def is_archive(path: str) -> Tuple[int, Any, str]: """Check if file in path is an ZIP, RAR or 7z file :param path: path to file :return: (zf, status, expected_extension) status: -1==Error/Retry, 0==OK, 1==Ignore """ if zipfile.is_zipfile(path): try: zf = zipfile.ZipFile(path) return 0, zf, ".zip" except: logging.info(T("Cannot read %s"), path, exc_info=True) return -1, None, "" elif rarfile.is_rarfile(path): try: # Set path to tool to open it rarfile.UNRAR_TOOL = sabnzbd.newsunpack.RAR_COMMAND zf = rarfile.RarFile(path) return 0, zf, ".rar" except: logging.info(T("Cannot read %s"), path, exc_info=True) return -1, None, "" elif sabnzbd.newsunpack.is_sevenfile(path): try: zf = sabnzbd.newsunpack.SevenZip(path) return 0, zf, ".7z" except: logging.info(T("Cannot read %s"), path, exc_info=True) return -1, None, "" else: logging.info("Archive %s is not a real archive!", os.path.basename(path)) return 1, None, ""
def is_archive(path): """ Check if file in path is an ZIP, RAR or 7z file :param path: path to file :return: (zf, status, expected_extension) status: -1==Error/Retry, 0==OK, 1==Ignore """ if zipfile.is_zipfile(path): try: zf = zipfile.ZipFile(path) return 0, zf, '.zip' except: logging.info(T('Cannot read %s'), path, exc_info=True) return -1, None, '' elif misc.is_rarfile(path): try: # Set path to tool to open it rarfile.UNRAR_TOOL = sabnzbd.newsunpack.RAR_COMMAND zf = rarfile.RarFile(path) return 0, zf, '.rar' except: logging.info(T('Cannot read %s'), path, exc_info=True) return -1, None, '' elif is_sevenfile(path): try: zf = SevenZip(path) return 0, zf, '.7z' except: logging.info(T('Cannot read %s'), path, exc_info=True) return -1, None, '' else: logging.info('Archive %s is not a real archive!', os.path.basename(path)) return 1, None, ''
def is_archive(path): """ Check if file in path is an ZIP, RAR or 7z file :param path: path to file :return: (zf, status, expected_extension) status: -1==Error/Retry, 0==OK, 1==Ignore """ if zipfile.is_zipfile(path): try: zf = zipfile.ZipFile(path) return 0, zf, '.zip' except: return -1, None, '' elif rarfile.is_rarfile(path): try: zf = rarfile.RarFile(path) # Set path to tool to open it rarfile.UNRAR_TOOL = sabnzbd.newsunpack.RAR_COMMAND return 0, zf, '.rar' except: return -1, None, '' elif is_sevenfile(path): try: zf = SevenZip(path) return 0, zf, '.7z' except: return -1, None, '' else: return 1, None, ''
def try_rar_check(nzo, workdir, setname): """ Attempt to verify set using the RARs Return True if verified, False when failed When setname is '', all RAR files will be used, otherwise only the matching one If no RAR's are found, returns True """ _, _, rars, _, _ = build_filelists(workdir) if setname: # Filter based on set rars = [ rar for rar in rars if os.path.basename(rar).startswith(setname) ] # Sort rars.sort(rar_sort) # Test if rars: nzo.status = Status.VERIFYING nzo.set_unpack_info('Repair', T('Trying RAR-based verification'), set=setname) nzo.set_action_line(T('Trying RAR-based verification'), '...') try: # Set path to unrar and open the file # Requires de-unicode for RarFile to work! rarfile.UNRAR_TOOL = sabnzbd.newsunpack.RAR_COMMAND zf = rarfile.RarFile(rars[0]) # Skip if it's encrypted if zf.needs_password(): msg = T('[%s] RAR-based verification failed: %s') % (unicoder( os.path.basename(rars[0])), T('Passworded')) nzo.set_unpack_info('Repair', msg, set=setname) return True # Will throw exception if something is wrong zf.testrar() # Success! msg = T('RAR files verified successfully') nzo.set_unpack_info('Repair', msg, set=setname) logging.info(msg) return True except rarfile.Error as e: nzo.fail_msg = T('RAR files failed to verify') msg = T('[%s] RAR-based verification failed: %s') % (unicoder( os.path.basename( rars[0])), unicoder(e.message.replace('\r\n', ' '))) nzo.set_unpack_info('Repair', msg, set=setname) logging.info(msg) return False else: # No rar-files, so just continue return True
