credsweeper 1.11.1__py3-none-any.whl → 1.11.3__py3-none-any.whl

credsweeper/scanner/scan_type/scan_type.py

@@ -38,13 +38,12 @@ class ScanType(ABC):
         raise NotImplementedError()
 
     @classmethod
-    def filtering(cls, config: Config, target: AnalysisTarget, line_data: LineData, filters: List[Filter]) -> bool:
+    def filtering(cls, target: AnalysisTarget, line_data: LineData, filters: List[Filter]) -> bool:
         """Check if line data should be removed based on filters.
 
         If `use_filters` option is false, always return False
 
         Args:
-            config: dict of credsweeper configuration
             target: AnalysisTarget from which `line_data` was obtained
             line_data: Line data to check with `filters`
             filters: Filters to use
@@ -112,7 +111,7 @@ class ScanType(ABC):
                     bypass_start = line_data.value_end
                     bypass_end = offset_end
 
-                if config.use_filters and cls.filtering(config, target, line_data, filters):
+                if config.use_filters and cls.filtering(target, line_data, filters):
                     if line_data.variable and 0 <= line_data.variable_start < line_data.variable_end:
                         # may be next matched item will be not filtered - let search it after variable
                         bypass_start = line_data.variable_end

credsweeper/scanner/scanner.py

@@ -146,7 +146,13 @@ class Scanner:
             # "cache" - YAPF and pycharm formatters ...
             matched_keyword = \
                 target_line_stripped_len >= self.min_keyword_len and (  #
-                        '=' in target_line_stripped or ':' in target_line_stripped)  #
+                        '=' in target_line_stripped
+                        or ':' in target_line_stripped
+                        or "set" in target_line_stripped
+                        or "#define" in target_line_stripped
+                        or "%define" in target_line_stripped
+                        or "%global" in target_line_stripped
+                )  #
             matched_pem_key = \
                 target_line_stripped_len >= self.min_pem_key_len \
                 and PEM_BEGIN_PATTERN in target_line_stripped and "PRIVATE" in target_line_stripped

credsweeper/secret/config.json

@@ -2,10 +2,13 @@
     "exclude": {
         "pattern": [],
         "containers": [
+            ".aar",
             ".apk",
             ".bz2",
             ".gz",
+            ".lzma",
             ".tar",
+            ".xz",
             ".zip"
         ],
         "documents": [
@@ -20,17 +23,20 @@
         ],
         "extension": [
             ".7z",
+            ".a",
             ".aac",
-            ".aar",
             ".avi",
+            ".bin",
             ".bmp",
             ".class",
             ".css",
             ".dmg",
             ".ear",
             ".eot",
+            ".elf",
             ".exe",
             ".gif",
+            ".gmo",
             ".ico",
             ".img",
             ".info",
@@ -45,6 +51,7 @@
             ".mp4",
             ".npy",
             ".npz",
+            ".obj",
             ".ogg",
             ".pak",
             ".png",
@@ -52,10 +59,13 @@
             ".pyc",
             ".pyd",
             ".pyo",
+            ".rar",
             ".rc",
             ".rc2",
             ".rar",
             ".realm",
+            ".res",
+            ".rpm",
             ".s7z",
             ".scss",
             ".so",
@@ -70,6 +80,7 @@
             ".wav",
             ".webm",
             ".webp",
+            ".wma",
             ".woff",
             ".yuv"
         ],
@@ -160,13 +171,13 @@
         "line_num",
         "path",
         "info",
-        "value",
-        "value_start",
-        "value_end",
         "variable",
         "variable_start",
         "variable_end",
-        "entropy_validation"
+        "value",
+        "value_start",
+        "value_end",
+        "entropy"
     ],
     "candidate_output": [
         "rule",

credsweeper/utils/pem_key_detector.py

@@ -4,12 +4,11 @@ import re
 import string
 from typing import List
 
-from credsweeper.common.constants import PEM_BEGIN_PATTERN, PEM_END_PATTERN, Chars
+from credsweeper.common.constants import PEM_BEGIN_PATTERN, PEM_END_PATTERN, ENTROPY_LIMIT_BASE64
 from credsweeper.config import Config
 from credsweeper.credentials import LineData
 from credsweeper.file_handler.analysis_target import AnalysisTarget
 from credsweeper.utils import Util
-from credsweeper.utils.entropy_validator import EntropyValidator
 
 logger = logging.getLogger(__name__)
 
