general: migrate modules to use 3.9 features

my/instagram/all.py

@@ -1,11 +1,10 @@
-from typing import Iterator
+from collections.abc import Iterator
 
-from my.core import Res, stat, Stats
+from my.core import Res, Stats, stat
 from my.core.source import import_source
 
 from .common import Message, _merge_messages
 
-
 src_gdpr = import_source(module_name='my.instagram.gdpr')
 @src_gdpr
 def _messages_gdpr() -> Iterator[Res[Message]]:

my/instagram/android.py

@@ -3,30 +3,29 @@ Bumble data from Android app database (in =/data/data/com.instagram.android/data
 """
 from __future__ import annotations
 
+import json
+import sqlite3
+from collections.abc import Iterator, Sequence
 from dataclasses import dataclass
 from datetime import datetime
-import json
 from pathlib import Path
-import sqlite3
-from typing import Iterator, Sequence, Optional, Dict, Union
 
 from my.core import (
-    get_files,
-    Paths,
-    make_config,
-    make_logger,
-    datetime_naive,
     Json,
+    Paths,
     Res,
     assert_never,
+    datetime_naive,
+    get_files,
+    make_config,
+    make_logger,
 )
-from my.core.common import unique_everseen
 from my.core.cachew import mcachew
+from my.core.common import unique_everseen
 from my.core.error import echain
-from my.core.sqlite import sqlite_connect_immutable, select
-
-from my.config import instagram as user_config
+from my.core.sqlite import select, sqlite_connect_immutable
 
+from my.config import instagram as user_config  # isort: skip
 
 logger = make_logger(__name__)
 
@@ -38,8 +37,8 @@ class instagram_android_config(user_config.android):
 
     # sadly doesn't seem easy to extract user's own handle/name from the db...
     # todo maybe makes more sense to keep in parent class? not sure...
-    username: Optional[str] = None
-    full_name: Optional[str] = None
+    username: str | None = None
+    full_name: str | None = None
 
 
 config = make_config(instagram_android_config)
@@ -101,13 +100,13 @@ class MessageError(RuntimeError):
         return self.rest == other.rest
 
 
-def _parse_message(j: Json) -> Optional[_Message]:
+def _parse_message(j: Json) -> _Message | None:
     id = j['item_id']
     t = j['item_type']
     tid = j['thread_key']['thread_id']
     uid = j['user_id']
     created = datetime.fromtimestamp(int(j['timestamp']) / 1_000_000)
-    text: Optional[str] = None
+    text: str | None = None
     if t == 'text':
         text = j['text']
     elif t == 'reel_share':
@@ -133,7 +132,7 @@ def _parse_message(j: Json) -> Optional[_Message]:
     )
 
 
-def _process_db(db: sqlite3.Connection) -> Iterator[Res[Union[User, _Message]]]:
+def _process_db(db: sqlite3.Connection) -> Iterator[Res[User | _Message]]:
     # TODO ugh. seems like no way to extract username?
     # sometimes messages (e.g. media_share) contain it in message field
     # but generally it's not present. ugh
@@ -175,7 +174,7 @@ def _process_db(db: sqlite3.Connection) -> Iterator[Res[Union[User, _Message]]]:
             yield e
 
 
-def _entities() -> Iterator[Res[Union[User, _Message]]]:
+def _entities() -> Iterator[Res[User | _Message]]:
     # NOTE: definitely need to merge multiple, app seems to recycle old messages
     # TODO: hmm hard to guarantee timestamp ordering when we use synthetic input data...
     # todo use TypedDict?
@@ -194,7 +193,7 @@ def _entities() -> Iterator[Res[Union[User, _Message]]]:
 
 @mcachew(depends_on=inputs)
 def messages() -> Iterator[Res[Message]]:
-    id2user: Dict[str, User] = {}
+    id2user: dict[str, User] = {}
     for x in unique_everseen(_entities):
         if isinstance(x, Exception):
             yield x

my/instagram/common.py

@@ -1,9 +1,10 @@
+from collections.abc import Iterator
 from dataclasses import replace
 from datetime import datetime
 from itertools import chain
-from typing import Iterator, Dict, Any, Protocol
+from typing import Any, Protocol
 
-from my.core import warn_if_empty, Res
+from my.core import Res, warn_if_empty
 
 
 class User(Protocol):
@@ -40,7 +41,7 @@ def _merge_messages(*sources: Iterator[Res[Message]]) -> Iterator[Res[Message]]:
     # ugh. seems that GDPR thread ids are completely uncorrelated to any android ids (tried searching over all sqlite dump)
     # so the only way to correlate is to try and match messages
     # we also can't use unique_everseen here, otherwise will never get a chance to unify threads
-    mmap: Dict[str, Message] = {}
+    mmap: dict[str, Message] = {}
     thread_map = {}
     user_map = {}
 
@@ -60,7 +61,7 @@ def _merge_messages(*sources: Iterator[Res[Message]]) -> Iterator[Res[Message]]:
                 user_map[m.user.id] = mm.user
         else:
             # not emitted yet, need to emit
-            repls: Dict[str, Any] = {}
+            repls: dict[str, Any] = {}
             tid = thread_map.get(m.thread_id)
             if tid is not None:
                 repls['thread_id'] = tid

my/instagram/gdpr.py

@@ -2,26 +2,27 @@
 Instagram data (uses [[https://www.instagram.com/download/request][official GDPR export]])
 """
 
+from __future__ import annotations
+
+import json
+from collections.abc import Iterator, Sequence
 from dataclasses import dataclass
 from datetime import datetime
-import json
 from pathlib import Path
-from typing import Iterator, Sequence, Dict, Union
 
 from more_itertools import bucket
 
 from my.core import (
-    get_files,
     Paths,
-    datetime_naive,
     Res,
     assert_never,
+    datetime_naive,
+    get_files,
     make_logger,
 )
 from my.core.common import unique_everseen
 
-from my.config import instagram as user_config
-
+from my.config import instagram as user_config  # isort: skip
 
 logger = make_logger(__name__)
 
@@ -70,7 +71,7 @@ def _decode(s: str) -> str:
     return s.encode('latin-1').decode('utf8')
 
 
-def _entities() -> Iterator[Res[Union[User, _Message]]]:
+def _entities() -> Iterator[Res[User | _Message]]:
     # it's worth processing all previous export -- sometimes instagram removes some metadata from newer ones
     # NOTE: here there are basically two options
     # - process inputs as is (from oldest to newest)
@@ -84,7 +85,7 @@ def _entities() -> Iterator[Res[Union[User, _Message]]]:
         yield from _entitites_from_path(path)
 
 
-def _entitites_from_path(path: Path) -> Iterator[Res[Union[User, _Message]]]:
+def _entitites_from_path(path: Path) -> Iterator[Res[User | _Message]]:
     # TODO make sure it works both with plan directory
     # idelaly get_files should return the right thing, and we won't have to force ZipPath/match_structure here
     # e.g. possible options are:
@@ -202,7 +203,7 @@ def _entitites_from_path(path: Path) -> Iterator[Res[Union[User, _Message]]]:
 
 # TODO basically copy pasted from android.py... hmm
 def messages() -> Iterator[Res[Message]]:
-    id2user: Dict[str, User] = {}
+    id2user: dict[str, User] = {}
     for x in unique_everseen(_entities):
         if isinstance(x, Exception):
             yield x