CHORES: pre-commit

.readthedocs.yaml

@@ -31,4 +31,3 @@ python:
     - requirements: docs/requirements.txt
     - method: pip
       path: .
-

src/algbench/benchmark.py

@@ -272,7 +272,7 @@ def clear(self):
     def delete_if(self, condition: typing.Callable[[typing.Dict], bool]):
         """
         Delete entries if a specific condition is met (return True).
-        Recreates the internal 'results' folder for this porpose. 
+        Recreates the internal 'results' folder for this porpose.
         Use `front` to get a preview on how an entry that is
         passed to the condition looks like.
 
@@ -289,14 +289,14 @@ def func(entry) -> typing.Optional[typing.Dict]:
 
     def apply(self, func: typing.Callable[[typing.Dict], typing.Optional[typing.Dict]]):
         """
-        Allows to modify all entries (in place !) inside this benchmark, 
+        Allows to modify all entries (in place !) inside this benchmark,
         based on the provided callable. It is being called for every
-        entry inside the database, and the returned entry will be stored 
+        entry inside the database, and the returned entry will be stored
         instead. If None is returned, the provided entry will be deleted
-        from the database. 
+        from the database.
 
         NOT THREAD-SAFE, execute this while no other instance is accessing
-        the file system. 
+        the file system.
         """
         self._db.apply(func)
         self.compress()
@@ -310,7 +310,7 @@ def __len__(self):
         However, this is not recommended, as it is slow.
         """
         return self._db.__len__()
-    
+
     def empty(self):
         """
         Return True if the database is empty, False otherwise.

src/algbench/benchmark_db.py

@@ -74,12 +74,16 @@ def clear(self):
         self._data.clear()
         self._env_data.clear()
 
-    def get_env_info(self, env_fingerprint, env_data: typing.Optional[NfsJsonDict]=None):
+    def get_env_info(
+        self, env_fingerprint, env_data: typing.Optional[NfsJsonDict] = None
+    ):
         if not env_data:
             env_data = self._env_data
         return env_data[env_fingerprint]
 
-    def _create_entry_with_env(self, entry: typing.Dict, env_data: typing.Optional[NfsJsonDict]=None):
+    def _create_entry_with_env(
+        self, entry: typing.Dict, env_data: typing.Optional[NfsJsonDict] = None
+    ):
         entry = entry.copy()
         try:
             if not env_data:
@@ -88,7 +92,7 @@ def _create_entry_with_env(self, entry: typing.Dict, env_data: typing.Optional[N
             return entry
         except KeyError:
             return None
-        
+
     def __iter__(self):
         for entry in self._data:
             entry_with_env = self._create_entry_with_env(entry)
@@ -125,4 +129,3 @@ def apply(self, func: typing.Callable[[typing.Dict], typing.Optional[typing.Dict
 
     def __len__(self):
         return len(self._arg_fingerprints)
-

src/algbench/db/nfs_json_dict.py

@@ -62,6 +62,6 @@ def delete(self):
     def move_directory(self, new_path: str):
         """
         Changes the internal directory of this NFSJsonDict.
-        Not thread safe! 
+        Not thread safe!
         """
-        self._db.move_directory(new_path)
+        self._db.move_directory(new_path)

src/algbench/db/nfs_json_list.py

@@ -9,7 +9,7 @@
 import socket
 import typing
 import zipfile
-from zipfile import ZipFile, BadZipFile
+from zipfile import BadZipFile, ZipFile
 
 from .json_serializer import to_json
 
@@ -22,7 +22,9 @@ class NfsJsonList:
     even for slurm pools with NFS.
     """
 
-    def __init__(self, path: typing.Union[str, pathlib.Path], file_split_mb: float=30):
+    def __init__(
+        self, path: typing.Union[str, pathlib.Path], file_split_mb: float = 30
+    ):
         self.path: typing.Union[str, pathlib.Path] = path
         if not os.path.exists(path):
             # Could fail in very few unlucky cases on an NFS (parallel creations)
@@ -100,13 +102,15 @@ def flush(self):
             if not os.path.isfile(path):
                 msg = "Could not write to disk for unknown reasons."
                 raise RuntimeError(msg)
-            
+
             if self._filesize > self._file_split_size:
                 new_unique_name = self._get_unique_name()
-                _log.info(f"File {self._subfile_path} exceeds {self._file_split_size} MB, starting new file {new_unique_name}.")
+                _log.info(
+                    f"File {self._subfile_path} exceeds {self._file_split_size} MB, starting new file {new_unique_name}."
+                )
                 self._subfile_path = new_unique_name
                 self._filesize = 0
-            
+
         _log.info(f"Wrote {len(self._cache)} entries to disk.")
         self._cache.clear()
 
