## Description
This PR adds generic OIDC as an authentication provider. This allows
personal users and companies potentially to define whatever IdP they
want as long as it supports the OIDC well known format. (Azure, Zitadel,
Authentik, KeyCloak, Google, etc. all support it)
## Related Issue
Fixes#1090
## Changes Made
- Adds OIDC buttons to the signin and registration pages
- Adds appropriate environment variables
- Adds migration to add OIDC to the `IdentityProvider` Enum
## Testing Performed
#### Zitadel
- Created application in Zitadel as an web app, with Client auth
- Enabled `User Info inside ID Token` in Token settings
- Copied client id and client secret to the new .ENV variables
- Copied the well-known URL from the URLs section to .ENV
- Created new account with OIDC provider button
- Verified email manually
- Signed into account with OIDC provider
- Logged out
- Signed into accounting again with OIDC provider
#### Authentik
- Created application in Authentik
- Copied client id and client secret to the new .ENV variables
- Copied the well-known URL from the URLs section to .ENV
- Created new account with OIDC provider button
- Verified email manually
- Signed into account with OIDC provider
- Logged out
- Signed into accounting again with OIDC provider
#### Azure AD
- Created application in Azure AD using OAuth2
- Copied client id and client secret to the new .ENV variables
- Copied the well-known URL from the URLs section to .ENV
- Created new account with OIDC provider button
- Verified email manually
- Signed into account with OIDC provider
- Logged out
- Signed into accounting again with OIDC provider
Adds fields to the Account model to support various pieces
of data returned by OIDC providers such as AzureAD and GitLab.
Additionally passes through the email verification status and handles
retrieving the email for providers such as AzureAD who use a different
claim instead.
Reverts the change to find-documents to use Kysely. I'd like to gain
confidence by using it in smaller pieces before commiting to doing
what is one of our most complicated queries in Documenso.
Introduces the ability to not send an email when sending
(publishing) a document using the API.
Additionally returns the signing link for each recipient
when working with recipient API endpoints and returns
the document object including recipients when sending
documents via API.
A series of changes to improve sealing robustness avoiding errors that
have appeared during monitoring.
Additionally handles the recently published CVE affecting the
`pdfjs-dist` library.
Bumps ReactPDF and pdfjs-dist to avoid the CVE that allows
for code execution in pdf's. This change doesn't specifically
upgrade to the latest pdfjs-dist due to issues with top level
await, instead disabling the evaluation of javascript within
the PDF.
Use Noto Sans to gracefully handle inserting custom text
on PDF's. Previously we were using Helvetica which is a
standard PDF font but that would fail for any character
that couldn't be encoded in WinANSI.
Noto Sans was chosen as it has support for a large number
of languages and glyphs with challenges now being adding
support for CJK glyphs.
Previously we used the form flattening method from PDF-Lib
but unfortunately when it encountered orphaned form items
or other PDF oddities it would throw an error.
Because of this certain documents would fail to seal and
be stuck in a pending state with no recourse available.
This change rewrites the form flattening handler to be
more lenient when coming across the unknown opting to skip
items it can't handle rather than abort.
## Description
Currently users who sign in via Google SSO do not get assigned a Stripe
customer account.
This enforces the Stripe customer requirement on sign in.
There might be a better place to put this so it's open to any
suggestions.
## Description
Improves the sealing process by being strict on how long certificate
generation can take, opting to fail generation and continue sealing.
Also changes the ordering of sealing so an error in the process won't
also cause a document to be "COMPLETED" since it hasn't been
cryptographically sealed yet.
The downside to this change is that documents that fail during sealing
will require manual intervention as a signer or owner won't be able to
*complete* the document.
## Testing Performed
- Modified code to force specific failure modes to occur and verified
that documents were either gracefully sealed without a certificate or
not sealed and not completed.
Improves the sealing process by being strict on how
long certificate generation can take, opting to fail
generation and continue sealing.
Also changes the ordering of sealing so an error in the
process won't also cause a document to be "COMPLETED"
since it hasn't been cryptographically sealed yet.
The downside to this change is that documents that fail
during sealing will require manual intervention as a signer
or owner won't be able to *complete* the document.
## Description
Automatically marks the document as completed if all the recipients are
CC.
## Changes Made
Added an if statement in the last form step (`onAddSubjectFormSubmit`)
that checks if all the recipients are CC. If so, the document status is
updated to `COMPLETED`.
## Testing Performed
Tested the changes and they work as expected.
## Checklist
<!--- Please check the boxes that apply to this pull request. -->
<!--- You can add or remove items as needed. -->
- [x] I have tested these changes locally and they work as expected.
