chore: Remove legacy chunk size configuration and clean up debug code

- Removed MAIN_LINES_PER_CHUNK parameter (replaced by token-based chunking)
- Changed language detection to auto-detect source from files and target from browser
- Removed deprecated chunk_size from CLI args, API config, and TranslationConfig
- Cleaned up 3 debug print statements in config_routes and translation_routes
- Updated output filename format to use parentheses style (file (lang).ext)
- Removed unused configuration validation code in CLI

Author: hydropix

.env.example

@@ -12,7 +12,7 @@ OUTPUT_DIR=translated_files  # Directory for translated output files
 LLM_PROVIDER=ollama
 # Your Google Gemini API key (required if using gemini provider)
 GEMINI_API_KEY=
-GEMINI_MODEL=gemini-3-flash
+GEMINI_MODEL=gemini-2.0-flash
 # Your OpenAI API key (required if using openai provider)
 OPENAI_API_KEY=
 
@@ -23,9 +23,8 @@ OPENROUTER_MODEL=anthropic/claude-4.5-haiku
 # See all models at https://openrouter.ai/models (text-only models are auto-filtered in the UI)
 
 # Translation Settings
-DEFAULT_SOURCE_LANGUAGE=English  # Default source language (can be any language name)
-DEFAULT_TARGET_LANGUAGE=Chinese  # Default target language (can be any language name)
-MAIN_LINES_PER_CHUNK=50  # Legacy parameter - kept for SRT grouping only (not used for text chunking)
+# Source language: Auto-detected from uploaded file (uses langdetect library)
+# Target language: Auto-detected from browser language (can be changed in UI)
 MAIN_CHUNK_SIZE=1000      # Maximum characters per chunk
 REQUEST_TIMEOUT=900       # API timeout in seconds
 
@@ -51,19 +50,6 @@ AUTO_ADJUST_CONTEXT=true  # Automatically adjust context/chunk size if prompt to
 # Advanced
 MAX_TRANSLATION_ATTEMPTS=3
 
-# =============================================================================
-# EPUB Translation - Token Alignment Fallback (Phase 2)
-# =============================================================================
-# When the LLM fails to preserve HTML placeholders correctly during EPUB
-# translation, this fallback uses word-level alignment to reinsert them at
-# semantically correct positions. This dramatically improves translation
-# quality by ensuring 100% placeholder integrity.
-#
-# Multi-phase fallback system:
-#   Phase 1: Normal translation with placeholders (retry logic)
-#   Phase 2: Token alignment fallback (translate clean, then reinsert)
-#   Phase 3: Untranslated fallback (preserve original if all else fails)
-
 EPUB_TOKEN_ALIGNMENT_ENABLED=true
 # Options: true (enable Phase 2 fallback), false (use old behavior with only Phase 1 + Phase 3)
 
@@ -76,18 +62,18 @@ EPUB_TOKEN_ALIGNMENT_METHOD=proportional
 SRT_LINES_PER_BLOCK=20
 SRT_MAX_CHARS_PER_BLOCK=2000
 
-# Translation Signature
-# This adds a discrete attribution to your translations (metadata for EPUB, footer for TXT, comment for SRT)
-# Please consider keeping this enabled to support the project and help others discover this free tool!
-# The signature is non-intrusive and placed at the end of files. Thank you for your support!
-SIGNATURE_ENABLED=true
-
 # Debug Mode
 # Enable verbose logging for troubleshooting configuration and connection issues.
 # Set to 'true' to see detailed logs about .env loading, API calls, and configuration values.
 # Useful when models are not detected or languages are not applied correctly.
 DEBUG_MODE=false
 
+# Translation Signature
+# This adds a discrete attribution to your translations (metadata for EPUB, footer for TXT, comment for SRT)
+# Please consider keeping this enabled to support the project and help others discover this free tool!
+# The signature is non-intrusive and placed at the end of files. Thank you for your support!
+SIGNATURE_ENABLED=true
+
 # TTS (Text-to-Speech) Configuration
 # Generate audio narration of translated documents
 # Requires ffmpeg installed on the system for Opus encoding
@@ -145,3 +131,5 @@ TTS_CFG_WEIGHT=0.5
 # cs (Czech), ar (Arabic), zh (Chinese), ja (Japanese), hu (Hungarian),
 # ko (Korean), hi (Hindi), vi (Vietnamese), sv (Swedish), da (Danish),
 # fi (Finnish), id (Indonesian), el (Greek)
+
+