@@ -68,10 +67,10 @@ class PemKeyDetector:
                 elif PEM_END_PATTERN in subline:
                     if "PGP" in target.line_strip:
                         # Check if entropy is high enough for base64 set with padding sign
-                        entropy_validator = EntropyValidator(key_data, Chars.BASE64STDPAD_CHARS)
-                        if entropy_validator.valid:
+                        entropy = Util.get_shannon_entropy(key_data)
+                        if ENTROPY_LIMIT_BASE64 <= entropy:
                             return line_data
-                        logger.debug("Filtered with entropy %f '%s'", entropy_validator.entropy, key_data)
+                        logger.debug("Filtered with entropy %f '%s'", entropy, key_data)
                     if "OPENSSH" in target.line_strip:
                         # Check whether the key is encrypted
                         with contextlib.suppress(Exception):

credsweeper/utils/util.py

@@ -12,13 +12,14 @@ from dataclasses import dataclass
 from pathlib import Path
 from typing import Any, Dict, List, Tuple, Optional, Union
 
+import numpy as np
 import whatthepatch
 import yaml
 from lxml import etree
 from typing_extensions import TypedDict
 
 from credsweeper.common.constants import DiffRowType, AVAILABLE_ENCODINGS, \
-    DEFAULT_ENCODING, LATIN_1, CHUNK_SIZE, MAX_LINE_LENGTH, CHUNK_STEP_SIZE
+    DEFAULT_ENCODING, LATIN_1, CHUNK_SIZE, MAX_LINE_LENGTH, CHUNK_STEP_SIZE, ASCII
 
 logger = logging.getLogger(__name__)
 
@@ -65,21 +66,17 @@ class Util:
         return result
 
     @staticmethod
-    def get_shannon_entropy(data: str, iterator: str) -> float:
+    def get_shannon_entropy(data: Union[str, bytes]) -> float:
         """Borrowed from http://blog.dkbza.org/2007/05/scanning-data-for-entropy-anomalies.html."""
         if not data:
-            return 0
-
-        entropy = 0.
-        data_len = float(len(data))
-        for x in iterator:
-            p_x = data.count(x) / data_len
-            if p_x > 0:
-                entropy += -p_x * math.log(p_x, 2)
-
+            return 0.
+        size = len(data)
+        _uniq, counts = np.unique(list(data), return_counts=True)
+        probabilities = counts / size
+        entropy = float(-np.sum(probabilities * np.log2(probabilities)))
         return entropy
 
-    """Precalculated data for speedup"""
+    # Precalculated data for speedup
     MIN_DATA_ENTROPY: Dict[int, float] = {
         16: 1.66973671780348,
         20: 2.07723544540831,
@@ -153,41 +150,39 @@ class Util:
         return entropy < min_entropy
 
     @staticmethod
-    def is_known(data: bytes) -> bool:
-        """
-        Returns true if any recognized binary format found
-        """
-        if Util.is_zip(data) \
-                or Util.is_gzip(data) \
-                or Util.is_tar(data) \
-                or Util.is_bzip2(data) \
-                or Util.is_com(data) \
-                or Util.is_pdf(data) \
-                or Util.is_elf(data):
-            return True
+    def is_known(data: Union[bytes, bytearray]) -> bool:
+        """Returns True if any known binary format is found to prevent extra scan a file without an extension."""
+        if isinstance(data, (bytes, bytearray)):
+            if 127 <= len(data) and data.startswith(b"\x7f\x45\x4c\x46"):
+                # https://en.wikipedia.org/wiki/Executable_and_Linkable_Format
+                # minimal ELF is 127 bytes https://github.com/tchajed/minimal-elf
+                return True
         return False
 
