From b355632f37e656ccce544bab7f5c07de537759db Mon Sep 17 00:00:00 2001
From: Chris Mutel <cmutel@gmail.com>
Date: Thu, 31 Oct 2024 11:33:48 +0100
Subject: [PATCH 1/6] Use Snowflake IDs for nodes and edges

---
 bw2data/backends/base.py    |  6 +++---
 bw2data/backends/proxies.py | 17 +++++++++++++++--
 bw2data/backends/schema.py  | 20 +++++++++++++++++---
 bw2data/backends/utils.py   | 11 ++++++++---
 bw2data/snowflake_ids.py    | 20 ++++++++++++++++++++
 tests/activity_proxy.py     | 20 ++++++++++++++++++++
 tests/compatibility.py      | 37 ++++++++++++++++++++++---------------
 tests/database.py           | 11 +++++------
 tests/ia.py                 | 16 ++++++++++++----
 tests/utils.py              | 33 +++++++++------------------------
 10 files changed, 131 insertions(+), 60 deletions(-)
 create mode 100644 bw2data/snowflake_ids.py

diff --git a/bw2data/backends/base.py b/bw2data/backends/base.py
index 296076ee..761b809e 100644
--- a/bw2data/backends/base.py
+++ b/bw2data/backends/base.py
@@ -28,7 +28,7 @@
 from bw2data.backends.utils import (
     check_exchange,
     dict_as_activitydataset,
-    dict_as_exchangedataset,
+    _dict_as_exchangedataset,
     get_csv_data_dict,
     retupleize_geo_strings,
 )
@@ -525,7 +525,7 @@ def _efficient_write_dataset(
 
             if "output" not in exchange:
                 exchange["output"] = (ds["database"], ds["code"])
-            exchanges.append(dict_as_exchangedataset(exchange))
+            exchanges.append(_dict_as_exchangedataset(exchange))
 
             # Query gets passed as INSERT INTO x VALUES ('?', '?'...)
             # SQLite3 has a limit of 999 variables,
@@ -542,7 +542,7 @@ def _efficient_write_dataset(
             check_activity_type(ds.get("type"))
             check_activity_keys(ds)
 
-        activities.append(dict_as_activitydataset(ds))
+        activities.append(dict_as_activitydataset(ds, add_snowflake_id=True))
 
         if len(activities) > 125:
             ActivityDataset.insert_many(activities).execute()
diff --git a/bw2data/backends/proxies.py b/bw2data/backends/proxies.py
index c947129e..be316292 100644
--- a/bw2data/backends/proxies.py
+++ b/bw2data/backends/proxies.py
@@ -14,7 +14,7 @@
     check_exchange_keys,
     check_exchange_type,
 )
-from bw2data.backends.utils import dict_as_activitydataset, dict_as_exchangedataset
+from bw2data.backends.utils import dict_as_activitydataset, _dict_as_exchangedataset
 from bw2data.configuration import labels
 from bw2data.errors import ValidityError
 from bw2data.logs import stdout_feedback_logger
@@ -338,6 +338,8 @@ def save(self, signal: bool = True, data_already_set: bool = False, force_insert
             check_activity_keys(self)
 
             for key, value in dict_as_activitydataset(self._data).items():
+                # ID value is either already in `._document` (update) or will be created by
+                # `SnowflakeIDBaseClass.save()`.
                 if key != "id":
                     setattr(self._document, key, value)
 
@@ -496,6 +498,8 @@ def new_edge(self, **kwargs):
         exc = Exchange()
         exc.output = self.key
         for key in kwargs:
+            if key == "id":
+                raise ValueError("`id` must be created automatically")
             exc[key] = kwargs[key]
         return exc
 
@@ -510,14 +514,23 @@ def copy(self, code: Optional[str] = None, signal: bool = True, **kwargs):
         activity = Activity()
         for key, value in self.items():
             if key != "id":
+                print(key, value)
                 activity[key] = value
         for k, v in kwargs.items():
+            if k == "id":
+                raise ValueError("`id` must be generated automatically")
             activity._data[k] = v
         activity._data["code"] = str(code or uuid.uuid4().hex)
         activity.save(signal=signal)
 
+        print("Copy Document ID", activity._document.id)
+        print("Original Document ID", self._document.id)
+
         for exc in self.exchanges():
             data = copy.deepcopy(exc._data)
+            if 'id' in data:
+                # New snowflake ID will be inserted by `_dict_as_exchangedataset`
+                del data['id']
             data["output"] = activity.key
             # Change `input` for production exchanges
             if exc["input"] == exc["output"]:
@@ -564,7 +577,7 @@ def save(self, signal: bool = True, data_already_set: bool = False, force_insert
             check_exchange_type(self._data.get("type"))
             check_exchange_keys(self)
 
-            for key, value in dict_as_exchangedataset(self._data).items():
+            for key, value in _dict_as_exchangedataset(self._data).items():
                 setattr(self._document, key, value)
 
         self._document.save(signal=signal, force_insert=force_insert)
diff --git a/bw2data/backends/schema.py b/bw2data/backends/schema.py
index 12ed3e7a..e32a1770 100644
--- a/bw2data/backends/schema.py
+++ b/bw2data/backends/schema.py
@@ -1,11 +1,25 @@
-from peewee import DoesNotExist, TextField
+from peewee import DoesNotExist, Model, TextField, IntegerField
 
 from bw2data.errors import UnknownObject
 from bw2data.signals import SignaledDataset
 from bw2data.sqlite import PickleField
+from bw2data.snowflake_ids import snowflake_id_generator
 
 
-class ActivityDataset(SignaledDataset):
+class SnowflakeIDBaseClass(SignaledDataset):
+    id = IntegerField(primary_key=True)
+
+    def save(self, **kwargs):
+        if self.id is None:
+            # If the primary key is already present, peewee will make an `UPDATE` query.
+            # This will have no effect if there isn't a matching row
+            # https://docs.peewee-orm.com/en/latest/peewee/models.html#id4
+            self.id = next(snowflake_id_generator)
+            kwargs['force_insert'] = True
+        super().save(**kwargs)
+
+
+class ActivityDataset(SnowflakeIDBaseClass):
     data = PickleField()  # Canonical, except for other C fields
     code = TextField()  # Canonical
     database = TextField()  # Canonical
@@ -19,7 +33,7 @@ def key(self):
         return (self.database, self.code)
 
 
-class ExchangeDataset(SignaledDataset):
+class ExchangeDataset(SnowflakeIDBaseClass):
     data = PickleField()  # Canonical, except for other C fields
     input_code = TextField()  # Canonical
     input_database = TextField()  # Canonical
diff --git a/bw2data/backends/utils.py b/bw2data/backends/utils.py
index afd14859..13797d9c 100644
--- a/bw2data/backends/utils.py
+++ b/bw2data/backends/utils.py
@@ -1,3 +1,4 @@
+from typing import Any
 import copy
 import warnings
 from typing import Optional
@@ -9,6 +10,7 @@
 from bw2data.configuration import labels
 from bw2data.errors import InvalidExchange, UntypedExchange
 from bw2data.meta import databases, methods
+from bw2data.snowflake_ids import snowflake_id_generator
 
 
 def get_csv_data_dict(ds):
@@ -66,8 +68,8 @@ def check_exchange(exc):
         raise ValueError("Invalid amount in exchange {}".format(exc))
 
