Resolve conflits and deploy

.github/workflows/docker-build-push-backend-container-on-tag.yml

@@ -38,5 +38,7 @@ jobs:
     - name: Run Trivy vulnerability scanner
       uses: aquasecurity/trivy-action@master
       with:
+        # To run locally: trivy image --severity HIGH,CRITICAL danswer/danswer-backend
         image-ref: docker.io/danswer/danswer-backend:${{ github.ref_name }}
         severity: 'CRITICAL,HIGH'
+        trivyignores: ./backend/.trivyignore

.github/workflows/pr-python-checks.yml

@@ -20,10 +20,12 @@ jobs:
         cache-dependency-path: |
           backend/requirements/default.txt
           backend/requirements/dev.txt
+          backend/requirements/model_server.txt
     - run: |
         python -m pip install --upgrade pip
         pip install -r backend/requirements/default.txt
         pip install -r backend/requirements/dev.txt
+        pip install -r backend/requirements/model_server.txt
 
     - name: Run MyPy
       run: |

CONTRIBUTING.md

@@ -85,6 +85,7 @@ Install the required python dependencies:
 ```bash
 pip install -r danswer/backend/requirements/default.txt
 pip install -r danswer/backend/requirements/dev.txt
+pip install -r danswer/backend/requirements/model_server.txt
 ```
 
 Install [Node.js and npm](https://docs.npmjs.com/downloading-and-installing-node-js-and-npm) for the frontend.
@@ -112,26 +113,24 @@ docker compose -f docker-compose.dev.yml -p danswer-stack up -d index relational
 (index refers to Vespa and relational_db refers to Postgres)
 
 #### Running Danswer
-
-Setup a folder to store config. Navigate to `danswer/backend` and run:
-```bash
-mkdir dynamic_config_storage
-```
-
 To start the frontend, navigate to `danswer/web` and run:
 ```bash
 npm run dev
 ```
 
-Package the Vespa schema. This will only need to be done when the Vespa schema is updated locally.
-
-Navigate to `danswer/backend/danswer/document_index/vespa/app_config` and run:
+Next, start the model server which runs the local NLP models.
+Navigate to `danswer/backend` and run:
 ```bash
-zip -r ../vespa-app.zip .
+uvicorn model_server.main:app --reload --port 9000
+```
+_For Windows (for compatibility with both PowerShell and Command Prompt):_
+```bash
+powershell -Command "
+    uvicorn model_server.main:app --reload --port 9000
+"
 ```
-- Note: If you don't have the `zip` utility, you will need to install it prior to running the above
 
-The first time running Danswer, you will also need to run the DB migrations for Postgres.
+The first time running Danswer, you will need to run the DB migrations for Postgres.
 After the first time, this is no longer required unless the DB models change.
 
 Navigate to `danswer/backend` and with the venv active, run:
@@ -149,17 +148,12 @@ python ./scripts/dev_run_background_jobs.py
 
 To run the backend API server, navigate back to `danswer/backend` and run:
 ```bash
-AUTH_TYPE=disabled \
-DYNAMIC_CONFIG_DIR_PATH=./dynamic_config_storage \
-VESPA_DEPLOYMENT_ZIP=./danswer/document_index/vespa/vespa-app.zip \
-uvicorn danswer.main:app --reload --port 8080
+AUTH_TYPE=disabled uvicorn danswer.main:app --reload --port 8080
 ```
 _For Windows (for compatibility with both PowerShell and Command Prompt):_
 ```bash
 powershell -Command "
     $env:AUTH_TYPE='disabled'
-    $env:DYNAMIC_CONFIG_DIR_PATH='./dynamic_config_storage'
-    $env:VESPA_DEPLOYMENT_ZIP='./danswer/document_index/vespa/vespa-app.zip'
     uvicorn danswer.main:app --reload --port 8080 
 "
 ```
@@ -178,20 +172,16 @@ pre-commit install
 
 Additionally, we use `mypy` for static type checking.
 Danswer is fully type-annotated, and we would like to keep it that way! 
-Right now, there is no automated type checking at the moment (coming soon), but we ask you to manually run it before
-creating a pull requests with `python -m mypy .` from the `danswer/backend` directory.
+To run the mypy checks manually, run `python -m mypy .` from the `danswer/backend` directory.
 
 
 #### Web
 We use `prettier` for formatting. The desired version (2.8.8) will be installed via a `npm i` from the `danswer/web` directory. 
 To run the formatter, use `npx prettier --write .` from the `danswer/web` directory.
-Like `mypy`, we have no automated formatting yet (coming soon), but we request that, for now,
-you run this manually before creating a pull request.
+Please double check that prettier passes before creating a pull request.
 
