: update error handling
* feat: update send_read_messages to mark received messages as not sent by the user
* test: controller spec
* test: channel listener
* test: channel and provider
---------
Co-authored-by: gabrieljablonski <contact@gabrieljablonski.com>
- Added an api endpoint for update message status ( available only for
api inboxes )
- Moved message status management to a service.
- Handles case where read status arrive before delivered
fixes: #10314 , #9962
- Refactor HandleStripeEventService to better manage features by plan
- Add constants for features available in each plan tier (Startup,
Business, Enterprise)
- Add channel_instagram to Startup plan features
- Improve downgrade handling to properly disable higher-tier features
- Clean up and optimize tests for maintainability
- Add comprehensive test coverage for plan upgrades and downgrades
---------
Co-authored-by: Muhsin Keloth <muhsinkeramam@gmail.com>