DOCKER.md

@@ -102,7 +102,7 @@ Docker automatically pulls the correct architecture for your system.
 | `PORT` | Web server port | `5000` |
 | `OLLAMA_NUM_CTX` | Context window size | `2048` |
 | `REQUEST_TIMEOUT` | API timeout (seconds) | `900` |
-| `MAIN_LINES_PER_CHUNK` | Lines per translation chunk | `25` |
+| `MAX_TOKENS_PER_CHUNK` | Tokens per translation chunk | `400` |
 | `SIGNATURE_ENABLED` | Add signature to translations | `true` |
 
 ## Volume Mounts

README.md

@@ -64,32 +64,32 @@ See [docs/PROVIDERS.md](docs/PROVIDERS.md) for detailed setup instructions.
 ## Command Line
 
 ```bash
-# Basic
-python translate.py -i book.epub -o book_zh.epub -sl English -tl Chinese
+# Basic (auto-generates "book (Chinese).epub")
+python translate.py -i book.epub -sl English -tl Chinese
 
 # With OpenRouter
-python translate.py -i book.txt -o book_fr.txt --provider openrouter \
-    --openrouter_api_key YOUR_KEY -m anthropic/claude-sonnet-4
+python translate.py -i book.txt --provider openrouter \
+    --openrouter_api_key YOUR_KEY -m anthropic/claude-sonnet-4 -tl French
 
 # With OpenAI
-python translate.py -i book.txt -o book_fr.txt --provider openai \
-    --openai_api_key YOUR_KEY -m gpt-4o
+python translate.py -i book.txt --provider openai \
+    --openai_api_key YOUR_KEY -m gpt-4o -tl French
 
 # With Gemini
-python translate.py -i book.txt -o book_fr.txt --provider gemini \
-    --gemini_api_key YOUR_KEY -m gemini-2.0-flash
+python translate.py -i book.txt --provider gemini \
+    --gemini_api_key YOUR_KEY -m gemini-2.0-flash -tl French
 
 # With local OpenAI-compatible server (llama.cpp, LM Studio, vLLM, etc.)
-python translate.py -i book.txt -o book_fr.txt --provider openai \
-    --api_endpoint http://localhost:8080/v1/chat/completions -m your-model
+python translate.py -i book.txt --provider openai \
+    --api_endpoint http://localhost:8080/v1/chat/completions -m your-model -tl French
 ```
 
 ### Main Options
 
 | Option | Description | Default |
 |--------|-------------|---------|
 | `-i, --input` | Input file | Required |
-| `-o, --output` | Output file | Auto |
+| `-o, --output` | Output file | Auto: `{name} ({lang}).{ext}` |
 | `-sl, --source_lang` | Source language | English |
 | `-tl, --target_lang` | Target language | Chinese |
 | `-m, --model` | Model name | mistral-small:24b |
@@ -117,8 +117,8 @@ OPENAI_API_KEY=sk-...
 GEMINI_API_KEY=...
 