- [ ] I have added/updated tests that prove the effectiveness of these
changes.
- [ ] I have updated the documentation to reflect these changes, if
applicable.
- [x] I have followed the project's coding style guidelines.
- [ ] I have addressed the code review feedback from the previous
submission, if applicable.
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
- **New Features**
- Enhanced document sending logic to update document status based on
recipient roles.
- **Bug Fixes**
- Removed redundant form submission handling in the document editing
feature.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
## Description
Upgrade playwright once again and use CDP for remote connections to
avoid version lock-in with `playwright.connect`
Resolves the issue where all CI is failing currently due to downgrading
playwright.
## Description
Currently if you create a team webhook, you can see it in your personal
webhooks.
However, interacting with them will throw errors because there is server
side logic preventing any interaction with them since they are outside
of the correct context.
So the solution is to either:
- Allow the user to edit the team webhooks that they created on their
personal webhooks page
- Hide team webhooks for personal webhooks pages
This PR goes with the second option, but is open to suggestions.
## Description
Added the ability for recipients to see fields from other recipients who
have completed the document when they are signing the document
Added the ability for the document owner to see fields from recipients
who have completed the field on the document page view (only visible
when the document is pending)
## 🚨🚨 Migrations🚨🚨
- Drop all `Fields` that do not have a `Recipient` set (not sure how it
was possible in the first place)
- Remove optional `Recipient` field on `Field` which doesn't make sense
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
## Summary by CodeRabbit
- **New Features**
- Enhanced document viewing by adding read-only fields based on document
status.
- Improved signing page by fetching and displaying completed fields for
tokens.
- Updated avatar component to show recipient status with tooltips for
better user interaction.
- **Bug Fixes**
- Made `recipientId` a required field in the database to ensure data
consistency.
- **Refactor**
- Optimized popover functionality in UI components for better
performance and user experience.
- **Documentation**
- Added detailed component and function descriptions for new features in
the system.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
## Description
Disable the document certificate download button when the document is
not complete
## Changes Made
- Disable UI button
- Disable TRPC API endpoint
## Testing Performed
Tested locally for pending, draft and completed documents
## Description
Adds the ability to prefill native PDF form fields via the API using
either normal documents or templates.
Since we won't always know when a document is uploaded and has forms we
opt to do this on creation for templates and on sending the document to
recipients in all cases. This means that for a created document it can
look a little funky since the form fields are missing the data until the
document is sent.
This should be improved in a later change but since we've scoped this to
an API only workflow for now we are less concerned with the visual
issues.
## Related Issue
N/A
## Changes Made
- Added the `formValues` field the document model
- Added a new method for finding and filling form fields based on a `key
| value` pair
- Updated the API input shapes to take the new field.
## Testing Performed
- Have created and tested a document using the API both for creation and
usage with a template.
- Have verified that the fields display as expected either during
creation or sending depending on the document type.
Added a dialog button at the bottom of the admin/documents/[id] page
with confirmation popup.
Confirmation popup have validation for reason to input.
On confirmation document is deleted, and an email is triggred to the
owner of document with the reason stated.
Let me know if there is any more requirement or correction is needed in
this pr. :) #1020
This change flattens and normalizes annotation and widget layers within
the PDF document removing items that can be accidentally modified after
signing which would void the signature attached to the document.
Initially this change was just to assign to an ArcoForm object in the
document catalog if it existed but quickly turned into the above.
When annotations aren't flattened Adobe PDF will say that the signature
needs to be validated and upon doing so will become invalid due to the
annotation layers being touched.
To resolve this I set out to flatten and remove the annotations by
pulling out their normal appearances if they are present, converting
them into xobjects and then drawing those using the drawObject operator.
This resolves a critical issue the users experienced during the signing
flow when they had marked up a document using annotations in pdf
editors.
## Details
Currently there's no way to have a proper PR template selector. Since we
rarely use the E2E template, I've opted to move the generic PR template
to be the default template.
Right now, the eslint command runs separately for each staged file. This
PR aims to change that by running just one eslint command for all the
files that have been changed.
## Description
Updates the `on-create.sh` script to use our `dx` command resolving
issues where it couldn't find the docker compose file due to changes
that have happened since publishing the Documenso docker image.
## Related Issue
Resolves#1026
## Test Details
N/A
## Checklist
- [x] I have written the new test and ensured it works as intended.
- [x] I have added necessary documentation to explain the purpose of the
test.
- [x] I have followed the project's testing guidelines and coding style.
- [x] I have addressed any review feedback from previous submissions, if
applicable.
## Additional Notes
N/A
Lucas Smith