Make all the env vars dynamic (#1092)

* Add truncate to task model (#1090) * retry text area height (#1091) * Make all the env vars dynamic * only check for db at runtime * remove secret from build step * improve dockerfile * Wrap db in singleton * add .env.local to dockerignore * changes to dockerfile * lint * aborted generations * move collections build check * Use a single dockerfile * lint * fixes * lint * don't return db during building * remove dev --------- Co-authored-by: Victor Muštar <[email protected]>
huggingface · May 3, 2024 · 552abe2 · 552abe2
1 parent 75b2833
commit 552abe2
Show file tree

Hide file tree

Showing 99 changed files with 590 additions and 649 deletions.
diff --git a/.dockerignore b/.dockerignore
@@ -8,4 +8,4 @@ node_modules/
 .svelte-kit/
 .env*
 !.env
-!.env.local
+.env.local
diff --git a/.env.template b/.env.template
@@ -231,6 +231,7 @@ MODELS=`[
       "parameters": {
         "temperature": 0.1,
         "stop": ["<|eot_id|>"],
+        "truncate": 1024,
       },
       "unlisted": true
     }

diff --git a/.github/workflows/build-image.yml b/.github/workflows/build-image.yml
@@ -8,7 +8,7 @@ on:
     branches:
       - "*"
     paths:
-      - "Dockerfile.local"
+      - "Dockerfile"
       - "entrypoint.sh"
   workflow_dispatch:
   release:
@@ -62,7 +62,7 @@ jobs:
         uses: docker/build-push-action@v5
         with:
           context: .
-          file: Dockerfile.local
+          file: Dockerfile
           push: ${{ github.event_name != 'pull_request' }}
           tags: ${{ steps.meta.outputs.tags }}
           labels: ${{ steps.meta.outputs.labels }}
@@ -116,7 +116,7 @@ jobs:
         uses: docker/build-push-action@v5
         with:
           context: .
-          file: Dockerfile.local
+          file: Dockerfile
           push: ${{ github.event_name != 'pull_request' }}
           tags: ${{ steps.meta.outputs.tags }}
           labels: ${{ steps.meta.outputs.labels }}

diff --git a/Dockerfile b/Dockerfile
@@ -1,6 +1,9 @@
 # syntax=docker/dockerfile:1
 # read the doc: https://huggingface.co/docs/hub/spaces-sdks-docker
 # you will also find guides on how best to write your Dockerfile
+ARG INCLUDE_DB=false
+
+# stage that install the dependencies
 FROM node:20 as builder-production
 
 WORKDIR /app
@@ -12,31 +15,74 @@ RUN --mount=type=cache,target=/app/.npm \
 
 FROM builder-production as builder
 
+ARG APP_BASE=
+ARG PUBLIC_APP_COLOR=blue
+
 RUN --mount=type=cache,target=/app/.npm \
         npm set cache /app/.npm && \
         npm ci
 
 COPY --link --chown=1000 . .
 
-RUN APP_BASE="/chat" npm run build
-RUN #npm run build
+RUN npm run build
 
-FROM node:20-slim
-RUN npm install -g vite-node
+# mongo image
+FROM mongo:latest as mongo
 
-RUN userdel -r node
+# image to be used if INCLUDE_DB is false
+FROM node:20-slim as local_db_false
+
+# image to be used if INCLUDE_DB is true
+FROM node:20-slim as local_db_true
+
+RUN apt-get update
+RUN apt-get install gnupg curl -y
+# copy mongo from the other stage
+COPY --from=mongo /usr/bin/mongo* /usr/bin/
+
+ENV MONGODB_URL=mongodb://localhost:27017
+RUN mkdir -p /data/db
+RUN chown -R 1000:1000 /data/db
+
+# final image
+FROM local_db_${INCLUDE_DB} as final
+
+# build arg to determine if the database should be included
+ARG INCLUDE_DB=false
+ENV INCLUDE_DB=${INCLUDE_DB}
+
+# svelte requires APP_BASE at build time so it must be passed as a build arg
+ARG APP_BASE=
+# tailwind requires the primary theme to be known at build time so it must be passed as a build arg
+ARG PUBLIC_APP_COLOR=blue
 
-RUN useradd -m -u 1000 user
 
+# install dotenv-cli
+RUN npm install -g dotenv-cli
+
+# switch to a user that works for spaces
+RUN userdel -r node
+RUN useradd -m -u 1000 user
 USER user
 