def try_rar_check(nzo, rars): """Attempt to verify set using the RARs Return True if verified, False when failed When setname is '', all RAR files will be used, otherwise only the matching one If no RAR's are found, returns True """ # Sort for better processing rars.sort(key=functools.cmp_to_key(rar_sort)) # Test if rars: setname = setname_from_path(rars[0]) nzo.status = Status.VERIFYING nzo.set_unpack_info("Repair", T("Trying RAR-based verification"), setname) nzo.set_action_line(T("Trying RAR-based verification"), "...") try: # Set path to unrar and open the file # Requires de-unicode for RarFile to work! rarfile.UNRAR_TOOL = sabnzbd.newsunpack.RAR_COMMAND zf = rarfile.RarFile(rars[0]) # Skip if it's encrypted if zf.needs_password(): msg = T("[%s] RAR-based verification failed: %s") % ( setname, T("Passworded")) nzo.set_unpack_info("Repair", msg) return True # Will throw exception if something is wrong zf.testrar() # Success! msg = T("RAR files verified successfully") nzo.set_unpack_info("Repair", msg, setname) logging.info(msg) return True except rarfile.Error as e: nzo.fail_msg = T("RAR files failed to verify") msg = T("[%s] RAR-based verification failed: %s") % (setname, e) nzo.set_unpack_info("Repair", msg, setname) logging.info(msg) return False else: # No rar-files, so just continue return True
def check_encrypted_and_unwanted_files(nzo, filepath): """ Combines check for unwanted and encrypted files to save on CPU and IO """ encrypted = False unwanted = None if (cfg.unwanted_extensions() and cfg.action_on_unwanted_extensions()) or ( nzo.encrypted == 0 and cfg.pause_on_pwrar()): # These checks should not break the assembler try: # Rarfile freezes on Windows special names, so don't try those! if sabnzbd.WIN32 and has_win_device(filepath): return encrypted, unwanted # Is it even a rarfile? if rarfile.is_rarfile(filepath): # Open the rar rarfile.UNRAR_TOOL = sabnzbd.newsunpack.RAR_COMMAND zf = rarfile.RarFile(filepath, all_names=True) # Check for encryption if nzo.encrypted == 0 and cfg.pause_on_pwrar() and ( zf.needs_password() or is_cloaked(nzo, filepath, zf.namelist())): # Load all passwords passwords = get_all_passwords(nzo) # Cloaked job? if is_cloaked(nzo, filepath, zf.namelist()): encrypted = True elif not sabnzbd.HAVE_CRYPTOGRAPHY and not passwords: # if no cryptography installed, only error when no password was set logging.info(T('%s missing'), 'Python Cryptography') nzo.encrypted = 1 encrypted = True elif sabnzbd.HAVE_CRYPTOGRAPHY: # Lets test if any of the password work password_hit = False for password in passwords: if password: logging.info( 'Trying password "%s" on job "%s"', password, nzo.final_name) try: zf.setpassword(password) except: # On weird passwords the setpassword() will fail # but the actual rartest() will work pass try: zf.testrar() password_hit = password break except rarfile.RarCRCError: # On CRC error we can continue! password_hit = password break except Exception as e: # Did we start from the right volume? if 'need to start extraction from a previous volume' in e[ 0]: return encrypted, unwanted # This one failed pass # Did any work? if password_hit: # We always trust the user's input if not nzo.password: nzo.password = password_hit # Don't check other files logging.info('Password "%s" matches for job "%s"', password_hit, nzo.final_name) nzo.encrypted = -1 encrypted = False else: # Encrypted and none of them worked nzo.encrypted = 1 encrypted = True else: # Don't check other files nzo.encrypted = -1 encrypted = False # Check for unwanted extensions if cfg.unwanted_extensions( ) and cfg.action_on_unwanted_extensions(): for somefile in zf.namelist(): logging.debug('File contains: %s', somefile) if get_ext(somefile).replace( '.', '').lower() in cfg.unwanted_extensions(): logging.debug('Unwanted file %s', somefile) unwanted = somefile zf.close() del zf except: logging.info('Error during inspection of RAR-file %s', filepath) logging.debug('Traceback: ', exc_info=True) return encrypted, unwanted
