Upload 48 files

app.py

@@ -0,0 +1,13 @@
+import streamlit as st
+import home
+import youtube
+
+PAGES = {
+    "Home": home,
+    "Youtube": youtube
+}
+st.sidebar.title('Navigation')
+
+selection = st.sidebar.selectbox("Go to", list(PAGES.keys()))
+page = PAGES[selection]
+page.app()

helper.py

@@ -0,0 +1,57 @@
+import re
+from dotenv import load_dotenv
+import os
+from googleapiclient.discovery import build
+
+load_dotenv()
+
+api_key = os.getenv('API_KEY')
+youtube = build('youtube', 'v3', developerKey=api_key)
+
+def get_all_comments(video_id):
+    comments = []
+    next_page_token = None
+
+    while True:
+        # Make API call to get comments
+        request = youtube.commentThreads().list(
+            part='snippet',
+            videoId=video_id,
+            textFormat='plainText',
+            pageToken=next_page_token  # Use pagination token for next set of comments
+        )
+        
+        # Execute the request
+        response = request.execute()
+
+        # Loop through the comments in the response
+        for item in response['items']:
+            comment = item['snippet']['topLevelComment']['snippet']['textDisplay']
+            author = item['snippet']['topLevelComment']['snippet']['authorDisplayName']
+            timestamp = item['snippet']['topLevelComment']['snippet']['publishedAt']
+            like_count = item['snippet']['topLevelComment']['snippet']['likeCount']
+            comments.append({
+                'author': author.strip(),
+                'comment': comment.strip(),
+                'timestamp': timestamp.strip(),
+                'like_count': like_count,
+            })
+        
+        # Check if there's another page of comments (pagination)
+        next_page_token = response.get('nextPageToken')
+
+        if not next_page_token or len(comments) >= 100:  # If no more pages, break the loop
+          break
+
+    return comments
+
+def extract_youtube_id(url_or_id):
+    pattern = r'(?:v=|\/)([a-zA-Z0-9_-]{11})(?:&|$)?'
+    
+    if re.fullmatch(r'[a-zA-Z0-9_-]{11}', url_or_id):
+        return url_or_id
+    
+    match = re.search(pattern, url_or_id)
+    if match:
+        return match.group(1)
+    return None  

home.py

@@ -0,0 +1,48 @@
+import streamlit as st
+from wordcloud import WordCloud
+import pandas as pd
+import matplotlib.pyplot as plt
+from PIL import Image
+import os
+import re
+import pkg_resources
+import streamlit as st
+
+# Get all installed packages and their versions
+packages = {dist.project_name: dist.version for dist in pkg_resources.working_set}
+
+# Display the packages in Streamlit
+
+
+def app():
+    st.header('Phone Brands Sentiment Analysis', divider='rainbow')
+    data = pd.read_csv("final_dataset_text_processed.csv")
+    data.dropna(inplace=True)
+    val = st.sidebar.pills("Choose Phone Word Cloud To Show", data["tipe_produk"].unique(), selection_mode="multi")
+    st.write("Installed Packages and Versions:")
+    st.json(packages)
+    for the_value in val:
+      the_product = data[data["tipe_produk"] == the_value]
+      if not the_product.empty:
+        text_data = " ".join(data["text_processed"])
+        wordcloud = WordCloud(width=800, height=400, background_color='white').generate(text_data)
+        regex = re.compile(r"^(.*)\.[^.]+$")
+        col1, col2 = st.columns([3, 7])
+        image = None
+        
+        for item in os.listdir("images"):
+          item_cleaned = regex.match(item).group(1)
+          if item_cleaned.lower().strip() == the_value.lower().strip():
+            image = Image.open(f"./images/{item}")
+            break
+        with col1:
+          if image:
+            st.write(the_value)
+            st.image(image)
+        plt.title(f"Wordcloud for phone {the_value}")
+        plt.figure(figsize=(10, 5))
+        plt.imshow(wordcloud, interpolation='bilinear')
+        plt.axis("off")
+        with col2:
+          st.write('<p style="visibility: hidden;">This is some text.</p>', unsafe_allow_html=True)
+          st.pyplot(plt)

requirements.txt

@@ -0,0 +1,13 @@
+wordcloud==1.9.4
+streamlit==1.40.1
+seaborn==0.13.2
+scikit-learn==1.5.2
+numpy==1.26.4
+nltk==3.9.1
+matplotlib==3.9.2
+joblib==1.4.2
+google-api-python-client==2.153.0
+google-auth==2.36.0
+google-auth-httplib2==0.2.0
+google-api-core==2.23.0
+googleapis-common-protos==1.66.0

youtube.py

@@ -0,0 +1,22 @@
+from helper import extract_youtube_id, get_all_comments
+import streamlit as st
+
+def app():
+  st.header('Phone Brands Sentiment Analysis', divider='rainbow')
+  user_input = st.text_input("Enter a youtube link for sentiment analysis")
+  if st.button('Submit', type="secondary"):
+    try:
+      the_youtube_id = extract_youtube_id(user_input)
+      if the_youtube_id:
+        with st.spinner("Please wait while we're loading the data..."):
+          the_data = get_all_comments(the_youtube_id)
+          st.write(f"Total Comments: {len(the_data)}")
+          for data in the_data:
+            st.write(data["comment"])
+      else:
+        st.write("Invalid youtube link.")
+    except:
+      st.write("Invalid youtube link.")
+  else:
+    st.write("Click the button to submit your link!")
+