GameSentenceMiner 2.18.8__tar.gz → 2.18.9__tar.gz

{gamesentenceminer-2.18.8 → gamesentenceminer-2.18.9}/GameSentenceMiner/ui/anki_confirmation.py

@@ -1,14 +1,18 @@
 import tkinter as tk
 from tkinter import scrolledtext
+from tkinter import messagebox
 from PIL import Image, ImageTk
 
 import ttkbootstrap as ttk
-from GameSentenceMiner.util.configuration import get_config, logger, gsm_state
+from GameSentenceMiner.util.configuration import get_config, logger, gsm_state, get_temporary_directory
 from GameSentenceMiner.util.audio_player import AudioPlayer
+from GameSentenceMiner.util.gsm_utils import make_unique_file_name
 
 import platform
 import subprocess
 import os
+import requests
+from urllib.parse import quote
 
 class AnkiConfirmationDialog(tk.Toplevel):
     """
@@ -20,13 +24,18 @@ class AnkiConfirmationDialog(tk.Toplevel):
         self.screenshot_timestamp = screenshot_timestamp
         self.translation_text = None
         self.sentence_text = None
+        self.sentence = sentence  # Store sentence text for TTS
         
         # Initialize screenshot_path here, will be updated by button if needed
-        self.screenshot_path = screenshot_path 
+        self.screenshot_path = screenshot_path
+        self.audio_path = audio_path  # Store audio path so it can be updated
 
         # Audio player management
         self.audio_player = AudioPlayer(finished_callback=self._audio_finished)
         self.audio_button = None  # Store reference to audio button
+        self.audio_path_label = None  # Store reference to audio path label
+        self.tts_button = None  # Store reference to TTS button
+        self.tts_status_label = None  # Store reference to TTS status label
 
         self.title("Confirm Anki Card Details")
         self.result = None  # This will store the user's choice
@@ -123,18 +132,36 @@ class AnkiConfirmationDialog(tk.Toplevel):
         # Audio Path
         if audio_path and os.path.isfile(audio_path):
             ttk.Label(main_frame, text="Audio Path:", font=("-weight bold")).grid(row=row, column=0, sticky="ne", padx=5, pady=2)
-            ttk.Label(main_frame, text=audio_path if audio_path else "No Audio", wraplength=400, justify="left").grid(row=row, column=1, sticky="w", padx=5, pady=2)
+            self.audio_path_label = ttk.Label(main_frame, text=audio_path if audio_path else "No Audio", wraplength=400, justify="left")
+            self.audio_path_label.grid(row=row, column=1, sticky="w", padx=5, pady=2)
             if audio_path and os.path.isfile(audio_path):
                 self.audio_button = ttk.Button(
-                    main_frame, 
-                    text="▶", 
-                    command=lambda: self._play_audio(audio_path),
+                    main_frame,
+                    text="▶",
+                    command=lambda: self._play_audio(self.audio_path),
                     bootstyle="outline-info",
                     width=12
                 )
                 self.audio_button.grid(row=row, column=2, sticky="w", padx=5, pady=2)
 
             row += 1
+            
+            # TTS Button - only show if TTS is enabled in config
+            if get_config().vad.use_tts_as_fallback and sentence:
+                self.tts_button = ttk.Button(
+                    main_frame,
+                    text="🔊 Generate TTS Audio",
+                    command=self._generate_tts_audio,
+                    bootstyle="info",
+                    width=20
+                )
+                self.tts_button.grid(row=row, column=1, sticky="w", padx=5, pady=2)
+                
+                # TTS Status Label
+                self.tts_status_label = ttk.Label(main_frame, text="", foreground="green")
+                self.tts_status_label.grid(row=row, column=2, sticky="w", padx=5, pady=2)
+                
+                row += 1
 
         # Action Buttons
         button_frame = ttk.Frame(main_frame)
@@ -215,6 +242,72 @@ class AnkiConfirmationDialog(tk.Toplevel):
             else:
                 self.audio_button.config(text="▶ Play Audio", bootstyle="outline-info")
         
