Merge branch 'main' of https://github.com/danswer-ai/danswer into bug…

…fix/celery_light_backoff

# Conflicts:
#	backend/danswer/background/celery/tasks/connector_deletion/tasks.py
#	backend/danswer/background/celery/tasks/indexing/tasks.py
#	backend/danswer/background/celery/tasks/pruning/tasks.py
#	backend/danswer/background/celery/tasks/shared/tasks.py

.github/pull_request_template.md

@@ -6,20 +6,24 @@
 [Describe the tests you ran to verify your changes]
 
 
-## Accepted Risk
-[Any know risks or failure modes to point out to reviewers]
+## Accepted Risk (provide if relevant)
+N/A
 
 
-## Related Issue(s)
-[If applicable, link to the issue(s) this PR addresses]
+## Related Issue(s) (provide if relevant)
+N/A
 
 
-## Checklist:
-- [ ] All of the automated tests pass
-- [ ] All PR comments are addressed and marked resolved
-- [ ] If there are migrations, they have been rebased to latest main
-- [ ] If there are new dependencies, they are added to the requirements
-- [ ] If there are new environment variables, they are added to all of the deployment methods
-- [ ] If there are new APIs that don't require auth, they are added to PUBLIC_ENDPOINT_SPECS
-- [ ] Docker images build and basic functionalities work
-- [ ] Author has done a final read through of the PR right before merge
+## Mental Checklist:
+- All of the automated tests pass
+- All PR comments are addressed and marked resolved
+- If there are migrations, they have been rebased to latest main
+- If there are new dependencies, they are added to the requirements
+- If there are new environment variables, they are added to all of the deployment methods
+- If there are new APIs that don't require auth, they are added to PUBLIC_ENDPOINT_SPECS
+- Docker images build and basic functionalities work
+- Author has done a final read through of the PR right before merge
+
+## Backporting (check the box to trigger backport action)
+Note: You have to check that the action passes, otherwise resolve the conflicts manually and tag the patches.
+- [ ] This PR should be backported (make sure to check that the backport attempt succeeds)

.github/workflows/nightly-close-stale-issues.yml

@@ -1,8 +1,13 @@
-name: 'Close stale issues and PRs'
+name: 'Nightly - Close stale issues and PRs'
 on:
   schedule:
     - cron: '0 11 * * *' # Runs every day at 3 AM PST / 4 AM PDT / 11 AM UTC
 
+permissions:
+  # contents: write # only for delete-branch option
+  issues: write
+  pull-requests: write
+
 jobs:
   stale:
     runs-on: ubuntu-latest