 ENV HOME=/home/user \
 	PATH=/home/user/.local/bin:$PATH
 
+WORKDIR /app
 
-COPY --from=builder-production --chown=1000 /app/node_modules /app/node_modules
-COPY --link --chown=1000 package.json /app/package.json
-COPY --from=builder --chown=1000 /app/build /app/build
+# add a .env.local if the user doesn't bind a volume to it
+RUN touch /app/.env.local
+
+# get the default config, the entrypoint script and the server script
+COPY --chown=1000 package.json /app/package.json
+COPY --chown=1000 .env /app/.env
+COPY --chown=1000 entrypoint.sh /app/entrypoint.sh
 COPY --chown=1000 gcp-*.json /app/
 
-CMD node /app/build/index.js
+#import the build & dependencies
+COPY --from=builder --chown=1000 /app/build /app/build
+COPY --from=builder --chown=1000 /app/node_modules /app/node_modules
+
+RUN chmod +x /app/entrypoint.sh
+
+CMD ["/bin/bash", "-c", "/app/entrypoint.sh"]
diff --git a/Dockerfile.local b/Dockerfile.local
diff --git a/entrypoint.sh b/entrypoint.sh
@@ -2,28 +2,16 @@ ENV_LOCAL_PATH=/app/.env.local
 
 if test -z "${DOTENV_LOCAL}" ; then
     if ! test -f "${ENV_LOCAL_PATH}" ; then
-        echo "DOTENV_LOCAL was not found in the ENV variables and .env.local is not set using a bind volume. We are using the default .env config."
+        echo "DOTENV_LOCAL was not found in the ENV variables and .env.local is not set using a bind volume. Make sure to set environment variables properly. "
     fi;
 else
     echo "DOTENV_LOCAL was found in the ENV variables. Creating .env.local file."
     cat <<< "$DOTENV_LOCAL" > ${ENV_LOCAL_PATH}
 fi;
 
 if [ "$INCLUDE_DB" = "true" ] ; then
-    echo "INCLUDE_DB is set to true."
-
-    MONGODB_CONFIG="MONGODB_URL=mongodb://localhost:27017"
-    if ! grep -q "^${MONGODB_CONFIG}$" ${ENV_LOCAL_PATH}; then
-      echo "Appending MONGODB_URL"
-      touch /app/.env.local
-      echo -e "\n${MONGODB_CONFIG}" >> ${ENV_LOCAL_PATH}
-    fi
-
-    mkdir -p /data/db
-    mongod &
     echo "Starting local MongoDB instance"
-
+    nohup mongod &
 fi;
 
-npm run build
-npm vite-node --options.transformMode.ssr='/.*/' /app/scripts/server.ts -- --host 0.0.0.0 --port 3000
+dotenv -e /app/.env -c -- node /app/build/index.js -- --host 0.0.0.0 --port 3000
diff --git a/package.json b/package.json
@@ -6,7 +6,7 @@
 	"scripts": {
 		"dev": "vite dev",
 		"build": "vite build",
-		"preview": "vite-node --options.transformMode.ssr='/.*/' scripts/server.ts",
+		"preview": "vite-node --options.transformMode.ssr='/.*/' scripts/server.,js",
 		"check": "svelte-kit sync && svelte-check --tsconfig ./tsconfig.json",
 		"check:watch": "svelte-kit sync && svelte-check --tsconfig ./tsconfig.json --watch",
 		"lint": "prettier --plugin-search-dir . --check . && eslint .",

diff --git a/scripts/populate.ts b/scripts/populate.ts
@@ -2,12 +2,13 @@ import readline from "readline";
 import minimist from "minimist";
 
 // @ts-expect-error: vite-node makes the var available but the typescript compiler doesn't see them
-import { MONGODB_URL } from "$env/static/private";
+import { env } from "$env/dynamic/private";
 
 import { faker } from "@faker-js/faker";
 import { ObjectId } from "mongodb";
 
-import { Database } from "$lib/server/database";
+// @ts-expect-error: vite-node makes the var available but the typescript compiler doesn't see them
+import { collections } from "$lib/server/database";
 import { models } from "../src/lib/server/models.ts";
 import type { User } from "../src/lib/types/User";
 import type { Assistant } from "../src/lib/types/Assistant";
@@ -17,6 +18,7 @@ import { defaultEmbeddingModel } from "../src/lib/server/embeddingModels.ts";
 import { Message } from "../src/lib/types/Message.ts";
 
 import { addChildren } from "../src/lib/utils/tree/addChildren.ts";
+import { generateSearchTokens } from "../src/lib/utils/searchTokens.ts";
 
 const rl = readline.createInterface({
 	input: process.stdin,
@@ -107,10 +109,10 @@ async function seed() {
 
 	if (flags.includes("reset")) {
 		console.log("Starting reset of DB");
-		await Database.getInstance().getCollections().users.deleteMany({});
-		await Database.getInstance().getCollections().settings.deleteMany({});
-		await Database.getInstance().getCollections().assistants.deleteMany({});
-		await Database.getInstance().getCollections().conversations.deleteMany({});
+		await collections.users.deleteMany({});
+		await collections.settings.deleteMany({});
+		await collections.assistants.deleteMany({});
+		await collections.conversations.deleteMany({});
 		console.log("Reset done");
 	}
 
