flow.record 3.15.dev7__tar.gz → 3.15.dev8__tar.gz

{flow.record-3.15.dev7/flow.record.egg-info → flow.record-3.15.dev8}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: flow.record
-Version: 3.15.dev7
+Version: 3.15.dev8
 Summary: A library for defining and creating structured data (called records) that can be streamed to disk or piped to other tools that use flow.record
 Author-email: Dissect Team <dissect@fox-it.com>
 License: Affero General Public License v3

{flow.record-3.15.dev7 → flow.record-3.15.dev8}/flow/record/__init__.py

@@ -2,6 +2,7 @@ import gzip
 import os
 
 from flow.record.base import (
+    IGNORE_FIELDS_FOR_COMPARISON,
     RECORD_VERSION,
     RECORDSTREAM_MAGIC,
     DynamicDescriptor,
@@ -20,6 +21,7 @@ from flow.record.base import (
     open_path,
     open_path_or_stream,
     open_stream,
+    set_ignored_fields_for_comparison,
     stream,
 )
 from flow.record.jsonpacker import JsonRecordPacker
@@ -35,6 +37,7 @@ from flow.record.stream import (
 )
 
 __all__ = [
+    "IGNORE_FIELDS_FOR_COMPARISON",
     "RECORD_VERSION",
     "RECORDSTREAM_MAGIC",
     "FieldType",
@@ -54,6 +57,7 @@ __all__ = [
     "open_path_or_stream",
     "open_path",
     "open_stream",
+    "set_ignored_fields_for_comparison",
     "stream",
     "dynamic_fieldtype",
     "DynamicDescriptor",

{flow.record-3.15.dev7 → flow.record-3.15.dev8}/flow/record/base.py

@@ -15,7 +15,17 @@ import warnings
 from datetime import datetime, timezone
 from itertools import zip_longest
 from pathlib import Path
-from typing import IO, Any, BinaryIO, Iterator, Mapping, Optional, Sequence, Union
+from typing import (
+    IO,
+    Any,
+    BinaryIO,
+    Iterable,
+    Iterator,
+    Mapping,
+    Optional,
+    Sequence,
+    Union,
+)
 from urllib.parse import parse_qsl, urlparse
 
 from flow.record.adapter import AbstractReader, AbstractWriter
@@ -96,6 +106,18 @@ class {name}(Record):
 """
 
 
+if env_excluded_fields := os.environ.get("FLOW_RECORD_IGNORE"):
+    IGNORE_FIELDS_FOR_COMPARISON = set(env_excluded_fields.split(","))
+else:
+    IGNORE_FIELDS_FOR_COMPARISON = set()
+
+
+def set_ignored_fields_for_comparison(ignored_fields: Iterable[str]) -> None:
+    """Can be used to update the IGNORE_FIELDS_FOR_COMPARISON from outside the flow.record package scope"""
+    global IGNORE_FIELDS_FOR_COMPARISON
+    IGNORE_FIELDS_FOR_COMPARISON = set(ignored_fields)
+
+
 class FieldType:
     def _typename(self):
         t = type(self)
@@ -117,14 +139,20 @@ class Record:
     def __eq__(self, other):
         if not isinstance(other, Record):
             return False
-        return self._pack() == other._pack()
 
-    def _pack(self, unversioned=False):
+        return self._pack(excluded_fields=IGNORE_FIELDS_FOR_COMPARISON) == other._pack(
+            excluded_fields=IGNORE_FIELDS_FOR_COMPARISON
+        )
+
+    def _pack(self, unversioned=False, excluded_fields: list = None):
         values = []
         for k in self.__slots__:
             v = getattr(self, k)
             v = v._pack() if isinstance(v, FieldType) else v
 
+            if excluded_fields and k in excluded_fields:
+                continue
+
             # Skip version field if requested (only for compatibility reasons)
             if unversioned and k == "_version" and v == 1:
                 continue
@@ -160,6 +188,9 @@ class Record:
             raise ValueError("Got unexpected field names: {kwds!r}".format(kwds=list(kwds)))
         return result
 