.github/workflows/pr-backport-autotrigger.yml

@@ -0,0 +1,92 @@
+name: Backport on Merge
+
+on:
+  pull_request:
+    types: [closed]
+
+jobs:
+  backport:
+    if: github.event.pull_request.merged == true
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v3
+        with:
+          fetch-depth: 0  # Fetch all history for all branches and tags
+
+      - name: Set up Git
+        run: |
+          git config user.name "github-actions[bot]"
+          git config user.email "github-actions[bot]@users.noreply.github.com"
+
+      - name: Check for Backport Checkbox
+        id: checkbox-check
+        run: |
+          PR_BODY="${{ github.event.pull_request.body }}"
+          if [[ "$PR_BODY" == *"[x] This PR should be backported"* ]]; then
+            echo "backport=true" >> $GITHUB_OUTPUT
+          else
+            echo "backport=false" >> $GITHUB_OUTPUT
+          fi
+
+      - name: List and sort release branches
+        id: list-branches
+        run: |
+          git fetch --all --tags
+          BRANCHES=$(git for-each-ref --format='%(refname:short)' refs/remotes/origin/release/* | sed 's|origin/release/||' | sort -Vr)
+          BETA=$(echo "$BRANCHES" | head -n 1)
+          STABLE=$(echo "$BRANCHES" | head -n 2 | tail -n 1)
+          echo "beta=$BETA" >> $GITHUB_OUTPUT
+          echo "stable=$STABLE" >> $GITHUB_OUTPUT
+          # Fetch latest tags for beta and stable
+          LATEST_BETA_TAG=$(git tag -l "v*.*.0-beta.*" | sort -Vr | head -n 1)
+          LATEST_STABLE_TAG=$(git tag -l "v*.*.*" | grep -v -- "-beta" | sort -Vr | head -n 1)
+          # Increment latest beta tag
+          NEW_BETA_TAG=$(echo $LATEST_BETA_TAG | awk -F '[.-]' '{print $1 "." $2 ".0-beta." ($NF+1)}')
+          # Increment latest stable tag
+          NEW_STABLE_TAG=$(echo $LATEST_STABLE_TAG | awk -F '.' '{print $1 "." $2 "." ($3+1)}')
+          echo "latest_beta_tag=$LATEST_BETA_TAG" >> $GITHUB_OUTPUT
+          echo "latest_stable_tag=$LATEST_STABLE_TAG" >> $GITHUB_OUTPUT
+          echo "new_beta_tag=$NEW_BETA_TAG" >> $GITHUB_OUTPUT
+          echo "new_stable_tag=$NEW_STABLE_TAG" >> $GITHUB_OUTPUT
+
+      - name: Echo branch and tag information
+        run: |
+          echo "Beta branch: ${{ steps.list-branches.outputs.beta }}"
+          echo "Stable branch: ${{ steps.list-branches.outputs.stable }}"
+          echo "Latest beta tag: ${{ steps.list-branches.outputs.latest_beta_tag }}"
+          echo "Latest stable tag: ${{ steps.list-branches.outputs.latest_stable_tag }}"
+          echo "New beta tag: ${{ steps.list-branches.outputs.new_beta_tag }}"
+          echo "New stable tag: ${{ steps.list-branches.outputs.new_stable_tag }}"
+
+      - name: Trigger Backport
+        if: steps.checkbox-check.outputs.backport == 'true'
+        run: |
+          set -e
+          echo "Backporting to beta ${{ steps.list-branches.outputs.beta }} and stable ${{ steps.list-branches.outputs.stable }}"
+          # Fetch all history for all branches and tags
+          git fetch --prune --unshallow
+          # Checkout the beta branch
+          git checkout ${{ steps.list-branches.outputs.beta }}
+          # Cherry-pick the merge commit from the merged PR
+          git cherry-pick -m 1 ${{ github.event.pull_request.merge_commit_sha }} || {
+            echo "Cherry-pick to beta failed due to conflicts."
+            exit 1
+          }
+          # Create new beta tag
+          git tag ${{ steps.list-branches.outputs.new_beta_tag }}
+          # Push the changes and tag to the beta branch
+          git push origin ${{ steps.list-branches.outputs.beta }}
+          git push origin ${{ steps.list-branches.outputs.new_beta_tag }}
+          # Checkout the stable branch
+          git checkout ${{ steps.list-branches.outputs.stable }}
+          # Cherry-pick the merge commit from the merged PR
+          git cherry-pick -m 1 ${{ github.event.pull_request.merge_commit_sha }} || {
+            echo "Cherry-pick to stable failed due to conflicts."
+            exit 1
+          }
+          # Create new stable tag
+          git tag ${{ steps.list-branches.outputs.new_stable_tag }}
+          # Push the changes and tag to the stable branch
+          git push origin ${{ steps.list-branches.outputs.stable }}
+          git push origin ${{ steps.list-branches.outputs.new_stable_tag }}

README.md

@@ -68,7 +68,7 @@ We also have built-in support for deployment on Kubernetes. Files for that can b
 
