Firebase separation

.firebaserc

@@ -0,0 +1,5 @@
+{
+  "projects": {
+    "default": "learbdb-cmuq"
+  }
+}

.github/workflows/firebase-hosting-merge.yml

@@ -0,0 +1,20 @@
+# This file was auto-generated by the Firebase CLI
+# https://github.com/firebase/firebase-tools
+
+name: Deploy to Firebase Hosting on merge
+'on':
+  push:
+    branches:
+      - main
+jobs:
+  build_and_deploy:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v3
+      - run: npm ci && npm run build
+      - uses: FirebaseExtended/action-hosting-deploy@v0
+        with:
+          repoToken: '${{ secrets.GITHUB_TOKEN }}'
+          firebaseServiceAccount: '${{ secrets.FIREBASE_SERVICE_ACCOUNT_LEARBDB_CMUQ }}'
+          channelId: live
+          projectId: learbdb-cmuq

.github/workflows/firebase-hosting-pull-request.yml

@@ -0,0 +1,17 @@
+# This file was auto-generated by the Firebase CLI
+# https://github.com/firebase/firebase-tools
+
+name: Deploy to Firebase Hosting on PR
+'on': pull_request
+jobs:
+  build_and_preview:
+    if: '${{ github.event.pull_request.head.repo.full_name == github.repository }}'
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v3
+      - run: npm ci && npm run build
+      - uses: FirebaseExtended/action-hosting-deploy@v0
+        with:
+          repoToken: '${{ secrets.GITHUB_TOKEN }}'
+          firebaseServiceAccount: '${{ secrets.FIREBASE_SERVICE_ACCOUNT_LEARBDB_CMUQ }}'
+          projectId: learbdb-cmuq

firebase.json

@@ -0,0 +1,37 @@
+{
+  "functions": [
+    {
+      "source": "functions",
+      "codebase": "default",
+      "ignore": [
+        "node_modules",
+        ".git",
+        "firebase-debug.log",
+        "firebase-debug.*.log"
+      ]
+    }
+  ],
+  "hosting": {
+    "public": "public",
+    "ignore": [
+      "firebase.json",
+      "**/.*",
+      "**/node_modules/**"
+    ],
+    "rewrites": [
+      {
+        "source": "**",
+        "destination": "/index.html"
+      }
+    ]
+  },
+  "emulators": {
+    "functions": {
+      "port": 5001
+    },
+    "ui": {
+      "enabled": true
+    },
+    "singleProjectMode": true
+  }
+}

