This PR updates the report pages for agents, inboxes, and teams by
replacing charts with aggregate values (under a feature flag). Users can
click on any item to view more details if needed. Most users seem to
prefer aggregate values, so this change will likely stay.
The PR also includes a few fixes:
- The summary reports now use the same logic for both the front-end and
CSV exports.
- Fixed an issue where a single quote was being added to values with
hyphens in CSV files. Now, ‘n/a’ is used when no value is available.
- Fixed a bug where the average value was calculated incorrectly when
multiple accounts were present.
These changes should make reports easier to use and more consistent.
### Agents:
<img width="1438" alt="Screenshot 2025-01-26 at 10 47 18 AM"
src="https://github.com/user-attachments/assets/bf2fcebc-6207-4701-9703-5c2110b7b8a0"
/>
### Inboxes
<img width="1438" alt="Screenshot 2025-01-26 at 10 47 10 AM"
src="https://github.com/user-attachments/assets/b83e1cf2-fd14-4e8e-8dcd-9033404a9f22"
/>
### Teams:
<img width="1436" alt="Screenshot 2025-01-26 at 10 47 01 AM"
src="https://github.com/user-attachments/assets/96b1ce07-f557-42ca-8143-546a111d6458"
/>
---------
Co-authored-by: Sivin Varghese <64252451+iamsivin@users.noreply.github.com>
Co-authored-by: Shivam Mishra <scm.mymail@gmail.com>
This PR introduces several improvements to the Captain AI dashboard
section:
- New billing page, with new colors, layout and meters for Captain usage
- Updated the base paywall component to use new colors
- Updated PageLayout.vue, it's more generic and can be used for other
pages as well
- Use flags to toggle empty state and loading state
- Add prop for `featureFlag` to show the paywall slot based on feature
enabled on account
- Update `useAccount` to add a `isCloudFeatureEnabled`
- **Removed feature flag checks from captain route definitions**, so the
captain entry will always be visible on the sidebar
- Add banner to Captain pages for the following cases
- Responses usage is over 80%
- Documents limit is fully exhausted
### Screenshots
<details><summary>Free plan</summary>
<p>
</p>
</details>
<details><summary>Paid plan</summary>
<p>
</p>
</details>
---------
Co-authored-by: Sojan Jose <sojan@pepalo.com>
Co-authored-by: Pranav <pranav@chatwoot.com>
Co-authored-by: Muhsin Keloth <muhsinkeramam@gmail.com>
# Pull Request Template
## Description
Fixes
https://linear.app/chatwoot/issue/CW-3853/vite-dev-build-fails-due-to-dart-sass
**Other issue**
| First Header | Second Header |
| ------------- | ------------- |
|
|
|
## Type of change
- [x] Bug fix (non-breaking change which fixes an issue)
## Checklist:
- [x] My code follows the style guidelines of this project
- [x] I have performed a self-review of my code
- [ ] I have commented on my code, particularly in hard-to-understand
areas
- [ ] I have made corresponding changes to the documentation
- [x] My changes generate no new warnings
- [ ] 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
- [ ] Any dependent changes have been merged and published in downstream
modules
# Pull Request Template
## Description
This PR fixes an issue where deleting the last item on the last page of
responses/documents, would show an empty state instead of loading the
previous page.
Fixes
> If you have pending responses spanning 2 or more pages .. and you
delete the last response in the last page.. instead of showing the
previous page the system show empty state.
## Type of change
- [x] Bug fix (non-breaking change which fixes an issue)
## How Has This Been Tested?
**Loom video**
https://www.loom.com/share/b0e89f774ccd45dab0e8dba2c34bd1ac?sid=d9923bcd-5030-42d9-9b7f-170df5297cfd
## Checklist:
- [x] My code follows the style guidelines of this project
- [x] I have performed a self-review of my code
- [ ] I have commented on my code, particularly in hard-to-understand
areas
- [ ] I have made corresponding changes to the documentation
- [x] My changes generate no new warnings
- [ ] 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
- [ ] Any dependent changes have been merged and published in downstream
modules
Currently, it’s unclear whether an FAQ item is generated from a
document, derived from a conversation, or added manually.
