Fixed save

Resources/captions.jsonl

@@ -1,5 +1,5 @@
-{"clean_text": "measure", "start_time": 0.0, "user_id": 11, "signer": 39, "file": "Aj9SDSAOXf4", "end_time": 1.435, "url": "www.youtube.com/watch?v=Aj9SDSAOXf4", "text": "measure"}
-{"clean_text": "apple", "start_time": 1.5, "user_id": 11, "signer": 39, "file": "Aj9SDSAOXf4", "end_time": 1.6, "url": "www.youtube.com/watch?v=Aj9SDSAOXf4", "text": "apple"}
-{"clean_text": "test", "start_time": 0.0, "user_id": 11, "signer": 39, "file": "c2ORbHSQ5pw", "end_time": 1.435, "url": "www.youtube.com/watch?v=c2ORbHSQ5pw", "text": "test"}
-{"clean_text": "aaaaaaaaaa", "start_time": 1.5, "user_id": 11, "signer": 39, "file": "c2ORbHSQ5pw", "end_time": 1.6, "url": "www.youtube.com/watch?v=c2ORbHSQ5pw", "text": "aaaaaaaaaa"}
-{"clean_text": "bbbbbbbb", "start_time": 1.5, "user_id": 11, "signer": 39, "file": "c2ORbHSQ5pw", "end_time": 1.6, "url": "www.youtube.com/watch?v=c2ORbHSQ5pw", "text": "bbbbbbbb"}
+{"clean_text":"measure","start_time":0.0,"user_id":"a","signer":39,"file":"Aj9SDSAOXf4","end_time":1.435,"url":"www.youtube.com\/watch?v=Aj9SDSAOXf4","text":"measure"}
+{"clean_text":"apple","start_time":1.5,"user_id":"a","signer":39,"file":"Aj9SDSAOXf4","end_time":1.6,"url":"www.youtube.com\/watch?v=Aj9SDSAOXf4","text":"apple"}
+{"clean_text":"test","start_time":0.0,"user_id":"sample_user","signer":39,"file":"c2ORbHSQ5pw","end_time":1.435,"url":"www.youtube.com\/watch?v=c2ORbHSQ5pw","text":"test"}
+{"clean_text":"aaaaaaaaaa","start_time":1.5,"user_id":"sample_user","signer":39,"file":"c2ORbHSQ5pw","end_time":1.6,"url":"www.youtube.com\/watch?v=c2ORbHSQ5pw","text":"aaaaaaaaaa"}
+{"clean_text":"bbbbbbbb","start_time":1.5,"user_id":"sample_user","signer":39,"file":"c2ORbHSQ5pw","end_time":1.6,"url":"www.youtube.com\/watch?v=c2ORbHSQ5pw","text":"lllllllll"}

Resources/captions2.jsonl

@@ -1,5 +0,0 @@
-{"clean_text":"measure","start_time":0.0,"user_id":11,"signer":39,"file":"Aj9SDSAOXf4","end_time":1.435,"url":"www.youtube.com\/watch?v=Aj9SDSAOXf4","text":"test"}
-{"clean_text":"apple","start_time":1.5,"user_id":11,"signer":39,"file":"Aj9SDSAOXf4","end_time":1.6,"url":"www.youtube.com\/watch?v=Aj9SDSAOXf4","text":"hhhh"}
-{"clean_text":"test","start_time":null,"user_id":11,"signer":39,"file":"c2ORbHSQ5pw","end_time":null,"url":"www.youtube.com\/watch?v=c2ORbHSQ5pw","text":null}
-{"clean_text":"aaaaaaaaaa","start_time":null,"user_id":11,"signer":39,"file":"c2ORbHSQ5pw","end_time":null,"url":"www.youtube.com\/watch?v=c2ORbHSQ5pw","text":null}
-{"clean_text":"bbbbbbbb","start_time":null,"user_id":11,"signer":39,"file":"c2ORbHSQ5pw","end_time":null,"url":"www.youtube.com\/watch?v=c2ORbHSQ5pw","text":null}

Resources/captions_backup.jsonl

@@ -0,0 +1,5 @@
+{"clean_text": "measure", "start_time": 0.0, "user_id": 11, "signer": 39, "file": "Aj9SDSAOXf4", "end_time": 1.435, "url": "www.youtube.com/watch?v=Aj9SDSAOXf4", "text": "measure"}
+{"clean_text": "apple", "start_time": 1.5, "user_id": 11, "signer": 39, "file": "Aj9SDSAOXf4", "end_time": 1.6, "url": "www.youtube.com/watch?v=Aj9SDSAOXf4", "text": "apple"}
+{"clean_text": "test", "start_time": 0.0, "user_id": 11, "signer": 39, "file": "c2ORbHSQ5pw", "end_time": 1.435, "url": "www.youtube.com/watch?v=c2ORbHSQ5pw", "text": "test"}
+{"clean_text": "aaaaaaaaaa", "start_time": 1.5, "user_id": 11, "signer": 39, "file": "c2ORbHSQ5pw", "end_time": 1.6, "url": "www.youtube.com/watch?v=c2ORbHSQ5pw", "text": "aaaaaaaaaa"}
+{"clean_text": "bbbbbbbb", "start_time": 1.5, "user_id": 11, "signer": 39, "file": "c2ORbHSQ5pw", "end_time": 1.6, "url": "www.youtube.com/watch?v=c2ORbHSQ5pw", "text": "bbbbbbbb"}