 
-def dict_as_activitydataset(ds):
-    return {
+def dict_as_activitydataset(ds: Any, add_snowflake_id: bool = False) -> dict:
+    val = {
         "data": ds,
         "database": ds["database"],
         "code": ds["code"],
@@ -76,9 +78,12 @@ def dict_as_activitydataset(ds):
         "product": ds.get("reference product"),
         "type": ds.get("type", labels.process_node_default),
     }
+    if add_snowflake_id:
+        val['id'] = next(snowflake_id_generator)
+    return val
 
 
-def dict_as_exchangedataset(ds):
+def _dict_as_exchangedataset(ds: Any) -> dict:
     return {
         "data": ds,
         "input_database": ds["input"][0],
diff --git a/bw2data/snowflake_ids.py b/bw2data/snowflake_ids.py
new file mode 100644
index 00000000..9a9bf082
--- /dev/null
+++ b/bw2data/snowflake_ids.py
@@ -0,0 +1,20 @@
+from snowflake import SnowflakeGenerator
+import uuid
+
+# Jan 1, 2024
+# from datetime import datetime
+# (datetime(2024, 1, 1) - datetime.utcfromtimestamp(0)).total_seconds() * 1000.0
+EPOCH_START_MS = 1704067200000
+
+# From https://softwaremind.com/blog/the-unique-features-of-snowflake-id-and-its-comparison-to-uuid/
+# Snowflake bits:
+# Sign bit: 1 bit. It will always be 0. This is reserved for future uses. It can potentially be used
+# to distinguish between signed and unsigned numbers.
+# Timestamp: 41 bits. Milliseconds since the epoch or custom epoch.
+# Datacenter ID: 5 bits, which gives us 2 ^ 5 = 32 datacenters.
+# Machine ID: 5 bits, which gives us 2 ^ 5 = 32 machines per datacenter.
+# However, `snowflake-id` lumps these two values datacenter and machine id values together into an
+# `instance` value with 2 ^ 10 = 1024 possible values.
+# Sequence number: 12 bits. For every ID generated on that machine/process, the sequence number is
+# incremented by 1. The number is reset to 0 every millisecond.
+snowflake_id_generator = SnowflakeGenerator(instance=uuid.getnode() % 1024, epoch=EPOCH_START_MS)
diff --git a/tests/activity_proxy.py b/tests/activity_proxy.py
index 2484bda8..55a897c0 100644
--- a/tests/activity_proxy.py
+++ b/tests/activity_proxy.py
@@ -179,6 +179,15 @@ def test_copy(activity):
     assert cp["name"] == "baz"
     assert cp["location"] == "bar"
     assert ExchangeDataset.select().count() == 2
+
+    for ds in ActivityDataset.select():
+        print(ds, ds.name, ds.code)
+
+    cp.save()
+
+    for ds in ActivityDataset.select():
+        print(ds, ds.name, ds.code)
+
     assert ActivityDataset.select().count() == 2
     assert (
         ActivityDataset.select()
@@ -232,6 +241,9 @@ def test_copy_with_kwargs(activity):
 
 @bw2test
 def test_delete_activity_parameters():
+    from bw2data import projects
+    print(projects.dir)
+
     db = DatabaseChooser("example")
     db.register()
 
@@ -241,6 +253,12 @@ def test_delete_activity_parameters():
     b.save()
     a.new_exchange(amount=0, input=b, type="technosphere", formula="foo * bar + 4").save()
 
+    print("Iterating over exchanges:")
+    for exc in ExchangeDataset.select():
+        print("Next:", exc, exc.id)
+
+    assert ExchangeDataset.select().count() == 1
+
     activity_data = [
         {
             "name": "reference_me",
@@ -258,6 +276,8 @@ def test_delete_activity_parameters():
     parameters.new_activity_parameters(activity_data, "my group")
     parameters.add_exchanges_to_group("my group", a)
 
+    assert ExchangeDataset.select().count() == 1
+
     assert ActivityParameter.select().count() == 2
     assert ParameterizedExchange.select().count() == 1
 
diff --git a/tests/compatibility.py b/tests/compatibility.py
index 25a1432a..54c20b45 100644
--- a/tests/compatibility.py
+++ b/tests/compatibility.py
@@ -9,6 +9,7 @@
     databases,
     geomapping,
     get_activity,
+    get_node,
     get_multilca_data_objs,
     get_node,
     methods,
@@ -65,27 +66,32 @@ def setup():
 def test_prepare_lca_inputs_basic(setup):
     d, objs, r = prepare_lca_inputs(demand={("food", "1"): 1}, method=("foo",))
     # ID is 3; two biosphere flows, then '1' is next written
-    assert d == {3: 1}
+    assert list(d.values()) == [1]
     assert {o.metadata["id"] for o in objs} == {o.datapackage().metadata["id"] for o in setup}
 
+    b1 = get_node(database="biosphere", code='1').id
+    b2 = get_node(database="biosphere", code='2').id
+    f1 = get_node(database="food", code='1').id
+    f2 = get_node(database="food", code='2').id
+
     remapping_expected = {
         "activity": {
-            1: ("biosphere", "1"),
-            2: ("biosphere", "2"),
-            3: ("food", "1"),
-            4: ("food", "2"),
+            b1: ("biosphere", "1"),
+            b2: ("biosphere", "2"),
+            f1: ("food", "1"),
+            f2: ("food", "2"),
         },
         "product": {
-            1: ("biosphere", "1"),
-            2: ("biosphere", "2"),
-            3: ("food", "1"),
-            4: ("food", "2"),
+            b1: ("biosphere", "1"),
+            b2: ("biosphere", "2"),
+            f1: ("food", "1"),
+            f2: ("food", "2"),
         },
         "biosphere": {
-            1: ("biosphere", "1"),
-            2: ("biosphere", "2"),
-            3: ("food", "1"),
-            4: ("food", "2"),
+            b1: ("biosphere", "1"),
+            b2: ("biosphere", "2"),
+            f1: ("food", "1"),
+            f2: ("food", "2"),
         },
     }
     assert r == remapping_expected
@@ -112,8 +118,9 @@ def test_prepare_lca_inputs_multiple_demands(setup):
     d, objs, r = prepare_lca_inputs(
         demands=[{("food", "1"): 1}, {("food", "2"): 10}], method=("foo",)
     )
-    # ID is 3; two biosphere flows, then '1' is next written
-    assert d == [{3: 1}, {4: 10}]
+    f1 = get_node(database="food", code='1').id
+    f2 = get_node(database="food", code='2').id
+    assert d == [{f1: 1}, {f2: 10}]
     assert {o.metadata["id"] for o in objs} == {o.datapackage().metadata["id"] for o in setup}
 
 
diff --git a/tests/database.py b/tests/database.py
index 0950c8e1..0070e222 100644
--- a/tests/database.py
+++ b/tests/database.py
@@ -19,6 +19,7 @@
 from bw2data.backends import Activity as PWActivity
 from bw2data.backends import sqlite3_lci_db
 from bw2data.database import Database
+from bw2data.snowflake_ids import EPOCH_START_MS
 from bw2data.errors import (
     DuplicateNode,
     InvalidExchange,
@@ -63,7 +64,7 @@ def test_get_code():
     activity = d.get("1")
     assert isinstance(activity, PWActivity)
     assert activity["name"] == "an emission"
-    assert activity.id == 1
+    assert activity.id > EPOCH_START_MS
 
