fix testcase

python-threatexchange/Dockerfile

@@ -8,5 +8,5 @@ RUN rm -rf /usr/src/threatexchange
 ENV TX_STATEDIR=/var/lib/threatexchange
 
 VOLUME ["/var/lib/threatexchange"]
-CMD ["threatexchange"]
-ENTRYPOINT ["/usr/local/bin/threatexchange"]
+# CMD ["threatexchange"]
+# ENTRYPOINT ["/usr/local/bin/threatexchange"]

python-threatexchange/threatexchange/content_type/preprocess/__init__.py

@@ -0,0 +1 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.

python-threatexchange/threatexchange/exchanges/impl/ncmec_api.py

@@ -41,13 +41,13 @@ class NCMECCheckpoint(
 
     # The biggest value of "to", and the next "from"
     get_entries_max_ts: int
-    # A url to fetch the next page of results.
+    # A url to fetch the next page of results
     # Only reference this value through get_paging_url_if_recent
-    paging_url: str
-    # a timestamp for the last fetch time, specifically used with a paging_url.
+    paging_url: str = ""
+    # a timestamp for the last fetch time, specifically used with a pagingpyth_url
     # NCMEC suggests not storing paging_urls long term so we consider them invalid
     # 12hr after the last_fetch_time
-    last_fetch_time: int
+    last_fetch_time: int = field(hash=True, default_factory=lambda: int(time.time()))
 
     def get_progress_timestamp(self) -> t.Optional[int]:
         return self.get_entries_max_ts
@@ -68,6 +68,14 @@ def __setstate__(self, d: t.Dict[str, t.Any]) -> None:
         ### field 'max_timestamp' renamed to 'get_entries_max_ts'
         if "max_timestamp" in d:
             d["get_entries_max_ts"] = d.pop("max_timestamp")
+
+        # 1.0.0 => 1.2.3:
+        # Add last_fetch_time
+        # the default_factory value was not being set correctly when
+        # reading from pickle
+        if not "last_fetch_time" in d:
+            d["last_fetch_time"] = int(time.time())
+
         self.__dict__ = d
 
 

python-threatexchange/threatexchange/exchanges/tests/test_state_compatibility.py

@@ -30,6 +30,7 @@ class SignalOpinionOwnerRemoved:
 import copy
 from dataclasses import dataclass, field
 import pickle
+import time
 import typing as t
 
 import pytest
@@ -148,7 +149,9 @@ def get_NCMECCheckpoint() -> t.Tuple[NCMECCheckpoint, t.Sequence[object]]:
     max_ts = 1197433091
 
     current = NCMECCheckpoint(
-        get_entries_max_ts=max_ts, paging_url="", last_fetch_time=0
+        get_entries_max_ts=max_ts,
+        paging_url="",
+        last_fetch_time=int(time.time()),
     )
 
     # 1.0.x
@@ -183,17 +186,21 @@ class NCMECCheckpointTsMoved(FetchCheckpointBase):
 
 
 @pytest.mark.parametrize(
-    ("current_version", "historical_versions"),
+    "get_checkpoint_func",
     [
-        get_SignalOpinion(),
-        get_FBThreatExchangeOpinion(),
-        get_NCMECOpinion(),
-        get_NCMECCheckpoint(),
+        get_SignalOpinion,
+        get_FBThreatExchangeOpinion,
+        get_NCMECOpinion,
+        get_NCMECCheckpoint,
     ],
 )
 def test_previous_pickle_state(
-    current_version: object, historical_versions: t.Sequence[object]
+    get_checkpoint_func: t.Callable[[], t.Tuple[object, t.Sequence[object]]],
+    monkeypatch: pytest.MonkeyPatch,
 ):
+    monkeypatch.setattr("time.time", lambda: 10**8)
+
+    current_version, historical_versions = get_checkpoint_func()
     # Sanity
     serialized = pickle.dumps(current_version)
     assert (