Raise errors for inputs that would cause blank outputs; handle length in CLI for variable length digest algorithms (#16)

* Raise errors for inputs that would cause blank outputs with custom exceptions for blank output errors and invalid length param use, with testing

* Add length parameter option for certain algs

* Refactor hash tests with parametrize

---------

Co-authored-by: Elizabeth Campolongo <[email protected]>

Author: thompsonmj

README.md

@@ -36,6 +36,8 @@ options:
                         Hash algorithm to use (default: md5; available: ripemd160, sha3_224, sha512_224, blake2b, sha384, sha256, sm3, sha3_256, shake_256, sha512, sha1, sha224, md5, md5-sha1, sha3_384, sha3_512, sha512_256, shake_128, blake2s)
 ```
 
+> Note: The available algorithms are determined by those available to `hashlib` and may vary depending on your system and OpenSSL version, so the set shown on your system with `sum-buddy -h` may be different from above. At a minimum, it should include: `{blake2s, blake2b, md5, sha1, sha224, sha256, sha384, sha512, sha3_224, sha3_256, sha3_384, sha3_512, shake_128, shake_256}`, which is given by `hashlib.algorithms_guaranteed`.
+
 #### CLI Examples
 
 - **Basic Usage:**

src/sumbuddy/__main__.py

@@ -2,13 +2,14 @@
 from sumbuddy.hasher import Hasher
 from sumbuddy.mapper import Mapper
 from sumbuddy.filter import Filter
+from sumbuddy.exceptions import EmptyInputDirectoryError, NoFilesAfterFilteringError, LengthUsedForFixedLengthHashError
 import csv
 import hashlib
 from tqdm import tqdm
 import sys
 import os
 
-def get_checksums(input_directory, output_filepath=None, ignore_file=None, include_hidden=False, algorithm='md5'):
+def get_checksums(input_directory, output_filepath=None, ignore_file=None, include_hidden=False, algorithm='md5', length=None):
     """
     Generate a CSV file with the filepath, filename, and checksum of all files in the input directory according to patterns to ignore. Checksum column is labeled by the selected algorithm (e.g., 'md5' or 'sha256').
     
@@ -19,10 +20,14 @@ def get_checksums(input_directory, output_filepath=None, ignore_file=None, inclu
     ignore_file - String [optional]. Filepath for the ignore patterns file.
     include_hidden - Boolean [optional]. Whether to include hidden files. Default is False.
     algorithm - String. Algorithm to use for checksums. Default: 'md5', see options with 'hashlib.algorithms_available'.
+    length - Integer [conditionally optional]. Length of the digest for SHAKE (required) and BLAKE (optional) algorithms in bytes.
     """
     mapper = Mapper()
-    file_paths = mapper.gather_file_paths(input_directory, ignore_file=ignore_file, include_hidden=include_hidden)
-    
+    try:
+        file_paths = mapper.gather_file_paths(input_directory, ignore_file=ignore_file, include_hidden=include_hidden)
+    except (EmptyInputDirectoryError, NoFilesAfterFilteringError) as e:
+        sys.exit(str(e))
+
     # Exclude the output file from being hashed
     if output_filepath:
         output_file_abs_path = os.path.abspath(output_filepath)
@@ -37,7 +42,7 @@ def get_checksums(input_directory, output_filepath=None, ignore_file=None, inclu
 
         disable_tqdm = output_filepath is None
         for file_path in tqdm(file_paths, desc=f"Calculating {algorithm} checksums on {input_directory}", disable=disable_tqdm):
-            checksum = hasher.checksum_file(file_path)
+            checksum = hasher.checksum_file(file_path, algorithm=algorithm, length=length)
             writer.writerow([file_path, os.path.basename(file_path), checksum])
 
     finally:
@@ -58,6 +63,7 @@ def main():
