CosmickVisions commited on
Commit
9c82681
·
verified ·
1 Parent(s): cdae590

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +109 -415
app.py CHANGED
@@ -1,31 +1,21 @@
1
- # Standard library imports
 
2
  import os
3
  import tempfile
4
  import uuid
 
 
 
 
 
5
  import base64
 
6
  import io
 
7
  import json
8
  import re
9
  from datetime import datetime, timedelta
10
 
11
- # Third-party imports
12
- import gradio as gr
13
- import groq
14
- import numpy as np
15
- import pandas as pd
16
- import openpyxl
17
- import requests
18
- import fitz # PyMuPDF
19
- from PIL import Image
20
- from dotenv import load_dotenv
21
- from transformers import AutoProcessor, AutoModelForVision2Seq
22
- import torch
23
-
24
- # LangChain imports
25
- from langchain_community.embeddings import HuggingFaceEmbeddings
26
- from langchain_community.vectorstores import FAISS
27
- from langchain.text_splitter import RecursiveCharacterTextSplitter
28
-
29
  # Load environment variables
30
  load_dotenv()
31
  client = groq.Client(api_key=os.getenv("GROQ_TECH_API_KEY"))
@@ -80,102 +70,6 @@ body { background-color: var(--light-background); font-family: 'Google Sans', 'R
80
  .qa-body { color: var(--dark-text); font-size: 0.95rem; margin-bottom: 10px; }
81
  .qa-meta { display: flex; justify-content: space-between; color: #5F6368; font-size: 0.85rem; }
82
  .tag { background-color: #E8F0FE; color: var(--primary-color); padding: 4px 8px; border-radius: 4px; font-size: 0.8rem; margin-right: 5px; display: inline-block; }
83
- .toggle-container { display: flex; align-items: center; margin-bottom: 15px; }
84
- .toggle-label { margin-right: 10px; font-weight: 500; }
85
- .search-toggle { margin-left: 5px; }
86
- .voice-btn { background-color: var(--primary-color) !important; border-radius: 50% !important; width: 44px !important; height: 44px !important; display: flex !important; align-items: center !important; justify-content: center !important; color: var(--white) !important; box-shadow: 0 2px 5px rgba(0,0,0,0.2) !important; }
87
- .speak-btn { background-color: var(--secondary-color) !important; border-radius: 24px !important; color: var(--white) !important; padding: 8px 16px !important; font-weight: 500 !important; margin-left: 10px !important; }
88
- .audio-controls { display: flex; align-items: center; margin-top: 10px; }
89
-
90
- /* Audio Visualization Elements */
91
- .audio-visualization {
92
- display: flex;
93
- align-items: center;
94
- justify-content: center;
95
- gap: 4px;
96
- height: 40px;
97
- padding: 10px;
98
- background-color: rgba(0,0,0,0.05);
99
- border-radius: 12px;
100
- margin: 10px 0;
101
- }
102
-
103
- .audio-bar {
104
- width: 3px;
105
- background-color: var(--accent-color);
106
- border-radius: 2px;
107
- height: 5px;
108
- transition: height 0.1s ease;
109
- }
110
-
111
- .audio-status {
112
- font-size: 0.85rem;
113
- color: var(--secondary-color);
114
- text-align: center;
115
- margin-top: 5px;
116
- font-style: italic;
117
- }
118
-
119
- .recording-indicator {
120
- width: 12px;
121
- height: 12px;
122
- border-radius: 50%;
123
- background-color: #ff4b4b;
124
- margin-right: 8px;
125
- animation: blink 1s infinite;
126
- }
127
-
128
- .playing-indicator {
129
- width: 12px;
130
- height: 12px;
131
- border-radius: 50%;
132
- background-color: #4bff4b;
133
- margin-right: 8px;
134
- animation: pulse 1s infinite;
135
- }
136
-
137
- @keyframes blink {
138
- 0% { opacity: 1; }
139
- 50% { opacity: 0.4; }
140
- 100% { opacity: 1; }
141
- }
142
-
143
- @keyframes pulse {
144
- 0% { transform: scale(1); }
145
- 50% { transform: scale(1.2); }
146
- 100% { transform: scale(1); }
147
- }
148
-
149
- .file-upload-enhancement .file-preview {
150
- max-height: 200px;
151
- overflow: auto;
152
- border: 1px solid var(--border-color);
153
- border-radius: 8px;
154
- padding: 10px;
155
- margin-top: 10px;
156
- background-color: rgba(0,0,0,0.02);
157
- }
158
-
159
- .excel-preview-table {
160
- width: 100%;
161
- border-collapse: collapse;
162
- font-size: 0.85rem;
163
- }
164
-
165
- .excel-preview-table th, .excel-preview-table td {
166
- border: 1px solid #ddd;
167
- padding: 4px 8px;
168
- text-align: left;
169
- }
170
-
171
- .excel-preview-table th {
172
- background-color: var(--secondary-color);
173
- color: white;
174
- }
175
-
176
- .excel-preview-table tr:nth-child(even) {
177
- background-color: rgba(0,0,0,0.03);
178
- }
179
  """
180
 