+    def __hash__(self) -> int:
+        return hash(self._pack(excluded_fields=IGNORE_FIELDS_FOR_COMPARISON))
+
     def __repr__(self):
         return "<{} {}>".format(
             self._desc.name, " ".join("{}={!r}".format(k, getattr(self, k)) for k in self._desc.fields)

{flow.record-3.15.dev7 → flow.record-3.15.dev8}/flow/record/version.py

@@ -12,5 +12,5 @@ __version__: str
 __version_tuple__: VERSION_TUPLE
 version_tuple: VERSION_TUPLE
 
-__version__ = version = '3.15.dev7'
-__version_tuple__ = version_tuple = (3, 15, 'dev7')
+__version__ = version = '3.15.dev8'
+__version_tuple__ = version_tuple = (3, 15, 'dev8')

{flow.record-3.15.dev7 → flow.record-3.15.dev8/flow.record.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: flow.record
-Version: 3.15.dev7
+Version: 3.15.dev8
 Summary: A library for defining and creating structured data (called records) that can be streamed to disk or piped to other tools that use flow.record
 Author-email: Dissect Team <dissect@fox-it.com>
 License: Affero General Public License v3

{flow.record-3.15.dev7 → flow.record-3.15.dev8}/tests/test_record.py

@@ -1,4 +1,7 @@
+import importlib
+import os
 import sys
+from unittest.mock import patch
 
 import pytest
 
@@ -15,7 +18,11 @@ from flow.record import (
     fieldtypes,
     record_stream,
 )
-from flow.record.base import merge_record_descriptors, normalize_fieldname
+from flow.record.base import (
+    merge_record_descriptors,
+    normalize_fieldname,
+    set_ignored_fields_for_comparison,
+)
 from flow.record.exceptions import RecordDescriptorError
 from flow.record.stream import RecordFieldRewriter
 
@@ -792,3 +799,60 @@ def test_normalize_fieldname():
     assert normalize_fieldname("my name (with) parentheses") == "my_name__with__parentheses"
     assert normalize_fieldname("_generated") == "_generated"
     assert normalize_fieldname("_source") == "_source"
+
+
+def test_compare_global_variable():
+    TestRecord = RecordDescriptor(
+        "test/record",
+        [
+            ("string", "firstname"),
+            ("string", "lastname"),
+        ],
+    )
+
+    same_same = TestRecord(firstname="James", lastname="Bond")
+    but_different = TestRecord(firstname="Ethan", lastname="Hunt")
+    but_still_same = TestRecord(firstname="Andrew", lastname="Bond")
+
+    records = [same_same, but_different, but_still_same]
+
+    assert same_same != but_still_same
+
+    set_ignored_fields_for_comparison({"_generated", "firstname"})
+    assert same_same == but_still_same
+    assert same_same != but_different
+    assert len(set(records)) == 2
+
+
+def test_compare_environment_variable():
+    with patch.dict(os.environ), patch.dict(sys.modules):
+        os.environ["FLOW_RECORD_IGNORE"] = "_generated,lastname"
+
+        # Force a re-import of flow.record so the global variable gets re-initialized based on the environment variable
+        keys = [key for key in sys.modules if key == "flow" or "flow." in key]
+        for key in keys:
+            del sys.modules[key]
+
+        importlib.import_module("flow.record")
+
+        from flow.record import IGNORE_FIELDS_FOR_COMPARISON, RecordDescriptor
+
+        assert IGNORE_FIELDS_FOR_COMPARISON == {"_generated", "lastname"}
+
+        TestRecord = RecordDescriptor(
+            "test/record",
+            [
+                ("string", "firstname"),
+                ("string", "lastname"),
+            ],
+        )
+
+        same_same = TestRecord(firstname="John", lastname="Rambo")
+        but_different = TestRecord(firstname="Johnny", lastname="English")
+        but_still_same = TestRecord(firstname="John", lastname="McClane")
+
+        records = [same_same, but_different, but_still_same]
+
+        assert same_same == but_still_same
+        assert same_same != but_different
+        assert len(set(records)) == 2