def rar_renamer(nzo, workdir): """ Deobfuscate rar file names: Use header and content information to give RAR-files decent names """ nzo.status = Status.VERIFYING nzo.set_unpack_info("Repair", T("Trying RAR-based verification")) nzo.set_action_line(T("Trying RAR-based verification"), "...") renamed_files = 0 # This is the most important datastructure (in case of mixed obfuscated rarsets) rarvolnr = {} # rarvolnr will contain per rar vol number the rarfilenames and their respective contents (and maybe other characteristics, like filesizes). # for example: rarvolnr[6]['somerandomfilename.rar']={'readme.txt', 'linux.iso'}, # which means 'somerandomfilename.rar' has rarvolnumber 6, and contents 'readme.txt' and 'linux.iso' # if we find a rarfile with rarvolnumber 7, and 'linux.iso' in it, we have a match! # The volume number and real extension of a (obfuscated) rar file # so volnrext['dfakjldfalkjdfl.blabla'] = (14, 'part014.rar') or (2, 'r000') # Not really needed, but handy to avoid a second lookup at the renaming volnrext = {} # Scan rar files in workdir, but not subdirs workdir_files = os.listdir(workdir) for file_to_check in workdir_files: file_to_check = os.path.join(workdir, file_to_check) # We only want files: if not (os.path.isfile(file_to_check)): continue # The function will check if it's a RAR-file # We do a sanity-check for the returned number rar_vol, new_extension = rarvolinfo.get_rar_extension(file_to_check) if 0 < rar_vol < 1000: logging.debug("Detected volume-number %s from RAR-header: %s ", rar_vol, file_to_check) volnrext[file_to_check] = (rar_vol, new_extension) # The files inside rar file rar_contents = rarfile.RarFile(os.path.join( workdir, file_to_check), single_file_check=True).filelist() try: rarvolnr[rar_vol] except: # does not yet exist, so create: rarvolnr[rar_vol] = {} rarvolnr[rar_vol][ file_to_check] = rar_contents # store them for matching (if needed) else: logging.debug("No RAR-volume-number found in %s", file_to_check) logging.debug("Deobfuscate: rarvolnr is: %s", rarvolnr) logging.debug("Deobfuscate: volnrext is: %s", volnrext) # Could be that there are no rar-files, we stop if not len(rarvolnr): return renamed_files # Check number of different obfuscated rar sets: numberofrarsets = len(rarvolnr[1]) if numberofrarsets == 1: # Just one obfuscated rarset logging.debug("Deobfuscate: Just one obfuscated rarset") for filename in volnrext: new_rar_name = "%s.%s" % (nzo.final_name, volnrext[filename][1]) new_rar_name = os.path.join(workdir, new_rar_name) new_rar_name = get_unique_filename(new_rar_name) logging.debug("Deobfuscate: Renaming %s to %s" % (filename, new_rar_name)) renamer(filename, new_rar_name) renamed_files += 1 else: # More than one obfuscated rarset, so we must do matching based of files inside the rar files logging.debug("Number of obfuscated rarsets: %s", numberofrarsets) # Assign (random) rar set names rarsetname = { } # in which rar set it should be, so rar set 'A', or 'B', or ... mychar = "A" # First things first: Assigning a rarsetname to the rar file which have volume number 1 for base_obfuscated_filename in rarvolnr[1]: rarsetname[ base_obfuscated_filename] = mychar + "--" + nzo.final_name mychar = chr(ord(mychar) + 1) logging.debug("Deobfuscate: rarsetname %s", rarsetname) # Do the matching, layer by layer (read: rarvolnumber) # So, all rar files with rarvolnr 1, find the contents (files inside the rar), # and match with rarfiles with rarvolnr 2, and put them in the correct rarset. # And so on, until the highest rarvolnr minus 1 matched against highest rarvolnr for n in range(1, len(rarvolnr.keys())): logging.debug( "Deobfuscate: Finding matches between rar sets %s and %s" % (n, n + 1)) for base_obfuscated_filename in rarvolnr[n]: matchcounter = 0 for next_obfuscated_filename in rarvolnr[n + 1]: # set() method with intersection (less strict): set(rarvolnr[n][base_obfuscated_filename]).intersection(set(rarvolnr[n+1][next_obfuscated_filename])) # check if the last filename inside the existing rar matches with the first filename in the following rar if rarvolnr[n][base_obfuscated_filename][-1] == rarvolnr[ n + 1][next_obfuscated_filename][0]: try: rarsetname[next_obfuscated_filename] = rarsetname[ base_obfuscated_filename] matchcounter += 1 except KeyError: logging.warning( T("No matching earlier rar file for %s"), next_obfuscated_filename) if matchcounter > 1: logging.info( "Deobfuscate: more than one match, so risk on false positive matching." ) # Do the renaming: for filename in rarsetname: new_rar_name = "%s.