extract_doc.py

"""
Extract text from old .doc (OLE2) file using olefile.
Usage: python extract_doc.py <path_to_doc>
"""
import sys, olefile, re, struct

path = sys.argv[1]
ole = olefile.OleFileIO(path)

# Read the WordDocument stream
doc_stream = ole.openstream('WordDocument').read()

# Read the Table stream or 0Table
for table_name in ['1Table', '0Table']:
    if ole.exists(table_name):
        table_stream = ole.openstream(table_name).read()
        break
else:
    table_stream = None

# Read Text pieces from the document
# The .doc format stores text in Unicode (UTF-16LE)
# Try to extract text by scanning for readable sequences

print("=" * 60)
print("文档内容提取")
print("=" * 60)

# Method 1: Extract UTF-16LE text
text_parts = []
i = 0
while i < len(doc_stream) - 1:
    # Read 2 bytes as UTF-16LE
    char = doc_stream[i:i+2]
    code = struct.unpack('<H', char)[0]
    if 0x4E00 <= code <= 0x9FFF or 0x3000 <= code <= 0x303F or \
       0x0020 <= code <= 0x007E or 0xFF00 <= code <= 0xFFEF or \
       code in (0x000D, 0x000A, 0x0009):
        text_parts.append(chr(code) if code < 0x10000 else '?')
    elif text_parts and text_parts[-1] != '\n':
        text_parts.append('\n')
    i += 2

text = ''.join(text_parts)

# Clean up the text
lines = []
for line in text.split('\n'):
    line = line.strip()
    if line and len(line) > 1:
        lines.append(line)

print('\n'.join(lines))

ole.close()