 
 @bw2test
@@ -73,7 +74,7 @@ def test_get_kwargs():
     activity = d.get(name="an emission")
     assert isinstance(activity, PWActivity)
     assert activity["name"] == "an emission"
-    assert activity.id == 1
+    assert activity.id > EPOCH_START_MS
 
 
 @bw2test
@@ -541,7 +542,7 @@ def test_processed_array_with_metadata():
                 "reference product": np.NaN,
                 "unit": "something",
                 "location": np.NaN,
-                "id": 1,
+                "id": df.id[0],
             }
         ]
     )
@@ -716,7 +717,7 @@ def test_process_without_exchanges_still_in_processed_array():
 
     package = database.datapackage()
     array = package.get_resource("a_database_technosphere_matrix.data")[0]
-    assert array[0] == 1
+    assert array[0] == 1.
     assert array.shape == (1,)
 
 
@@ -774,8 +775,6 @@ def test_new_node_error():
 
     with pytest.raises(DuplicateNode):
         database.new_node("foo")
-    with pytest.raises(DuplicateNode):
-        database.new_node(code="bar", id=act.id)
 
 
 @bw2test
diff --git a/tests/ia.py b/tests/ia.py
index c51b9fce..279aa5ad 100644
--- a/tests/ia.py
+++ b/tests/ia.py
@@ -278,8 +278,11 @@ def test_method_geocollection():
         }
     )
 
+    f1 = get_node(code="1").id
+    f2 = get_node(code="2").id
+
     m = Method(("foo",))
-    m.write([(1, 2, "RU"), (3, 4, ("foo", "bar"))])
+    m.write([(f1, 2, "RU"), (f2, 4, ("foo", "bar"))])
     assert m.metadata["geocollections"] == ["foo", "world"]
 
 
@@ -294,11 +297,14 @@ def test_method_geocollection_missing_ok():
         }
     )
 
+    f1 = get_node(code="1").id
+    f3 = get_node(code="3").id
+
     m = Method(("foo",))
     m.write(
         [
-            (1, 2, None),
-            (3, 4),
+            (f1, 2, None),
+            (f3, 4),
         ]
     )
     assert m.metadata["geocollections"] == ["world"]
@@ -313,10 +319,12 @@ def test_method_geocollection_warning():
         }
     )
 
+    f1 = get_node(code="1").id
+
     m = Method(("foo",))
     m.write(
         [
-            (1, 2, "Russia"),
+            (f1, 2, "Russia"),
         ]
     )
     assert m.metadata["geocollections"] == []
diff --git a/tests/utils.py b/tests/utils.py
index 48203ba0..7190b485 100644
--- a/tests/utils.py
+++ b/tests/utils.py
@@ -6,6 +6,7 @@
 from bw2data.backends import Activity as PWActivity
 from bw2data.errors import MultipleResults, UnknownObject, ValidityError
 from bw2data.tests import BW2DataTest, bw2test
+from bw2data.snowflake_ids import EPOCH_START_MS
 from bw2data.utils import (
     as_uncertainty_dict,
     combine_methods,
@@ -208,7 +209,7 @@ def test_as_uncertainty_dict_set_negative():
 def test_get_node_normal():
     Database("biosphere").write(biosphere)
     node = get_node(name="an emission")
-    assert node.id == 1
+    assert node.id > EPOCH_START_MS
     assert isinstance(node, PWActivity)
 
 
@@ -229,7 +230,7 @@ def test_get_node_key():
 def test_get_node_multiple_filters():
     Database("biosphere").write(biosphere)
     node = get_node(name="an emission", type="emission")
-    assert node.id == 1
+    assert node.id > EPOCH_START_MS
     assert isinstance(node, PWActivity)
 
 
@@ -277,7 +278,7 @@ def test_get_node_extended_search():
 @bw2test
 def test_get_activity_activity():
     Database("biosphere").write(biosphere)
-    node = get_node(id=1)
+    node = get_node(code="1")
     found = get_activity(node)
     assert found is node
 
@@ -285,33 +286,17 @@ def test_get_activity_activity():
 @bw2test
 def test_get_activity_id():
     Database("biosphere").write(biosphere)
-    node = get_activity(1)
-    assert node.id == 1
+    node = get_activity(code="1")
+    node = get_activity(node.id)
+    assert node.id > EPOCH_START_MS
     assert isinstance(node, PWActivity)
 
 
-@bw2test
-def test_get_activity_id_different_ints():
-    Database("biosphere").write(biosphere)
-    different_ints = [
-        int(1),
-        np.intp(1),
-        np.int8(1),
-        np.int16(1),
-        np.int32(1),
-        np.int64(1),
-    ]
-    for i in different_ints:
-        node = get_activity(i)
-        assert node.id == i
-        assert isinstance(node, PWActivity)
-
-
 @bw2test
 def test_get_activity_key():
     Database("biosphere").write(biosphere)
     node = get_activity(("biosphere", "1"))
-    assert node.id == 1
+    assert node.id > EPOCH_START_MS
     assert isinstance(node, PWActivity)
 
 
@@ -319,7 +304,7 @@ def test_get_activity_key():
 def test_get_activity_kwargs():
     Database("biosphere").write(biosphere)
     node = get_activity(name="an emission", type="emission")
-    assert node.id == 1
+    assert node.id > EPOCH_START_MS
     assert isinstance(node, PWActivity)
 
 

From 46f06aa8a7834d1949c7235f67c296864085c47d Mon Sep 17 00:00:00 2001
From: Chris Mutel <cmutel@gmail.com>
Date: Thu, 31 Oct 2024 11:39:44 +0100
Subject: [PATCH 2/6] Continued progress and cleanups
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Update compatibility.py

Remove debugging print statements

Clarify some comments

Improve error message

Cleanup error messages

Co-Authored-By: João Gonçalves <jsvgoncalves@gmail.com>
---
 bw2data/backends/base.py    |  3 +++
 bw2data/backends/proxies.py | 21 +++++++++------------
 bw2data/backends/schema.py  |  6 ++++--
 bw2data/backends/utils.py   |  2 ++
 bw2data/snowflake_ids.py    |  4 ++--
 tests/activity_proxy.py     | 13 -------------
 tests/compatibility.py      |  2 +-
 7 files changed, 21 insertions(+), 30 deletions(-)

diff --git a/bw2data/backends/base.py b/bw2data/backends/base.py
index 761b809e..2eef82bb 100644
--- a/bw2data/backends/base.py
+++ b/bw2data/backends/base.py
@@ -687,6 +687,9 @@ def new_node(self, code: str = None, **kwargs):
             kwargs.pop("database")
         obj["database"] = self.name
 