181
  # Function to process PDF files
@@ -215,147 +109,8 @@ def process_pdf(pdf_file):
215
  os.unlink(pdf_path)
216
  return None, f"Error processing PDF: {str(e)}", {"page_images": [], "total_pages": 0, "total_words": 0}
217
 
218
- # New function to process Excel files
219
- def process_excel(excel_file):
220
- if excel_file is None:
221
- return None, "No file uploaded", {"data_preview": "", "total_sheets": 0, "total_rows": 0}
222
-
223
- try:
224
- session_id = str(uuid.uuid4())
225
- with tempfile.NamedTemporaryFile(suffix=".xlsx", delete=False) as temp_file:
226
- temp_file.write(excel_file)
227
- excel_path = temp_file.name
228
-
229
- # Read Excel file with pandas
230
- excel_data = pd.ExcelFile(excel_path)
231
- sheet_names = excel_data.sheet_names
232
- all_texts = []
233
- total_rows = 0
234
-
235
- # Process each sheet
236
- for sheet in sheet_names:
237
- df = pd.read_excel(excel_path, sheet_name=sheet)
238
- total_rows += len(df)
239
-
240
- # Convert dataframe to text for vectorization
241
- sheet_text = f"Sheet: {sheet}\n"
242
- sheet_text += df.to_string(index=False)
243
- all_texts.append(sheet_text)
244
-
245
- # Generate HTML preview of first sheet
246
- first_df = pd.read_excel(excel_path, sheet_name=0)
247
- preview_rows = min(10, len(first_df))
248
- data_preview = first_df.head(preview_rows).to_html(classes="excel-preview-table", index=False)
249
-
250
- # Process for vectorstore
251
- text_splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=200)
252
- chunks = text_splitter.create_documents(all_texts)
253
- vectorstore = FAISS.from_documents(chunks, embeddings)
254
- index_path = os.path.join(FAISS_INDEX_DIR, session_id)
255
- vectorstore.save_local(index_path)
256
- user_vectorstores[session_id] = vectorstore
257
-
258
- os.unlink(excel_path)
259
- excel_state = {"data_preview": data_preview, "total_sheets": len(sheet_names), "total_rows": total_rows}
260
- return session_id, f"✅ Successfully processed {len(chunks)} text chunks from Excel file", excel_state
261
- except Exception as e:
262
- if "excel_path" in locals() and os.path.exists(excel_path):
263
- os.unlink(excel_path)
264
- return None, f"Error processing Excel file: {str(e)}", {"data_preview": "", "total_sheets": 0, "total_rows": 0}
265
-
266
- # Function to analyze image using SmolDocling
267
- def analyze_image(image_file):
268
- """
269
- Basic image analysis function that doesn't rely on external models
270
- """
271
- if image_file is None:
272
- return "No image uploaded. Please upload an image to analyze."
273
-
274
- try:
275
- image = Image.open(image_file)
276
- width, height = image.size
277
- format = image.format
278
- mode = image.mode
279
-
280
- analysis = f"""## Technical Document Analysis
281
-
282
- **Image Properties:**
283
- - Dimensions: {width}x{height} pixels
284
- - Format: {format}
285
- - Color Mode: {mode}