This PR resolves the issue by providing visibility into the source of
each FAQ. Users can now see whether an FAQ was generated or manually
added and, if applicable, by whom.
- Move the document_id to a polymorphic relation (documentable).
- Updated the APIs to accommodate the change.
- Update the service to add corresponding references.
- Updated the specs.
<img width="1007" alt="Screenshot 2025-01-15 at 11 27 56 PM"
src="https://github.com/user-attachments/assets/7d58f798-19c0-4407-b3e2-748a919d14af"
/>
---------
Co-authored-by: Sivin Varghese <64252451+iamsivin@users.noreply.github.com>
This PR introduces support for an assistant filter on the documents page.
- Moved the existing assistant filter functionality to a standalone, reusable component.
- Updated the documents page and responses page to use the component
This PR introduces a review step for generated FAQs, allowing a human to
validate and approve them before use in customer interactions. While
hallucinations are minimal, this step ensures accurate and reliable FAQs
for Captain to use during LLM calls when responding to customers.
- Added a status field for the FAQ
- Allow the filter on the UI.
<img width="1072" alt="Screenshot 2025-01-15 at 6 39 26 PM"
src="https://github.com/user-attachments/assets/81dfc038-31e9-40e6-8a09-586ebc4e8384"
/>
Previously, the Reports API fetched data based on event triggers. For
example, when an event occurred on an account, the system would
automatically retrieve and display updated information. However, this
approach was designed under the assumption that reports would be
accessed by a small number of users and on an infrequent basis (e.g.,
once daily or weekly).
In scenarios where large customers have multiple team members actively
monitoring reports, this event-driven approach led to an excessive
number of requests, significantly straining the system.
This PR introduces a interval-based fetching of reports instead of the
event-driven model.
Migration Guide: https://chwt.app/v4/migration
This PR imports all the work related to Captain into the EE codebase. Captain represents the AI-based features in Chatwoot and includes the following key components:
- Assistant: An assistant has a persona, the product it would be trained on. At the moment, the data at which it is trained is from websites. Future integrations on Notion documents, PDF etc. This PR enables connecting an assistant to an inbox. The assistant would run the conversation every time before transferring it to an agent.
- Copilot for Agents: When an agent is supporting a customer, we will be able to offer additional help to lookup some data or fetch information from integrations etc via copilot.
- Conversation FAQ generator: When a conversation is resolved, the Captain integration would identify questions which were not in the knowledge base.
- CRM memory: Learns from the conversations and identifies important information about the contact.
---------
Co-authored-by: Vishnu Narayanan <vishnu@chatwoot.com>
Co-authored-by: Sojan <sojan@pepalo.com>
Co-authored-by: iamsivin <iamsivin@gmail.com>
Co-authored-by: Sivin Varghese <64252451+iamsivin@users.noreply.github.com>
# Pull Request Template
## Description
Changed to make use of nullish coalescing operator to only short circuit
in cases when the `metric` variable is not zero. This change also begs
the question as to whether the `stringToFloat` test function should
exist - to me it seems interesting to have test code embedded into
production code for the frontend?
Fixes#10640
## Type of change
Please delete options that are not relevant.
- [x] Bug fix (non-breaking change which fixes an issue)
- [ ] 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?
Tested by opening the report overview page and reviewing that this shows
`---` for agents with no conversations assigned to them.
## 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
- [ ] I have added tests that prove my fix is effective or that my
feature works
- [ ] 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: Shivam Mishra <scm.mymail@gmail.com>
- Increase the sidebar and chatlist width on screens bigger than 2xl
- Fix background of the copilot sidebar
- Increase the breakpoint for hiding chatlist from 1024 to 1200
Added the possibility to mark as a single conversation in the API type
inbox. This allows the conversation builder to search for the last
conversation.