@@ -126,11 +128,11 @@ async function seed() {
 			avatarUrl: faker.image.avatar(),
 		}));
 
-		await Database.getInstance().getCollections().users.insertMany(newUsers);
+		await collections.users.insertMany(newUsers);
 		console.log("Done creating users.");
 	}
 
-	const users = await Database.getInstance().getCollections().users.find().toArray();
+	const users = await collections.users.find().toArray();
 	if (flags.includes("settings") || flags.includes("all")) {
 		console.log("Updating settings for all users");
 		users.forEach(async (user) => {
@@ -145,7 +147,7 @@ async function seed() {
 				customPrompts: {},
 				assistants: [],
 			};
-			await Database.getInstance().getCollections().settings.updateOne(
+			await collections.settings.updateOne(
 				{ userId: user._id },
 				{ $set: { ...settings } },
 				{ upsert: true }
@@ -158,10 +160,11 @@ async function seed() {
 		console.log("Creating assistants for all users");
 		await Promise.all(
 			users.map(async (user) => {
+				const name = faker.animal.insect();
 				const assistants = faker.helpers.multiple<Assistant>(
 					() => ({
 						_id: new ObjectId(),
-						name: faker.animal.insect(),
+						name,
 						createdById: user._id,
 						createdByName: user.username,
 						createdAt: faker.date.recent({ days: 30 }),
@@ -174,11 +177,13 @@ async function seed() {
 						exampleInputs: faker.helpers.multiple(() => faker.lorem.sentence(), {
 							count: faker.number.int({ min: 0, max: 4 }),
 						}),
+						searchTokens: generateSearchTokens(name),
+						last24HoursCount: faker.number.int({ min: 0, max: 1000 }),
 					}),
 					{ count: faker.number.int({ min: 3, max: 10 }) }
 				);
-				await Database.getInstance().getCollections().assistants.insertMany(assistants);
-				await Database.getInstance().getCollections().settings.updateOne(
+				await collections.assistants.insertMany(assistants);
+				await collections.settings.updateOne(
 					{ userId: user._id },
 					{ $set: { assistants: assistants.map((a) => a._id.toString()) } },
 					{ upsert: true }
@@ -194,7 +199,7 @@ async function seed() {
 			users.map(async (user) => {
 				const conversations = faker.helpers.multiple(
 					async () => {
-						const settings = await Database.getInstance().getCollections().settings.findOne<Settings>({ userId: user._id });
+						const settings = await collections.settings.findOne<Settings>({ userId: user._id });
 
 						const assistantId =
 							settings?.assistants && settings.assistants.length > 0 && faker.datatype.boolean(0.1)
@@ -203,7 +208,7 @@ async function seed() {
 
 						const preprompt =
 							(assistantId
-								? await Database.getInstance().getCollections().assistants
+								? await collections.assistants
 										.findOne({ _id: assistantId })
 										.then((assistant: Assistant) => assistant?.preprompt ?? "")
 								: faker.helpers.maybe(() => faker.hacker.phrase(), { probability: 0.5 })) ?? "";
@@ -229,7 +234,7 @@ async function seed() {
 					{ count: faker.number.int({ min: 10, max: 200 }) }
 				);
 
-				await Database.getInstance().getCollections().conversations.insertMany(await Promise.all(conversations));
+				await collections.conversations.insertMany(await Promise.all(conversations));
 			})
 		);
 		console.log("Done creating conversations.");
@@ -241,7 +246,7 @@ async function seed() {
 	try {
 		rl.question(
 			"You're about to run a seeding script on the following MONGODB_URL: \x1b[31m" +
-				MONGODB_URL +
+				env.MONGODB_URL +
 				"\x1b[0m\n\n With the following flags: \x1b[31m" +
 				flags.join("\x1b[0m , \x1b[31m") +
 				"\x1b[0m\n \n\n Are you sure you want to continue? (yes/no): ",

diff --git a/scripts/server.ts b/scripts/server.ts