286
-
287
- **Technical Analysis:**
288
- 1. Document Quality:
289
- - Resolution: {'High' if width > 2000 or height > 2000 else 'Medium' if width > 1000 or height > 1000 else 'Low'}
290
- - Color Depth: {mode}
291
-
292
- 2. Recommendations:
293
- - For text extraction, consider using PDF format
294
- - For technical diagrams, ensure high resolution
295
- - Consider OCR for text content
296
-
297
- **Note:** For detailed technical analysis, please convert to PDF format
298
- """
299
- return analysis
300
- except Exception as e:
301
- return f"Error analyzing image: {str(e)}\n\nPlease try using PDF format instead."
302
-
303
- # Function to handle different file types
304
- def process_file(file_data, file_type):
305
- if file_data is None:
306
- return None, "No file uploaded", None
307
-
308
- if file_type == "pdf":
309
- return process_pdf(file_data)
310
- elif file_type == "excel":
311
- return process_excel(file_data)
312
- elif file_type == "image":
313
- # For image files, we'll just use them directly for analysis
314
- # But we'll return a session ID to maintain consistency
315
- session_id = str(uuid.uuid4())
316
- return session_id, "✅ Image file ready for analysis", None
317
- else:
318
- return None, "Unsupported file type", None
319
-
320
- # Function for speech-to-text conversion
321
- def speech_to_text():
322
- try:
323
- r = sr.Recognizer()
324
- with sr.Microphone() as source:
325
- r.adjust_for_ambient_noise(source)
326
- audio = r.listen(source)
327
- text = r.recognize_google(audio)
328
- return text
329
- except sr.UnknownValueError:
330
- return "Could not understand audio. Please try again."
331
- except sr.RequestError as e:
332
- return f"Error with speech recognition service: {e}"
333
- except Exception as e:
334
- return f"Error converting speech to text: {str(e)}"
335
-
336
- # Function for text-to-speech conversion
337
- def text_to_speech(text, history):
338
- if not text or not history:
339
- return None
340
-
341
- try:
342
- # Get the last bot response
343
- last_response = history[-1][1]
344
-
345
- # Convert text to speech
346
- tts = pyttsx3.init()
347
- tts.setProperty('rate', 150)
348
- tts.setProperty('volume', 0.9)
349
- tts.save_to_file(last_response, "temp_output.mp3")
350
- tts.runAndWait()
351
-
352
- return "temp_output.mp3"
353
- except Exception as e:
354
- print(f"Error in text-to-speech: {e}")
355
- return None
356
-
357
  # Function to generate chatbot responses with Tech theme
358
- def generate_response(message, session_id, model_name, history, web_search_enabled=True):
359
  if not message:
360
  return history
361
  try:
@@ -366,8 +121,8 @@ def generate_response(message, session_id, model_name, history, web_search_enabl
366
  if docs:
367
  context = "\n\nRelevant information from uploaded PDF:\n" + "\n".join(f"- {doc.page_content}" for doc in docs)
368
 