 # Performance
-MAIN_LINES_PER_CHUNK=25
 REQUEST_TIMEOUT=900
+MAX_TOKENS_PER_CHUNK=400  # Token-based chunking (default: 400 tokens)
 ```
 
 ---

deployment/.env.docker.example

@@ -47,7 +47,6 @@ DEFAULT_SOURCE_LANGUAGE=English
 DEFAULT_TARGET_LANGUAGE=Chinese
 
 # Chunking configuration
-MAIN_LINES_PER_CHUNK=25
 MAIN_CHUNK_SIZE=1000
 
 # API request timeout (in seconds)

deployment/docker-compose.yml

@@ -36,7 +36,6 @@ services:
       # Translation Settings
       - DEFAULT_SOURCE_LANGUAGE=${DEFAULT_SOURCE_LANGUAGE:-English}
       - DEFAULT_TARGET_LANGUAGE=${DEFAULT_TARGET_LANGUAGE:-Chinese}
-      - MAIN_LINES_PER_CHUNK=${MAIN_LINES_PER_CHUNK:-25}
       - MAIN_CHUNK_SIZE=${MAIN_CHUNK_SIZE:-1000}
       - REQUEST_TIMEOUT=${REQUEST_TIMEOUT:-900}
 

docs/CLI.md

@@ -24,7 +24,7 @@ python translate.py -i input_file -o output_file
 
 | Option | Description | Default |
 |--------|-------------|---------|
-| `-o, --output` | Output file path | Auto-generated |
+| `-o, --output` | Output file path | Auto-generated as `{original} ({target_lang}).{ext}` |
 
 ### Languages
 
@@ -70,14 +70,17 @@ python translate.py -i input_file -o output_file
 ### Basic Translation
 
 ```bash
-# Text file
-python translate.py -i book.txt -o book_fr.txt -sl English -tl French
+# Text file (auto-generates "book (French).txt")
+python translate.py -i book.txt -sl English -tl French
 
-# Subtitles
-python translate.py -i movie.srt -o movie_fr.srt -tl French
+# Subtitles (auto-generates "movie (French).srt")
+python translate.py -i movie.srt -tl French
 
-# EPUB
-python translate.py -i novel.epub -o novel_fr.epub -tl French
+# EPUB (auto-generates "novel (French).epub")
+python translate.py -i novel.epub -tl French
+
+# Custom output filename
+python translate.py -i book.txt -o my_custom_name.txt -tl French
 ```
 
 ### With Different Providers
@@ -142,8 +145,8 @@ OPENAI_API_KEY=sk-...
 GEMINI_API_KEY=...
 
 # Performance
-MAIN_LINES_PER_CHUNK=25
 REQUEST_TIMEOUT=900
+MAX_TOKENS_PER_CHUNK=400  # Token-based chunking (default: 400 tokens)
 
 # Languages
 DEFAULT_SOURCE_LANGUAGE=English

docs/DOCKER_DEPLOYMENT.md

@@ -539,7 +539,7 @@ For larger chunks and better translations:
 