     @staticmethod
-    def is_binary(data: bytes) -> bool:
+    def is_binary(data: Union[bytes, bytearray]) -> bool:
         """
-        Returns True when two zeroes sequence is found which never exists in text format (UTF-8, UTF-16)
-        UTF-32 is not supported
+        Returns True when two zeroes sequence is found in begin of data.
+        The sequence never exists in text format (UTF-8, UTF-16). UTF-32 is not supported.
         """
         if 0 <= data.find(b"\0\0", 0, MAX_LINE_LENGTH):
             return True
-        non_ascii_cnt = 0
-        for n in range(min([len(data), MAX_LINE_LENGTH])):
-            i = data[n]
-            if 0x20 > i and i not in (0x09, 0x0A, 0x0D) or 0x7E < i < 0xA0:
-                # less than space and not tab, line feed, line end
-                non_ascii_cnt += 1
+        else:
+            return False
+
+    NOT_LATIN1_PRINTABLE_SET = (set(range(0,
+                                          256)).difference(set(x for x in string.printable.encode(ASCII))).difference(
+                                              set(x for x in range(0xA0, 0x100))))
+
+    @staticmethod
+    def is_latin1(data: Union[bytes, bytearray]) -> bool:
+        """Returns True when data looks like LATIN-1 for first MAX_LINE_LENGTH bytes."""
+        result = False
         if data:
+            non_latin1_cnt = sum(1 for x in data[:MAX_LINE_LENGTH] if x in Util.NOT_LATIN1_PRINTABLE_SET)
             # experiment for 255217 binary files shown avg = 0.268264 ± 0.168767, so let choose minimal
-            chunk_len = float(MAX_LINE_LENGTH if MAX_LINE_LENGTH < len(data) else len(data))
-            result = 0.1 < non_ascii_cnt / chunk_len
-        else:
-            # empty data case
-            result = False
+            chunk_len = min(MAX_LINE_LENGTH, len(data))
+            result = 0.1 > non_latin1_cnt / chunk_len
         return result
 
     @staticmethod
@@ -231,7 +226,7 @@ class Util:
             encodings = AVAILABLE_ENCODINGS
         for encoding in encodings:
             try:
-                if binary_suggest and LATIN_1 == encoding and (Util.is_known(content) or Util.is_binary(content)):
+                if binary_suggest and LATIN_1 == encoding and (Util.is_binary(content) or not Util.is_latin1(content)):
                     # LATIN_1 may convert data (bytes in range 0x80:0xFF are transformed)
                     # so skip this encoding when checking binaries
                     logger.warning("Binary file detected")
@@ -374,7 +369,7 @@ class Util:
             line = change["line"]
             if isinstance(line, str):
                 rows_data.extend(Util.preprocess_diff_rows(change.get("new"), change.get("old"), line))
-            elif isinstance(line, bytes):
+            elif isinstance(line, (bytes, bytearray)):
                 logger.warning("The feature is available with the deep scan option")
             else:
                 logger.error(f"Unknown type of line {type(line)}")
@@ -382,9 +377,9 @@ class Util:
         return rows_data
 
     @staticmethod
-    def is_zip(data: bytes) -> bool:
+    def is_zip(data: Union[bytes, bytearray]) -> bool:
         """According https://en.wikipedia.org/wiki/List_of_file_signatures"""
-        if isinstance(data, bytes) and 3 < len(data):
+        if isinstance(data, (bytes, bytearray)) and 3 < len(data):
             # PK
             if data.startswith(b"PK"):
                 if 0x03 == data[2] and 0x04 == data[3]:
@@ -398,18 +393,18 @@ class Util:
         return False
 
     @staticmethod
-    def is_com(data: bytes) -> bool:
+    def is_com(data: Union[bytes, bytearray]) -> bool:
         """According https://en.wikipedia.org/wiki/List_of_file_signatures"""
-        if isinstance(data, bytes) and 8 < len(data):
+        if isinstance(data, (bytes, bytearray)) and 8 < len(data):
             if data.startswith(b"\xD0\xCF\x11\xE0\xA1\xB1\x1A\xE1"):
                 # Compound File Binary Format: doc, xls, ppt, msi, msg
                 return True
         return False
 
     @staticmethod
-    def is_tar(data: bytes) -> bool:
+    def is_tar(data: Union[bytes, bytearray]) -> bool:
         """According https://en.wikipedia.org/wiki/List_of_file_signatures"""
-        if isinstance(data, bytes) and 512 <= len(data):
+        if isinstance(data, (bytes, bytearray)) and 512 <= len(data):
             if 0x75 == data[257] and 0x73 == data[258] and 0x74 == data[259] \
                     and 0x61 == data[260] and 0x72 == data[261] and (
                     0x00 == data[262] and 0x30 == data[263] and 0x30 == data[264]
@@ -425,9 +420,9 @@ class Util:
         return False
 
     @staticmethod
-    def is_bzip2(data: bytes) -> bool:
+    def is_bzip2(data: Union[bytes, bytearray]) -> bool:
         """According https://en.wikipedia.org/wiki/Bzip2"""
-        if isinstance(data, bytes) and 10 <= len(data):
+        if isinstance(data, (bytes, bytearray)) and 10 <= len(data):
             if data.startswith(b"\x42\x5A\x68") \
                     and 0x31 <= data[3] <= 0x39 \
                     and 0x31 == data[4] and 0x41 == data[5] and 0x59 == data[6] \
@@ -436,34 +431,41 @@ class Util:
         return False
 