369
- # Check if it's a GitHub repo search and web search is enabled
370
- if web_search_enabled and re.match(r'^/github\s+.+', message, re.IGNORECASE):
371
  query = re.sub(r'^/github\s+', '', message, flags=re.IGNORECASE)
372
  repo_results = search_github_repos(query)
373
  if repo_results:
@@ -384,8 +139,8 @@ def generate_response(message, session_id, model_name, history, web_search_enabl
384
  history.append((message, "No GitHub repositories found for your query."))
385
  return history
386
 
387
- # Check if it's a Stack Overflow search and web search is enabled
388
- if web_search_enabled and re.match(r'^/stack\s+.+', message, re.IGNORECASE):
389
  query = re.sub(r'^/stack\s+', '', message, flags=re.IGNORECASE)
390
  qa_results = search_stackoverflow(query)
391
  if qa_results:
@@ -682,155 +437,106 @@ def perform_stack_search(query, tag, sort_by):
682
  with gr.Blocks(css=custom_css, theme=gr.themes.Soft()) as demo:
683
  current_session_id = gr.State(None)
684
  pdf_state = gr.State({"page_images": [], "total_pages": 0, "total_words": 0})
685
- excel_state = gr.State({"data_preview": "", "total_sheets": 0, "total_rows": 0})
686
- file_type = gr.State("none")
687
- audio_status = gr.State("Ready")
688
-
689
  gr.HTML("""
690
  <div class="header">
691
- <div class="header-title">Tech-Vision Enhanced</div>
692
- <div class="header-subtitle">Analyze technical documents, spreadsheets, and images with AI</div>
693
  </div>
694
  """)
695
  with gr.Row(elem_classes="container"):
696
  with gr.Column(scale=1, min_width=300):
697
- with gr.Tabs():
698
- with gr.TabItem("PDF"):
699
- pdf_file = gr.File(label="Upload PDF Document", file_types=[".pdf"], type="binary")
700
- pdf_upload_button = gr.Button("Process PDF", variant="primary")
701
-
702
- with gr.TabItem("Excel"):
703
- excel_file = gr.File(label="Upload Excel File", file_types=[".xlsx", ".xls"], type="binary")
704
- excel_upload_button = gr.Button("Process Excel", variant="primary")
705
-
706
- with gr.TabItem("Image"):
707
- image_input = gr.File(
708
- label="Upload Image",
709
- file_types=["image"],
710
- type="filepath"
711
- )
712
- analyze_btn = gr.Button("Analyze Image")
713
-
714
- file_status = gr.Markdown("No file uploaded yet")
715
-
716
- # Model selector
717
  model_dropdown = gr.Dropdown(
718
  choices=["llama3-70b-8192", "llama3-8b-8192", "mixtral-8x7b-32768", "gemma-7b-it"],
719
  value="llama3-70b-8192",
720
  label="Select Groq Model"
721
  )
722
-
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
723
  with gr.Column(scale=2, min_width=600):
724
  with gr.Tabs():
725
  with gr.TabItem("PDF Viewer"):
726
  with gr.Column(elem_classes="pdf-viewer-container"):
727
  page_slider = gr.Slider(minimum=1, maximum=1, step=1, label="Page Number", value=1)
728
  pdf_image = gr.Image(label="PDF Page", type="pil", elem_classes="pdf-viewer-image")
729
- pdf_stats = gr.Markdown("No PDF uploaded yet", elem_classes="stats-box")
730
 
731
- with gr.TabItem("Excel Viewer"):
732
- excel_preview = gr.HTML(label="Excel Preview", elem_classes="file-preview")
733
- excel_stats = gr.Markdown("No Excel file uploaded yet", elem_classes="stats-box")
734
 
735
- with gr.TabItem("Image Analysis"):
736
- image_preview = gr.Image(label="Image Preview", type="pil")
737
- image_analysis_results = gr.Markdown("Upload an image and click 'Analyze Image' to see analysis results")
 
 
738
 
739
- # Audio visualization elements
740
- with gr.Row(elem_classes="container"):
741
- with gr.Column():
742
- audio_vis = gr.HTML("""
743
- <div class="audio-visualization">
744
- <div class="audio-bar" style="height: 5px;"></div>
745
- <div class="audio-bar" style="height: 12px;"></div>
746
- <div class="audio-bar" style="height: 18px;"></div>
747
- <div class="audio-bar" style="height: 15px;"></div>
748
- <div class="audio-bar" style="height: 10px;"></div>
749
- <div class="audio-bar" style="height: 20px;"></div>
750
- <div class="audio-bar" style="height: 14px;"></div>
751
- <div class="audio-bar" style="height: 8px;"></div>
752
- </div>
753
- """, visible=False)
754
- audio_status_display = gr.Markdown("", elem_classes="audio-status")
755
-
756
- # Chat interface
757
  with gr.Row(elem_classes="container"):
758
  with gr.Column(scale=2, min_width=600):
759
- chatbot = gr.Chatbot(
760
- height=400,
761
- show_copy_button=True,
762
- elem_classes="chat-container",
763
- type="messages" # Use the new messages format
764
- )
765
  with gr.Row():
766
- msg = gr.Textbox(
767
- show_label=False,
768
- placeholder="Ask about your document or click the microphone to speak...",
769
- scale=5
770
- )
771
- voice_btn = gr.Button("🎤", elem_classes="voice-btn")
772
  send_btn = gr.Button("Send", scale=1)
773
-
774
- with gr.Row(elem_classes="audio-controls"):
775
- clear_btn = gr.Button("Clear Conversation")
776
- speak_btn = gr.Button("🔊 Speak Response", elem_classes="speak-btn")
777
- audio_player = gr.Audio(label="Response Audio", type="filepath", visible=False)
778
 