 ```env
 OLLAMA_NUM_CTX=16384
-MAIN_LINES_PER_CHUNK=50
+MAX_TOKENS_PER_CHUNK=800  # Larger token chunks for better context
 ```
 
 ### Adjust Timeouts

docs/TROUBLESHOOTING.md

@@ -43,7 +43,7 @@ Solutions to common problems with TBL.
 **Cause**: Chunk is too large for model's context window.
 
 **Solutions**:
-1. Reduce chunk size: `-cs 15` or `MAIN_LINES_PER_CHUNK=15`
+1. Reduce chunk size: `MAX_TOKENS_PER_CHUNK=200` (default: 400)
 2. Increase context window: `OLLAMA_NUM_CTX=8192`
 
 ---
@@ -56,7 +56,7 @@ Solutions to common problems with TBL.
 
 **Solutions**:
 1. Increase timeout: `REQUEST_TIMEOUT=1800` (30 min)
-2. Reduce chunk size: `MAIN_LINES_PER_CHUNK=15`
+2. Reduce chunk size: `MAX_TOKENS_PER_CHUNK=200` (default: 400)
 3. Try a smaller model
 4. Try a cloud provider
 

src/api/blueprints/config_routes.py

@@ -22,7 +22,6 @@ def get_config_path():
 from src.config import (
     API_ENDPOINT as DEFAULT_OLLAMA_API_ENDPOINT,
     DEFAULT_MODEL,
-    MAIN_LINES_PER_CHUNK,
     REQUEST_TIMEOUT,
     OLLAMA_NUM_CTX,
     MAX_TRANSLATION_ATTEMPTS,
@@ -113,7 +112,6 @@ def mask_api_key(key):
         config_response = {
             "api_endpoint": DEFAULT_OLLAMA_API_ENDPOINT,
             "default_model": DEFAULT_MODEL,
-            "chunk_size": MAIN_LINES_PER_CHUNK,
             "timeout": REQUEST_TIMEOUT,
             "context_window": OLLAMA_NUM_CTX,
             "max_attempts": MAX_TRANSLATION_ATTEMPTS,
@@ -124,15 +122,12 @@ def mask_api_key(key):
             "openrouter_api_key": mask_api_key(OPENROUTER_API_KEY),
             "gemini_api_key_configured": bool(GEMINI_API_KEY),
             "openai_api_key_configured": bool(OPENAI_API_KEY),
-            "openrouter_api_key_configured": bool(OPENROUTER_API_KEY),
-            "default_source_language": DEFAULT_SOURCE_LANGUAGE,
-            "default_target_language": DEFAULT_TARGET_LANGUAGE
+            "openrouter_api_key_configured": bool(OPENROUTER_API_KEY)
+            # Languages are no longer sent from server - handled by browser detection
         }
 
         if DEBUG_MODE:
             logger.debug(f"📤 /api/config response:")
-            logger.debug(f"   default_source_language: {DEFAULT_SOURCE_LANGUAGE}")
-            logger.debug(f"   default_target_language: {DEFAULT_TARGET_LANGUAGE}")
             logger.debug(f"   api_endpoint: {DEFAULT_OLLAMA_API_ENDPOINT}")
             logger.debug(f"   default_model: {DEFAULT_MODEL}")
 
@@ -207,7 +202,6 @@ def _get_openrouter_models(provided_api_key=None):
                 })
 
         except Exception as e:
-            print(f"❌ Error retrieving OpenRouter models: {e}")
             return jsonify({
                 "models": [],
                 "model_names": [],
@@ -353,7 +347,6 @@ def _get_gemini_models(provided_api_key=None):
                 })
 
         except Exception as e:
-            print(f"❌ Error retrieving Gemini models: {e}")
             return jsonify({
                 "models": [],
                 "default": default_model,
@@ -558,9 +551,9 @@ def save_settings():
             'OPENROUTER_MODEL',
             'DEFAULT_MODEL',
             'LLM_PROVIDER',
-            'DEFAULT_SOURCE_LANGUAGE',
-            'DEFAULT_TARGET_LANGUAGE',
             'API_ENDPOINT'
+            # DEFAULT_SOURCE_LANGUAGE and DEFAULT_TARGET_LANGUAGE removed
+            # Languages are now auto-detected (source) and browser-detected (target)
         }
 
         try:
@@ -607,9 +600,8 @@ def get_settings():
             "openrouter_api_key_configured": bool(OPENROUTER_API_KEY),
             "default_model": DEFAULT_MODEL or "",
             "llm_provider": os.getenv('LLM_PROVIDER', 'ollama'),
-            "default_source_language": DEFAULT_SOURCE_LANGUAGE or "English",
-            "default_target_language": DEFAULT_TARGET_LANGUAGE or "Chinese",
             "api_endpoint": DEFAULT_OLLAMA_API_ENDPOINT or ""
+            # Languages are no longer stored in .env - auto-detected per session
         })
 
     return bp

src/api/blueprints/translation_routes.py

@@ -7,7 +7,6 @@
 from flask import Blueprint, request, jsonify
 
 from src.config import (
-    MAIN_LINES_PER_CHUNK,
     REQUEST_TIMEOUT,
     OLLAMA_NUM_CTX
 )
@@ -64,15 +63,11 @@ def start_translation_request():
         # Generate unique translation ID
         translation_id = f"trans_{int(time.time() * 1000)}"
 
-        # Debug: Log received prompt_options
-        print(f"[DEBUG] Received prompt_options: {data.get('prompt_options', {})}")
-
         # Build configuration
         config = {
             'source_language': data['source_language'],
             'target_language': data['target_language'],
             'model': data['model'],
-            'chunk_size': int(data.get('chunk_size', MAIN_LINES_PER_CHUNK)),
             'llm_api_endpoint': data['llm_api_endpoint'],
             'request_timeout': int(data.get('timeout', REQUEST_TIMEOUT)),
             'context_window': int(data.get('context_window', OLLAMA_NUM_CTX)),