functions/.gitignore

@@ -0,0 +1 @@
+node_modules/

functions/index.js

@@ -0,0 +1,146 @@
+const fetch = require('node-fetch'); // Import for webscraping in fetchContentFromURL()
+import { OpenAIApi, Configuration } from 'openai';
+// const { Configuration, OpenAIApi } = require('openai');
+
+// Function to fetch content from URL using a web scraping service
+async function fetchContentFromURL(url) {
+    try {
+        const response = await fetch(url);
+        if (!response.ok) {
+            throw new Error(`HTTP error! status: ${response.status}`);
+        }
+        return await response.text();
+    } catch (error) {
+        console.error(`Could not fetch content from URL: ${error}`);
+        throw error;
+    }
+}
+
+function simplifyContent(content) {
+    // Denote headings with a marker (like '###') and add a line break
+    content = content.replace(/<h[1-6].*?>(.*?)<\/h[1-6]>/g, '\n### $1\n');
+    // Remove script and style elements and their content
+    let simplifiedContent = content.replace(/<script.*?>.*?<\/script>/gms, '');
+    simplifiedContent = simplifiedContent.replace(/<style.*?>.*?<\/style>/gms, '');
+    // Remove all remaining HTML tags, leaving the inner text
+    simplifiedContent = simplifiedContent.replace(/<[^>]+>/g, '');
+    // Manually replace common HTML entities
+    simplifiedContent = simplifiedContent
+        .replace(/&amp;/g, '&')
+        .replace(/&lt;/g, '<')
+        .replace(/&gt;/g, '>')
+        .replace(/&quot;/g, '"')
+        .replace(/&#39;/g, "'");
+    // Remove inline CSS and JavaScript event handlers
+    simplifiedContent = simplifiedContent.replace(/style\s*=\s*'.*?'/gi, '');
+    simplifiedContent = simplifiedContent.replace(/on\w+\s*=\s*".*?"/gi, '');
+    // Normalize whitespace without removing sentence punctuation
+    simplifiedContent = simplifiedContent.replace(/\s+/g, ' ').trim();
+    // Condense multiple line breaks into a single one
+    simplifiedContent = simplifiedContent.replace(/(\r\n|\r|\n){2,}/g, '\n');
+    return simplifiedContent;
+}
+
+
+// Placeholder function to perform GPT analysis for media type and topics using Mistral-7b via OpenRouter
+async function performGPTAnalysis(simplifiedContent, apiKey) {
+    // Implement logic to send content to Mistral-7b via OpenRouter for GPT analysis
+    // Send content and receive GPT analysis response
+
+    // this is the code that we tried to use for the GPT Analysis 
+    // try {
+    //     const configuration = new Configuration({
+    //         apiKey: apiKey,  // Use the provided API key
+    //         baseURL: "https://openrouter.ai/api/v1" // Your custom API endpoint
+    //     });
+
+    //     const openai = new OpenAIApi(configuration);
+
+    //     // Using the specified prompt
+    //     const prompt = `Analyze the following text and provide the media type and key topics: ${simplifiedContent}`;
+
+    //     const completion = await openai.createCompletion({
+    //         model: "mistralai/mistral-7b-instruct",
+    //         prompt: prompt,
+    //         max_tokens: 150 // Adjust as needed
+    //     });
+
+    //     //return completion.data.choices[0].text.trim();
+    //     return inferredMediaType;
+    // } catch (error) {
+    //     console.error('Error with OpenAI completion:', error);
+    //     throw error;
+    // }
+    // however, it gives the error below: 
+    // { "error": "Something went wrong", "details": "Configuration is not a constructor" }
+
+    // Placeholder code
+    const inferredMediaType = ["article"];
+    const extractedTopics = ["topic1", "topic2"];
+    return { inferredMediaType, extractedTopics };
+}
+
+// Placeholder function to map inferred values to predefined formats and topics
+function mapInferredValues(mediaType, topics) {
+    // Implement logic to map inferred media type and topics to predefined formats and topics
+    // Match inferred values with predefined taxonomy
+    // Placeholder code
+    const predefinedMediaType = "Article";
+    const predefinedTopics = ["Topic 1", "Topic 2"];
+    return { predefinedMediaType, predefinedTopics };
+}
+
+// Placeholder function to format the response
+function formatResponse(predefinedMediaType, predefinedTopics) {
+    // Implement logic to format the extracted metadata into the desired response structure
+    // Construct the response object
+    // Placeholder code
+    const response = {
+        format: predefinedMediaType,
+        topics: predefinedTopics,
+    };
+    return response;
+}
+
+export async function handler(event) {
+    try {
+        // Extract URL and API Key from the request body
+        const { url, apiKey } = JSON.parse(event.body);
+
+        // Validate if URL and API Key are present
+        if (!url || !apiKey) {
+            return {
+                statusCode: 400,
+                body: JSON.stringify({ error: 'URL and API Key are required' }),
+            };
+        }
+
+        // Step 1: Fetch content from the URL using a web scraping service
+        const fetchedContent = await fetchContentFromURL(url);
+
+        // Step 2: Simplify the fetched content for GPT analysis
+        const simplifiedContent = simplifyContent(fetchedContent);
+
+        // Step 3: Perform GPT analysis for media type and topics
+        const { inferredMediaType, extractedTopics } = await performGPTAnalysis(simplifiedContent, apiKey);
+
+        // Step 4: Map inferred values to predefined formats and topics
+        const { predefinedMediaType, predefinedTopics } = mapInferredValues(inferredMediaType, extractedTopics);
+
+        // Step 5: Format the response
+        const formattedResponse = formatResponse(predefinedMediaType, predefinedTopics);
+
+        // Return the formatted response
+        return {
+            statusCode: 200,
+            // returning the output of the simplifyContent function, to test the function
+            body: JSON.stringify(simplifiedContent),
+        };
+    } catch (error) {
+        console.error('Error occurred:', error.message);
+        return {
+            statusCode: 500,
+            body: JSON.stringify({ error: 'Something went wrong', details: error.message }),
+        };
+    }
+}

functions/package.json

@@ -0,0 +1,23 @@
+{
+  "name": "functions",
+  "description": "Cloud Functions for Firebase",
+  "scripts": {
+    "serve": "firebase emulators:start --only functions",
+    "shell": "firebase functions:shell",
+    "start": "npm run shell",
+    "deploy": "firebase deploy --only functions",
+    "logs": "firebase functions:log"
+  },
+  "engines": {
+    "node": "18"
+  },
+  "main": "index.js",
+  "dependencies": {
+    "firebase-admin": "^11.8.0",
+    "firebase-functions": "^4.3.1"
+  },
+  "devDependencies": {
+    "firebase-functions-test": "^3.1.0"
+  },
+  "private": true
+}

netlify.toml

@@ -0,0 +1,2 @@
+[build]
+  functions = "netlify/functions/"