 
 ### Release Process
 Danswer follows the semver versioning standard.
 A set of Docker containers will be pushed automatically to DockerHub with every tag.
 You can see the containers [here](https://hub.docker.com/search?q=danswer%2F).
-
-As pre-1.0 software, even patch releases may contain breaking or non-backwards-compatible changes.

README.md

@@ -22,11 +22,12 @@
 </a>
 </p>
 
-<strong>[Danswer](https://www.danswer.ai/)</strong> is the ChatGPT for teams. Danswer provides a Chat interface and plugs into any LLM of
-your choice. Danswer can be deployed anywhere and for any scale - on a laptop, on-premise, or to cloud. Since you own
-the deployment, your user data and chats are fully in your own control. Danswer is MIT licensed and designed to be
-modular and easily extensible. The system also comes fully ready for production usage with user authentication, role
-management (admin/basic users), chat persistence, and a UI for configuring Personas (AI Assistants) and their Prompts.
+<strong>[Danswer](https://www.danswer.ai/)</strong> is the AI Assistant connected to your company's docs, apps, and people. 
+Danswer provides a Chat interface and plugs into any LLM of your choice. Danswer can be deployed anywhere and for any 
+scale - on a laptop, on-premise, or to cloud. Since you own the deployment, your user data and chats are fully in your 
+own control. Danswer is MIT licensed and designed to be modular and easily extensible. The system also comes fully ready 
+for production usage with user authentication, role management (admin/basic users), chat persistence, and a UI for 
+configuring Personas (AI Assistants) and their Prompts.
 
 Danswer also serves as a Unified Search across all common workplace tools such as Slack, Google Drive, Confluence, etc.
 By combining LLMs and team specific knowledge, Danswer becomes a subject matter expert for the team. Imagine ChatGPT if

backend/.trivyignore

@@ -0,0 +1,46 @@
+# https://github.com/madler/zlib/issues/868
+# Pulled in with base Debian image, it's part of the contrib folder but unused
+# zlib1g is fine
+# Will be gone with Debian image upgrade
+# No impact in our settings
+CVE-2023-45853
+
+# krb5 related, worst case is denial of service by resource exhaustion
+# Accept the risk
+CVE-2024-26458
+CVE-2024-26461
+CVE-2024-26462
+CVE-2024-26458
+CVE-2024-26461
+CVE-2024-26462
+CVE-2024-26458
+CVE-2024-26461
+CVE-2024-26462
+CVE-2024-26458
+CVE-2024-26461
+CVE-2024-26462
+
+# Specific to Firefox which we do not use
+# No impact in our settings
+CVE-2024-0743
+
+# bind9 related, worst case is denial of service by CPU resource exhaustion
+# Accept the risk
+CVE-2023-50387
+CVE-2023-50868
+CVE-2023-50387
+CVE-2023-50868
+
+# libexpat1, XML parsing resource exhaustion
+# We don't parse any user provided XMLs
+# No impact in our settings
+CVE-2023-52425
+CVE-2024-28757
+
+# sqlite, only used by NLTK library to grab word lemmatizer and stopwords
+# No impact in our settings
+CVE-2023-7104
+
+# libharfbuzz0b, O(n^2) growth, worst case is denial of service
+# Accept the risk
+CVE-2023-25193

backend/Dockerfile

@@ -12,7 +12,9 @@ RUN echo "DANSWER_VERSION: ${DANSWER_VERSION}"
 # zip for Vespa step futher down
 # ca-certificates for HTTPS
 RUN apt-get update && \
-    apt-get install -y cmake curl zip ca-certificates libgnutls30=3.7.9-2+deb12u2 && \
+    apt-get install -y cmake curl zip ca-certificates libgnutls30=3.7.9-2+deb12u2 \
+    libblkid1=2.38.1-5+deb12u1 libmount1=2.38.1-5+deb12u1 libsmartcols1=2.38.1-5+deb12u1 \
+    libuuid1=2.38.1-5+deb12u1 && \
     rm -rf /var/lib/apt/lists/* && \
     apt-get clean
 
@@ -29,15 +31,16 @@ RUN pip install --no-cache-dir --upgrade -r /tmp/requirements.txt && \
 # xserver-common and xvfb included by playwright installation but not needed after
 # perl-base is part of the base Python Debian image but not needed for Danswer functionality
 # perl-base could only be removed with --allow-remove-essential
-RUN apt-get remove -y --allow-remove-essential perl-base xserver-common xvfb cmake libldap-2.5-0 libldap-2.5-0 && \
+RUN apt-get remove -y --allow-remove-essential perl-base xserver-common xvfb cmake \
+    libldap-2.5-0 libldap-2.5-0 && \
     apt-get autoremove -y && \
     rm -rf /var/lib/apt/lists/* && \
     rm /usr/local/lib/python3.11/site-packages/tornado/test/test.key
 