+        if "id" in kwargs:
+            raise ValueError(f"`id` must be created automatically, but `id={kwargs['id']}` given.")
+
         if code is None:
             obj["code"] = uuid.uuid4().hex
         else:
diff --git a/bw2data/backends/proxies.py b/bw2data/backends/proxies.py
index be316292..677c01cd 100644
--- a/bw2data/backends/proxies.py
+++ b/bw2data/backends/proxies.py
@@ -497,10 +497,10 @@ def new_edge(self, **kwargs):
         """Create a new exchange linked to this activity"""
         exc = Exchange()
         exc.output = self.key
-        for key in kwargs:
+        for key, value in kwargs.items():
             if key == "id":
-                raise ValueError("`id` must be created automatically")
-            exc[key] = kwargs[key]
+                raise ValueError(f"`id` must be created automatically, but `id={value}` given.")
+            exc[key] = value
         return exc
 
     def copy(self, code: Optional[str] = None, signal: bool = True, **kwargs):
@@ -514,22 +514,19 @@ def copy(self, code: Optional[str] = None, signal: bool = True, **kwargs):
         activity = Activity()
         for key, value in self.items():
             if key != "id":
-                print(key, value)
                 activity[key] = value
-        for k, v in kwargs.items():
-            if k == "id":
-                raise ValueError("`id` must be generated automatically")
-            activity._data[k] = v
+        for key, value in kwargs.items():
+            if key == "id":
+                raise ValueError(f"`id` must be created automatically, but `id={value}` given.")
+            activity._data[key] = value
         activity._data["code"] = str(code or uuid.uuid4().hex)
         activity.save(signal=signal)
 
-        print("Copy Document ID", activity._document.id)
-        print("Original Document ID", self._document.id)
-
         for exc in self.exchanges():
             data = copy.deepcopy(exc._data)
             if 'id' in data:
-                # New snowflake ID will be inserted by `_dict_as_exchangedataset`
+                # New snowflake ID will be inserted by `.save()`; shouldn't be copied over
+                # or specified manually
                 del data['id']
             data["output"] = activity.key
             # Change `input` for production exchanges
diff --git a/bw2data/backends/schema.py b/bw2data/backends/schema.py
index e32a1770..78087e1d 100644
--- a/bw2data/backends/schema.py
+++ b/bw2data/backends/schema.py
@@ -11,8 +11,10 @@ class SnowflakeIDBaseClass(SignaledDataset):
 
     def save(self, **kwargs):
         if self.id is None:
-            # If the primary key is already present, peewee will make an `UPDATE` query.
-            # This will have no effect if there isn't a matching row
+            # If the primary key column data is already present (even if the object doesn't exist in
+            # the database), peewee will make an `UPDATE` query. This will have no effect if there
+            # isn't a matching row. Need for force an `INSERT` query instead as we generate the ids
+            # ourselves.
             # https://docs.peewee-orm.com/en/latest/peewee/models.html#id4
             self.id = next(snowflake_id_generator)
             kwargs['force_insert'] = True
diff --git a/bw2data/backends/utils.py b/bw2data/backends/utils.py
index 13797d9c..3060dab4 100644
--- a/bw2data/backends/utils.py
+++ b/bw2data/backends/utils.py
@@ -78,6 +78,8 @@ def dict_as_activitydataset(ds: Any, add_snowflake_id: bool = False) -> dict:
         "product": ds.get("reference product"),
         "type": ds.get("type", labels.process_node_default),
     }
+    # Use during `insert_many` calls as these skip auto id generation because they don't call
+    # `.save()`
     if add_snowflake_id:
         val['id'] = next(snowflake_id_generator)
     return val
diff --git a/bw2data/snowflake_ids.py b/bw2data/snowflake_ids.py
index 9a9bf082..45c0b889 100644
--- a/bw2data/snowflake_ids.py
+++ b/bw2data/snowflake_ids.py
@@ -13,8 +13,8 @@
 # Timestamp: 41 bits. Milliseconds since the epoch or custom epoch.
 # Datacenter ID: 5 bits, which gives us 2 ^ 5 = 32 datacenters.
 # Machine ID: 5 bits, which gives us 2 ^ 5 = 32 machines per datacenter.
-# However, `snowflake-id` lumps these two values datacenter and machine id values together into an
-# `instance` value with 2 ^ 10 = 1024 possible values.
+# However, `snowflake-id` lumps the two datacenter and machine id values together into an
+# `instance` parameter with 2 ^ 10 = 1024 possible values.
 # Sequence number: 12 bits. For every ID generated on that machine/process, the sequence number is
 # incremented by 1. The number is reset to 0 every millisecond.
 snowflake_id_generator = SnowflakeGenerator(instance=uuid.getnode() % 1024, epoch=EPOCH_START_MS)
diff --git a/tests/activity_proxy.py b/tests/activity_proxy.py
index 55a897c0..a16af223 100644
--- a/tests/activity_proxy.py
+++ b/tests/activity_proxy.py
@@ -180,14 +180,8 @@ def test_copy(activity):
     assert cp["location"] == "bar"
     assert ExchangeDataset.select().count() == 2
 
-    for ds in ActivityDataset.select():
-        print(ds, ds.name, ds.code)
-
     cp.save()
 
-    for ds in ActivityDataset.select():
-        print(ds, ds.name, ds.code)
-
     assert ActivityDataset.select().count() == 2
     assert (
         ActivityDataset.select()
@@ -241,9 +235,6 @@ def test_copy_with_kwargs(activity):
 
 @bw2test
 def test_delete_activity_parameters():
-    from bw2data import projects
-    print(projects.dir)
-
     db = DatabaseChooser("example")
     db.register()
 
@@ -253,10 +244,6 @@ def test_delete_activity_parameters():
     b.save()
     a.new_exchange(amount=0, input=b, type="technosphere", formula="foo * bar + 4").save()
 
-    print("Iterating over exchanges:")
-    for exc in ExchangeDataset.select():
-        print("Next:", exc, exc.id)
-
     assert ExchangeDataset.select().count() == 1
 
     activity_data = [
diff --git a/tests/compatibility.py b/tests/compatibility.py
index 54c20b45..dc6616fb 100644
--- a/tests/compatibility.py
+++ b/tests/compatibility.py
@@ -110,7 +110,7 @@ def test_prepare_lca_inputs_multiple_demands_data_types(setup):
     first = get_node(database="food", code="1")
     second = get_node(database="food", code="2")
     d, objs, r = prepare_lca_inputs(demands=[{first: 1}, {second.id: 10}], method=("foo",))
-    assert d == [{3: 1}, {4: 10}]
+    assert d == [{first.id: 1}, {second.id: 10}]
     assert {o.metadata["id"] for o in objs} == {o.datapackage().metadata["id"] for o in setup}
 
 

From 9780dfadeaddc93efb3bda6548d19922bc682a57 Mon Sep 17 00:00:00 2001
From: Chris Mutel <cmutel@gmail.com>
Date: Fri, 8 Nov 2024 10:52:06 +0100
Subject: [PATCH 3/6] Resolve util func naming discrepancy

