More typing fixes

src/rmscene/crdt_sequence.py

@@ -10,7 +10,10 @@
 from .tagged_block_common import CrdtId
 
 
-_T = tp.TypeVar("_T")
+# If the type constraint is for a CrdtSequenceItem[Superclass], then a
+# CrdtSequenceItem[Subclass] would do, so it is covariant.
+
+_T = tp.TypeVar("_T", covariant=True)
 
 
 @dataclass
@@ -22,7 +25,11 @@ class CrdtSequenceItem(tp.Generic[_T]):
     value: _T
 
 
-class CrdtSequence(tp.Generic[_T]):
+# As a mutable container, CrdtSequence is invariant.
+_Ti = tp.TypeVar("_Ti", covariant=False)
+
+
+class CrdtSequence(tp.Generic[_Ti]):
     """Ordered CRDT Sequence container.
 
     The Sequence contains `CrdtSequenceItem`s, each of which has an ID and
@@ -57,27 +64,27 @@ def keys(self) -> list[CrdtId]:
         """Return CrdtIds in order."""
         return list(self)
 
-    def values(self) -> list[_T]:
+    def values(self) -> list[_Ti]:
         """Return list of sorted values."""
         return [self[item_id] for item_id in self]
 
-    def items(self) -> Iterable[tuple[CrdtId, _T]]:
+    def items(self) -> Iterable[tuple[CrdtId, _Ti]]:
         """Return list of sorted key, value pairs."""
         return [(item_id, self[item_id]) for item_id in self]
 
-    def __getitem__(self, key: CrdtId) -> _T:
+    def __getitem__(self, key: CrdtId) -> _Ti:
         """Return item with key"""
         return self._items[key].value
 
     ## Access SequenceItems
 
-    def sequence_items(self) -> list[CrdtSequenceItem[_T]]:
+    def sequence_items(self) -> list[CrdtSequenceItem[_Ti]]:
         """Iterate through CrdtSequenceItems."""
         return list(self._items.values())
 
     ## Modify sequence
 
-    def add(self, item: CrdtSequenceItem[_T]):
+    def add(self, item: CrdtSequenceItem[_Ti]):
         if item.item_id in self._items:
             raise ValueError("Already have item %s" % item.item_id)
         self._items[item.item_id] = item

src/rmscene/text.py

@@ -17,7 +17,9 @@
 _logger = logging.getLogger(__name__)
 
 
-def expand_text_item(item: CrdtSequenceItem[str | int]) -> Iterable[CrdtSequenceItem[str | int]]:
+def expand_text_item(
+    item: CrdtSequenceItem[str | int],
+) -> tp.Iterator[CrdtSequenceItem[str | int]]:
     """Expand TextItem into single-character TextItems.
 
     Text is stored as strings in TextItems, each with an associated ID for the
@@ -56,7 +58,9 @@ def expand_text_item(item: CrdtSequenceItem[str | int]) -> Iterable[CrdtSequence
     yield CrdtSequenceItem(item_id, left_id, item.right_id, deleted_length, chars[-1])
 
 
-def expand_text_items(items: Iterable[CrdtSequenceItem[str | int]]) -> Iterable[CrdtSequenceItem[str | int]]:
+def expand_text_items(
+    items: Iterable[CrdtSequenceItem[str | int]],
+) -> tp.Iterator[CrdtSequenceItem[str | int]]:
     """Expand a sequence of TextItems into single-character TextItems."""
     for item in items:
         yield from expand_text_item(item)
@@ -74,6 +78,7 @@ def __str__(self):
 @dataclass
 class TextSpan:
     """Base class for text spans with formatting."""
+
     contents: list[tp.Union["TextSpan", CrdtStr]]
 
 
@@ -88,6 +93,7 @@ class ItalicSpan(TextSpan):
 @dataclass
 class Paragraph:
     """Paragraph of text."""
+
     contents: list[TextSpan]
     start_id: CrdtId
     style: LwwValue[si.ParagraphStyle]
@@ -98,7 +104,6 @@ def __str__(self):
 
 @dataclass
 class TextDocument:
-
     contents: list[Paragraph]
 
     @classmethod
@@ -146,8 +151,12 @@ def parse_paragraph_contents():
                     elif char in span_end_codes:
                         span_type, nested = stack.pop()
                         if span_type is not span_end_codes[char]:
-                            _logger.error("Unexpected end of span at %s: got %s, expected %s",
-                                          k, span_end_codes[char], span_type)
+                            _logger.error(
+                                "Unexpected end of span at %s: got %s, expected %s",
+                                k,
+                                span_end_codes[char],
+                                span_type,
+                            )
                         if span_type is not None:
                             stack[-1][1].append(span_type(nested))
                     else:
@@ -172,7 +181,9 @@ def parse_paragraph_contents():
 
         paragraphs = []
         while keys:
-            style = text.styles.get(last_linebreak, LwwValue(CrdtId(0, 0), si.ParagraphStyle.PLAIN))
+            style = text.styles.get(
+                last_linebreak, LwwValue(CrdtId(0, 0), si.ParagraphStyle.PLAIN)
+            )
             contents = parse_paragraph_contents()
             p = Paragraph(contents, last_linebreak, style)
             paragraphs += [p]