 ## 🚧 Roadmap
 * Chat/Prompt sharing with specific teammates and user groups.
-* Multi-Model model support, chat with images, video etc.
+* Multimodal model support, chat with images, video etc.
 * Choosing between LLMs and parameters during chat session.
 * Tool calling and agent configurations options.
 * Organizational understanding and ability to locate and suggest experts from your team.

backend/alembic/env.py

@@ -14,6 +14,7 @@
 from danswer.db.models import Base
 from celery.backends.database.session import ResultModelBase  # type: ignore
 from danswer.db.engine import get_all_tenant_ids
+from shared_configs.configs import POSTGRES_DEFAULT_SCHEMA
 
 # Alembic Config object
 config = context.config
@@ -57,11 +58,15 @@ def get_schema_options() -> tuple[str, bool, bool]:
             if "=" in pair:
                 key, value = pair.split("=", 1)
                 x_args[key.strip()] = value.strip()
-    schema_name = x_args.get("schema", "public")
+    schema_name = x_args.get("schema", POSTGRES_DEFAULT_SCHEMA)
     create_schema = x_args.get("create_schema", "true").lower() == "true"
     upgrade_all_tenants = x_args.get("upgrade_all_tenants", "false").lower() == "true"
 
-    if MULTI_TENANT and schema_name == "public" and not upgrade_all_tenants:
+    if (
+        MULTI_TENANT
+        and schema_name == POSTGRES_DEFAULT_SCHEMA
+        and not upgrade_all_tenants
+    ):
         raise ValueError(
             "Cannot run default migrations in public schema when multi-tenancy is enabled. "
             "Please specify a tenant-specific schema."

backend/danswer/__init__.py

@@ -1,3 +1,3 @@
 import os
 
-__version__ = os.environ.get("DANSWER_VERSION", "") or "0.3-dev"
+__version__ = os.environ.get("DANSWER_VERSION", "") or "Development"

backend/danswer/access/models.py

@@ -70,3 +70,12 @@ def build(
             user_groups=set(user_groups),
             is_public=is_public,
         )
+
+
+default_public_access = DocumentAccess(
+    external_user_emails=set(),
+    external_user_group_ids=set(),
+    user_emails=set(),
+    user_groups=set(),
+    is_public=True,
+)

backend/danswer/auth/users.py

@@ -93,7 +93,8 @@
 from danswer.utils.telemetry import optional_telemetry
 from danswer.utils.telemetry import RecordType
 from danswer.utils.variable_functionality import fetch_versioned_implementation
-from shared_configs.configs import current_tenant_id
+from shared_configs.configs import CURRENT_TENANT_ID_CONTEXTVAR
+from shared_configs.configs import POSTGRES_DEFAULT_SCHEMA
 
 logger = setup_logger()
 
@@ -187,7 +188,7 @@ def verify_email_domain(email: str) -> None:
 
