File size: 34,705 Bytes
8418b54 |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 |
import os
import json
import tempfile
import subprocess
from typing import Optional, Dict, Any, List
import PyPDF2
import docx
from openpyxl import load_workbook
import pandas as pd
from pptx import Presentation
from PIL import Image
import zipfile
import csv
import chardet
class FileProcessor:
"""Enhanced file processor for various document types with improved error handling"""
def __init__(self):
self.supported_extensions = {
'pdf': self._process_pdf,
'docx': self._process_docx,
'doc': self._process_doc,
'pptx': self._process_pptx,
'ppt': self._process_ppt,
'xlsx': self._process_xlsx,
'xls': self._process_xls,
'csv': self._process_csv,
'txt': self._process_txt,
'json': self._process_json,
'rtf': self._process_rtf,
'odt': self._process_odt,
'ods': self._process_ods,
'odp': self._process_odp
}
# Enhanced processing limits
self.max_file_size = 100 * 1024 * 1024 # 100MB
self.max_text_length = 1000000 # 1M characters
self.max_pages_pdf = 500
self.max_sheets_excel = 50
def process_file(self, file_path: str, extension: str = None) -> Optional[str]:
"""Process a file and extract its text content with enhanced error handling"""
try:
if not os.path.exists(file_path):
print(f"File not found: {file_path}")
return None
# Check file size
file_size = os.path.getsize(file_path)
if file_size > self.max_file_size:
print(f"File too large: {file_size} bytes (max: {self.max_file_size})")
return f"File too large for processing: {file_size / (1024*1024):.1f}MB"
# Determine extension if not provided
if not extension:
extension = file_path.split('.')[-1].lower() if '.' in file_path else ''
extension = extension.lower().strip('.')
if extension not in self.supported_extensions:
print(f"Unsupported file extension: {extension}")
return f"Unsupported file type: .{extension}"
# Process file based on extension
processor = self.supported_extensions[extension]
content = processor(file_path)
if content:
# Truncate if too long
if len(content) > self.max_text_length:
content = content[:self.max_text_length] + "\n[Content truncated due to length limit]"
print(f"Successfully processed {extension.upper()} file: {os.path.basename(file_path)}")
return content
else:
print(f"No content extracted from: {os.path.basename(file_path)}")
return f"Could not extract content from {extension.upper()} file"
except Exception as e:
print(f"Error processing file {file_path}: {e}")
return f"Error processing file: {str(e)}"
def _process_pdf(self, file_path: str) -> Optional[str]:
"""Extract text from PDF files with enhanced handling"""
try:
text_content = []
with open(file_path, 'rb') as file:
pdf_reader = PyPDF2.PdfReader(file)
num_pages = len(pdf_reader.pages)
if num_pages > self.max_pages_pdf:
print(f"PDF too long ({num_pages} pages), processing first {self.max_pages_pdf}")
num_pages = self.max_pages_pdf
for page_num, page in enumerate(pdf_reader.pages[:num_pages]):
try:
page_text = page.extract_text()
if page_text.strip():
text_content.append(f"--- Page {page_num + 1} ---")
text_content.append(page_text)
text_content.append("")
except Exception as e:
text_content.append(f"--- Page {page_num + 1} (Error reading) ---")
print(f"Error reading PDF page {page_num + 1}: {e}")
if not text_content:
# Try alternative PDF processing with pdfplumber if available
try:
import pdfplumber
with pdfplumber.open(file_path) as pdf:
for page_num, page in enumerate(pdf.pages[:self.max_pages_pdf]):
page_text = page.extract_text()
if page_text:
text_content.append(f"--- Page {page_num + 1} ---")
text_content.append(page_text)
text_content.append("")
except ImportError:
return "PDF contains non-text content or requires advanced processing"
return "\n".join(text_content) if text_content else None