779
- # Event Handlers for PDF processing
780
- pdf_upload_button.click(
781
- lambda x: ("pdf", x),
782
- inputs=[pdf_file],
783
- outputs=[file_type, file_status]
784
- ).then(
785
  process_pdf,
786
  inputs=[pdf_file],
787
- outputs=[current_session_id, file_status, pdf_state]
788
  ).then(
789
  update_pdf_viewer,
790
  inputs=[pdf_state],
791
- outputs=[page_slider, pdf_image, pdf_stats]
792
- )
793
-
794
- # Event Handlers for Excel processing
795
- def update_excel_preview(state):
796
- if not state:
797
- return "", "No Excel file uploaded yet"
798
- preview = state.get("data_preview", "")
799
- sheets = state.get("total_sheets", 0)
800
- rows = state.get("total_rows", 0)
801
- stats = f"**Excel Statistics:**\nSheets: {sheets}\nTotal Rows: {rows}"
802
- return preview, stats
803
-
804
- excel_upload_button.click(
805
- lambda x: ("excel", x),
806
- inputs=[excel_file],
807
- outputs=[file_type, file_status]
808
- ).then(
809
- process_excel,
810
- inputs=[excel_file],
811
- outputs=[current_session_id, file_status, excel_state]
812
- ).then(
813
- update_excel_preview,
814
- inputs=[excel_state],
815
- outputs=[excel_preview, excel_stats]
816
- )
817
-
818
- # Event Handlers for Image Analysis
819
- analyze_btn.click(
820
- lambda x: ("image", x),
821
- inputs=[image_input],
822
- outputs=[file_type, file_status]
823
- ).then(
824
- analyze_image,
825
- inputs=[image_input],
826
- outputs=[image_analysis_results]
827
- ).then(
828
- lambda x: Image.open(x) if x else None,
829
- inputs=[image_input],
830
- outputs=[image_preview]
831
  )
832
 
833
- # Chat message handling
834
  msg.submit(
835
  generate_response,
836
  inputs=[msg, current_session_id, model_dropdown, chatbot],
@@ -843,55 +549,43 @@ with gr.Blocks(css=custom_css, theme=gr.themes.Soft()) as demo:
843
  outputs=[chatbot]
844
  ).then(lambda: "", None, [msg])
845
 
846
- # Improved speech-to-text with visual feedback
847
- voice_btn.click(
848
- speech_to_text,
849
- inputs=[audio_status],
850
- outputs=[audio_status_display, audio_vis, msg]
851
- )
852
-
853
- # Improved text-to-speech with visual feedback
854
- speak_btn.click(
855
- text_to_speech,
856
- inputs=[audio_status, chatbot],
857
- outputs=[audio_status_display, audio_vis, audio_player]
858
- ).then(
859
- lambda x: gr.update(visible=True) if x else gr.update(visible=False),
860
- inputs=[audio_player],
861
- outputs=[audio_player]
862
  )
863
 
864
- # Page navigation for PDF
865
  page_slider.change(
866
  update_image,
867
  inputs=[page_slider, pdf_state],
868
  outputs=[pdf_image]
869
  )
870
 
871
- # Clear conversation and reset UI
872
- clear_btn.click(
873
- lambda: (
874
- [], None, "No file uploaded yet",
875
- {"page_images": [], "total_pages": 0, "total_words": 0},
876
- {"data_preview": "", "total_sheets": 0, "total_rows": 0},
877
- "none", 0, None, "No PDF uploaded yet", "",
878
- "No Excel file uploaded yet", None,
879
- "Upload an image and click 'Analyze Image' to see results", None,
880
- gr.update(visible=False), "Ready"
881
- ),
882
- None,
883
- [chatbot, current_session_id, file_status, pdf_state, excel_state,
884
- file_type, page_slider, pdf_image, pdf_stats, excel_preview,
885
- excel_stats, image_preview, image_analysis_results, audio_player,
886
- audio_vis, audio_status_display]
 
887
  )
888
 
889
- # Add footer with creator attribution
890
- gr.HTML("""
891
- <div style="text-align: center; margin-top: 20px; padding: 10px; color: #666; font-size: 0.8rem; border-top: 1px solid #eee;">
892
- Created by Calvin Allen Crawford
893
- </div>
894
- """)
895
 
896
  # Launch the app
897
  if __name__ == "__main__":
 
1
+ import gradio as gr
2
+ import groq
3
  import os
4
  import tempfile
5
  import uuid
6
+ from dotenv import load_dotenv
7
+ from langchain.text_splitter import RecursiveCharacterTextSplitter
8
+ from langchain.vectorstores import FAISS
9
+ from langchain.embeddings import HuggingFaceEmbeddings
10
+ import fitz # PyMuPDF
11
  import base64
12
+ from PIL import Image
13
  import io
14
+ import requests
15
  import json
16
  import re
17
  from datetime import datetime, timedelta
18
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
19
  # Load environment variables
20
  load_dotenv()
21
  client = groq.Client(api_key=os.getenv("GROQ_TECH_API_KEY"))
 
