credsweeper 1.11.2__tar.gz → 1.11.4__tar.gz

{credsweeper-1.11.2 → credsweeper-1.11.4}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: credsweeper
-Version: 1.11.2
+Version: 1.11.4
 Summary: Credential Sweeper
 Project-URL: Homepage, https://github.com/Samsung/CredSweeper
 Project-URL: Bug Tracker, https://github.com/Samsung/CredSweeper/issues

{credsweeper-1.11.2 → credsweeper-1.11.4}/credsweeper/__init__.py

@@ -18,4 +18,4 @@ __all__ = [
     '__version__'
 ]
 
-__version__ = "1.11.2"
+__version__ = "1.11.4"

{credsweeper-1.11.2 → credsweeper-1.11.4}/credsweeper/__main__.py

@@ -4,6 +4,7 @@ import os
 import sys
 import time
 from argparse import ArgumentParser, ArgumentTypeError, Namespace, BooleanOptionalAction
+from pathlib import Path
 from typing import Any, Union, Dict
 
 from credsweeper import __version__
@@ -62,7 +63,7 @@ def logger_levels(log_level: str) -> str:
     Returns True if log_level UPPERCASE is one of keys
     """
     val = log_level.upper()
-    if any(val == i for i in Logger.LEVELS.keys()):
+    if val in Logger.LEVELS:
         return val
     raise ArgumentTypeError(f"Log level provided: {log_level} -- must be one of: {' | '.join(Logger.LEVELS.keys())}")
 
@@ -88,10 +89,11 @@ def check_integrity() -> int:
     Returns CRC32 of files in integer
     """
     crc32 = 0
-    for root, dirs, files in os.walk(APP_PATH):
-        for file_path in files:
-            if Util.get_extension(file_path) in [".py", ".json", ".txt", ".yaml", ".onnx"]:
-                data = Util.read_data(os.path.join(root, file_path))
+    for root, _dirs, files in os.walk(APP_PATH):
+        for file_name in files:
+            if Util.get_extension(file_name) in [".py", ".json", ".txt", ".yaml", ".onnx"]:
+                file_path = Path(root) / file_name
+                data = Util.read_data(file_path)
                 if data:
                     crc32 ^= binascii.crc32(data)
     return crc32

{credsweeper-1.11.2 → credsweeper-1.11.4}/credsweeper/app.py

@@ -11,15 +11,17 @@ from colorama import Style
 # Directory of credsweeper sources MUST be placed before imports to avoid circular import error
 APP_PATH = Path(__file__).resolve().parent
 
-from credsweeper.common.constants import Severity, ThresholdPreset, DiffRowType
+from credsweeper.common.constants import Severity, ThresholdPreset, DiffRowType, DEFAULT_ENCODING
 from credsweeper.config import Config
 from credsweeper.credentials import Candidate, CredentialManager, CandidateKey
 from credsweeper.deep_scanner.deep_scanner import DeepScanner
+from credsweeper.file_handler.content_provider import ContentProvider
 from credsweeper.file_handler.diff_content_provider import DiffContentProvider
 from credsweeper.file_handler.file_path_extractor import FilePathExtractor
 from credsweeper.file_handler.abstract_provider import AbstractProvider
 from credsweeper.file_handler.text_content_provider import TextContentProvider
 from credsweeper.scanner import Scanner
+from credsweeper.ml_model.ml_validator import MlValidator
 from credsweeper.utils import Util
 
 logger = logging.getLogger(__name__)
@@ -94,7 +96,7 @@ class CredSweeper:
             log_level: str - level for pool initializer according logging levels (UPPERCASE)
 
         """
-        self.pool_count: int = int(pool_count) if int(pool_count) > 1 else 1
+        self.pool_count: int = max(1, int(pool_count))
         if not (_severity := Severity.get(severity)):
             raise RuntimeError(f"Severity level provided: {severity}"
                                f" -- must be one of: {' | '.join([i.value for i in Severity])}")
@@ -123,9 +125,9 @@ class CredSweeper:
         self.ml_config = ml_config
         self.ml_model = ml_model
         self.ml_providers = ml_providers
-        self.ml_validator = None
         self.__thrifty = thrifty
         self.__log_level = log_level
+        self.__ml_validator: Optional[MlValidator] = None
 
     # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # #
 
@@ -182,35 +184,22 @@ class CredSweeper:
 
     # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # #
 
-    # the import cannot be done on top due
-    # TypeError: cannot pickle 'onnxruntime.capi.onnxruntime_pybind11_state.InferenceSession' object
-    from credsweeper.ml_model import MlValidator
-
-    # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # #
-
     @property
     def ml_validator(self) -> MlValidator:
         """ml_validator getter"""
-        from credsweeper.ml_model import MlValidator
         if not self.__ml_validator:
-            self.__ml_validator: MlValidator = MlValidator(
+            self.__ml_validator = MlValidator(
                 threshold=self.ml_threshold,  #
                 ml_config=self.ml_config,  #
                 ml_model=self.ml_model,  #
                 ml_providers=self.ml_providers,  #
             )
-        assert self.__ml_validator, "self.__ml_validator was not initialized"
+        if not self.__ml_validator:
+            raise RuntimeError("MlValidator was not initialized!")
         return self.__ml_validator
 
     # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # #
 
-    @ml_validator.setter
-    def ml_validator(self, _ml_validator: Optional[MlValidator]) -> None:
-        """ml_validator setter"""
-        self.__ml_validator = _ml_validator
-
-    # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # #
-
     @staticmethod
     def pool_initializer(log_kwargs) -> None:
         """Ignore SIGINT in child processes."""
@@ -219,20 +208,6 @@ class CredSweeper:
 
     # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # #
 
-    @property
-    def config(self) -> Config:
-        """config getter"""
-        return self.__config
-
-    # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # #
-
-    @config.setter
-    def config(self, config: Config) -> None:
-        """config setter"""
-        self.__config = config
-
-    # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # #
-
     def run(self, content_provider: AbstractProvider) -> int:
         """Run an analysis of 'content_provider' object.
 
@@ -241,9 +216,10 @@ class CredSweeper:
 
         """
         _empty_list: Sequence[Union[DiffContentProvider, TextContentProvider]] = []
-        file_extractors: Sequence[Union[DiffContentProvider, TextContentProvider]] = \
-            content_provider.get_scannable_files(self.config) if content_provider else _empty_list
-        logger.info(f"Start Scanner for {len(file_extractors)} providers")
+        file_extractors = content_provider.get_scannable_files(self.config) if content_provider else _empty_list
+        if not file_extractors:
+            logger.info(f"No scannable targets for {len(content_provider.paths)} paths")
+            return 0
         self.scan(file_extractors)
         self.post_processing()
         # PatchesProvider has the attribute. Circular import error appears with using the isinstance
@@ -260,7 +236,7 @@ class CredSweeper:
             content_providers: file objects to scan
 
         """
-        if 1 < self.pool_count:
+        if 1 < self.pool_count and 1 < len(content_providers):
             self.__multi_jobs_scan(content_providers)
         else:
             self.__single_job_scan(content_providers)
@@ -269,6 +245,7 @@ class CredSweeper:
 
     def __single_job_scan(self, content_providers: Sequence[Union[DiffContentProvider, TextContentProvider]]) -> None:
         """Performs scan in main thread"""
+        logger.info(f"Scan for {len(content_providers)} providers")
         all_cred = self.files_scan(content_providers)
         self.credential_manager.set_credentials(all_cred)
 
@@ -284,12 +261,14 @@ class CredSweeper:
             if "SILENCE" == self.__log_level:
                 logging.addLevelName(60, "SILENCE")
             log_kwargs["level"] = self.__log_level
-        with multiprocessing.get_context("spawn").Pool(processes=self.pool_count,
-                                                       initializer=self.pool_initializer,
+        pool_count = min(self.pool_count, len(content_providers))
+        logger.info(f"Scan in {pool_count} processes for {len(content_providers)} providers")
+        with multiprocessing.get_context("spawn").Pool(processes=pool_count,
+                                                       initializer=CredSweeper.pool_initializer,
                                                        initargs=(log_kwargs, )) as pool:
             try:
-                for scan_results in pool.imap_unordered(self.files_scan, (content_providers[x::self.pool_count]
-                                                                          for x in range(self.pool_count))):
+                for scan_results in pool.imap_unordered(self.files_scan,
+                                                        (content_providers[x::pool_count] for x in range(pool_count))):
                     for cred in scan_results:
                         self.credential_manager.add_credential(cred)
             except KeyboardInterrupt:
@@ -301,9 +280,7 @@ class CredSweeper:
 
     # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # #
 
-    def files_scan(
-            self,  #
-            content_providers: Sequence[Union[DiffContentProvider, TextContentProvider]]) -> List[Candidate]:
+    def files_scan(self, content_providers: Sequence[ContentProvider]) -> List[Candidate]:
         """Auxiliary method for scan one sequence"""
         all_cred: List[Candidate] = []
         for provider in content_providers:
@@ -316,7 +293,7 @@ class CredSweeper:
 
     # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # #
 
-    def file_scan(self, content_provider: Union[DiffContentProvider, TextContentProvider]) -> List[Candidate]:
+    def file_scan(self, content_provider: ContentProvider) -> List[Candidate]:
         """Run scanning of file from 'file_provider'.
 
         Args:
@@ -415,7 +392,7 @@ class CredSweeper:
             if isinstance(change_type, DiffRowType):
                 # add suffix for appropriated reports to create two files for the patch scan
                 json_path = json_path.with_suffix(f".{change_type.value}{json_path.suffix}")
-            with open(json_path, 'w') as f:
+            with open(json_path, 'w', encoding=DEFAULT_ENCODING) as f:
                 # use the approach to reduce total memory usage in case of huge data
                 first_item = True
                 f.write('[\n')
@@ -446,8 +423,12 @@ class CredSweeper:
             for credential in credentials:
                 for line_data in credential.line_data_list:
                     # bright rule name and path or info
+                    if isinstance(credential.ml_probability, float):
+                        ml_probability_info = f" {credential.ml_probability:.6f}"
+                    else:
+                        ml_probability_info = ""
                     print(Style.BRIGHT + credential.rule_name +
-                          f" {line_data.info or line_data.path}:{line_data.line_num} {credential.ml_probability}" +
+                          f" {line_data.info or line_data.path}:{line_data.line_num}{ml_probability_info}" +
                           Style.RESET_ALL)
                     print(line_data.get_colored_line(hashed=self.hashed, subtext=self.subtext))
 

{credsweeper-1.11.2 → credsweeper-1.11.4}/credsweeper/common/constants.py

@@ -96,10 +96,6 @@ class Chars(Enum):
     ASCII_PRINTABLE = string.printable
 
 
-ENTROPY_LIMIT_BASE64 = 4.5
-ENTROPY_LIMIT_BASE3x = 3
-
-
 class GroupType(Enum):
     """Group type - used in Group constructor for load predefined set of filters"""
     KEYWORD = "keyword"
@@ -148,7 +144,8 @@ OVERLAP_SIZE = 1000
 CHUNK_STEP_SIZE = CHUNK_SIZE - OVERLAP_SIZE
 # ML hunk size to limit of variable or value size and get substring near value
 ML_HUNK = 80
-""" values according https://docs.python.org/3/library/codecs.html """
+
+# values according https://docs.python.org/3/library/codecs.html
 UTF_8 = "utf_8"
 UTF_16 = "utf_16"
 LATIN_1 = "latin_1"

