File size: 10,273 Bytes
8b21729
 
 
 
 
cfeb3a6
 
 
8b21729
cfeb3a6
8b21729
cfeb3a6
 
8b21729
 
cfeb3a6
 
8b21729
 
 
90b0a17
8b21729
 
 
 
 
 
 
 
 
 
 
 
 
cfeb3a6
8b21729
 
cfeb3a6
8b21729
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
cfeb3a6
 
8b21729
90b0a17
 
 
8b21729
 
 
17e3d1d
8b21729
 
 
 
 
 
 
 
 
 
 
 
 
 
90b0a17
8b21729
90b0a17
8b21729
 
 
 
 
90b0a17
 
 
8b21729
 
 
90b0a17
 
8b21729
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
90b0a17
8b21729
 
 
90b0a17
8b21729
 
90b0a17
8b21729
 
90b0a17
8b21729
 
 
 
 
90b0a17
 
8b21729
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
90b0a17
 
8b21729
90b0a17
 
 
 
 
 
 
 
 
 
8b21729
 
 
 
 
 
90b0a17
 
cfeb3a6
 
8b21729
cfeb3a6
8b21729
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
"""
Configuration settings for Data Extractor Using Gemini
Optimized for Gemini-only model usage with robust directory management
"""

import os
from pathlib import Path
from dotenv import load_dotenv
import logging

# Load environment variables
load_dotenv()

logger = logging.getLogger(__name__)


class Settings:
    """Configuration settings with Gemini-only model support and robust directory management."""
    
    # === GEMINI MODEL CONFIGURATION ===
    GOOGLE_API_KEY = os.getenv("GOOGLE_API_KEY")
    
    # Gemini model specifications - using gemini-2.5-flash (supports thinking budget)
    DATA_EXTRACTOR_MODEL = os.getenv("DATA_EXTRACTOR_MODEL", "gemini-2.5-pro")
    DATA_ARRANGER_MODEL = os.getenv("DATA_ARRANGER_MODEL", "gemini-2.5-pro") 
    CODE_GENERATOR_MODEL = os.getenv("CODE_GENERATOR_MODEL", "gemini-2.5-flash")
    
    # Thinking budgets optimized for each task type
    DATA_EXTRACTOR_MODEL_THINKING_BUDGET = int(os.getenv("DATA_EXTRACTOR_THINKING_BUDGET", "4096"))
    DATA_ARRANGER_MODEL_THINKING_BUDGET = int(os.getenv("DATA_ARRANGER_THINKING_BUDGET", "4096"))
    CODE_GENERATOR_MODEL_THINKING_BUDGET = int(os.getenv("CODE_GENERATOR_THINKING_BUDGET", "4096"))
    
    # === FILE PROCESSING CONFIGURATION ===
    MAX_FILE_SIZE_MB = int(os.getenv("MAX_FILE_SIZE_MB", "50"))
    SUPPORTED_FILE_TYPES = [
        "pdf", "txt", "docx", "xlsx", "csv", "md", "json", "xml", "html",
        "png", "jpg", "jpeg", "doc", "xls", "ppt", "pptx"
    ]
    
    # === DIRECTORY MANAGEMENT ===
    # Centralized working directory - all operations happen within this directory
    WORKING_DIR = Path(os.getenv("WORKING_DIR", "/tmp/data_extractor_gemini"))
    
    # Subdirectories within working directory
    TEMP_DIR = WORKING_DIR / "temp"
    INPUT_DIR = WORKING_DIR / "input" 
    OUTPUT_DIR = WORKING_DIR / "output"
    CACHE_DIR = WORKING_DIR / "cache"
    LOGS_DIR = WORKING_DIR / "logs"
    
    # === WORKFLOW CONFIGURATION ===
    # Retry and timeout settings
    MAX_RETRIES = int(os.getenv("MAX_RETRIES", "3"))
    RETRY_DELAY_SECONDS = int(os.getenv("RETRY_DELAY_SECONDS", "5"))
    AGENT_TIMEOUT_SECONDS = int(os.getenv("AGENT_TIMEOUT_SECONDS", "300"))
    
    # Cache settings
    ENABLE_CACHING = os.getenv("ENABLE_CACHING", "true").lower() == "true"
    CACHE_TTL_HOURS = int(os.getenv("CACHE_TTL_HOURS", "24"))
    
    @classmethod
    def initialize_directories(cls):
        """Initialize all required directories with proper permissions."""
        directories = [
            cls.WORKING_DIR,
            cls.TEMP_DIR,
            cls.INPUT_DIR,
            cls.OUTPUT_DIR,
            cls.CACHE_DIR,
            cls.LOGS_DIR
        ]
        
        created_dirs = []
        for directory in directories:
            try:
                directory.mkdir(parents=True, exist_ok=True)
                
                # Test write permissions
                test_file = directory / ".write_test"
                test_file.write_text("test")
                test_file.unlink()
                
                created_dirs.append(str(directory))
                logger.debug(f"Directory initialized: {directory}")
                
            except Exception as e:
                logger.error(f"Failed to initialize directory {directory}: {e}")
                raise RuntimeError(f"Cannot create or write to directory {directory}: {e}")
        
        logger.info(f"Successfully initialized {len(created_dirs)} directories")
        return created_dirs
    
    @classmethod
    def validate_config(cls):
        """Comprehensive configuration validation with detailed error reporting."""
        errors = []
        warnings = []
        
        # === CRITICAL VALIDATIONS ===
        
        # Google API Key validation
        if not cls.GOOGLE_API_KEY:
            errors.append("GOOGLE_API_KEY is required. Get it from https://aistudio.google.com/app/apikey")
        elif len(cls.GOOGLE_API_KEY) < 30:
            warnings.append("GOOGLE_API_KEY appears to be too short - verify it's correct")
        