     @staticmethod
-    def is_gzip(data: bytes) -> bool:
+    def is_gzip(data: Union[bytes, bytearray]) -> bool:
         """According https://www.rfc-editor.org/rfc/rfc1952"""
-        if isinstance(data, bytes) and 3 <= len(data):
+        if isinstance(data, (bytes, bytearray)) and 3 <= len(data):
             if data.startswith(b"\x1F\x8B\x08"):
                 return True
         return False
 
     @staticmethod
-    def is_pdf(data: bytes) -> bool:
+    def is_pdf(data: Union[bytes, bytearray]) -> bool:
         """According https://en.wikipedia.org/wiki/List_of_file_signatures - pdf"""
-        if isinstance(data, bytes) and 5 <= len(data):
+        if isinstance(data, (bytes, bytearray)) and 5 <= len(data):
             if data.startswith(b"\x25\x50\x44\x46\x2D"):
                 return True
         return False
 
     @staticmethod
-    def is_jks(data: bytes) -> bool:
+    def is_jks(data: Union[bytes, bytearray]) -> bool:
         """According https://en.wikipedia.org/wiki/List_of_file_signatures - jks"""
-        if isinstance(data, bytes) and 4 <= len(data):
+        if isinstance(data, (bytes, bytearray)) and 4 <= len(data):
             if data.startswith(b"\xFE\xED\xFE\xED"):
                 return True
         return False
 
     @staticmethod
-    def is_asn1(data: bytes) -> bool:
+    def is_lzma(data: Union[bytes, bytearray]) -> bool:
+        """According https://en.wikipedia.org/wiki/List_of_file_signatures - lzma also xz"""
+        if isinstance(data, (bytes, bytearray)) and 6 <= len(data):
+            if data.startswith(b"\xFD\x37\x7A\x58\x5A\x00") or data.startswith(b"\x5D\x00\x00"):
+                return True
+        return False
+
+    @staticmethod
+    def is_asn1(data: Union[bytes, bytearray]) -> bool:
         """Only sequence type 0x30 and size correctness is checked"""
-        data_length = len(data)
-        if isinstance(data, bytes) and 4 <= data_length:
+        if isinstance(data, (bytes, bytearray)) and 4 <= len(data):
             # sequence
             if 0x30 == data[0]:
                 # https://www.oss.com/asn1/resources/asn1-made-simple/asn1-quick-reference/basic-encoding-rules.html#Lengths
@@ -471,7 +473,7 @@ class Util:
                 byte_len = (0x7F & length)
                 if 0x80 == length and data.endswith(b"\x00\x00"):
                     return True
-                elif 0x80 < length and 1 < byte_len < data_length:  # additional check
+                elif 0x80 < length and 1 < byte_len < len(data):  # additional check
                     len_bytes = data[2:2 + byte_len]
                     try:
                         long_size = struct.unpack(">h", len_bytes)
@@ -482,26 +484,17 @@ class Util:
                     length = data[2]
                 else:
                     byte_len = 0
-                return data_length == length + 2 + byte_len
-        return False
-
-    @staticmethod
-    def is_elf(data: Union[bytes, bytearray]) -> bool:
-        """According to https://en.wikipedia.org/wiki/Executable_and_Linkable_Format use only 5 bytes"""
-        if isinstance(data, (bytes, bytearray)) and 127 <= len(data):
-            # minimal is 127 bytes https://github.com/tchajed/minimal-elf
-            if data.startswith(b"\x7f\x45\x4c\x46") and (0x01 == data[5] or 0x02 == data[5]):
-                return True
+                return len(data) == length + 2 + byte_len
         return False
 