except Exception as e:
print(f"Error processing PDF: {e}")
return None
def _process_docx(self, file_path: str) -> Optional[str]:
"""Extract text from DOCX files with enhanced table handling"""
try:
doc = docx.Document(file_path)
text_content = []
# Extract paragraphs
for paragraph in doc.paragraphs:
if paragraph.text.strip():
text_content.append(paragraph.text)
# Extract tables with improved formatting
for table_num, table in enumerate(doc.tables, 1):
text_content.append(f"\n--- Table {table_num} ---")
for row_num, row in enumerate(table.rows):
row_text = []
for cell in row.cells:
cell_text = cell.text.strip().replace('\n', ' ').replace('\t', ' ')
row_text.append(cell_text)
if any(row_text): # Only add non-empty rows
text_content.append(" | ".join(row_text))
text_content.append("--- End Table ---\n")
return "\n".join(text_content) if text_content else None
except Exception as e:
print(f"Error processing DOCX: {e}")
return None
def _process_doc(self, file_path: str) -> Optional[str]:
"""Extract text from DOC files using available tools"""
try:
# Try with docx2txt first
try:
import docx2txt
text = docx2txt.process(file_path)
return text if text.strip() else None
except ImportError:
pass
# Try with antiword (if available on system)
try:
result = subprocess.run(
['antiword', file_path],
capture_output=True,
text=True,
timeout=30
)
if result.returncode == 0:
return result.stdout
except (subprocess.SubprocessError, FileNotFoundError):
pass
# Try with LibreOffice (if available)
try:
temp_dir = tempfile.mkdtemp()
result = subprocess.run([
'libreoffice', '--headless', '--convert-to', 'txt',
'--outdir', temp_dir, file_path
], capture_output=True, timeout=60)
if result.returncode == 0:
txt_file = os.path.join(temp_dir, os.path.splitext(os.path.basename(file_path))[0] + '.txt')
if os.path.exists(txt_file):
with open(txt_file, 'r', encoding='utf-8') as f:
content = f.read()
# Clean up
import shutil
shutil.rmtree(temp_dir, ignore_errors=True)
return content
except (subprocess.SubprocessError, FileNotFoundError):
pass
return "DOC file processing requires additional tools (docx2txt, antiword, or LibreOffice)"
except Exception as e:
print(f"Error processing DOC: {e}")
return None
def _process_pptx(self, file_path: str) -> Optional[str]:
"""Extract text from PPTX files with enhanced slide handling"""
try:
presentation = Presentation(file_path)
text_content = []
for slide_num, slide in enumerate(presentation.slides, 1):
slide_text = []
slide_text.append(f"--- Slide {slide_num} ---")
# Extract text from shapes
for shape in slide.shapes:
if hasattr(shape, "text") and shape.text.strip():
# Check if it's a title or content
if hasattr(shape, 'placeholder_format') and shape.placeholder_format:
if shape.placeholder_format.type == 1: # Title
slide_text.append(f"TITLE: {shape.text}")
else:
slide_text.append(shape.text)
else:
slide_text.append(shape.text)
# Extract table content if present
for shape in slide.shapes:
if shape.has_table:
slide_text.append("TABLE:")
table = shape.table
for row in table.rows:
row_text = [cell.text.strip() for cell in row.cells]
slide_text.append(" | ".join(row_text))
# Extract notes
if slide.has_notes_slide:
notes_text = slide.notes_slide.notes_text_frame.text
if notes_text.strip():
slide_text.append(f"NOTES: {notes_text}")
if len(slide_text) > 1: # More than just the slide header
text_content.extend(slide_text)
text_content.append("")
return "\n".join(text_content) if text_content else None
except Exception as e:
print(f"Error processing PPTX: {e}")
return None
def _process_ppt(self, file_path: str) -> Optional[str]:
"""Extract text from PPT files using LibreOffice if available"""
try:
# Try with LibreOffice
try:
temp_dir = tempfile.mkdtemp()
result = subprocess.run([