70
  .qa-body { color: var(--dark-text); font-size: 0.95rem; margin-bottom: 10px; }
71
  .qa-meta { display: flex; justify-content: space-between; color: #5F6368; font-size: 0.85rem; }
72
  .tag { background-color: #E8F0FE; color: var(--primary-color); padding: 4px 8px; border-radius: 4px; font-size: 0.8rem; margin-right: 5px; display: inline-block; }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
73
  """
74
 
75
  # Function to process PDF files
 
109
  os.unlink(pdf_path)
110
  return None, f"Error processing PDF: {str(e)}", {"page_images": [], "total_pages": 0, "total_words": 0}
111
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
112
  # Function to generate chatbot responses with Tech theme
113
+ def generate_response(message, session_id, model_name, history):
114
  if not message:
115
  return history
116
  try:
 
121
  if docs:
122
  context = "\n\nRelevant information from uploaded PDF:\n" + "\n".join(f"- {doc.page_content}" for doc in docs)
123
 
124
+ # Check if it's a GitHub repo search
125
+ if re.match(r'^/github\s+.+', message, re.IGNORECASE):
126
  query = re.sub(r'^/github\s+', '', message, flags=re.IGNORECASE)
127
  repo_results = search_github_repos(query)
128
  if repo_results:
 
139
  history.append((message, "No GitHub repositories found for your query."))
140
  return history
141
 
142
+ # Check if it's a Stack Overflow search
143
+ if re.match(r'^/stack\s+.+', message, re.IGNORECASE):
144
  query = re.sub(r'^/stack\s+', '', message, flags=re.IGNORECASE)
145
  qa_results = search_stackoverflow(query)
146
  if qa_results:
 
437
  with gr.Blocks(css=custom_css, theme=gr.themes.Soft()) as demo:
438
  current_session_id = gr.State(None)
439
  pdf_state = gr.State({"page_images": [], "total_pages": 0, "total_words": 0})
 
 
 
 
440
  gr.HTML("""
441
  <div class="header">
442
+ <div class="header-title">Tech-Vision</div>
443
+ <div class="header-subtitle">Analyze technical documents with Groq's LLM API.</div>
444
  </div>
445
  """)
446
  with gr.Row(elem_classes="container"):
447
  with gr.Column(scale=1, min_width=300):
448
+ pdf_file = gr.File(label="Upload PDF Document", file_types=[".pdf"], type="binary")
449
+ upload_button = gr.Button("Process PDF", variant="primary")
450
+ pdf_status = gr.Markdown("No PDF uploaded yet")
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
451
  model_dropdown = gr.Dropdown(
452
  choices=["llama3-70b-8192", "llama3-8b-8192", "mixtral-8x7b-32768", "gemma-7b-it"],
453
  value="llama3-70b-8192",
454
  label="Select Groq Model"
455
  )
456
+
457
+ # Tech Tools Section
458
+ gr.Markdown("### Developer Tools", elem_classes="tool-title")
459
+ with gr.Group(elem_classes="tool-container"):
460
+ with gr.Tabs():
461
+ with gr.TabItem("GitHub Search"):
462
+ repo_query = gr.Textbox(label="Search Query", placeholder="Enter keywords to search for repositories")
463
+ with gr.Row():
464
+ language = gr.Dropdown(
465
+ choices=["any", "JavaScript", "Python", "Java", "C++", "TypeScript", "Go", "Rust", "PHP", "C#"],
466
+ value="any",
467
+ label="Language"
468
+ )
469
+ min_stars = gr.Dropdown(
470
+ choices=["0", "10", "50", "100", "1000", "10000"],
471
+ value="0",
472
+ label="Min Stars"
473
+ )
474
+ sort_by = gr.Dropdown(
475
+ choices=["stars", "forks", "updated"],
476
+ value="stars",
477
+ label="Sort By"
478
+ )
479
+ repo_search_btn = gr.Button("Search Repositories")
480
+
481
+ with gr.TabItem("Stack Overflow"):
482
+ stack_query = gr.Textbox(label="Search Query", placeholder="Enter your technical question")
483
+ with gr.Row():
484
+ tag = gr.Dropdown(
485
+ choices=["any", "python", "javascript", "java", "c++", "react", "node.js", "android", "ios", "sql"],
486
+ value="any",
487
+ label="Tag"
488
+ )
489
+ so_sort_by = gr.Dropdown(
490
+ choices=["votes", "newest", "activity"],
491
+ value="votes",
492
+ label="Sort By"
493
+ )
494
+ so_search_btn = gr.Button("Search Stack Overflow")
495
+
496
+ with gr.TabItem("Code Explainer"):
497
+ code_input = gr.Textbox(
498
+ label="Code to Explain",
499
+ placeholder="Paste your code here...",
500
+ lines=10
501
+ )
502
+ explain_btn = gr.Button("Explain Code")
503
+
504
  with gr.Column(scale=2, min_width=600):
505
  with gr.Tabs():
506
  with gr.TabItem("PDF Viewer"):
507
  with gr.Column(elem_classes="pdf-viewer-container"):
508
  page_slider = gr.Slider(minimum=1, maximum=1, step=1, label="Page Number", value=1)
509
  pdf_image = gr.Image(label="PDF Page", type="pil", elem_classes="pdf-viewer-image")
510
+ stats_display = gr.Markdown("No PDF uploaded yet", elem_classes="stats-box")
511
 