I thought about searching for the last conversation with created_at:
desc order, as is done in some channels... but I didn't change the way
the conversation is searched.
Fixes: #7726
Co-authored-by: Sojan Jose <sojan@pepalo.com>
This PR resolves the issue with updating the profile picture in the profile settings.
**Cause of issue**
The issue can be reproduced with the old `ProfileAvatar.vue` component.
While the exact reason is unclear, it seems related to cases where the
file might be `null`.
**Solution**
Replaced the old `ProfileAvatar.vue` with `Avatar.vue` and tested it. It
works fine. I’ve attached a loom video below.
Fixes https://linear.app/chatwoot/issue/CW-3768/profile-picture-bug
Co-authored-by: Pranav <pranav@chatwoot.com>
Co-authored-by: Pranav <pranavrajs@gmail.com>
### The problem
Writing in the text editor can be very frustrating, the reason is that
the editor had a debounced save method which would push the article to
the backend and update the current state. This however is a bad idea,
since the can take anywhere between 100-300ms depending on network
conditions.
While this would be in progress, the article is still being edited by
the user. So at the end of the network request, the state returned from
the backend and the current state in the editor is diverged. But since
the update happens anyway, the editor would prepend older context.
```
Time -->
User Action: [Edit 1] ---> [Edit 2] ---> [Edit 3]
Backend Save: Save Req (Edit 1) ----> Response (Edit 1)
Resulting Editor State: [Edit 3] + [Edit 1] (Outdated state prepended)
```
### The solution
The solution is to unbind the article from the backend state, ensuring
that the article editor is the source of truth and ignoring the
responses. This pull request does this by adding an asynchronous save
functionality. The changes include adding a new `saveArticleAsync` event
and ensuring that the local state is not updated unnecessarily during
asynchronous saves.
```
Time -->
User Action: [Edit 1] ---> [Edit 2] ---> [Edit 3]
Backend Save: Save Req (Edit 1) ----> Response (ignored)
Resulting Editor State: [Edit 3] (Consistent and up-to-date)
```
Added the following two debounced methods
These complementary debounce methods prevent unnecessary re-renders
while ensuring backend is in sync. `saveArticleAsync` preserves the
editor as the source of truth, while `saveArticle` manages periodic
state updates from the backend with a delay large enough to safely
assume that the user has stopped typing
Method | Delay | Behavior
-- | -- | --
`saveArticleAsync` | 400ms | Sends data to backend and ignores the
response
`saveArticle` | 2.5s | Sends data and updates local state with the
backend response
### How to test
1. Remove the following line
dc042f6ddc/app/javascript/dashboard/components-next/HelpCenter/Pages/ArticleEditorPage/ArticleEditor.vue (L64)
1. Update the latency here to 400 (P.S. the diff shows the latency to be
600, but that was added as a stop-gap solution)
dc042f6ddc/app/javascript/dashboard/components-next/HelpCenter/Pages/ArticleEditorPage/ArticleEditor.vue (L51)
1. Set the browser network latency to Slow 3G or 3G
1. Start writing on the editor, try fixing typos with backspace or
moving around with the cursor
---------
Co-authored-by: Sivin Varghese <64252451+iamsivin@users.noreply.github.com>
Co-authored-by: Pranav <pranav@chatwoot.com>
Ensures users are seamlessly directed to the first available menu item upon opening a group, improving UX by reducing unnecessary clicks. This change enhances navigation flow within groups.
Co-authored-by: Pranav <pranavrajs@gmail.com>
This PR has the initial version of the new sidebar targeted for the next major redesign of the app. This PR includes the following changes
- Components in the `layouts-next` and `base-next` directories in `dashboard/components`
- Two generic components `Avatar` and `Icon`
- `SidebarGroup` component to manage expandable sidebar groups with nested navigation items. This includes handling active states, transitions, and permissions.
- `SidebarGroupHeader` component to display the header of each navigation group with optional icons and active state indication.
- `SidebarGroupLeaf` component for individual navigation items within a group, supporting icons and active state.