@@ -192,10 +196,10 @@ def delete(self):
 
     def move_directory(self, new_path: str):
         """
-        Changes the internal directory where the db files are 
+        Changes the internal directory where the db files are
         stored while keeping all files (renames the directory).
         This operation is not thread safe, especially not
-        over other nodes or instances of this script. 
+        over other nodes or instances of this script.
         """
 
         if os.path.exists(new_path) or os.path.isfile(new_path):
@@ -206,4 +210,4 @@ def move_directory(self, new_path: str):
         _log.info(f"Moved NFSJsonList from {self.path} to {new_path}.")
         self._subfile_path = self._get_unique_name()
         self._filesize = 0
-        self.path = new_path
+        self.path = new_path

src/algbench/db/nfs_json_set.py

@@ -51,10 +51,10 @@ def delete(self):
 
     def __len__(self):
         return len(self._values)
-    
+
     def move_directory(self, new_path: str):
         """
-        Changes the internal directory of this NFSJsonSet (essentially 
-        moves the directory). Not thread safe! 
+        Changes the internal directory of this NFSJsonSet (essentially
+        moves the directory). Not thread safe!
         """
         self._db.move_directory(new_path)

src/algbench/environment.py

@@ -27,6 +27,7 @@ def get_git_revision() -> typing.Optional[str]:
         label = None
     return label
 
+
 def get_python_file() -> typing.Optional[str]:
     """
     Return the path of the calling python file.

tests/test_apply.py

@@ -1,31 +1,32 @@
-import os
 import typing
+
 from algbench import Benchmark
 
-def test_apply():
 
+def test_apply():
     benchmark = Benchmark("./test_apply")
 
-    def generate_entry(args): 
-        return {"entry_index" : args["index"], "data" : f"Data for entry {args['index']}"}
+    def generate_entry(args):
+        return {"entry_index": args["index"], "data": f"Data for entry {args['index']}"}
+
     for i in range(20):
-        benchmark.add(generate_entry, {"index" : i})
-    
+        benchmark.add(generate_entry, {"index": i})
+
     benchmark.compress()
 
     def db_count_entries(b: Benchmark) -> int:
         return len([0 for _ in b])
 
     assert db_count_entries(benchmark) == 20
 
-
     def func(entry: typing.Dict):
         if entry["result"]["entry_index"] % 2 == 1:
             return None
         else:
             del entry["timestamp"]
             del entry["runtime"]
             return entry
+
     benchmark.apply(func)
 
     assert db_count_entries(benchmark) == 10
@@ -34,7 +35,7 @@ def func(entry: typing.Dict):
         assert "timestamp" not in entry
         assert "result" in entry
 
-    benchmark.repair() # implies both a delete_if() and apply() call
+    benchmark.repair()  # implies both a delete_if() and apply() call
     assert db_count_entries(benchmark) == 10
 
-    benchmark.delete()
+    benchmark.delete()

tests/test_basics.py

@@ -5,6 +5,9 @@
 
 def test_simple():
     benchmark = Benchmark("./test_benchmark")
+    if not benchmark.empty():
+        benchmark.delete()
+        benchmark = Benchmark("./test_benchmark")
 
     def f(x, _test=2, default="default"):
         print(x)

tests/test_extensive.py

@@ -9,7 +9,7 @@ def test_extensive_use_case():
     def f(x, _test=2, default="default"):
         print(x)
         return {"r1": x, "r2": "test"}
-    
+
     for x in range(500):
         benchmark.add(f, x, _test=None)
 
@@ -32,5 +32,6 @@ def f(x, _test=2, default="default"):
     assert len(benchmark_) == 250
     benchmark.delete()
 
-if __name__=="__main__":
-    test_extensive_use_case()
+
+if __name__ == "__main__":
+    test_extensive_use_case()

tests/test_file_splitting.py

@@ -1,24 +1,25 @@
 import os
+
 from algbench import Benchmark
 
-def test_file_splitting():
 
+def test_file_splitting():
     benchmark = Benchmark("./test_file_split_benchmark")
 
     # random data generation taken from https://stackoverflow.com/questions/16308989/
-    def generate_one_mb_results(args): 
-        offset = ord(b'A')
+    def generate_one_mb_results(args):
+        offset = ord(b"A")
         rand_bytes = os.urandom(1024 * 1024)
         array = bytearray(rand_bytes)
         for i, bt in enumerate(array):
             array[i] = offset + bt % 26
-        return {str(args["index"]) : bytes(array).decode()}
-        
+        return {str(args["index"]): bytes(array).decode()}
+
     for i in range(100):
-        benchmark.add(generate_one_mb_results, {"index" : i})
+        benchmark.add(generate_one_mb_results, {"index": i})
 
-    # Hardcoded for now. The file splits are hardcoded in nfs_json_list to be made at 30 MB. 
-    # For the 100MB + some curly brackets and environments, this should produce 4 files. 
-    assert len(os.listdir("./test_file_split_benchmark/results")) == 4 
+    # Hardcoded for now. The file splits are hardcoded in nfs_json_list to be made at 30 MB.
+    # For the 100MB + some curly brackets and environments, this should produce 4 files.
+    assert len(os.listdir("./test_file_split_benchmark/results")) == 4
 
-    benchmark.delete()
+    benchmark.delete()