---
 bw2data/backends/base.py    | 4 ++--
 bw2data/backends/proxies.py | 4 ++--
 bw2data/backends/utils.py   | 2 +-
 3 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/bw2data/backends/base.py b/bw2data/backends/base.py
index 2eef82bb..84bda8fd 100644
--- a/bw2data/backends/base.py
+++ b/bw2data/backends/base.py
@@ -28,7 +28,7 @@
 from bw2data.backends.utils import (
     check_exchange,
     dict_as_activitydataset,
-    _dict_as_exchangedataset,
+    dict_as_exchangedataset,
     get_csv_data_dict,
     retupleize_geo_strings,
 )
@@ -525,7 +525,7 @@ def _efficient_write_dataset(
 
             if "output" not in exchange:
                 exchange["output"] = (ds["database"], ds["code"])
-            exchanges.append(_dict_as_exchangedataset(exchange))
+            exchanges.append(dict_as_exchangedataset(exchange))
 
             # Query gets passed as INSERT INTO x VALUES ('?', '?'...)
             # SQLite3 has a limit of 999 variables,
diff --git a/bw2data/backends/proxies.py b/bw2data/backends/proxies.py
index 677c01cd..ba085c36 100644
--- a/bw2data/backends/proxies.py
+++ b/bw2data/backends/proxies.py
@@ -14,7 +14,7 @@
     check_exchange_keys,
     check_exchange_type,
 )
-from bw2data.backends.utils import dict_as_activitydataset, _dict_as_exchangedataset
+from bw2data.backends.utils import dict_as_activitydataset, dict_as_exchangedataset
 from bw2data.configuration import labels
 from bw2data.errors import ValidityError
 from bw2data.logs import stdout_feedback_logger
@@ -574,7 +574,7 @@ def save(self, signal: bool = True, data_already_set: bool = False, force_insert
             check_exchange_type(self._data.get("type"))
             check_exchange_keys(self)
 
-            for key, value in _dict_as_exchangedataset(self._data).items():
+            for key, value in dict_as_exchangedataset(self._data).items():
                 setattr(self._document, key, value)
 
         self._document.save(signal=signal, force_insert=force_insert)
diff --git a/bw2data/backends/utils.py b/bw2data/backends/utils.py
index 3060dab4..0e4a55cd 100644
--- a/bw2data/backends/utils.py
+++ b/bw2data/backends/utils.py
@@ -85,7 +85,7 @@ def dict_as_activitydataset(ds: Any, add_snowflake_id: bool = False) -> dict:
     return val
 
 
-def _dict_as_exchangedataset(ds: Any) -> dict:
+def dict_as_exchangedataset(ds: Any) -> dict:
     return {
         "data": ds,
         "input_database": ds["input"][0],

From e4fc87bea37b4f89706a0d652a3f7bd16b592ac8 Mon Sep 17 00:00:00 2001
From: Chris Mutel <cmutel@gmail.com>
Date: Fri, 8 Nov 2024 10:52:20 +0100
Subject: [PATCH 4/6] Single source for snowflake ids

---
 bw2data/revisions.py                |  4 ++--
 tests/unit/test_node_edge_events.py | 18 +++++++++---------
 2 files changed, 11 insertions(+), 11 deletions(-)

diff --git a/bw2data/revisions.py b/bw2data/revisions.py
index e418e308..ffcdacfc 100644
--- a/bw2data/revisions.py
+++ b/bw2data/revisions.py
@@ -2,8 +2,8 @@
 from typing import Any, Optional, Sequence, TypeVar
 
 import deepdiff
-from snowflake import SnowflakeGenerator as sfg
 
+from bw2data.snowflake_ids import snowflake_id_generator
 from bw2data.backends.proxies import Activity, Exchange
 from bw2data.backends.schema import ActivityDataset, ExchangeDataset, SignaledDataset
 from bw2data.backends.utils import dict_as_activitydataset, dict_as_exchangedataset
@@ -186,7 +186,7 @@ def generate_metadata(
 ) -> dict[str, Any]:
     metadata = metadata or {}
     metadata["parent_revision"] = parent_revision
-    metadata["revision"] = revision or next(sfg(0))
+    metadata["revision"] = revision or next(snowflake_id_generator)
     metadata["authors"] = metadata.get("authors", "Anonymous")
     metadata["title"] = metadata.get("title", "Untitled revision")
     metadata["description"] = metadata.get("description", "No description")
diff --git a/tests/unit/test_node_edge_events.py b/tests/unit/test_node_edge_events.py
index 205adfe5..810fbbb6 100644
--- a/tests/unit/test_node_edge_events.py
+++ b/tests/unit/test_node_edge_events.py
@@ -1,13 +1,13 @@
 import json
 
 import pytest
-from snowflake import SnowflakeGenerator as sfg
 
 from bw2data import get_node
 from bw2data.backends.schema import ExchangeDataset
 from bw2data.database import DatabaseChooser
 from bw2data.project import projects
 from bw2data.tests import bw2test
+from bw2data.snowflake_ids import snowflake_id_generator
 
 
 @bw2test
@@ -68,7 +68,7 @@ def test_node_revision_apply_create():
     database = DatabaseChooser("db")
     database.register()
 
-    revision_id = next(sfg(0))
+    revision_id = next(snowflake_id_generator)
     revision = {
         "data": [
             {
@@ -172,7 +172,7 @@ def test_node_revision_apply_delete():
     node.save()
     assert len(database) == 1
 
-    revision_id = next(sfg(0))
+    revision_id = next(snowflake_id_generator)
 
     revision = {
         "data": [
@@ -261,7 +261,7 @@ def test_node_revision_apply_update():
     node = database.new_node(code="A", name="A", location="kalamazoo")
     node.save()
 
-    revision_id = next(sfg(0))
+    revision_id = next(snowflake_id_generator)
 
     revision = {
         "data": [
@@ -355,7 +355,7 @@ def test_node_revision_apply_activity_database_change():
     node.new_edge(input=node, type="production", amount=1.0).save()
     assert len(node.exchanges()) == 2
 
-    revision_id = next(sfg(0))
+    revision_id = next(snowflake_id_generator)
 
     num_revisions_before = len([
         fp
@@ -459,7 +459,7 @@ def test_node_revision_apply_activity_code_change():
     node.new_edge(input=node, type="production", amount=1.0).save()
     assert len(node.exchanges()) == 2
 
-    revision_id = next(sfg(0))
+    revision_id = next(snowflake_id_generator)
 
     num_revisions_before = len([
         fp
@@ -660,9 +660,9 @@ def test_node_revision_apply_activity_copy():
 
     projects.dataset.set_sourced()
 
-    revision_id_1 = next(sfg(0))
-    revision_id_2 = next(sfg(0))
-    revision_id_3 = next(sfg(0))
+    revision_id_1 = next(snowflake_id_generator)
+    revision_id_2 = next(snowflake_id_generator)
+    revision_id_3 = next(snowflake_id_generator)
 
     num_revisions_before = len([
         fp

From 29d615a3f77592fec69de75d36f8604c228aa9de Mon Sep 17 00:00:00 2001
From: Chris Mutel <cmutel@gmail.com>
Date: Fri, 8 Nov 2024 11:42:59 +0100
Subject: [PATCH 5/6] Fix ORM class inheritance hierarchy