- `SidebarGroupSeparator` component to visually separate nested navigation items. (They look a lot like header)
- `SidebarGroupEmptyLeaf` component to render empty state of any navigation groups.
----
Co-authored-by: Pranav <pranav@chatwoot.com>
Co-authored-by: Pranav <pranavrajs@gmail.com>
When moving form using Gmail Legacy auth to using OAuth, we need the
email address that will be used to connect. This is because we need to
store this email address in the cache and reuse when we get the callback
to find the associated inbox.
However there are cases where the imap login might be
`support@company.com` and the email used to communicate will be
`contact@company.com` (Probably an alias) In that case, we need to send
the correct email address to Chatwoot when re-authenticating
At the moment, we used the inbox email. This PR adds a check that
defaults to to `imap_login` if that is available and imap is enabled
This PR also fixes an unrelated problem where the email inbox creation
flow was not working
---
Tested it, it is working correctly
On production on multiple instances it may happen that the UI is
rendered in correctly, with a lot of options in the sidebar not
available. On further investigation I found out that the feature flag
checks were disabling multiple of those, and also we could see many
correlated errors that pointed towards missing information.
So, there were two problems here
1. The `vuex-router-sync` was not very reliable in some cases
2. In `App.vue` the watch on `currentAccountId` didn't always trigger.
## Fix Tested on Staging
Basically tried to reload the page ~50 times with cache enabled,
disabled, throttling, navigating different pages.
https://www.loom.com/share/1bb27294aa364ac4acfb647780d6385a?sid=87e31330-8cb7-4ded-8616-5e95e2ae3516
<details><summary>
#### What I thought was the fix
</summary>
<p>
### My chain of actions
Replacing vuex-router-sync at first worked fine, but then I saw it was
still failing in some cases, I assumed (I was half-correct tho) that the
rendering of the `App.vue` and syncing of the route to the store was not
happening in a synchronous pattern. So I decided, let's not rely on the
store when the route is directly available in the App context.
Following this, I refactored `useAccount` composable to use `useRoute`
directly, instead of the store, and then replaced the getter inside
`App.vue`. What this did was surface the issue but more consistently 🤯
I saw the watcher, added some console logs, and turns out it was not
getting triggered in all those cases. So I added an `immediate` to it.
And viola, it works!
At the moment, this is deployed to staging and seems to be working
correctly. But we still need to verify it for sure, since how this issue
was surfaced is still a mystery. All we know is that it shows up when
the widget is also loaded alongside the app (if it loads before or after
the app, it works fine)
### What about the route in the store?
Well I have used the `route` usage there with fallback to the store
state. Since Vuex exists in the app context, the route should always be
available to it. But after today I have lost all trust in JavaScript and
will worship rails until end of my life, so I added that in a
`try-catch` block, logged the error to Sentry
</p>
</details>
## Here's the real fix
If you read the explanation I wrote earlier, I thought I fixed the
issue, but then the chat list navigation completely broke. So I removed
the custom route sync implementation and added the original package
back. Turns out the vuex-router-sync earlier was placed after the app
was initalized, however for it to work, the vue app context is not
required. And it's best to run it before the app is even bootstrapped,
so I added it back and placed it correctly.
So the following changes fixes this problem
1. Hoisting the `sync` function call to before we call `createApp` this
ensures that the stores and route hooks are in place before even the app
is created
2. Ensuring the `initializeAccount` is run immediately when watching
`currentAccountId`
4. Source `currentAccountId` for critical top of the tree components
directly from the route instead of the store
There were two warnings showing up on new teams page
1. `errorMessage` prop was getting invalid value, this was because were
short circuting the error message using `&&`, fixed it by using ternary
operator
2. `vue-i18n` has deprecated [rails
syntax](https://vue-i18n.intlify.dev/guide/essentials/syntax#rails-i18n-format),
fixed that warning by removing `%` before `{}` for interpolation
> Note: the `vue-i18n` deprecation needs to be handled, but we can do it
later