{credsweeper-1.11.2 → credsweeper-1.11.4}/credsweeper/common/keyword_pattern.py

@@ -3,25 +3,30 @@ import re
 
 class KeywordPattern:
     """Pattern set of keyword types"""
-    key_left = r"(\\[nrt]|%[0-9a-f]{2})?" \
+    directive = r"(?P<directive>(?:(?:[#%]define|%global)(?:\s|\\t)|\bset))?"
+    key_left = r"(?:\\[nrt]|%[0-9a-f]{2}|\s)*" \
                r"(?P<variable>(([`'\"]{1,8}[^:='\"`}<>\\/&?]*|[^:='\"`}<>\s()\\/&?;,%]*)" \
                r"(?P<keyword>"
     # there will be inserted a keyword
     key_right = r")" \
-                r"[^%:='\"`<>{?!&;\n]*" \
+                r"[^%:='\"`<>({?!&;\n]*" \
                 r")" \
                 r"(&(quot|apos);|%[0-9a-f]{2}|[`'\"])*" \
                 r")"  # <variable>
-    separator = r"(\s|\\{1,8}[tnr])*\]?(\s|\\{1,8}[tnr])*" \
-                r"(?P<separator>:(\s[a-z]{3,9}[?]?\s)?=|:(?!:)|=(>|&gt;|(\\\\*u00|%)26gt;)|!==|!=|===|==|=|%3d)" \
+    separator = r"(?(directive)|(\s|\\{1,8}[tnr])*\]?(\s|\\{1,8}[tnr])*)" \
+                r"(?P<separator>:(\s[a-z]{3,9}[?]?\s)?=|:(?!:)|=(>|&gt;|(\\\\*u00|%)26gt;)|!==|!=|===|==|=~|=" \
+                r"|(?(directive)(\\t|\s|\((?!\))){1,80}|%3d))" \
                 r"(\s|\\{1,8}[tnr])*"
     # might be curly, square or parenthesis with words before
     wrap = r"(?P<wrap>(" \
            r"(new(\s|\\{1,8}[tnr]|byte|char|string|\[\]){1,8})?" \
+           r"(?P<get>([_a-z][0-9a-z_.\[\]]*\.)get|(os\.)?getenv)?" \
            r"([0-9a-z_.]|::|-(>|&gt;))*" \
-           r"[\[\(\{]" \
+           r"\s*" \
+           r"(\[(?!\])|\((?!\))|\{(?!\}))" \
            r"(\s|\\{1,8}[tnr])*" \
-           r"([0-9a-z_]{1,32}[:=]\s*)?" \
+           r"(?(get)('[^']+'|\"[^\"]+\")\s*,\s*|)" \
+           r"([0-9a-z_]{1,32}\s*[:=]\s*)?" \
            r"){1,8})?"
     string_prefix = r"(((b|r|br|rb|u|f|rf|fr|l|@)(?=(\\*[`'\"])))?"
     left_quote = r"(?P<value_leftquote>((?P<esq>\\{1,8})?([`'\"]|&(quot|apos);)){1,4}))?"
@@ -39,7 +44,7 @@ class KeywordPattern:
             r"(?P<url_esc>%[0-9a-f]{2})" \
             r"|" \
             r"(?(url_esc)[^\s`'\",;\\&]|[^\s`'\",;\\])" \
-            r")"\
+            r")" \
             r"){4,8000}" \
             r"|" \
             r"(<[^>]{4,8000}>)" \
@@ -48,18 +53,19 @@ class KeywordPattern:
             r"|" \
             r"(\$?\{{1,3}[^}]{4,8000}\}{1,3})" \
             r"|" \
-            r"(?(wrap)(?(value_leftquote)(?!\\(?P=value_leftquote))|[^\]\)\}]){16,8000})"\
+            r"(?(wrap)(?(value_leftquote)(?!\\(?P=value_leftquote))|[^\]\)\}]){16,8000})" \
             r")"  # <value>
     right_quote = r"(?(value_leftquote)" \
                   r"(?P<value_rightquote>(?<!\\)(?P=value_leftquote)|\\$|(?<=[0-9a-z+_/-])$)" \
                   r"|" \
-                  r"(?(wrap)(\]|\)|\}|,|;|\\|$))" \
+                  r"(?(wrap)(\]|\)|\}|;|\\|$))" \
                   r")"
 
     @classmethod
     def get_keyword_pattern(cls, keyword: str) -> re.Pattern:
         """Returns compiled regex pattern"""
         expression = ''.join([  #
+            cls.directive,  #
             cls.key_left,  #
             keyword,  #
             cls.key_right,  #

{credsweeper-1.11.2 → credsweeper-1.11.4}/credsweeper/common/morpheme_checklist.txt

@@ -885,7 +885,7 @@ mbler
 mean
 measur
 medi
-medusa
+medus
 meet
 mem_
 memb
@@ -925,7 +925,7 @@ month
 morp
 mory
 mote
-motorola
+motor
 mount
 move
 mpeg
@@ -1005,6 +1005,7 @@ origin
 orithm
 ormat
 orph
+otorola
 ottle
 ously
 out
@@ -1485,6 +1486,7 @@ up_
 updat
 upgrade
 url
+usa
 usb
 use
 usin

