Fall 2022 Work

.github/workflows/add_collaborators.yml

@@ -1,51 +1,51 @@
-# Author: Rishab 
-name: Add Collaborators v1.1
+# # Author: Rishab 
+# name: Add Collaborators v1.1
 
-on: 
-  push:
-    branches:
-      - master
-      - main
-      - dev
+# on: 
+#   push:
+#     branches:
+#       - master
+#       - main
+#       - dev
 
-jobs:
-  add-collaborators:
-    runs-on: ubuntu-latest
-    steps:
-      - name: Checkout Repo
-        uses: actions/checkout@v2
-      - name: Add Collaborators
-        uses: actions/[email protected]
-        with:
-          github-token: ${{secrets.ADMIN_GITHUB_TOKEN}}
-          script: |
-            const fs = require('fs');
-            const readline = require('readline');
-            const readInterface = readline.createInterface({
-              input: fs.createReadStream('./COLLABORATORS'),
-              output: process.stdout,
-              console: false
-            });
-            readInterface.on('line', async function(line) {
-              try {
-                await github.repos.checkCollaborator({
-                            owner: context.repo.owner,
-                            repo: context.repo.repo,
-                            username: line,
-                            });
-              } catch(err) {
-                console.log(err)
-                if(err.toString() === "HttpError: Not Found"){
-                  await github.repos.addCollaborator({
-                            owner: context.repo.owner,
-                            repo: context.repo.repo,
-                            username: line,
-                            permission: 'push'
-                            });
-                } else { 
-                  core.setFailed(err.toString())
-                }
+# jobs:
+#   add-collaborators:
+#     runs-on: ubuntu-latest
+#     steps:
+#       - name: Checkout Repo
+#         uses: actions/checkout@v2
+#       - name: Add Collaborators
+#         uses: actions/[email protected]
+#         with:
+#           github-token: ${{secrets.ADMIN_GITHUB_TOKEN}}
+#           script: |
+#             const fs = require('fs');
+#             const readline = require('readline');
+#             const readInterface = readline.createInterface({
+#               input: fs.createReadStream('./COLLABORATORS'),
+#               output: process.stdout,
+#               console: false
+#             });
+#             readInterface.on('line', async function(line) {
+#               try {
+#                 await github.repos.checkCollaborator({
+#                             owner: context.repo.owner,
+#                             repo: context.repo.repo,
+#                             username: line,
+#                             });
+#               } catch(err) {
+#                 console.log(err)
+#                 if(err.toString() === "HttpError: Not Found"){
+#                   await github.repos.addCollaborator({
+#                             owner: context.repo.owner,
+#                             repo: context.repo.repo,
+#                             username: line,
+#                             permission: 'push'
+#                             });
+#                 } else { 
+#                   core.setFailed(err.toString())
+#                 }
 
-              }
-            });
+#               }
+#             });
 

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

@@ -3,7 +3,7 @@
 
 name: Deploy to Firebase Hosting on merge
 "on":
-  push:
+  merge:
     branches:
       - main
 jobs:

.github/workflows/push-test.yml

@@ -0,0 +1,32 @@
+# Author: Haowei Li
+# (It seems like a trend for yml that people put names, so I do it too)
+
+# The purpose of this yml file is to run some tests on the client to see if everything runs as supposed to
+# client is based on React.JS
+name: Unit Test
+on:
+  push:
+    branches:
+       - main
+       - dev
+defaults:
+  run:
+    working-directory: client
+jobs:
+  tests:
+    runs-on: ubuntu-latest
+
+    steps: 
+      - uses: actions/checkout@v2
+
+      - name: Set up Node.js version
+        uses: actions/setup-node@v1
+        with:
+          node-version: '14.x'
+
+      - name: npm install, build, and test
+        run: |
+          npm install
+          npm install [email protected]
+          npm test
+#npm install [email protected]

.gitignore

@@ -4,7 +4,7 @@
 */__pycache__
 */tmp
 package-lock.json
-settings.js
 serviceaccount.json
-client/src/components/settings.js
+*/coverage
+client/package-lock.json
 client/src/components/settings.js

