GameSentenceMiner 2.16.4__tar.gz → 2.16.5__tar.gz

{gamesentenceminer-2.16.4 → gamesentenceminer-2.16.5}/GameSentenceMiner/config_gui.py

@@ -440,6 +440,7 @@ class ConfigApp:
         default_category_config = getattr(self.default_settings, category)
 
         setattr(self.settings, category, default_category_config)
+        self.create_vars()  # Recreate variables to reflect default values
         recreate_tab()
         self.save_settings(profile_change=False)
         self.reload_settings()
@@ -1146,6 +1147,9 @@ class ConfigApp:
         HoverInfoLabelWidget(vad_frame, text=use_cpu_i18n.get('label', 'Force CPU'), tooltip=use_cpu_i18n.get('tooltip', 'Even if CUDA is installed, use CPU for Whisper'), row=self.current_row, column=0)
         ttk.Checkbutton(vad_frame, variable=self.use_cpu_for_inference_value, bootstyle="round-toggle").grid(row=self.current_row, column=1, sticky='W', pady=2)
         self.current_row += 1
+        
+        # Add Reset Button
+        self.add_reset_button(vad_frame, "vad", self.current_row, column=0, recreate_tab=self.create_vad_tab)
 
     @new_tab
     def create_paths_tab(self):

{gamesentenceminer-2.16.4 → gamesentenceminer-2.16.5}/GameSentenceMiner/util/ffmpeg.py

@@ -453,7 +453,7 @@ def trim_audio_based_on_last_line(untrimmed_audio, video_path, game_line, next_l
     ffmpeg_command = ffmpeg_base_command_list + [
         "-i", untrimmed_audio,
         "-ss", str(start_trim_time)]
-    if next_line and next_line > game_line.time:
+    if next_line and next_line > game_line.time and total_seconds:
         end_trim_seconds = total_seconds + (next_line - game_line.time).total_seconds() + get_config().audio.pre_vad_end_offset
         end_trim_time = f"{end_trim_seconds:.3f}"
         ffmpeg_command.extend(['-to', end_trim_time])
@@ -495,9 +495,9 @@ def get_video_timings(video_path, game_line, anki_card_creation_time=None):
     total_seconds = file_length - time_delta.total_seconds()
     total_seconds_after_offset = total_seconds + get_config().audio.beginning_offset
     if total_seconds < 0 or total_seconds >= file_length:
-        logger.error("Line mined is outside of the replay buffer! Defaulting to the beginning of the replay buffer. ")
+        logger.error("Line mined is outside of the replay buffer! Defaulting to the last 30 seconds of the replay buffer.")
         logger.info("Recommend either increasing replay buffer length in OBS Settings or mining faster.")
-        return 0, 0, 0, file_length
+        return max(file_length - 30, 0), 0, max(file_length - 30, 0), file_length
 
     return total_seconds_after_offset, total_seconds, total_seconds_after_offset, file_length
 

{gamesentenceminer-2.16.4 → gamesentenceminer-2.16.5}/GameSentenceMiner/web/database_api.py

@@ -1,11 +1,14 @@
+import copy
 import datetime
 import re
 import csv
 import io
 from collections import defaultdict
+import time
 
 import flask
 from flask import request, jsonify
+import regex
 
 from GameSentenceMiner.util.db import GameLinesTable
 from GameSentenceMiner.util.configuration import logger, get_config, save_current_config
@@ -685,6 +688,7 @@ def register_database_api_routes(app):
         Provides aggregated, cumulative stats for charting.
         Accepts optional 'year' parameter to filter heatmap data.
         """
+        punctionation_regex = regex.compile(r'[\p{P}\p{S}]')
         # Get optional year filter parameter
         filter_year = request.args.get('year', None)
         
@@ -697,13 +701,25 @@ def register_database_api_routes(app):
         # 2. Process data into daily totals for each game
         # Structure: daily_data[date_str][game_name] = {'lines': N, 'chars': N}
         daily_data = defaultdict(lambda: defaultdict(lambda: {'lines': 0, 'chars': 0}))
+        
+        
+        # start_time = time.perf_counter()
+        # for line in all_lines:
+        #     day_str = datetime.date.fromtimestamp(float(line.timestamp)).strftime('%Y-%m-%d')
+        #     game = line.game_name or "Unknown Game"
+        #     daily_data[day_str][game]['lines'] += 1
+        #     daily_data[day_str][game]['chars'] += len(line.line_text) if line.line_text else 0
+        # end_time = time.perf_counter()
+        # logger.info(f"Without Punctuation removal and daily aggregation took {end_time - start_time:.4f} seconds for {len(all_lines)} lines")
 
         for line in all_lines:
             day_str = datetime.date.fromtimestamp(float(line.timestamp)).strftime('%Y-%m-%d')
             game = line.game_name or "Unknown Game"
-            
+            # Remove punctuation and symbols from line text before counting characters
+            clean_text = punctionation_regex.sub('', line.line_text) if line.line_text else ''
+            line.line_text = clean_text  # Update line text to cleaned version for future use
             daily_data[day_str][game]['lines'] += 1
-            daily_data[day_str][game]['chars'] += len(line.line_text) if line.line_text else 0
+            daily_data[day_str][game]['chars'] += len(clean_text)
 
         # 3. Create cumulative datasets for Chart.js
         sorted_days = sorted(daily_data.keys())

{gamesentenceminer-2.16.4 → gamesentenceminer-2.16.5}/GameSentenceMiner.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: GameSentenceMiner
-Version: 2.16.4
+Version: 2.16.5
 Summary: A tool for mining sentences from games. Update: Overlay?
 Author-email: Beangate <bpwhelan95@gmail.com>
 License: MIT License

{gamesentenceminer-2.16.4 → gamesentenceminer-2.16.5}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: GameSentenceMiner
-Version: 2.16.4
+Version: 2.16.5
 Summary: A tool for mining sentences from games. Update: Overlay?
 Author-email: Beangate <bpwhelan95@gmail.com>
 License: MIT License

{gamesentenceminer-2.16.4 → gamesentenceminer-2.16.5}/pyproject.toml

@@ -7,7 +7,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "GameSentenceMiner"
-version = "2.16.4"
+version = "2.16.5"
 description = "A tool for mining sentences from games. Update: Overlay?"
 readme = "README.md"
 requires-python = ">=3.10"