%s" % (rarsetname[filename], volnrext[filename][1]) new_rar_name = os.path.join(workdir, new_rar_name) new_rar_name = get_unique_filename(new_rar_name) logging.debug("Deobfuscate: Renaming %s to %s" % (filename, new_rar_name)) renamer(filename, new_rar_name) renamed_files += 1 # Done: The obfuscated rar files have now been renamed to regular formatted filenames return renamed_files
def check_encrypted_and_unwanted_files(nzo: NzbObject, filepath: str) -> Tuple[bool, Optional[str]]: """ Combines check for unwanted and encrypted files to save on CPU and IO """ encrypted = False unwanted = None if (cfg.unwanted_extensions() and cfg.action_on_unwanted_extensions()) or ( nzo.encrypted == 0 and cfg.pause_on_pwrar() ): # These checks should not break the assembler try: # Rarfile freezes on Windows special names, so don't try those! if sabnzbd.WIN32 and has_win_device(filepath): return encrypted, unwanted # Is it even a rarfile? if rarfile.is_rarfile(filepath): # Open the rar rarfile.UNRAR_TOOL = sabnzbd.newsunpack.RAR_COMMAND zf = rarfile.RarFile(filepath, single_file_check=True) # Check for encryption if ( nzo.encrypted == 0 and cfg.pause_on_pwrar() and (zf.needs_password() or is_cloaked(nzo, filepath, zf.namelist())) ): # Load all passwords passwords = get_all_passwords(nzo) # Cloaked job? if is_cloaked(nzo, filepath, zf.namelist()): encrypted = True elif not passwords: # Only error when no password was set nzo.encrypted = 1 encrypted = True else: # Lets test if any of the password work password_hit = False for password in passwords: if password: logging.info('Trying password "%s" on job "%s"', password, nzo.final_name) try: zf.setpassword(password) except rarfile.Error: # On weird passwords the setpassword() will fail # but the actual testrar() will work pass try: zf.testrar() password_hit = password break except rarfile.RarWrongPassword: # This one really didn't work pass except rarfile.RarCRCError as e: # CRC errors can be thrown for wrong password or # missing the next volume (with correct password) if "cannot find volume" in str(e).lower(): # We assume this one worked! password_hit = password break # This one didn't work pass except: # All the other errors we skip, they might be fixable in post-proc. # For example starting from the wrong volume, or damaged files # This will cause the check to be performed again for the next rar, might # be disk-intensive! Could be removed later and just accept the password. return encrypted, unwanted # Did any work? if password_hit: # We always trust the user's input if not nzo.password: nzo.password = password_hit # Don't check other files logging.info('Password "%s" matches for job "%s"', password_hit, nzo.final_name) nzo.encrypted = -1 encrypted = False else: # Encrypted and none of them worked nzo.encrypted = 1 encrypted = True # Check for unwanted extensions if cfg.unwanted_extensions() and cfg.action_on_unwanted_extensions(): for somefile in zf.namelist(): logging.debug("File contains: %s", somefile) if get_ext(somefile).replace(".", "").lower() in cfg.unwanted_extensions(): logging.debug("Unwanted file %s", somefile) unwanted = somefile zf.close() del zf except: logging.info("Error during inspection of RAR-file %s", filepath) logging.debug("Traceback: ", exc_info=True) return encrypted, unwanted