     group.add_argument("-i", "--ignore-file", help="Filepath for the ignore patterns file")
     group.add_argument("-H", "--include-hidden", action="store_true", help="Include hidden files")
     parser.add_argument("-a", "--algorithm", default="md5", help=f"Hash algorithm to use (default: md5; available: {available_algorithms})")
+    parser.add_argument("-l", "--length", type=int, help="Length of the digest for SHAKE (required) or BLAKE (optional) algorithms in bytes")
 
     args = parser.parse_args()
 
@@ -69,7 +75,11 @@ def main():
         if overwrite.lower() != 'y':
             sys.exit("Exited without executing")
 
-    get_checksums(args.input_dir, args.output_file, args.ignore_file, args.include_hidden, args.algorithm)
+    try:
+        get_checksums(args.input_dir, args.output_file, args.ignore_file, args.include_hidden, args.algorithm, args.length)
+    except (LengthUsedForFixedLengthHashError) as e:
+        sys.exit(str(e))
+
 
 if __name__ == "__main__":
     main()

src/sumbuddy/exceptions.py

@@ -0,0 +1,14 @@
+class EmptyInputDirectoryError(Exception):
+    def __init__(self, input_directory):
+        message = f"The directory {input_directory} and subdirectories (if any) contain no files. \nPlease provide a directory with files."
+        super().__init__(message)
+
+class NoFilesAfterFilteringError(Exception):
+    def __init__(self, input_directory, ignore_file):
+        message = f"The directory {input_directory} contains files, but all are filtered out. \nCheck patterns in your {ignore_file} file and/or hidden files settings."
+        super().__init__(message)
+
+class LengthUsedForFixedLengthHashError(Exception):
+    def __init__(self, algorithm):
+        message = f"Length paremeter is not applicable for fixed-length algorithm '{algorithm}'."
+        super().__init__(message)

src/sumbuddy/hasher.py

@@ -1,4 +1,5 @@
 import hashlib
+from sumbuddy.exceptions import LengthUsedForFixedLengthHashError
 
 class Hasher:
     def __init__(self, algorithm='md5'):
@@ -51,7 +52,7 @@ def checksum_file(self, file_path, algorithm=None, length=None):
         # Other algorithms
         else:
             if length is not None:
-                raise ValueError(f"Length parameter is not applicable for fixed-length algorithm '{algorithm}'")
+                raise LengthUsedForFixedLengthHashError(algorithm)
             hash_func = hashlib.new(algorithm)
 
         # Read the file and update the hash function

src/sumbuddy/mapper.py

@@ -1,5 +1,6 @@
 import os
 from sumbuddy.filter import Filter
+from sumbuddy.exceptions import EmptyInputDirectoryError, NoFilesAfterFilteringError
 
 class Mapper:
     def __init__(self):
@@ -42,11 +43,19 @@ def gather_file_paths(self, input_directory, ignore_file=None, include_hidden=Fa
 
         file_paths = []
         root_directory = os.path.abspath(input_directory)
+        has_files = False
 
         for root, dirs, files in os.walk(input_directory):
+            if files:
+                has_files = True
             for name in files:
                 file_path = os.path.join(root, name)
                 if self.filter_manager.should_include(file_path, root_directory):
                     file_paths.append(file_path)
 
+        if not has_files:
+            raise EmptyInputDirectoryError(input_directory)
+        if not file_paths:
+            raise NoFilesAfterFilteringError(input_directory, ignore_file)
+
         return file_paths

tests/test_getChecksums.py

@@ -21,7 +21,7 @@ def setUp(self):
     @patch('os.path.exists', return_value=True)
     @patch('builtins.open', new_callable=mock_open)