app.py

@@ -3,7 +3,7 @@ import firebase_admin
 from firebase_admin import db
 from auth_functions import auth_function
 from video_player_functions import youtube_link_to_id, get_video_embed_by_id, get_video_link_by_pointer
-from caption_editor_functions import get_captions_by_video_id, save, captions
+from caption_editor_functions import get_captions_by_video_id, save_dataframe
 from Resources.css import css
 
 cred_obj = firebase_admin.credentials.Certificate('Resources/key.json')
@@ -14,12 +14,22 @@ videos_ref = db.reference("/Videos")
 users_ref = db.reference("/Users")
 
 next_video_pointer = 0
+user = ""
 
 
 def auth(username, password):
     return auth_function(username, password, db_ref=users_ref)
 
 
+def get_username(request: gr.Request):
+    global user
+    user = request.username
+
+
+def save(df, video_id):
+    return save_dataframe(df, video_id, user)
+
+
 def get_next_components():
     global next_video_pointer
     next_video_link = get_video_link_by_pointer(next_video_pointer, db_ref=videos_ref)
@@ -32,13 +42,15 @@ def get_next_components():
 
     next_video = get_video_embed_by_id(next_video_id)
     next_captions = get_captions_by_video_id(next_video_id)
-    return next_video, next_captions
+
+    return next_video, next_captions, next_video_id
 
 
-(start_video, start_captions) = get_next_components()
+(start_video, start_captions, start_video_id) = get_next_components()
 
 with gr.Blocks(css=css) as app:
     gr.Markdown("## Caption Editor")
+    current_video_id = gr.Textbox(value=start_video_id, visible=False, interactive=False)
     with gr.Row():
         with gr.Column():
             caption_editor = gr.DataFrame(interactive=True,
@@ -52,7 +64,11 @@ with gr.Blocks(css=css) as app:
             video_embed = gr.HTML(value=start_video)
             next_video_button = gr.Button("Next")
 
-    next_video_button.click(fn=get_next_components, outputs=[video_embed, caption_editor])
-    save_button.click(fn=save, inputs=caption_editor, outputs=save_result)
+    next_video_button.click(fn=get_next_components,
+                            outputs=[video_embed, caption_editor, current_video_id])
+    save_button.click(fn=save,
+                      inputs=[caption_editor, current_video_id],
+                      outputs=save_result)
+    app.load(get_username)
 
 app.launch(auth=auth)

caption_editor_functions.py

@@ -1,24 +1,39 @@
 import pandas as pd
-
-with open("Resources/captions.jsonl") as file:
-    captions = pd.read_json(file, lines=True)
+import numpy as np
 
 
 def get_captions_by_video_id(video_id):
-    global captions
+    with open("Resources/captions.jsonl") as file:
+        captions = pd.read_json(file, lines=True)
+
     captions_edit = captions[captions['file'] == video_id]
     captions_edit = captions_edit[['start_time', 'text', 'end_time']]
     captions_edit.columns = ["Start", "Text", "End"]
     return captions_edit
 
 
-def save(df):
+def save_dataframe(df, video_id, user):
     try:
-        global captions
-        captions['start_time'] = df['Start'].apply(lambda x: float(x))
-        captions['text'] = df['Text']
-        captions['end_time'] = df['End'].apply(lambda x: float(x))
-        captions.to_json('Resources/captions2.jsonl', orient='records', lines=True)
+        with open("Resources/captions.jsonl") as file:
+            captions = pd.read_json(file, lines=True)
+
+        other_captions = captions[captions['file'] != video_id].copy()
+        new_captions = captions[captions['file'] == video_id].copy()
+
+        new_captions['start_time'] = np.where(df['Start'].isnull(),
+                                              new_captions['start_time'],
+                                              df['Start'].apply(lambda x: float(x)))
+        new_captions['text'] = np.where(df['Text'].isnull(),
+                                        new_captions['text'],
+                                        df['Text'])
+        new_captions['end_time'] = np.where(df['End'].isnull(),
+                                            new_captions['end_time'],
+                                            df['End'].apply(lambda x: float(x)))
+        new_captions['user_id'] = user
+
+        all_captions = pd.concat([other_captions, new_captions], ignore_index=True)
+
+        all_captions.to_json('Resources/captions.jsonl', orient='records', lines=True)
         return "Save successful!"
     except ValueError:
         return "Save failed: Incorrect input format"