README.md

@@ -104,11 +104,12 @@ gcloud run deploy api-dev \
 
 ### Next Steps
 
+- Create different user types, with support for different user permissions
+  - Motivation: so patients and caretakers can both upload/manage files on behalf of the patient
+  - Suggested direction
+    - Currently, the ERD of the system looks as follows: [link](https://excalidraw.com/#json=-LCSG-ShDmak9AprUI9LT,zhR7TQiJovH9fbLHI2MJsA)
+    - A modification of the database and user flow to follow this general guideline would allow for these features to be built, as this would allow both patients and caretakers to upload/retrieve from the same document within the newly created collections: [link](https://excalidraw.com/#json=21EzZvSgTpRM558zRtxWx,e5qdQqTUEmp2myNCfwgo-g)
 - Develop additional audio processing features: 
   - Background noise reduction
   - Improve the clarity of slurred subject voice
-- Other features: 
-  - Folder layout in processed files pages
-  - User is able to delete processed files
-  - User input in trimming/splitting process
-
+  - User input in trimming/splitting process

backend/app.py

@@ -8,7 +8,6 @@
 import os
 import datetime
 from flask import Flask, request, jsonify
-from flask_cors import CORS
 from google.cloud import firestore
 import google.auth.transport.requests
 from google.oauth2 import id_token
@@ -158,6 +157,7 @@ def upload_audios():
         return 'Unauthorized', 401
     user_ref = users_collection.document(claims['sub'])
     doc = user_ref.get()
+    # return {str(doc)}, 200
     if doc.exists:
         doc = doc.to_dict()
         for file in files:
@@ -171,30 +171,28 @@ def upload_audios():
             # upload processed audio
             for path in processedFilePaths:
                 dest_processed_file = f'Audio{uuid.uuid1()}.wav'
-                processedFileName = path.split('/tmp/')[1]
+                originalFileName = path.split('/tmp/')[1]
                 blob = bucket.blob(dest_processed_file)
                 blob.upload_from_filename(path, content_type='audio/wav')
+                created = False
                 if "audio" in doc:
-                    doc["audio"].append({processedFileName: destination_file_name, "processed": {dest_processed_file}})
+                    # check if file already exists
+                    for audio in doc["audio"]:
+                        if destination_file_name in str(list(audio.values())):
+                            created = True
+                            audio["processed"].append(dest_processed_file)
+                    if not created:
+                        doc["audio"].append({originalFileName: destination_file_name, "processed": [dest_processed_file]})
                     user_ref.update({"audio": doc["audio"]})
                 else:
-                    user_ref.update({"audio": [{processedFileName: destination_file_name, "processed "+ destination_file_name: {dest_processed_file}}]})
-                    doc = user_ref.get()
-                    doc = doc.to_dict()
-            # if "audio" in doc:
-            #     doc["audio"].append({file.filename: destination_file_name})
-            #     user_ref.update({"audio": doc["audio"]})
-            # else:
-            #     user_ref.update({"audio": [{file.filename: destination_file_name}]})
+                    user_ref.update({"audio": [{originalFileName: destination_file_name, "processed": [dest_processed_file]}]})
         return {'message': 'Files uploaded successfully'}, 200
     else:
         print(u'No such document!')
         return {'message': 'No such document'}, 400
 
 
 # Get Audio
-
-
 @app.route('/retrieve_audio', methods=['POST'])
 def retrieve_audio():
     fileName = request.json['fileName']
@@ -208,7 +206,78 @@ def retrieve_audio():
     print("Generated GET signed URL:")
     return url
 
+@app.route('/delete_unprocessed_audio', methods=['DELETE'])
+def delete_unprocessed_audio():
+    # In firebase, under audio, there is a list of uploaded audio,
+    # each uploaded audio has two entries: the original audio file name and "processed".
+    # The original audio file is by itself like "name":"address", 
+    # while the "processed" is a list containing the names of the processed audio files like "processed":"[address1, address2, address2]".
+
+    cloud_storage_filename = request.json['cloudStorageFileName']
+    auth_header = request.json['Authorization']
+
+    # delete the audio from cloud storage
+    blob = bucket.blob(cloud_storage_filename)
+    blob.delete()
+
+    # delete the audio from firebase, and the associated processed audios from cloud storage
+    idtoken = auth_header.split(' ').pop()
+    claims = id_token.verify_firebase_token(
+        idtoken, HTTP_REQUEST, audience=os.environ.get('GOOGLE_CLOUD_PROJECT'))
+    user_ref = users_collection.document(claims['sub'])
+
+    doc = user_ref.get()
+    if doc.exists:
+        doc = doc.to_dict()
+        new_audios = []
+        for audio in doc["audio"]:
+            if cloud_storage_filename not in list(audio.values()):
+                new_audios.append(audio)
+            else:
+                # delete the processed audios from cloud storage
+                for processed_audio in audio["processed"]:
+                    blob = bucket.blob(processed_audio)
+                    blob.delete()
+
+        user_ref.update({'audio': new_audios})
+        return {'message': "Success"}, 200
+    else:
+        print(u'No such document!')
+        return {'message': 'No such document'}, 400
+
 