        # Model name validation
        gemini_models = [cls.DATA_EXTRACTOR_MODEL, cls.DATA_ARRANGER_MODEL, cls.CODE_GENERATOR_MODEL]
        for i, model in enumerate(gemini_models):
            model_names = ["DATA_EXTRACTOR_MODEL", "DATA_ARRANGER_MODEL", "CODE_GENERATOR_MODEL"]
            if not model:
                errors.append(f"{model_names[i]} cannot be empty")
            elif not model.startswith("gemini-"):
                errors.append(f"{model_names[i]} must be a Gemini model (starts with 'gemini-'), got: {model}")
        
        # Directory validation
        try:
            cls.initialize_directories()
        except Exception as e:
            errors.append(f"Directory initialization failed: {e}")
        
        # === MODERATE VALIDATIONS ===
        
        # File size validation
        if cls.MAX_FILE_SIZE_MB <= 0:
            errors.append("MAX_FILE_SIZE_MB must be positive")
        elif cls.MAX_FILE_SIZE_MB > 100:
            warnings.append(f"MAX_FILE_SIZE_MB ({cls.MAX_FILE_SIZE_MB}) is very large - may cause memory issues")
        
        # Supported file types validation
        if not cls.SUPPORTED_FILE_TYPES:
            errors.append("SUPPORTED_FILE_TYPES cannot be empty")
        
        # Thinking budget validation
        budgets = [
            (cls.DATA_EXTRACTOR_MODEL_THINKING_BUDGET, "DATA_EXTRACTOR_MODEL_THINKING_BUDGET"),
            (cls.DATA_ARRANGER_MODEL_THINKING_BUDGET, "DATA_ARRANGER_MODEL_THINKING_BUDGET"),
            (cls.CODE_GENERATOR_MODEL_THINKING_BUDGET, "CODE_GENERATOR_MODEL_THINKING_BUDGET")
        ]
        
        for budget, name in budgets:
            if budget < 1024:
                warnings.append(f"{name} ({budget}) is quite low - may affect model performance")
            elif budget > 8192:
                warnings.append(f"{name} ({budget}) is very high - may be unnecessary")
        
        # Retry configuration validation
        if cls.MAX_RETRIES < 1:
            warnings.append("MAX_RETRIES should be at least 1")
        elif cls.MAX_RETRIES > 10:
            warnings.append("MAX_RETRIES is very high - may cause long delays")
        
        # === RESULT PROCESSING ===
        
        if errors:
            error_msg = "❌ Configuration validation failed:\n"
            error_msg += "\n".join(f"  • {error}" for error in errors)
            
            if warnings:
                error_msg += "\n\n⚠️  Warnings:\n"
                error_msg += "\n".join(f"  • {warning}" for warning in warnings)
            
            raise ValueError(error_msg)
        
        if warnings:
            logger.warning("Configuration warnings detected:")
            for warning in warnings:
                logger.warning(f"  • {warning}")
        
        logger.info("✅ Configuration validation successful")
        return True
    
    @classmethod
    def get_session_directories(cls, session_id: str):
        """Get session-specific directory structure."""
        session_base = cls.WORKING_DIR / session_id
        
        return {
            "base": session_base,
            "input": session_base / "input",
            "output": session_base / "output", 
            "temp": session_base / "temp",
            "cache": session_base / "cache"
        }
    
    @classmethod
    def create_session_directories(cls, session_id: str):
        """Create and validate session-specific directories."""
        session_dirs = cls.get_session_directories(session_id)
        
        created = []
        for name, directory in session_dirs.items():
            try:
                directory.mkdir(parents=True, exist_ok=True)
                
                # Test write permissions
                test_file = directory / ".write_test"
                test_file.write_text("test")
                test_file.unlink()
                
                created.append(str(directory))
                
            except Exception as e:
                logger.error(f"Failed to create session directory {name}: {e}")
                raise RuntimeError(f"Cannot create session directory {directory}: {e}")
        
        logger.info(f"Created {len(created)} session directories for {session_id}")
        return session_dirs
    
    @classmethod
    def cleanup_session(cls, session_id: str, keep_output: bool = True):
        """Clean up session directories with option to preserve output."""
        session_dirs = cls.get_session_directories(session_id)
        
        import shutil
        cleaned = []
        
        for name, directory in session_dirs.items():
            if keep_output and name == "output":
                continue
                
            if directory.exists():
                try:
                    shutil.rmtree(directory)
                    cleaned.append(str(directory))
                except Exception as e:
                    logger.warning(f"Could not clean {name} directory: {e}")
        
        logger.info(f"Cleaned {len(cleaned)} session directories for {session_id}")
        return cleaned
    
    @classmethod
    def get_debug_info(cls):
        """Get comprehensive debug information about current configuration."""
        import platform
        import sys
        
        return {
            "python_version": sys.version,
            "platform": platform.platform(),
            "temp_dir": str(cls.TEMP_DIR),
            "temp_dir_exists": cls.TEMP_DIR.exists(),
            "models": {
                "data_extractor": cls.DATA_EXTRACTOR_MODEL,
                "data_arranger": cls.DATA_ARRANGER_MODEL,
                "code_generator": cls.CODE_GENERATOR_MODEL,
            },
            "api_keys": {
                "google_api_key_present": bool(cls.GOOGLE_API_KEY),
                "google_api_key_length": len(cls.GOOGLE_API_KEY) if cls.GOOGLE_API_KEY else 0
            }
        }


# Global settings instance
settings = Settings()

# Auto-initialize directories on import
try:
    settings.initialize_directories()
    logger.debug("Settings initialized successfully")
except Exception as e:
    logger.error(f"Failed to initialize settings: {e}")
    # Don't raise here to allow import to succeed