---
 bw2data/backends/base.py            |  4 +-
 bw2data/backends/proxies.py         |  4 +-
 bw2data/backends/schema.py          | 20 +-------
 bw2data/backends/utils.py           |  8 ++--
 bw2data/project.py                  |  8 +---
 bw2data/revisions.py                |  8 ++--
 bw2data/snowflake_ids.py            | 21 +++++++-
 tests/compatibility.py              | 13 +++--
 tests/conftest.py                   |  1 +
 tests/database.py                   | 12 ++---
 tests/fixtures/__init__.py          |  2 +-
 tests/test_schema_migrations.py     | 19 ++++----
 tests/unit/test_node_edge_events.py | 74 +++++++++++++++++------------
 tests/unit/test_schema_sourcing.py  |  6 +--
 tests/utils.py                      |  2 +-
 15 files changed, 106 insertions(+), 96 deletions(-)

diff --git a/bw2data/backends/base.py b/bw2data/backends/base.py
index 84bda8fd..39f448c3 100644
--- a/bw2data/backends/base.py
+++ b/bw2data/backends/base.py
@@ -856,7 +856,9 @@ def _add_inventory_geomapping_to_datapackage(self, dp: Datapackage) -> None:
             dict_iterator=(
                 {
                     "row": row[0],
-                    "col": geomapping[location_mapper(retupleize_geo_strings(row[1]) or config.global_location)],
+                    "col": geomapping[
+                        location_mapper(retupleize_geo_strings(row[1]) or config.global_location)
+                    ],
                     "amount": 1,
                 }
                 for row in inv_mapping_qs.tuples()
diff --git a/bw2data/backends/proxies.py b/bw2data/backends/proxies.py
index ba085c36..ab300903 100644
--- a/bw2data/backends/proxies.py
+++ b/bw2data/backends/proxies.py
@@ -524,10 +524,10 @@ def copy(self, code: Optional[str] = None, signal: bool = True, **kwargs):
 
         for exc in self.exchanges():
             data = copy.deepcopy(exc._data)
-            if 'id' in data:
+            if "id" in data:
                 # New snowflake ID will be inserted by `.save()`; shouldn't be copied over
                 # or specified manually
-                del data['id']
+                del data["id"]
             data["output"] = activity.key
             # Change `input` for production exchanges
             if exc["input"] == exc["output"]:
diff --git a/bw2data/backends/schema.py b/bw2data/backends/schema.py
index 78087e1d..a92a5579 100644
--- a/bw2data/backends/schema.py
+++ b/bw2data/backends/schema.py
@@ -1,24 +1,8 @@
-from peewee import DoesNotExist, Model, TextField, IntegerField
+from peewee import DoesNotExist, TextField
 
 from bw2data.errors import UnknownObject
-from bw2data.signals import SignaledDataset
+from bw2data.snowflake_ids import SnowflakeIDBaseClass
 from bw2data.sqlite import PickleField
-from bw2data.snowflake_ids import snowflake_id_generator
-
-
-class SnowflakeIDBaseClass(SignaledDataset):
-    id = IntegerField(primary_key=True)
-
-    def save(self, **kwargs):
-        if self.id is None:
-            # If the primary key column data is already present (even if the object doesn't exist in
-            # the database), peewee will make an `UPDATE` query. This will have no effect if there
-            # isn't a matching row. Need for force an `INSERT` query instead as we generate the ids
-            # ourselves.
-            # https://docs.peewee-orm.com/en/latest/peewee/models.html#id4
-            self.id = next(snowflake_id_generator)
-            kwargs['force_insert'] = True
-        super().save(**kwargs)
 
 
 class ActivityDataset(SnowflakeIDBaseClass):
diff --git a/bw2data/backends/utils.py b/bw2data/backends/utils.py
index 0e4a55cd..d3287564 100644
--- a/bw2data/backends/utils.py
+++ b/bw2data/backends/utils.py
@@ -1,15 +1,15 @@
-from typing import Any
 import copy
 import warnings
-from typing import Optional
+from typing import Any, Optional
 
 import numpy as np
 
 from bw2data import config
-from bw2data.backends.schema import SignaledDataset, get_id
+from bw2data.backends.schema import get_id
 from bw2data.configuration import labels
 from bw2data.errors import InvalidExchange, UntypedExchange
 from bw2data.meta import databases, methods
+from bw2data.signals import SignaledDataset
 from bw2data.snowflake_ids import snowflake_id_generator
 
 
@@ -81,7 +81,7 @@ def dict_as_activitydataset(ds: Any, add_snowflake_id: bool = False) -> dict:
     # Use during `insert_many` calls as these skip auto id generation because they don't call
     # `.save()`
     if add_snowflake_id:
-        val['id'] = next(snowflake_id_generator)
+        val["id"] = next(snowflake_id_generator)
     return val
 
 
diff --git a/bw2data/project.py b/bw2data/project.py
index 618f99e6..994fa4e3 100644
--- a/bw2data/project.py
+++ b/bw2data/project.py
@@ -7,7 +7,7 @@
 from copy import copy
 from functools import partial
 from pathlib import Path
-from typing import TYPE_CHECKING, Any, Optional, Sequence, TypeVar
+from typing import Any, Optional, Sequence, TypeVar
 
 import wrapt
 from bw_processing import safe_filename
@@ -23,12 +23,6 @@
 from bw2data.utils import maybe_path
 
 
-if TYPE_CHECKING:
-    from bw2data import revisions
-    from bw2data.backends import schema
-    SD = TypeVar("SD", bound=schema.SignaledDataset)
-
-
 READ_ONLY_PROJECT = """
 ***Read only project***
 
diff --git a/bw2data/revisions.py b/bw2data/revisions.py
index ffcdacfc..6148c1a5 100644
--- a/bw2data/revisions.py
+++ b/bw2data/revisions.py
@@ -3,11 +3,12 @@
 
 import deepdiff
 
-from bw2data.snowflake_ids import snowflake_id_generator
 from bw2data.backends.proxies import Activity, Exchange
-from bw2data.backends.schema import ActivityDataset, ExchangeDataset, SignaledDataset
+from bw2data.backends.schema import ActivityDataset, ExchangeDataset
 from bw2data.backends.utils import dict_as_activitydataset, dict_as_exchangedataset
 from bw2data.errors import DifferentObjects, IncompatibleClasses, InconsistentData
+from bw2data.signals import SignaledDataset
+from bw2data.snowflake_ids import snowflake_id_generator
 from bw2data.utils import get_node
 
 try:
@@ -16,9 +17,6 @@
     from typing_extensions import Self
 
 
-SD = TypeVar("SD", bound=SignaledDataset)
-
-
 class RevisionGraph:
     """Graph of revisions, edges are based on `metadata.parent_revision`."""
 
diff --git a/bw2data/snowflake_ids.py b/bw2data/snowflake_ids.py
index 45c0b889..3872cafe 100644
--- a/bw2data/snowflake_ids.py
+++ b/bw2data/snowflake_ids.py
@@ -1,6 +1,10 @@
-from snowflake import SnowflakeGenerator
 import uuid
 