'libreoffice', '--headless', '--convert-to', 'txt',
'--outdir', temp_dir, file_path
], capture_output=True, timeout=60)
if result.returncode == 0:
txt_file = os.path.join(temp_dir, os.path.splitext(os.path.basename(file_path))[0] + '.txt')
if os.path.exists(txt_file):
with open(txt_file, 'r', encoding='utf-8') as f:
content = f.read()
# Clean up
import shutil
shutil.rmtree(temp_dir, ignore_errors=True)
return content
except (subprocess.SubprocessError, FileNotFoundError):
pass
return "PPT file processing requires LibreOffice or conversion to PPTX format"
except Exception as e:
print(f"Error processing PPT: {e}")
return None
def _process_xlsx(self, file_path: str) -> Optional[str]:
"""Extract text from XLSX files with enhanced sheet handling"""
try:
workbook = load_workbook(file_path, data_only=True)
text_content = []
sheet_count = 0
for sheet_name in workbook.sheetnames:
if sheet_count >= self.max_sheets_excel:
text_content.append(f"[Additional {len(workbook.sheetnames) - sheet_count} sheets truncated]")
break
sheet = workbook[sheet_name]
text_content.append(f"--- Sheet: {sheet_name} ---")
# Get data dimensions
max_row = min(sheet.max_row, 1000) # Limit rows
max_col = min(sheet.max_column, 100) # Limit columns
# Extract data efficiently
data = []
for row in sheet.iter_rows(min_row=1, max_row=max_row, min_col=1, max_col=max_col, values_only=True):
if any(cell is not None for cell in row):
row_data = [str(cell) if cell is not None else "" for cell in row]
# Remove trailing empty cells
while row_data and not row_data[-1]:
row_data.pop()
if row_data:
data.append(row_data)
if data:
# Format as table
for row in data[:100]: # Limit to first 100 rows per sheet
text_content.append(" | ".join(row))
else:
text_content.append("[Empty sheet]")
text_content.append("")
sheet_count += 1
return "\n".join(text_content) if text_content else None
except Exception as e:
print(f"Error processing XLSX: {e}")
return None
def _process_xls(self, file_path: str) -> Optional[str]:
"""Extract text from XLS files with enhanced error handling"""
try:
# Use pandas to read XLS files
xl_file = pd.ExcelFile(file_path)
text_content = []
sheet_count = 0
for sheet_name in xl_file.sheet_names:
if sheet_count >= self.max_sheets_excel:
text_content.append(f"[Additional {len(xl_file.sheet_names) - sheet_count} sheets truncated]")
break
text_content.append(f"--- Sheet: {sheet_name} ---")
try:
df = pd.read_excel(file_path, sheet_name=sheet_name)
# Convert DataFrame to string representation
if not df.empty:
# Limit rows and columns for processing
limited_df = df.head(100).iloc[:, :20] # First 100 rows, 20 columns
text_content.append(limited_df.to_string(index=False))
if len(df) > 100:
text_content.append(f"[{len(df) - 100} additional rows not shown]")
else:
text_content.append("[Empty sheet]")
except Exception as e:
text_content.append(f"[Error reading sheet: {e}]")
text_content.append("")
sheet_count += 1
return "\n".join(text_content) if text_content else None
except Exception as e:
print(f"Error processing XLS: {e}")
return None
def _process_csv(self, file_path: str) -> Optional[str]:
"""Extract text from CSV files with enhanced encoding detection"""
try:
# Detect encoding
encoding = 'utf-8'
try:
with open(file_path, 'rb') as f:
raw_data = f.read(10000) # Read first 10KB
detected = chardet.detect(raw_data)
if detected['encoding'] and detected['confidence'] > 0.7:
encoding = detected['encoding']
except:
pass
# Try reading with detected encoding
try:
df = pd.read_csv(file_path, encoding=encoding)
except UnicodeDecodeError:
# Fallback encodings
for fallback_encoding in ['latin-1', 'cp1252', 'iso-8859-1']:
try:
df = pd.read_csv(file_path, encoding=fallback_encoding)
break
except UnicodeDecodeError:
continue
else:
return "Could not decode CSV file with any encoding"
text_content = []
text_content.append("--- CSV Data ---")
text_content.append(f"Columns ({len(df.columns)}): {', '.join(df.columns.astype(str).tolist())}")
text_content.append(f"Total rows: {len(df)}")
text_content.append("")
# Show first 100 rows and limit columns if too many
limited_df = df.head(100)
if len(df.columns) > 20:
limited_df = limited_df.iloc[:, :20]
text_content.append(f"[Showing first 20 of {len(df.columns)} columns]")
text_content.append(limited_df.to_string(index=False))
if len(df) > 100:
text_content.append(f"\n[{len(df) - 100} additional rows not shown]")
return "\n".join(text_content)
except Exception as e:
print(f"Error processing CSV: {e}")
return None
def _process_txt(self, file_path: str) -> Optional[str]:
"""Extract text from TXT files with encoding detection"""
try:
# Detect encoding
encoding = 'utf-8'
try:
with open(file_path, 'rb') as f:
raw_data = f.read()
detected = chardet.detect(raw_data)
if detected['encoding'] and detected['confidence'] > 0.7:
encoding = detected['encoding']
except:
pass
# Try reading with detected encoding first
encodings_to_try = [encoding, 'utf-8', 'utf-16', 'latin-1', 'cp1252']
for enc in encodings_to_try:
try:
with open(file_path, 'r', encoding=enc) as file:
content = file.read()
return content if content.strip() else None
except UnicodeDecodeError:
continue
# If all encodings fail, try with error handling
with open(file_path, 'r', encoding='utf-8', errors='replace') as file:
return file.read()
except Exception as e:
print(f"Error processing TXT: {e}")
return None
def _process_json(self, file_path: str) -> Optional[str]:
"""Extract text from JSON files with pretty formatting"""
try:
with open(file_path, 'r', encoding='utf-8') as file:
data = json.load(file)
# Convert JSON to readable text format
if isinstance(data, dict):
text_content = ["--- JSON Object ---"]
text_content.append(json.dumps(data, indent=2, ensure_ascii=False)[:50000]) # Limit size
elif isinstance(data, list):
text_content = ["--- JSON Array ---"]
text_content.append(f"Array with {len(data)} items:")
sample_items = min(10, len(data))
text_content.append(json.dumps(data[:sample_items], indent=2, ensure_ascii=False))
if len(data) > sample_items:
text_content.append(f"... and {len(data) - sample_items} more items")
else:
text_content = [str(data)]
return "\n".join(text_content)
except Exception as e:
print(f"Error processing JSON: {e}")
return None
def _process_rtf(self, file_path: str) -> Optional[str]:
"""Extract text from RTF files"""
try:
# Try with striprtf if available
try:
from striprtf.striprtf import rtf_to_text
with open(file_path, 'r', encoding='utf-8') as file:
rtf_content = file.read()
return rtf_to_text(rtf_content)
except ImportError:
pass
# Try with LibreOffice
try:
temp_dir = tempfile.mkdtemp()
result = subprocess.run([
'libreoffice', '--headless', '--convert-to', 'txt',
'--outdir', temp_dir, file_path
], capture_output=True, timeout=60)
if result.returncode == 0:
txt_file = os.path.join(temp_dir, os.path.splitext(os.path.basename(file_path))[0] + '.txt')
if os.path.exists(txt_file):
with open(txt_file, 'r', encoding='utf-8') as f:
content = f.read()
# Clean up
import shutil
shutil.rmtree(temp_dir, ignore_errors=True)
return content
except (subprocess.SubprocessError, FileNotFoundError):
pass
return "RTF file processing requires striprtf package or LibreOffice"
except Exception as e:
print(f"Error processing RTF: {e}")
return None
def _process_odt(self, file_path: str) -> Optional[str]:
"""Extract text from ODT files using LibreOffice or zip extraction"""
try:
# Try with LibreOffice first
try:
temp_dir = tempfile.mkdtemp()
result = subprocess.run([
'libreoffice', '--headless', '--convert-to', 'txt',