def rar_renamer(nzo: NzbObject, workdir): """ Deobfuscate rar file names: Use header and content information to give RAR-files decent names """ nzo.status = Status.VERIFYING nzo.set_unpack_info("Repair", T("Trying RAR renamer")) nzo.set_action_line(T("Trying RAR renamer"), "...") renamed_files = 0 # This is the most important datastructure (in case of mixed obfuscated rarsets) rarvolnr = {} # rarvolnr will contain per rar vol number the rarfilenames and their respective contents (and maybe other characteristics, like filesizes). # for example: rarvolnr[6]['somerandomfilename.rar']={'readme.txt', 'linux.iso'}, # which means 'somerandomfilename.rar' has rarvolnumber 6, and contents 'readme.txt' and 'linux.iso' # if we find a rarfile with rarvolnumber 7, and 'linux.iso' in it, we have a match! # The volume number and real extension of a (obfuscated) rar file # so volnrext['dfakjldfalkjdfl.blabla'] = (14, 'part014.rar') or (2, 'r000') # Not really needed, but handy to avoid a second lookup at the renaming volnrext = {} # Scan rar files in workdir, but not subdirs workdir_files = os.listdir(workdir) for file_to_check in workdir_files: file_to_check = os.path.join(workdir, file_to_check) # We only want files: if not (os.path.isfile(file_to_check)): continue # The function will check if it's a RAR-file # We do a sanity-check for the returned number rar_vol, new_extension = rarvolinfo.get_rar_extension(file_to_check) if 0 < rar_vol < 1000: logging.debug("Detected volume-number %s from RAR-header: %s ", rar_vol, file_to_check) volnrext[file_to_check] = (rar_vol, new_extension) # The files inside rar file rar_contents = rarfile.RarFile(os.path.join( workdir, file_to_check), single_file_check=True).filelist() try: rarvolnr[rar_vol] except: # does not yet exist, so create: rarvolnr[rar_vol] = {} rarvolnr[rar_vol][ file_to_check] = rar_contents # store them for matching (if needed) else: logging.debug("No RAR-volume-number found in %s", file_to_check) logging.debug("Deobfuscate: rarvolnr is: %s", rarvolnr) logging.debug("Deobfuscate: volnrext is: %s", volnrext) # Could be that there are no rar-files, we stop if not len(rarvolnr): return renamed_files # this can probably done with a max-key-lambda oneliner, but ... how? numberofrarsets = 0 for mykey in rarvolnr.keys(): numberofrarsets = max(numberofrarsets, len(rarvolnr[mykey])) logging.debug("Number of rarset is %s", numberofrarsets) if numberofrarsets == 1: # Just one obfuscated rarset ... that's easy logging.debug("Deobfuscate: Just one obfuscated rarset") for filename in volnrext: new_rar_name = "%s.%s" % (nzo.final_name, volnrext[filename][1]) new_rar_name = os.path.join(workdir, new_rar_name) new_rar_name = get_unique_filename(new_rar_name) logging.debug("Deobfuscate: Renaming %s to %s" % (filename, new_rar_name)) renamer(filename, new_rar_name) renamed_files += 1 return renamed_files # numberofrarsets bigger than 1, so a mixed rar set, so we need pre-checking # Sanity check of the rar set # Get the highest rar part number (that's the upper limit): highest_rar = sorted(rarvolnr.keys())[-1] # A staircase check: number of rarsets should no go up, but stay the same or go down how_many_previous = 1000 # 1000 rarset mixed ... should be enough ... typical is 1, 2 or maybe 3 # Start at part001.rar and go the highest for rar_set_number in range(1, highest_rar + 1): try: how_many_here = len(rarvolnr[rar_set_number]) except: # rarset does not exist at all logging.warning( "rarset %s is missing completely, so I can't deobfuscate.", rar_set_number) return 0 # OK, it exists, now let's check it's not higher if how_many_here > how_many_previous: # this should not happen: higher number of rarset than previous number of rarset logging.warning( "no staircase! rarset %s is higher than previous, so I can't deobfuscate.", rar_set_number) return 0 how_many_previous = how_many_here # OK, that looked OK (a declining staircase), so we can safely proceed # More than one obfuscated rarset, so we must do matching based of files inside the rar files # Assign (random) rar set names, first come first serve basis rarsetname = { } # in which rar set it should be, so rar set 'A', or 'B', or ... mychar = "A" # First things first: Assigning a rarsetname to the rar file which have volume number 1 