     @patch('sumbuddy.Mapper.gather_file_paths', return_value=['file1.txt', 'file2.txt'])
-    @patch('sumbuddy.Hasher.checksum_file', side_effect=lambda x: 'dummychecksum')
+    @patch('sumbuddy.Hasher.checksum_file', side_effect=lambda x, **kwargs: 'dummychecksum')
     def test_get_checksums_to_file(self, mock_checksum, mock_gather, mock_open, mock_exists, mock_abspath):
         get_checksums(self.input_directory, self.output_filepath, ignore_file=None, include_hidden=False, algorithm=self.algorithm)
 
@@ -35,7 +35,7 @@ def test_get_checksums_to_file(self, mock_checksum, mock_gather, mock_open, mock
     @patch('os.path.exists', return_value=True)
     @patch('builtins.open', new_callable=mock_open)
     @patch('sumbuddy.Mapper.gather_file_paths', return_value=['file1.txt', 'file2.txt'])
-    @patch('sumbuddy.Hasher.checksum_file', side_effect=lambda x: 'dummychecksum')
+    @patch('sumbuddy.Hasher.checksum_file', side_effect=lambda x, **kwargs: 'dummychecksum')
     def test_get_checksums_to_stdout(self, mock_checksum, mock_gather, mock_open, mock_exists, mock_abspath):
         output_stream = StringIO()
         with patch('sys.stdout', new=output_stream):
@@ -50,7 +50,7 @@ def test_get_checksums_to_stdout(self, mock_checksum, mock_gather, mock_open, mo
     @patch('os.path.exists', return_value=True)
     @patch('builtins.open', new_callable=mock_open)
     @patch('sumbuddy.Mapper.gather_file_paths', return_value=['file1.txt', 'file2.txt'])
-    @patch('sumbuddy.Hasher.checksum_file', side_effect=lambda x: 'dummychecksum')
+    @patch('sumbuddy.Hasher.checksum_file', side_effect=lambda x, **kwargs: 'dummychecksum')
     def test_get_checksums_with_ignore_file(self, mock_checksum, mock_gather, mock_open, mock_exists, mock_abspath):
         get_checksums(self.input_directory, output_filepath=None, ignore_file=self.ignore_file, include_hidden=False, algorithm=self.algorithm)
         mock_gather.assert_called_with(self.input_directory, ignore_file=self.ignore_file, include_hidden=False)
@@ -59,7 +59,7 @@ def test_get_checksums_with_ignore_file(self, mock_checksum, mock_gather, mock_o
     @patch('os.path.exists', return_value=True)
     @patch('builtins.open', new_callable=mock_open)
     @patch('sumbuddy.Mapper.gather_file_paths', return_value=['file1.txt', 'file2.txt', '.hidden_file'])
-    @patch('sumbuddy.Hasher.checksum_file', side_effect=lambda x: 'dummychecksum')
+    @patch('sumbuddy.Hasher.checksum_file', side_effect=lambda x, **kwargs: 'dummychecksum')
     def test_get_checksums_include_hidden(self, mock_checksum, mock_gather, mock_open, mock_exists, mock_abspath):
         get_checksums(self.input_directory, output_filepath=None, ignore_file=None, include_hidden=True, algorithm=self.algorithm)
         mock_gather.assert_called_with(self.input_directory, ignore_file=None, include_hidden=True)
@@ -68,7 +68,7 @@ def test_get_checksums_include_hidden(self, mock_checksum, mock_gather, mock_ope
     @patch('os.path.exists', return_value=True)
     @patch('builtins.open', new_callable=mock_open)
     @patch('sumbuddy.Mapper.gather_file_paths', return_value=['file1.txt', 'file2.txt'])
-    @patch('sumbuddy.Hasher.checksum_file', side_effect=lambda x: 'dummychecksum')
+    @patch('sumbuddy.Hasher.checksum_file', side_effect=lambda x, **kwargs: 'dummychecksum')
     def test_get_checksums_different_algorithm(self, mock_checksum, mock_gather, mock_open, mock_exists, mock_abspath):
         algorithm = 'sha256'
         get_checksums(self.input_directory, output_filepath=None, ignore_file=None, include_hidden=False, algorithm=algorithm)