{credsweeper-1.11.2 → credsweeper-1.11.4}/credsweeper/credentials/candidate_key.py

@@ -24,7 +24,7 @@ class CandidateKey:
         return self.key == other.key
 
     def __ne__(self, other):
-        return not (self == other)
+        return not bool(self == other)
 
     def __repr__(self) -> str:
         return f"{self.key}:{self.__line}"

{credsweeper-1.11.2 → credsweeper-1.11.4}/credsweeper/credentials/credential_manager.py

@@ -110,7 +110,8 @@ class CredentialManager:
                 # Match by file path+line num+value. Value required so two different credentials still be
                 #  processed independently
                 candidate_key = CandidateKey(line_data)
-                if candidate_key not in groups:
-                    groups[candidate_key] = list()
-                groups[candidate_key].append(credential_candidate)
+                if candidate_key in groups:
+                    groups[candidate_key].append(credential_candidate)
+                else:
+                    groups[candidate_key] = [credential_candidate]
         return groups

{credsweeper-1.11.2 → credsweeper-1.11.4}/credsweeper/credentials/line_data.py

@@ -10,7 +10,6 @@ from colorama import Fore, Style
 from credsweeper.common.constants import MAX_LINE_LENGTH, UTF_8, StartEnd, ML_HUNK
 from credsweeper.config import Config
 from credsweeper.utils import Util
-from credsweeper.utils.entropy_validator import EntropyValidator
 
 
 class LineData:
@@ -32,7 +31,7 @@ class LineData:
     """
 
     quotation_marks = ('"', "'", '`')
-    comment_starts = ("//", "* ", "#", "/*", "<!––", "%{", "%", "...", "(*", "--", "--[[", "#=")
+    comment_starts = ("//", "* ", "# ", "/*", "<!––", "%{", "%", "...", "(*", "--", "--[[", "#=")
     bash_param_split = re.compile("\\s+(\\-|\\||\\>|\\w+?\\>|\\&)")
     line_endings = re.compile(r"\\{1,8}[nr]")
     # https://en.wikipedia.org/wiki/Percent-encoding
@@ -87,8 +86,9 @@ class LineData:
         self.url_part = False
         self.wrap = None
         self._3d_escaped_separator = False
-
         self.initialize(match_obj)
+        # the line is very useful for debug breakpoint
+        pass  # pylint: disable=W0107
 
     def compare(self, other: 'LineData') -> bool:
         """Comparison method - skip whole line and checks only when variable and value are the same"""
@@ -327,11 +327,8 @@ class LineData:
             True if file require quotation, False otherwise
 
         """
-        if not self.path:
-            return False
-        if Util.get_extension(self.path) in self.config.source_quote_ext:
-            return True
-        return False
+        file_type = self.file_type or Util.get_extension(self.path)
+        return bool(file_type) and file_type in self.config.source_quote_ext
 
     @staticmethod
     def get_hash_or_subtext(
@@ -373,10 +370,10 @@ class LineData:
     def to_str(self, subtext: bool = False, hashed: bool = False) -> str:
         """Represent line_data with subtext or|and hashed values"""
         cut_pos = StartEnd(self.variable_start, self.value_end) if subtext else None
-        return f"line: '{self.get_hash_or_subtext(self.line, hashed, cut_pos)}'" \
-               f" | line_num: {self.line_num} | path: {self.path}" \
+        return f"path: {self.path}" \
+               f" | line_num: {self.line_num}" \
                f" | value: '{self.get_hash_or_subtext(self.value, hashed)}'" \
-               f" | entropy_validation: {EntropyValidator(self.value)}"
+               f" | line: '{self.get_hash_or_subtext(self.line, hashed, cut_pos)}'"
 