'--outdir', temp_dir, file_path
], capture_output=True, timeout=60)
if result.returncode == 0:
txt_file = os.path.join(temp_dir, os.path.splitext(os.path.basename(file_path))[0] + '.txt')
if os.path.exists(txt_file):
with open(txt_file, 'r', encoding='utf-8') as f:
content = f.read()
# Clean up
import shutil
shutil.rmtree(temp_dir, ignore_errors=True)
return content
except (subprocess.SubprocessError, FileNotFoundError):
pass
# Try manual XML extraction
try:
with zipfile.ZipFile(file_path, 'r') as zip_file:
if 'content.xml' in zip_file.namelist():
content_xml = zip_file.read('content.xml').decode('utf-8')
# Basic XML tag removal (not perfect but functional)
import re
text = re.sub(r'<[^>]+>', ' ', content_xml)
text = re.sub(r'\s+', ' ', text)
return text.strip() if text.strip() else None
except Exception:
pass
return "ODT file processing requires LibreOffice"
except Exception as e:
print(f"Error processing ODT: {e}")
return None
def _process_ods(self, file_path: str) -> Optional[str]:
"""Extract text from ODS files"""
try:
# Try with LibreOffice
try:
temp_dir = tempfile.mkdtemp()
result = subprocess.run([
'libreoffice', '--headless', '--convert-to', 'csv',
'--outdir', temp_dir, file_path
], capture_output=True, timeout=60)
if result.returncode == 0:
csv_file = os.path.join(temp_dir, os.path.splitext(os.path.basename(file_path))[0] + '.csv')
if os.path.exists(csv_file):
content = self._process_csv(csv_file)
# Clean up
import shutil
shutil.rmtree(temp_dir, ignore_errors=True)
return content
except (subprocess.SubprocessError, FileNotFoundError):
pass
return "ODS file processing requires LibreOffice"
except Exception as e:
print(f"Error processing ODS: {e}")
return None
def _process_odp(self, file_path: str) -> Optional[str]:
"""Extract text from ODP files"""
try:
# Try with LibreOffice
try:
temp_dir = tempfile.mkdtemp()
result = subprocess.run([
'libreoffice', '--headless', '--convert-to', 'txt',
'--outdir', temp_dir, file_path
], capture_output=True, timeout=60)
if result.returncode == 0:
txt_file = os.path.join(temp_dir, os.path.splitext(os.path.basename(file_path))[0] + '.txt')
if os.path.exists(txt_file):
with open(txt_file, 'r', encoding='utf-8') as f:
content = f.read()
# Clean up
import shutil
shutil.rmtree(temp_dir, ignore_errors=True)
return content
except (subprocess.SubprocessError, FileNotFoundError):
pass
return "ODP file processing requires LibreOffice"
except Exception as e:
print(f"Error processing ODP: {e}")
return None
def get_file_info(self, file_path: str) -> Dict[str, Any]:
"""Get comprehensive information about a file"""
try:
stat = os.stat(file_path)
extension = file_path.split('.')[-1].lower() if '.' in file_path else ''
return {
'filename': os.path.basename(file_path),
'size': stat.st_size,
'size_mb': round(stat.st_size / (1024 * 1024), 2),
'size_human': self._format_file_size(stat.st_size),
'extension': extension,
'supported': extension in self.supported_extensions,
'modified': stat.st_mtime,
'type': self._get_file_type(extension),
'processing_complexity': self._get_processing_complexity(extension, stat.st_size)
}
except Exception as e:
return {
'filename': os.path.basename(file_path) if file_path else 'unknown',
'error': str(e),
'supported': False,
'type': 'unknown'
}
def _format_file_size(self, size_bytes: int) -> str:
"""Format file size in human readable format"""
for unit in ['B', 'KB', 'MB', 'GB']:
if size_bytes < 1024.0:
return f"{size_bytes:.1f} {unit}"
size_bytes /= 1024.0
return f"{size_bytes:.1f} TB"
def _get_file_type(self, extension: str) -> str:
"""Get file type category"""
document_types = {'pdf', 'docx', 'doc', 'txt', 'rtf', 'odt'}
spreadsheet_types = {'xlsx', 'xls', 'csv', 'ods'}
presentation_types = {'pptx', 'ppt', 'odp'}
data_types = {'json', 'xml'}