+from peewee import IntegerField
+from snowflake import SnowflakeGenerator
+
+from bw2data.signals import SignaledDataset
+
 # Jan 1, 2024
 # from datetime import datetime
 # (datetime(2024, 1, 1) - datetime.utcfromtimestamp(0)).total_seconds() * 1000.0
@@ -18,3 +22,18 @@
 # Sequence number: 12 bits. For every ID generated on that machine/process, the sequence number is
 # incremented by 1. The number is reset to 0 every millisecond.
 snowflake_id_generator = SnowflakeGenerator(instance=uuid.getnode() % 1024, epoch=EPOCH_START_MS)
+
+
+class SnowflakeIDBaseClass(SignaledDataset):
+    id = IntegerField(primary_key=True)
+
+    def save(self, **kwargs):
+        if self.id is None:
+            # If the primary key column data is already present (even if the object doesn't exist in
+            # the database), peewee will make an `UPDATE` query. This will have no effect if there
+            # isn't a matching row. Need for force an `INSERT` query instead as we generate the ids
+            # ourselves.
+            # https://docs.peewee-orm.com/en/latest/peewee/models.html#id4
+            self.id = next(snowflake_id_generator)
+            kwargs["force_insert"] = True
+        super().save(**kwargs)
diff --git a/tests/compatibility.py b/tests/compatibility.py
index dc6616fb..d7777fae 100644
--- a/tests/compatibility.py
+++ b/tests/compatibility.py
@@ -9,7 +9,6 @@
     databases,
     geomapping,
     get_activity,
-    get_node,
     get_multilca_data_objs,
     get_node,
     methods,
@@ -69,10 +68,10 @@ def test_prepare_lca_inputs_basic(setup):
     assert list(d.values()) == [1]
     assert {o.metadata["id"] for o in objs} == {o.datapackage().metadata["id"] for o in setup}
 
-    b1 = get_node(database="biosphere", code='1').id
-    b2 = get_node(database="biosphere", code='2').id
-    f1 = get_node(database="food", code='1').id
-    f2 = get_node(database="food", code='2').id
+    b1 = get_node(database="biosphere", code="1").id
+    b2 = get_node(database="biosphere", code="2").id
+    f1 = get_node(database="food", code="1").id
+    f2 = get_node(database="food", code="2").id
 
     remapping_expected = {
         "activity": {
@@ -118,8 +117,8 @@ def test_prepare_lca_inputs_multiple_demands(setup):
     d, objs, r = prepare_lca_inputs(
         demands=[{("food", "1"): 1}, {("food", "2"): 10}], method=("foo",)
     )
-    f1 = get_node(database="food", code='1').id
-    f2 = get_node(database="food", code='2').id
+    f1 = get_node(database="food", code="1").id
+    f2 = get_node(database="food", code="2").id
     assert d == [{f1: 1}, {f2: 10}]
     assert {o.metadata["id"] for o in objs} == {o.datapackage().metadata["id"] for o in setup}
 
diff --git a/tests/conftest.py b/tests/conftest.py
index db094fa3..c8cf5bcb 100644
--- a/tests/conftest.py
+++ b/tests/conftest.py
@@ -1,4 +1,5 @@
 """Fixtures for bw2data"""
+
 import sqlite3
 
 # import pytest
diff --git a/tests/database.py b/tests/database.py
index 0070e222..7ac20476 100644
--- a/tests/database.py
+++ b/tests/database.py
@@ -11,15 +11,14 @@
     calculation_setups,
     databases,
     geomapping,
-    projects,
     get_activity,
     get_id,
     get_node,
+    projects,
 )
 from bw2data.backends import Activity as PWActivity
 from bw2data.backends import sqlite3_lci_db
 from bw2data.database import Database
-from bw2data.snowflake_ids import EPOCH_START_MS
 from bw2data.errors import (
     DuplicateNode,
     InvalidExchange,
@@ -33,6 +32,7 @@
     ParameterizedExchange,
     parameters,
 )
+from bw2data.snowflake_ids import EPOCH_START_MS
 from bw2data.tests import bw2test
 
 from .fixtures import biosphere
@@ -105,10 +105,10 @@ def test_copy(food):
 
 def test_copy_metadata(food):
     d = Database("food")
-    d.metadata['custom'] = "something"
+    d.metadata["custom"] = "something"
     d.copy("repas")
     assert "repas" in databases
-    assert databases['repas']['custom'] == 'something'
+    assert databases["repas"]["custom"] == "something"
 
 
 @bw2test
@@ -458,7 +458,7 @@ def test_geomapping_array_includes_only_processes():
 @bw2test
 def test_geomapping_array_normalization():
     database = Database("a database")
-    database.register(location_normalization={'RoW': 'GLO'})
+    database.register(location_normalization={"RoW": "GLO"})
     database.write(
         {
             ("a database", "foo"): {
@@ -717,7 +717,7 @@ def test_process_without_exchanges_still_in_processed_array():
 
     package = database.datapackage()
     array = package.get_resource("a_database_technosphere_matrix.data")[0]
-    assert array[0] == 1.
+    assert array[0] == 1.0
     assert array.shape == (1,)
 
 
diff --git a/tests/fixtures/__init__.py b/tests/fixtures/__init__.py
index b7becf7f..06353416 100644
--- a/tests/fixtures/__init__.py
+++ b/tests/fixtures/__init__.py
@@ -1 +1 @@
-from .basic import get_naughty, food2, food, biosphere, lcia
+from .basic import biosphere, food, food2, get_naughty, lcia
diff --git a/tests/test_schema_migrations.py b/tests/test_schema_migrations.py
index 2799bc32..a2efd04e 100644
--- a/tests/test_schema_migrations.py
+++ b/tests/test_schema_migrations.py
@@ -1,9 +1,10 @@
-from pathlib import Path
-from bw2data.project import add_sourced_columns, projects, ProjectDataset, config
-from bw2data.tests import bw2test
 import shutil
+from pathlib import Path
+
 from peewee import SqliteDatabase
 
+from bw2data.project import ProjectDataset, add_sourced_columns, config, projects
+from bw2data.tests import bw2test
 
 original_projects_db = Path(__file__).parent / "fixtures" / "projects.db"
 
@@ -27,13 +28,13 @@ def test_add_sourced_columns(tmp_path):
 
     columns = {o.name: o for o in db.get_columns("projectdataset")}
     assert "is_sourced" in columns
-    assert columns['is_sourced'].data_type.upper() == 'INTEGER'
-    assert columns['is_sourced'].default == '0'
-    assert columns['is_sourced'].null is True
+    assert columns["is_sourced"].data_type.upper() == "INTEGER"
+    assert columns["is_sourced"].default == "0"
+    assert columns["is_sourced"].null is True
     assert "revision" in columns
-    assert columns['revision'].data_type.upper() == 'INTEGER'
-    assert columns['revision'].default is None
-    assert columns['revision'].null is True
+    assert columns["revision"].data_type.upper() == "INTEGER"
+    assert columns["revision"].default is None
+    assert columns["revision"].null is True
 
     db = SqliteDatabase(tmp_path / "projects.backup-is-sourced.db")
     db.connect()
diff --git a/tests/unit/test_node_edge_events.py b/tests/unit/test_node_edge_events.py
index 810fbbb6..3400db6a 100644
--- a/tests/unit/test_node_edge_events.py
+++ b/tests/unit/test_node_edge_events.py
@@ -6,8 +6,8 @@
 from bw2data.backends.schema import ExchangeDataset
 from bw2data.database import DatabaseChooser
 from bw2data.project import projects
-from bw2data.tests import bw2test
 from bw2data.snowflake_ids import snowflake_id_generator
+from bw2data.tests import bw2test
 