     @staticmethod
     def is_html(data: Union[bytes, bytearray]) -> bool:
         """Used to detect html format. Suppose, invocation of is_xml() was True before."""
         if isinstance(data, (bytes, bytearray)):
-            for opening_tag, closing_tag in [(b"<html>", b"</html>"), (b"<table", b"</table>"), (b"<p>", b"</p>"),
-                                             (b"<span>", b"</span>"), (b"<div>", b"</div>"), (b"<li>", b"</li>"),
-                                             (b"<ol>", b"</ol>"), (b"<ul>", b"</ul>"), (b"<th>", b"</th>"),
-                                             (b"<tr>", b"</tr>"), (b"<td>", b"</td>")]:
+            for opening_tag, closing_tag in [(b"<html", b"</html>"), (b"<body", b"</body>"), (b"<table", b"</table>"),
+                                             (b"<p>", b"</p>"), (b"<span>", b"</span>"), (b"<div>", b"</div>"),
+                                             (b"<li>", b"</li>"), (b"<ol>", b"</ol>"), (b"<ul>", b"</ul>"),
+                                             (b"<th>", b"</th>"), (b"<tr>", b"</tr>"), (b"<td>", b"</td>")]:
                 opening_pos = data.find(opening_tag, 0, MAX_LINE_LENGTH)
                 if 0 <= opening_pos < data.find(closing_tag, opening_pos):
                     # opening and closing tags were found - suppose it is an HTML
@@ -658,81 +651,11 @@ class Util:
         except Exception as exc:
             logging.error(f"Failed to write: {file_path} {exc}")
 
-    @staticmethod
-    def __extract_value(node: Any, value: Any) -> List[Any]:
-        result = []
-        for i in getattr(node, "targets"):
-            if hasattr(i, "id"):
-                result.append({getattr(i, "id"): value})
-            else:
-                logger.error(f"{str(i)} has no 'id'")
-        return result
-
-    @staticmethod
-    def __extract_assign(node: Any) -> List[Any]:
-        result = []
-        if hasattr(node, "value") and hasattr(node, "targets"):
-            value = getattr(node, "value")
-            if hasattr(value, "value"):
-                # python 3.8 - 3.10
-                result.extend(Util.__extract_value(node, getattr(value, "value")))
-            else:
-                logger.error(f"value.{value} has no 'value' {dir(value)}")
-        else:
-            logger.error(f"{str(node)} has no 'value' {dir(node)}")
-        return result
-
-    @staticmethod
-    def ast_to_dict(node: Any) -> List[Any]:
-        """Recursive parsing AST tree of python source to list with strings"""
-        result: List[Any] = []
-        if hasattr(node, "value") and isinstance(node.value, str):
-            result.append(node.value)
-
-        if isinstance(node, ast.Module) \
-                or isinstance(node, ast.FunctionDef):
-            if hasattr(node, "body"):
-                for i in node.body:
-                    x = Util.ast_to_dict(i)
-                    if x:
-                        result.extend(x)
-        elif isinstance(node, ast.Import):
-            logger.debug("Import:%s", str(node))
-        elif isinstance(node, ast.Assign):
-            result.extend(Util.__extract_assign(node))
-        elif isinstance(node, ast.Expr) \
-                or isinstance(node, ast.AnnAssign) \
-                or isinstance(node, ast.AugAssign) \
-                or isinstance(node, ast.Call) \
-                or isinstance(node, ast.JoinedStr) \
-                or isinstance(node, ast.Return) \
-                or isinstance(node, ast.ImportFrom) \
-                or isinstance(node, ast.Assert) \
-                or isinstance(node, ast.Pass) \
-                or isinstance(node, ast.Raise) \
-                or isinstance(node, ast.Str) \
-                or isinstance(node, ast.Name) \
-                or isinstance(node, ast.FormattedValue) \
-                or isinstance(node, ast.Global):
-            if hasattr(node, "value"):
-                result.extend(Util.ast_to_dict(getattr(node, "value")))
-            if hasattr(node, "args"):
-                for i in getattr(node, "args"):
-                    result.extend(Util.ast_to_dict(i))
-            if hasattr(node, "values"):
-                for i in getattr(node, "values"):
-                    result.extend(Util.ast_to_dict(i))
-            else:
-                logger.debug(f"skip:{str(node)}")
-        else:
-            logger.debug(f"unknown:{str(node)}")
-        return result
-
     @staticmethod
     def parse_python(source: str) -> List[Any]:
-        """Parse python source to list of strings and assignments"""
+        """Parse python source and back to remove strings merge and line wrap"""
         src = ast.parse(source)
-        result = Util.ast_to_dict(src)
+        result = ast.unparse(src).splitlines()
         return result
 
     @staticmethod

{credsweeper-1.11.1.dist-info → credsweeper-1.11.3.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: credsweeper
-Version: 1.11.1
+Version: 1.11.3
 Summary: Credential Sweeper
 Project-URL: Homepage, https://github.com/Samsung/CredSweeper
 Project-URL: Bug Tracker, https://github.com/Samsung/CredSweeper/issues