512
+ with gr.TabItem("GitHub Results"):
513
+ repo_results = gr.Markdown("Search for repositories to see results here")
 
514
 
515
+ with gr.TabItem("Stack Overflow Results"):
516
+ stack_results = gr.Markdown("Search for questions to see results here")
517
+
518
+ with gr.TabItem("Code Explanation"):
519
+ code_explanation = gr.Markdown("Paste your code and click 'Explain Code' to see an explanation here")
520
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
521
  with gr.Row(elem_classes="container"):
522
  with gr.Column(scale=2, min_width=600):
523
+ chatbot = gr.Chatbot(height=500, bubble_full_width=False, show_copy_button=True, elem_classes="chat-container")
 
 
 
 
 
524
  with gr.Row():
525
+ msg = gr.Textbox(show_label=False, placeholder="Ask about your document, type /github to search repos, or /stack to search Stack Overflow...", scale=5)
 
 
 
 
 
526
  send_btn = gr.Button("Send", scale=1)
527
+ clear_btn = gr.Button("Clear Conversation")
 
 
 
 
528
 
529
+ # Event Handlers
530
+ upload_button.click(
 
 
 
 
531
  process_pdf,
532
  inputs=[pdf_file],
533
+ outputs=[current_session_id, pdf_status, pdf_state]
534
  ).then(
535
  update_pdf_viewer,
536
  inputs=[pdf_state],
537
+ outputs=[page_slider, pdf_image, stats_display]
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
538
  )
539
 
 
540
  msg.submit(
541
  generate_response,
542
  inputs=[msg, current_session_id, model_dropdown, chatbot],
 
549
  outputs=[chatbot]
550
  ).then(lambda: "", None, [msg])
551
 
552
+ clear_btn.click(
553
+ lambda: ([], None, "No PDF uploaded yet", {"page_images": [], "total_pages": 0, "total_words": 0}, 0, None, "No PDF uploaded yet"),
554
+ None,
555
+ [chatbot, current_session_id, pdf_status, pdf_state, page_slider, pdf_image, stats_display]
 
 
 
 
 
 
 
 
 
 
 
 
556
  )
557
 
 
558
  page_slider.change(
559
  update_image,
560
  inputs=[page_slider, pdf_state],
561
  outputs=[pdf_image]
562
  )
563
 
564
+ # Tech tool handlers
565
+ repo_search_btn.click(
566
+ perform_repo_search,
567
+ inputs=[repo_query, language, sort_by, min_stars],
568
+ outputs=[repo_results]
569
+ )
570
+
571
+ so_search_btn.click(
572
+ perform_stack_search,
573
+ inputs=[stack_query, tag, so_sort_by],
574
+ outputs=[stack_results]
575
+ )
576
+
577
+ explain_btn.click(
578
+ explain_code,
579
+ inputs=[code_input],
580
+ outputs=[code_explanation]
581
  )
582
 
583
+ # Add footer with attribution
584
+ gr.HTML("""
585
+ <div style="text-align: center; margin-top: 20px; padding: 10px; color: #666; font-size: 0.8rem; border-top: 1px solid #eee;">
586
+ Created by Calvin Allen Crawford
587
+ </div>
588
+ """)
589
 
590
  # Launch the app
591
  if __name__ == "__main__":