if extension in document_types:
return 'document'
elif extension in spreadsheet_types:
return 'spreadsheet'
elif extension in presentation_types:
return 'presentation'
elif extension in data_types:
return 'data'
else:
return 'unknown'
def _get_processing_complexity(self, extension: str, file_size: int) -> str:
"""Estimate processing complexity"""
if extension in ['txt', 'csv', 'json']:
return 'low'
elif extension in ['docx', 'xlsx', 'pptx'] and file_size < 10 * 1024 * 1024:
return 'medium'
elif extension in ['pdf', 'doc', 'xls', 'ppt'] or file_size > 10 * 1024 * 1024:
return 'high'
else:
return 'medium'
def batch_process_files(self, file_paths: List[str]) -> Dict[str, Any]:
"""Process multiple files and return comprehensive results"""
results = {
'successful': [],
'failed': [],
'combined_content': [],
'total_files': len(file_paths),
'total_size': 0,
'processing_time': 0,
'file_types': {}
}
import time
start_time = time.time()
for file_path in file_paths:
try:
file_info = self.get_file_info(file_path)
results['total_size'] += file_info.get('size', 0)
file_type = file_info.get('type', 'unknown')
results['file_types'][file_type] = results['file_types'].get(file_type, 0) + 1
if file_info.get('supported', False):
content = self.process_file(file_path)
if content:
results['successful'].append({
'filename': file_info['filename'],
'content': content,
'size_mb': file_info['size_mb'],
'type': file_type,
'complexity': file_info.get('processing_complexity', 'unknown')
})
results['combined_content'].append(f"=== {file_info['filename']} ===")
results['combined_content'].append(content)
results['combined_content'].append("")
else:
results['failed'].append({
'filename': file_info['filename'],
'reason': 'No content extracted',
'type': file_type
})
else:
results['failed'].append({
'filename': file_info['filename'],
'reason': 'Unsupported file type',
'type': file_type
})
except Exception as e:
results['failed'].append({
'filename': os.path.basename(file_path) if file_path else 'unknown',
'reason': str(e),
'type': 'unknown'
})
results['processing_time'] = time.time() - start_time
results['combined_text'] = "\n".join(results['combined_content'])
results['success_rate'] = len(results['successful']) / len(file_paths) if file_paths else 0
results['total_size_mb'] = results['total_size'] / (1024 * 1024)
return results
def validate_file(self, file_path: str) -> Dict[str, Any]:
"""Validate a file before processing"""
validation_result = {
'valid': False,
'errors': [],
'warnings': [],
'info': {}
}
try:
if not os.path.exists(file_path):
validation_result['errors'].append("File does not exist")
return validation_result
file_info = self.get_file_info(file_path)
validation_result['info'] = file_info
# Check file size
if file_info['size'] > self.max_file_size:
validation_result['errors'].append(f"File too large: {file_info['size_human']} (max: {self._format_file_size(self.max_file_size)})")
if file_info['size'] == 0:
validation_result['errors'].append("File is empty")
# Check file type
if not file_info['supported']:
validation_result['errors'].append(f"Unsupported file type: .{file_info['extension']}")
# Check processing complexity
complexity = file_info.get('processing_complexity', 'unknown')
if complexity == 'high':
validation_result['warnings'].append("File may require significant processing time")
# Additional format-specific validation
extension = file_info['extension']
if extension == 'pdf' and file_info['size'] > 50 * 1024 * 1024:
validation_result['warnings'].append("Large PDF files may have incomplete text extraction")
validation_result['valid'] = len(validation_result['errors']) == 0
except Exception as e:
validation_result['errors'].append(f"Validation error: {str(e)}")
return validation_result |