 def get_tenant_id_for_email(email: str) -> str:
     if not MULTI_TENANT:
-        return "public"
+        return POSTGRES_DEFAULT_SCHEMA
     # Implement logic to get tenant_id from the mapping table
     with Session(get_sqlalchemy_engine()) as db_session:
         result = db_session.execute(
@@ -233,35 +234,62 @@ async def create(
         safe: bool = False,
         request: Optional[Request] = None,
     ) -> User:
-        verify_email_is_invited(user_create.email)
-        verify_email_domain(user_create.email)
-        if hasattr(user_create, "role"):
-            user_count = await get_user_count()
-            if user_count == 0 or user_create.email in get_default_admin_user_emails():
-                user_create.role = UserRole.ADMIN
-            else:
-                user_create.role = UserRole.BASIC
-        user = None
         try:
-            user = await super().create(user_create, safe=safe, request=request)  # type: ignore
-        except exceptions.UserAlreadyExists:
-            user = await self.get_by_email(user_create.email)
-            # Handle case where user has used product outside of web and is now creating an account through web
-            if (
-                not user.has_web_login
-                and hasattr(user_create, "has_web_login")
-                and user_create.has_web_login
-            ):
-                user_update = UserUpdate(
-                    password=user_create.password,
-                    has_web_login=True,
-                    role=user_create.role,
-                    is_verified=user_create.is_verified,
-                )
-                user = await self.update(user_update, user)
-            else:
-                raise exceptions.UserAlreadyExists()
-        return user
+            tenant_id = (
+                get_tenant_id_for_email(user_create.email)
+                if MULTI_TENANT
+                else POSTGRES_DEFAULT_SCHEMA
+            )
+        except exceptions.UserNotExists:
+            raise HTTPException(status_code=401, detail="User not found")
+
+        if not tenant_id:
+            raise HTTPException(
+                status_code=401, detail="User does not belong to an organization"
+            )
+
+        async with get_async_session_with_tenant(tenant_id) as db_session:
+            token = CURRENT_TENANT_ID_CONTEXTVAR.set(tenant_id)
+
+            verify_email_is_invited(user_create.email)
+            verify_email_domain(user_create.email)
+            if MULTI_TENANT:
+                tenant_user_db = SQLAlchemyUserAdminDB(db_session, User, OAuthAccount)
+                self.user_db = tenant_user_db
+                self.database = tenant_user_db
+
+            if hasattr(user_create, "role"):
+                user_count = await get_user_count()
+                if (
+                    user_count == 0
+                    or user_create.email in get_default_admin_user_emails()
+                ):
+                    user_create.role = UserRole.ADMIN
+                else:
+                    user_create.role = UserRole.BASIC
+            user = None
+            try:
+                user = await super().create(user_create, safe=safe, request=request)  # type: ignore
+            except exceptions.UserAlreadyExists:
+                user = await self.get_by_email(user_create.email)
+                # Handle case where user has used product outside of web and is now creating an account through web
+                if (
+                    not user.has_web_login
+                    and hasattr(user_create, "has_web_login")
+                    and user_create.has_web_login
+                ):
+                    user_update = UserUpdate(
+                        password=user_create.password,
+                        has_web_login=True,
+                        role=user_create.role,
+                        is_verified=user_create.is_verified,
+                    )
+                    user = await self.update(user_update, user)
+                else:
+                    raise exceptions.UserAlreadyExists()
+
+            CURRENT_TENANT_ID_CONTEXTVAR.reset(token)
+            return user
 
     async def on_after_login(
         self,
@@ -302,7 +330,9 @@ async def oauth_callback(
         # Get tenant_id from mapping table
         try:
             tenant_id = (
-                get_tenant_id_for_email(account_email) if MULTI_TENANT else "public"
+                get_tenant_id_for_email(account_email)
+                if MULTI_TENANT
+                else POSTGRES_DEFAULT_SCHEMA
             )
         except exceptions.UserNotExists:
             raise HTTPException(status_code=401, detail="User not found")
@@ -312,15 +342,15 @@ async def oauth_callback(
 
         token = None
         async with get_async_session_with_tenant(tenant_id) as db_session:
-            token = current_tenant_id.set(tenant_id)
+            token = CURRENT_TENANT_ID_CONTEXTVAR.set(tenant_id)
 
             verify_email_in_whitelist(account_email, tenant_id)
             verify_email_domain(account_email)
 
             if MULTI_TENANT:
                 tenant_user_db = SQLAlchemyUserAdminDB(db_session, User, OAuthAccount)
                 self.user_db = tenant_user_db
-                self.database = tenant_user_db
+                self.database = tenant_user_db  # type: ignore
 
             oauth_account_dict = {
                 "oauth_name": oauth_name,
@@ -402,7 +432,7 @@ async def oauth_callback(
                 user.oidc_expiry = None  # type: ignore
 
             if token:
-                current_tenant_id.reset(token)
+                CURRENT_TENANT_ID_CONTEXTVAR.reset(token)
 
             return user