+    def _generate_tts_audio(self):
+        """Generate TTS audio from the sentence text"""
+        try:
+            # Get the current sentence text from the widget
+            sentence_text = self.sentence_text.get("1.0", tk.END).strip()
+            
+            if not sentence_text:
+                messagebox.showerror("TTS Error", "No sentence text available for TTS generation.")
+                return
+            
+            # URL-encode the sentence text
+            encoded_text = quote(sentence_text)
+            
+            # Build the TTS URL by replacing $s with the encoded text
+            tts_url = get_config().vad.tts_url.replace("$s", encoded_text)
+            
+            logger.info(f"Fetching TTS audio from: {tts_url}")
+            
+            # Fetch TTS audio from the URL
+            response = requests.get(tts_url, timeout=10)
+            
+            if not response.ok:
+                error_msg = f"Failed to fetch TTS audio: HTTP {response.status_code}"
+                logger.error(error_msg)
+                messagebox.showerror("TTS Error", f"{error_msg}\n\nIs your TTS service running?")
+                return
+            
+            # Save TTS audio to GSM temporary directory with game name
+            game_name = gsm_state.current_game if gsm_state.current_game else "tts"
+            filename = f"{game_name}_tts_audio.opus"
+            tts_audio_path = make_unique_file_name(
+                os.path.join(get_temporary_directory(), filename)
+            )
+            with open(tts_audio_path, 'wb') as f:
+                f.write(response.content)
+            
+            logger.info(f"TTS audio saved to: {tts_audio_path}")
+            
+            # Update the audio path
+            self.audio_path = tts_audio_path
+            
+            # Update the audio path label
+            if self.audio_path_label:
+                self.audio_path_label.config(text=tts_audio_path)
+            
+            # Update the audio button command to use the new path
+            if self.audio_button:
+                self.audio_button.config(command=lambda: self._play_audio(self.audio_path))
+            
+            # Update status label to show success
+            if self.tts_status_label:
+                self.tts_status_label.config(text="✓ TTS Audio Generated", foreground="green")
+            
+        except requests.exceptions.Timeout:
+            error_msg = "TTS request timed out. Please check if your TTS service is running."
+            logger.error(error_msg)
+            messagebox.showerror("TTS Error", error_msg)
+        except requests.exceptions.RequestException as e:
+            error_msg = f"Failed to connect to TTS service: {str(e)}"
+            logger.error(error_msg)
+            messagebox.showerror("TTS Error", f"{error_msg}\n\nPlease check your TTS URL configuration.")
+        except Exception as e:
+            error_msg = f"Unexpected error generating TTS: {str(e)}"
+            logger.error(error_msg)
+            messagebox.showerror("TTS Error", error_msg)
+    
     def _cleanup_audio(self):
         """Clean up audio stream resources"""
         self.audio_player.cleanup()

{gamesentenceminer-2.18.8 → gamesentenceminer-2.18.9}/GameSentenceMiner/util/get_overlay_coords.py

@@ -106,10 +106,10 @@ class OverlayThread(threading.Thread):
         while True:
             if overlay_server_thread.has_clients():
                 if get_config().overlay.periodic:
-                    await self.overlay_processor.find_box_and_send_to_overlay('')
+                    await self.overlay_processor.find_box_and_send_to_overlay('', True)
                     await asyncio.sleep(get_config().overlay.periodic_interval)
                 elif self.first_time_run:
-                    await self.overlay_processor.find_box_and_send_to_overlay('')
+                    await self.overlay_processor.find_box_and_send_to_overlay('', False)
                     self.first_time_run = False
                 else:
                     await asyncio.sleep(3)
@@ -161,7 +161,7 @@ class OverlayProcessor:
             self.lens = None
             self.regex = None
             
-    async def find_box_and_send_to_overlay(self, sentence_to_check: str = None):
+    async def find_box_and_send_to_overlay(self, sentence_to_check: str = None, check_against_last: bool = False):
         """
         Sends the detected text boxes to the overlay via WebSocket.
         Cancels any running OCR task before starting a new one.
@@ -175,7 +175,7 @@ class OverlayProcessor:
                 logger.info("Previous OCR task was cancelled")
         
