local version of interface

interface/.firebaserc

@@ -1,5 +1,19 @@
 {
   "projects": {
     "default": "codeonline-ee37b"
-  }
+  },
+  "targets": {
+    "codeonline-ee37b": {
+      "hosting": {
+        "test": [
+          "proactive-test"
+        ],
+        "proactive": [
+          "code-proactive-chat"
+        ]
+      }
+    }
+  },
+  "etags": {},
+  "dataconnectEmulatorConfig": {}
 }

interface/.gitignore

@@ -1,5 +1,5 @@
 # See https://help.github.com/articles/ignoring-files/ for more about ignoring files.
-
+.vscode
 # dependencies
 /node_modules
 /.pnp

interface/README.md

@@ -1,49 +1,32 @@
 
-# Getting Started
+# RealHumanEval Interface
 
-RealHumanEval is built as a React app using Google Firebase for hosting, data storage, and API calling. However, it can be run locally by configuring certain flags and inputting the API keys.
+This is a local version of RealHumanEval, however, by replacing the API calls to be non-local in [app/functions/cloud_functions_helper.js](app/functions/cloud_functions_helper.js) and task loading in [app/functions/task_logic.js](app/functions/task_logic.js) and [app/page.tsx](app/page.tsx) you can host this with Firebase as we did for the actual user study.
 
-## Requirements
+The repo is organized as follows:
 
+- [app](app): Contains the frontend code for the project.
 
-For the required packages:
-```
-npm install next react react-dom
-npm install firebase
-npm install axios
-```
-You will need API keys for the following services to run the app:
-
-OpenAI: https://openai.com/api/
-Together.ai: https://api.together.xyz/
-One compiler: https://rapidapi.com/onecompiler-onecompiler-default/api/onecompiler-apis
-
-
-### Firebase
-
-To recreate the same functionality, you need to link your web app with a Firebase project.
+- [app/components](app/components): Contains the React components for the project.
 
-Once you create your Firebase project, make sure to enable Cloud Firestore. In Firestore, create an empty collection called “responses”. Then create a collection called “tasks”, each entry in tasks must have three items: task_descriptions, function_signatures and unit_tests. Each of the three items are a list of strings. We include a copy of our tasks collection. 
+- [app/functions](app/functions): Contains the helper functions for the project.
 
-In the Firebase console, under “Project Settings”, make sure to create an app, then copy the Firebase config to initialize_firebase.js in interface/app/functions.
+- The main page is [app/page.tsx](app/page.tsx).
 
-Make sure to login to your firebase project under the folder of interface, following https://firebase.google.com/docs/web/setup.
+### Installation
+This is a [Next.js](https://nextjs.org/) project bootstrapped with [`create-next-app`](https://github.com/vercel/next.js/tree/canary/packages/create-next-app).
 
-Now in react, for firebase functionality, you need to enable it:
-```
-firebase experiments:enable webframeworks
-```
+Follow the instructions below to install the necessary dependencies.
 
-Then you need to set the API keys one by one:
 
 ```
-firebase functions:secrets:set OPENAI_KEY
-firebase functions:secrets:set TOGETHER_KEY
-firebase functions:secrets:set RAPIDAPI_KEY
+npm install next react react-dom
+npm install axios
+npm install openai
 ```
 
 
-### Deployment
+You need to your API keys to [app/components/settings.tsx](app/components/settings.tsx) file. For now this is well tested with OpenAI API and OneCompiler API. 
 
 
 First, run the development server:
@@ -60,34 +43,9 @@ bun dev
 
 Open [http://localhost:3000](http://localhost:3000) with your browser to see the result.
 
-To deploy online using firebase:
-
-```bash
- Firebase deploy
-```
-
-
-
-
-# Roadmap of Features
-
-- Adding multiple tabs to editor: currently it is easy to add tabs by setting the "useTabs" variable to true in Editor.tsx, however, you cannot import code from one tab to another. Moreover, you cannot choose which tab to execute. 
-
-- Adding a console: currently you can only run the code in the editor, but you cannot use interactive input "input()".
-
-- Moving beyond OneCompiler: it would be good to have a simple python environment as a console to allow for installing different packages and so on.
-
-- Adding a file system: it would be good to have a file system to allow for saving and loading files.
-
-- Display online statistics about AI usage and participant performance (e.g., acceptance rate of suggestions, percentage of upvoted messages, …)
-
-
-
-
-
-
-
-
+This is what you will see:
 
+![image](public/final_interface.png)
 
 
+![image](public/integrate_proactive1.gif)