 
 @bw2test
@@ -357,11 +357,13 @@ def test_node_revision_apply_activity_database_change():
 
     revision_id = next(snowflake_id_generator)
 
-    num_revisions_before = len([
-        fp
-        for fp in (projects.dataset.dir / "revisions").iterdir()
-        if fp.stem.lower() != "head" and fp.is_file()
-    ])
+    num_revisions_before = len(
+        [
+            fp
+            for fp in (projects.dataset.dir / "revisions").iterdir()
+            if fp.stem.lower() != "head" and fp.is_file()
+        ]
+    )
 
     revision = {
         "data": [
@@ -393,11 +395,13 @@ def test_node_revision_apply_activity_database_change():
         assert exc.input == node
         assert exc.output == node
 
-    num_revisions_after = len([
-        fp
-        for fp in (projects.dataset.dir / "revisions").iterdir()
-        if fp.stem.lower() != "head" and fp.is_file()
-    ])
+    num_revisions_after = len(
+        [
+            fp
+            for fp in (projects.dataset.dir / "revisions").iterdir()
+            if fp.stem.lower() != "head" and fp.is_file()
+        ]
+    )
     assert num_revisions_after == num_revisions_before
 
 
@@ -461,11 +465,13 @@ def test_node_revision_apply_activity_code_change():
 
     revision_id = next(snowflake_id_generator)
 
-    num_revisions_before = len([
-        fp
-        for fp in (projects.dataset.dir / "revisions").iterdir()
-        if fp.stem.lower() != "head" and fp.is_file()
-    ])
+    num_revisions_before = len(
+        [
+            fp
+            for fp in (projects.dataset.dir / "revisions").iterdir()
+            if fp.stem.lower() != "head" and fp.is_file()
+        ]
+    )
 
     revision = {
         "data": [
@@ -496,11 +502,13 @@ def test_node_revision_apply_activity_code_change():
         assert exc.input == node
         assert exc.output == node
 
-    num_revisions_after = len([
-        fp
-        for fp in (projects.dataset.dir / "revisions").iterdir()
-        if fp.stem.lower() != "head" and fp.is_file()
-    ])
+    num_revisions_after = len(
+        [
+            fp
+            for fp in (projects.dataset.dir / "revisions").iterdir()
+            if fp.stem.lower() != "head" and fp.is_file()
+        ]
+    )
     assert num_revisions_after == num_revisions_before
 
 
@@ -664,11 +672,13 @@ def test_node_revision_apply_activity_copy():
     revision_id_2 = next(snowflake_id_generator)
     revision_id_3 = next(snowflake_id_generator)
 
-    num_revisions_before = len([
-        fp
-        for fp in (projects.dataset.dir / "revisions").iterdir()
-        if fp.stem.lower() != "head" and fp.is_file()
-    ])
+    num_revisions_before = len(
+        [
+            fp
+            for fp in (projects.dataset.dir / "revisions").iterdir()
+            if fp.stem.lower() != "head" and fp.is_file()
+        ]
+    )
 
     revisions = [
         {
@@ -798,11 +808,13 @@ def test_node_revision_apply_activity_copy():
         assert exc.output == new_node
         assert exc["amount"] == 0.1
 
-    num_revisions_after = len([
-        fp
-        for fp in (projects.dataset.dir / "revisions").iterdir()
-        if fp.stem.lower() != "head" and fp.is_file()
-    ])
+    num_revisions_after = len(
+        [
+            fp
+            for fp in (projects.dataset.dir / "revisions").iterdir()
+            if fp.stem.lower() != "head" and fp.is_file()
+        ]
+    )
     assert num_revisions_after == num_revisions_before
 
 
diff --git a/tests/unit/test_schema_sourcing.py b/tests/unit/test_schema_sourcing.py
index 0eb4861e..7cd90a92 100644
--- a/tests/unit/test_schema_sourcing.py
+++ b/tests/unit/test_schema_sourcing.py
@@ -12,7 +12,7 @@
 
 
 @bw2test
-@patch("bw2data.backends.schema.SignaledDataset.save")
+@patch("bw2data.signals.SignaledDataset.save")
 def test_signaleddataset_save_is_called(*mocks: Mock):
     # On saving an `Activity`, the generic `SignaledDataset.save` method is called
     projects.set_current("activity-event")
@@ -26,7 +26,7 @@ def test_signaleddataset_save_is_called(*mocks: Mock):
 
 @bw2test
 @patch("bw2data.signals.signaleddataset_on_save.send")
-@patch("bw2data.backends.schema.SignaledDataset.get")
+@patch("bw2data.signals.SignaledDataset.get")
 def test_signal_is_sent(*mocks: Mock):
     # On saving an `Activity`, the signal is sent
     projects.set_current("activity-event")
@@ -48,7 +48,7 @@ def test_signal_received(*mocks: Mock):
     db.register()
     a = db.new_node(code="A", name="A")
     a.save()
-    bw2data.project._signal_dataset_saved("test", old=None, new=a._document)
+    bw2data.project.signal_dispatcher("test", old=None, new=a._document)
     for m in mocks:
         assert m.called
 
diff --git a/tests/utils.py b/tests/utils.py
index 7190b485..b6a82556 100644
--- a/tests/utils.py
+++ b/tests/utils.py
@@ -5,8 +5,8 @@
 from bw2data import Database, Method, labels, methods
 from bw2data.backends import Activity as PWActivity
 from bw2data.errors import MultipleResults, UnknownObject, ValidityError
-from bw2data.tests import BW2DataTest, bw2test
 from bw2data.snowflake_ids import EPOCH_START_MS
+from bw2data.tests import BW2DataTest, bw2test
 from bw2data.utils import (
     as_uncertainty_dict,
     combine_methods,

From 209475a3a9416a2ce5136d0bb6a04e7d40ebb9b6 Mon Sep 17 00:00:00 2001
From: Chris Mutel <cmutel@gmail.com>
Date: Fri, 8 Nov 2024 11:51:37 +0100
Subject: [PATCH 6/6] fixup! Fix ORM class inheritance hierarchy

---
 bw2data/project.py | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/bw2data/project.py b/bw2data/project.py
index 994fa4e3..f2a3de9b 100644
--- a/bw2data/project.py
+++ b/bw2data/project.py
@@ -7,7 +7,7 @@
 from copy import copy
 from functools import partial
 from pathlib import Path
-from typing import Any, Optional, Sequence, TypeVar
+from typing import TYPE_CHECKING, Any, Optional, Sequence
 
 import wrapt
 from bw_processing import safe_filename
@@ -23,6 +23,10 @@
 from bw2data.utils import maybe_path
 
 
+if TYPE_CHECKING:
+    from bw2data import revisions
+
+
 READ_ONLY_PROJECT = """
 ***Read only project***