for base_obfuscated_filename in rarvolnr[1]: rarsetname[base_obfuscated_filename] = mychar + "--" + nzo.final_name mychar = chr(ord(mychar) + 1) logging.debug("Deobfuscate: rarsetname %s", rarsetname) # Do the matching, layer by layer (read: rarvolnumber) # So, all rar files with rarvolnr 1, find the contents (files inside the rar), # and match with rarfiles with rarvolnr 2, and put them in the correct rarset. # And so on, until the highest rarvolnr minus 1 matched against highest rarvolnr for n in range(1, len(rarvolnr)): logging.debug( "Deobfuscate: Finding matches between rar sets %s and %s" % (n, n + 1)) for base_obfuscated_filename in rarvolnr[n]: matchcounter = 0 for next_obfuscated_filename in rarvolnr[n + 1]: # set() method with intersection (less strict): set(rarvolnr[n][base_obfuscated_filename]).intersection(set(rarvolnr[n+1][next_obfuscated_filename])) # check if the last filename inside the existing rar matches with the first filename in the following rar if rarvolnr[n][base_obfuscated_filename][-1] == rarvolnr[ n + 1][next_obfuscated_filename][0]: try: rarsetname[next_obfuscated_filename] = rarsetname[ base_obfuscated_filename] matchcounter += 1 except KeyError: logging.warning( T("No matching earlier rar file for %s"), next_obfuscated_filename) if matchcounter > 1: logging.info( "Deobfuscate: more than one match, so risk on false positive matching." ) # Do the renaming: for filename in rarsetname: new_rar_name = "%s.%s" % (rarsetname[filename], volnrext[filename][1]) new_rar_name = os.path.join(workdir, new_rar_name) new_rar_name = get_unique_filename(new_rar_name) logging.debug("Deobfuscate: Renaming %s to %s" % (filename, new_rar_name)) renamer(filename, new_rar_name) renamed_files += 1 # Done: The obfuscated rar files have now been renamed to regular formatted filenames return renamed_files
def check_encrypted_and_unwanted_files(nzo, filepath): """ Combines check for unwanted and encrypted files to save on CPU and IO """ encrypted = False unwanted = None if cfg.unwanted_extensions() or (nzo.encrypted == 0 and cfg.pause_on_pwrar()): # Safe-format for Windows # RarFile requires de-unicoded filenames for zf.testrar() filepath_split = os.path.split(filepath) workdir_short = short_path(filepath_split[0]) filepath = deunicode(os.path.join(workdir_short, filepath_split[1])) # Is it even a rarfile? if rarfile.is_rarfile(filepath): try: zf = rarfile.RarFile(filepath, all_names=True) # Check for encryption if nzo.encrypted == 0 and cfg.pause_on_pwrar() and (zf.needs_password() or is_cloaked(filepath, zf.namelist())): # Load all passwords passwords = get_all_passwords(nzo) # if no cryptography installed, only error when no password was set if not sabnzbd.HAVE_CRYPTOGRAPHY and not passwords: logging.info(T('%s missing'), 'Python Cryptography') nzo.encrypted = 1 encrypted = True elif sabnzbd.HAVE_CRYPTOGRAPHY: # Lets test if any of the password work password_hit = False rarfile.UNRAR_TOOL = sabnzbd.newsunpack.RAR_COMMAND for password in passwords: if password: logging.info('Trying password "%s" on job "%s"', password, nzo.final_name) try: zf.setpassword(password) zf.testrar() password_hit = password break except rarfile.RarCRCError: # On CRC error we can continue! password_hit = password break except: pass # Did any work? if password_hit: # Don't check other files logging.info('Password "%s" matches for job "%s"', password_hit, nzo.final_name) nzo.encrypted = -1 encrypted = False else: # Encrypted and none of them worked nzo.encrypted = 1 encrypted = True else: # Don't check other files nzo.encrypted = -1 encrypted = False # Check for unwanted extensions if cfg.unwanted_extensions(): for somefile in zf.namelist(): logging.debug('File contains: %s', somefile) if os.path.splitext(somefile)[1].replace('.', '').lower() in cfg.unwanted_extensions(): logging.debug('Unwanted file %s', somefile) unwanted = somefile zf.close() zf.close() del zf except: logging.debug('RAR file %s cannot be inspected', filepath) return encrypted, unwanted