         # Start new task
-        self.current_task = asyncio.create_task(self.find_box_for_sentence(sentence_to_check))
+        self.current_task = asyncio.create_task(self.find_box_for_sentence(sentence_to_check, check_against_last))
         try:
             await self.current_task
         except asyncio.CancelledError:
@@ -183,7 +183,7 @@ class OverlayProcessor:
         # logger.info(f"Sending {len(boxes)} boxes to overlay.")
         # await send_word_coordinates_to_overlay(boxes)
 
-    async def find_box_for_sentence(self, sentence_to_check: str = None) -> List[Dict[str, Any]]:
+    async def find_box_for_sentence(self, sentence_to_check: str = None, check_against_last: bool = False) -> List[Dict[str, Any]]:
         """
         Public method to perform OCR and find text boxes for a given sentence.
         
@@ -191,7 +191,7 @@ class OverlayProcessor:
         error handling.
         """
         try:
-            return await self._do_work(sentence_to_check)
+            return await self._do_work(sentence_to_check, check_against_last=check_against_last)
         except Exception as e:
             logger.error(f"Error during OCR processing: {e}", exc_info=True)
             return []
@@ -304,7 +304,7 @@ class OverlayProcessor:
         
         return composite_img
 
-    async def _do_work(self, sentence_to_check: str = None) -> Tuple[List[Dict[str, Any]], int]:
+    async def _do_work(self, sentence_to_check: str = None, check_against_last: bool = False) -> Tuple[List[Dict[str, Any]], int]:
         """The main OCR workflow with cancellation support."""
         if not self.lens:
             logger.error("OCR engines are not initialized. Cannot perform OCR for Overlay.")
@@ -348,7 +348,7 @@ class OverlayProcessor:
             text_str = "".join([text for text in text if self.regex.match(text)])
             
             # RapidFuzz fuzzy match 90% to not send the same results repeatedly
-            if self.last_oneocr_result:
+            if self.last_oneocr_result and check_against_last:
                 
                 score = fuzz.ratio(text_str, self.last_oneocr_result)
                 if score >= 80:
@@ -403,7 +403,7 @@ class OverlayProcessor:
         text_str = "".join([text for text in text_list if self.regex.match(text)])
         
         # RapidFuzz fuzzy match 90% to not send the same results repeatedly
-        if self.last_lens_result:
+        if self.last_lens_result and check_against_last:
             score = fuzz.ratio(text_str, self.last_lens_result)
             if score >= 80:
                 logger.info("Google Lens results are similar to the last results (score: %d). Skipping overlay update.", score)
@@ -596,7 +596,7 @@ async def main_run_ocr():
     """
     overlay_processor = OverlayProcessor()
     while True:
-        await overlay_processor.find_box_and_send_to_overlay('')
+        await overlay_processor.find_box_and_send_to_overlay('', False)
         await asyncio.sleep(10)
 
 

{gamesentenceminer-2.18.8 → gamesentenceminer-2.18.9}/GameSentenceMiner.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: GameSentenceMiner
-Version: 2.18.8
+Version: 2.18.9
 Summary: A tool for mining sentences from games. Update: Overlay?
 Author-email: Beangate <bpwhelan95@gmail.com>
 License: MIT License

{gamesentenceminer-2.18.8 → gamesentenceminer-2.18.9}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: GameSentenceMiner
-Version: 2.18.8
+Version: 2.18.9
 Summary: A tool for mining sentences from games. Update: Overlay?
 Author-email: Beangate <bpwhelan95@gmail.com>
 License: MIT License

{gamesentenceminer-2.18.8 → gamesentenceminer-2.18.9}/pyproject.toml

@@ -7,7 +7,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "GameSentenceMiner"
-version = "2.18.8"
+version = "2.18.9"
 description = "A tool for mining sentences from games. Update: Overlay?"
 readme = "README.md"
 requires-python = ">=3.10"