     def __str__(self):
         return self.to_str()
@@ -393,6 +390,10 @@ class LineData:
         """
         cut_pos = StartEnd(self.variable_start if 0 <= self.variable_start else self.value_start,
                            self.value_end) if subtext else None
+        if isinstance(self.value, str):
+            entropy = round(Util.get_shannon_entropy(self.value), 5)
+        else:
+            entropy = None
         full_output = {
             "key": self.key,
             "line": self.get_hash_or_subtext(self.line, hashed, cut_pos),
@@ -401,18 +402,18 @@ class LineData:
             # info may contain variable name - so let it be hashed if requested
             "info": self.get_hash_or_subtext(self.info, hashed),
             "pattern": self.pattern.pattern,
+            "variable": self.get_hash_or_subtext(self.variable, hashed),
+            "variable_start": self.variable_start,
+            "variable_end": self.variable_end,
             "separator": self.separator,
             "separator_start": self.separator_start,
             "separator_end": self.separator_end,
             "value": self.get_hash_or_subtext(self.value, hashed),
             "value_start": self.value_start,
             "value_end": self.value_end,
-            "variable": self.get_hash_or_subtext(self.variable, hashed),
-            "variable_start": self.variable_start,
-            "variable_end": self.variable_end,
+            "entropy": entropy,
             "value_leftquote": self.value_leftquote,
             "value_rightquote": self.value_rightquote,
-            "entropy_validation": EntropyValidator(self.value).to_dict()
         }
         reported_output = {k: v for k, v in full_output.items() if k in self.config.line_data_output}
         return reported_output

{credsweeper-1.11.2 → credsweeper-1.11.4}/credsweeper/deep_scanner/abstract_scanner.py

@@ -1,5 +1,5 @@
 from abc import abstractmethod, ABC
-from typing import List
+from typing import List, Optional
 
 from credsweeper.config import Config
 from credsweeper.credentials import Candidate
@@ -40,3 +40,12 @@ class AbstractScanner(ABC):
             recursive_limit_size: int) -> List[Candidate]:
         """Abstract method to be defined in DeepScanner"""
         raise NotImplementedError(__name__)
+
+    @abstractmethod
+    def data_scan(
+            self,  #
+            data_provider: DataContentProvider,  #
+            depth: int,  #
+            recursive_limit_size: int) -> Optional[List[Candidate]]:
+        """Abstract method to be defined in DeepScanner"""
+        raise NotImplementedError(__name__)

credsweeper-1.11.4/credsweeper/deep_scanner/deb_scanner.py

@@ -0,0 +1,48 @@
+import logging
+from abc import ABC
+from typing import List, Optional
+
+from credsweeper.common.constants import ASCII, MIN_DATA_LEN
+from credsweeper.credentials import Candidate
+from credsweeper.deep_scanner.abstract_scanner import AbstractScanner
+from credsweeper.file_handler.data_content_provider import DataContentProvider
+
+logger = logging.getLogger(__name__)
+
+
+class DebScanner(AbstractScanner, ABC):
+    """Implements deb (ar) scanning"""
+
+    def data_scan(
+            self,  #
+            data_provider: DataContentProvider,  #
+            depth: int,  #
+            recursive_limit_size: int) -> Optional[List[Candidate]]:
+        """Extracts data file from .ar (debian) archive and launches data_scan"""
+        candidates: Optional[List[Candidate]] = None
+        offset = 8  # b"!<arch>\n"
+        while offset < len(data_provider.data):
+            try:
+                file_size_data = data_provider.data[offset + 48:offset + 58]
+                file_size = int(file_size_data.decode(ASCII))
+                offset += 60
+                if file_size < MIN_DATA_LEN:
+                    offset += file_size
+                    continue
+                data = data_provider.data[offset:offset + file_size]
+                deb_content_provider = DataContentProvider(data=data,
+                                                           file_path=data_provider.file_path,
+                                                           file_type=data_provider.file_type,
+                                                           info=f"{data_provider.info}|DEB:0x{offset:x}")
+                new_limit = recursive_limit_size - file_size
+                deb_candidates = self.recursive_scan(deb_content_provider, depth, new_limit)
+                if deb_candidates is not None:
+                    if candidates:
+                        candidates.extend(deb_candidates)
+                    else:
+                        candidates = deb_candidates
+                # data padding = 2
+                offset += 1 + file_size if 1 & file_size else file_size
+            except Exception as exc:
+                logger.error(exc)
+        return candidates