+@app.route('/delete_processed_audio', methods=['DELETE'])
+def delete_processed_audio():
+    # get request data
+    cloud_storage_filename = request.json['cloudStorageFileName']
+    auth_header = request.json['Authorization']
+
+    # delete the audio from cloud storage
+    blob = bucket.blob(cloud_storage_filename)
+    blob.delete()
+
+    # delete the audio from firebase, and the associated processed audios from cloud storage
+    idtoken = auth_header.split(' ').pop()
+    claims = id_token.verify_firebase_token(
+        idtoken, HTTP_REQUEST, audience=os.environ.get('GOOGLE_CLOUD_PROJECT'))
+    user_ref = users_collection.document(claims['sub'])
+
+    doc = user_ref.get()
+    if doc.exists:        
+        doc = doc.to_dict()
+        for audio in doc["audio"]:
+            if cloud_storage_filename in audio['processed']:
+                audio['processed'].remove(cloud_storage_filename)
+
+        user_ref.update({'audio': doc['audio']})
+        return {'message': 'File deleted successfully'}, 200
+
+    else:
+        print(u'No such document!')
+        return {'message': 'No such document'}, 400
+
+
+
 port = int(os.environ.get('PORT', 8080))
 if __name__ == '__main__':
-    app.run(threaded=True, host='0.0.0.0', port=port)
+    app.run(threaded=True, host='0.0.0.0', port=port, debug=True)

backend/audio.py

@@ -21,8 +21,8 @@ def deadSpace(filePath):
 
     sample_rate = 48000
 
-    prepPath = '/tmp/' + re.split('.wav|.WAV', (filePath.split('/tmp/')[1]))[0] + 'Edited.WAV'
-
+    #prepPath = '/tmp/' + re.split('.wav|.WAV', (filePath.split('/tmp/')[1]))[0] + 'Edited.WAV' # the original way that append "Edited" each chunk
+    prepPath = '/tmp/' + re.split('.wav|.WAV', (filePath.split('/tmp/')[1]))[0] + '.WAV'
     # Preparation of wav file for vad function
     # pySeg = AudioSegment.from_wav(filePath).set_channels(1).set_frame_rate(sample_rate).export(prepPath, format='wav')
     pySeg = AudioSegment.from_file(filePath).set_channels(1).set_frame_rate(sample_rate).export(prepPath, format='wav')
@@ -40,7 +40,8 @@ def deadSpace(filePath):
 
     # Writes voice audio chunks + appends valid files
     for i, segment in enumerate(segments):
-        path = prepPath.split('.WAV')[0] + 'chunk-'+ str(i) + '.WAV'
+        # path = prepPath.split('.WAV')[0] + 'chunk-'+ str(i) + '.WAV' # the original way that indexes each chunk
+        path = prepPath.split('.WAV')[0] + '.WAV'
         print(' Writing %s' % (path))
         vadfuncs.write_wave(path, segment, sample_rate)
 

backend/requirements.txt

@@ -5,5 +5,7 @@ google-auth
 google-cloud-firestore
 google-cloud-storage
 pydub
+ffprobe
+ffmpeg
 webrtcvad
 

client/babel.config.json

@@ -0,0 +1,4 @@
+// Reference: https://forum.codewithmosh.com/t/7-social-tests-still-referenceerror-regeneratorruntime-is-not-defined-resolved/1978
+{
+    "plugins": ["@babel/plugin-transform-runtime"]
+}

client/myModuleResolver.js

@@ -0,0 +1,19 @@
+// my-module-resolve.js
+module.exports = (request, options) => {
+    // Call the defaultResolver, so we leverage its cache, error handling, etc.
+    return options.defaultResolver(request, {
+      ...options,
+      // Use packageFilter to process parsed `package.json` before the resolution (see https://www.npmjs.com/package/resolve#resolveid-opts-cb)
+      packageFilter: pkg => {
+        if(pkg.name.startsWith('@firebase')) {
+          return {
+            ...pkg,
+            // Alter the value of `main` before resolving the package
+            main: pkg.esm5 || pkg.module,
+          };
+        }
+
+        return pkg;
+      },
+    });
+  };