 # Set up application files
 WORKDIR /app
 COPY ./danswer /app/danswer
-COPY ./shared_models /app/shared_models
+COPY ./shared_configs /app/shared_configs
 COPY ./alembic /app/alembic
 COPY ./alembic.ini /app/alembic.ini
 COPY supervisord.conf /usr/etc/supervisord.conf

backend/Dockerfile.model_server

@@ -13,23 +13,14 @@ RUN apt-get remove -y --allow-remove-essential perl-base && \
 
 WORKDIR /app
 
-# Needed for model configs and defaults
-COPY ./danswer/configs /app/danswer/configs
-COPY ./danswer/dynamic_configs /app/danswer/dynamic_configs
-
 # Utils used by model server
 COPY ./danswer/utils/logger.py /app/danswer/utils/logger.py
-COPY ./danswer/utils/timing.py /app/danswer/utils/timing.py
-COPY ./danswer/utils/telemetry.py /app/danswer/utils/telemetry.py
 
 # Place to fetch version information
 COPY ./danswer/__init__.py /app/danswer/__init__.py
 
-# Shared implementations for running NLP models locally
-COPY ./danswer/search/search_nlp_models.py /app/danswer/search/search_nlp_models.py
-
-# Request/Response models
-COPY ./shared_models /app/shared_models
+# Shared between Danswer Backend and Model Server
+COPY ./shared_configs /app/shared_configs
 
 # Model Server main code
 COPY ./model_server /app/model_server

backend/alembic/versions/173cae5bba26_port_config_store.py

@@ -0,0 +1,29 @@
+"""Port Config Store
+
+Revision ID: 173cae5bba26
+Revises: e50154680a5c
+Create Date: 2024-03-19 15:30:44.425436
+
+"""
+from alembic import op
+import sqlalchemy as sa
+from sqlalchemy.dialects import postgresql
+
+# revision identifiers, used by Alembic.
+revision = "173cae5bba26"
+down_revision = "e50154680a5c"
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    op.create_table(
+        "key_value_store",
+        sa.Column("key", sa.String(), nullable=False),
+        sa.Column("value", postgresql.JSONB(astext_type=sa.Text()), nullable=False),
+        sa.PrimaryKeyConstraint("key"),
+    )
+
+
+def downgrade() -> None:
+    op.drop_table("key_value_store")

backend/alembic/versions/38eda64af7fe_add_chat_session_sharing.py

@@ -0,0 +1,41 @@
+"""Add chat session sharing
+
+Revision ID: 38eda64af7fe
+Revises: 776b3bbe9092
+Create Date: 2024-03-27 19:41:29.073594
+
+"""
+from alembic import op
+import sqlalchemy as sa
+
+# revision identifiers, used by Alembic.
+revision = "38eda64af7fe"
+down_revision = "776b3bbe9092"
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    op.add_column(
+        "chat_session",
+        sa.Column(
+            "shared_status",
+            sa.Enum(
+                "PUBLIC",
+                "PRIVATE",
+                name="chatsessionsharedstatus",
+                native_enum=False,
+            ),
+            nullable=True,
+        ),
+    )
+    op.execute("UPDATE chat_session SET shared_status='PRIVATE'")
+    op.alter_column(
+        "chat_session",
+        "shared_status",
+        nullable=False,
+    )
+
+
+def downgrade() -> None:
+    op.drop_column("chat_session", "shared_status")

backend/alembic/versions/4738e4b3bae1_pg_file_store.py

@@ -0,0 +1,28 @@
+"""PG File Store
+
+Revision ID: 4738e4b3bae1
+Revises: e91df4e935ef
+Create Date: 2024-03-20 18:53:32.461518
+
+"""
+from alembic import op
+import sqlalchemy as sa
+
+# revision identifiers, used by Alembic.
+revision = "4738e4b3bae1"
+down_revision = "e91df4e935ef"
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    op.create_table(
+        "file_store",
+        sa.Column("file_name", sa.String(), nullable=False),
+        sa.Column("lobj_oid", sa.Integer(), nullable=False),
+        sa.PrimaryKeyConstraint("file_name"),
+    )
+
+
+def downgrade() -> None:
+    op.drop_table("file_store")