Switching from our own Peekable implementation to more-itertools.peekable. (frequenz-floss#1376)

Addresses
frequenz-floss#1325 and
frequenz-floss#1326.

Switching to more-itertools.peekable as suggested in the [original
thread](frequenz-floss#1295 (comment)).

Author: simonvoelcker

pyproject.toml

@@ -38,6 +38,7 @@ dependencies = [
   "typing_extensions >= 4.14.1, < 5",
   "marshmallow >= 3.19.0, < 5",
   "marshmallow_dataclass >= 8.7.1, < 9",
+  "more-itertools >= 10.8.0, < 11",
 ]
 dynamic = ["version"]
 

src/frequenz/sdk/timeseries/formulas/_lexer.py

@@ -8,11 +8,11 @@
 
 from collections.abc import Iterator
 
+from more_itertools import peekable
 from typing_extensions import override
 
 from . import _token
 from ._exceptions import FormulaSyntaxError
-from ._peekable import Peekable
 
 
 class Lexer(Iterator[_token.Token]):
@@ -25,33 +25,40 @@ def __init__(self, formula: str):
             formula: The formula string to lex.
         """
         self._formula: str = formula
-        self._iter: Peekable[tuple[int, str]] = Peekable(enumerate(iter(formula)))
+        self._iter: peekable[tuple[int, str]] = peekable(enumerate(iter(formula)))
+
+    def _peek_char(self) -> tuple[int, str] | None:
+        """Return the next position and character, or None if _iter is at its end."""
+        try:
+            return self._iter.peek()
+        except StopIteration:
+            return None
 
     def _read_integer(self) -> str:
         num_str = ""
-        peek = self._iter.peek()
+        peek = self._peek_char()
         while peek is not None and peek[1].isdigit():
             _, char = next(self._iter)
             num_str += char
-            peek = self._iter.peek()
+            peek = self._peek_char()
         return num_str
 
     def _read_number(self) -> str:
         num_str = ""
-        peek = self._iter.peek()
+        peek = self._peek_char()
         while peek is not None and (peek[1].isdigit() or peek[1] == "."):
             _, char = next(self._iter)
             num_str += char
-            peek = self._iter.peek()
+            peek = self._peek_char()
         return num_str
 
     def _read_symbol(self) -> str:
         word_str = ""
-        peek = self._iter.peek()
+        peek = self._peek_char()
         while peek is not None and peek[1].isalnum():
             _, char = next(self._iter)
             word_str += char
-            peek = self._iter.peek()
+            peek = self._peek_char()
         return word_str
 
     @override
@@ -62,10 +69,10 @@ def __iter__(self) -> Lexer:
     @override
     def __next__(self) -> _token.Token:  # pylint: disable=too-many-branches
         """Return the next token from the formula string."""
-        peek = self._iter.peek()
+        peek = self._peek_char()
         while peek is not None and peek[1].isspace():
             _ = next(self._iter)
-            peek = self._iter.peek()
+            peek = self._peek_char()
 
         if peek is None:
             raise StopIteration

src/frequenz/sdk/timeseries/formulas/_parser.py

@@ -12,6 +12,7 @@
 from frequenz.channels import Receiver
 from frequenz.client.common.microgrid.components import ComponentId
 from frequenz.quantities import Quantity
+from more_itertools import peekable
 
 from frequenz.sdk.timeseries import Sample
 from frequenz.sdk.timeseries._base_types import QuantityT
@@ -22,7 +23,6 @@
 from ._formula import Formula
 from ._functions import FunCall, Function
 from ._lexer import Lexer
-from ._peekable import Peekable
 from ._resampled_stream_fetcher import ResampledStreamFetcher
 
 _logger = logging.getLogger(__name__)
@@ -67,10 +67,17 @@ def __init__(
         """Initialize the parser."""
         self._name: str = name
         self._formula: str = formula
-        self._lexer: Peekable[_token.Token] = Peekable(Lexer(formula))
+        self._lexer: peekable[_token.Token] = peekable(Lexer(formula))
         self._telemetry_fetcher: ResampledStreamFetcher = telemetry_fetcher
         self._create_method: Callable[[float], QuantityT] = create_method
 
+    def _peek_next_token(self) -> _token.Token | None:
+        """Get the next token from the lexer, or None if there is None."""
+        try:
+            return self._lexer.peek()
+        except StopIteration:
+            return None
+
     def _parse_term(self) -> AstNode[QuantityT] | None:
         """Parse a term.
 
@@ -81,7 +88,7 @@ def _parse_term(self) -> AstNode[QuantityT] | None:
         if factor is None:
             return None
 
-        token: _token.Token | None = self._lexer.peek()
+        token: _token.Token | None = self._peek_next_token()
         while token is not None and isinstance(token, (_token.Plus, _token.Minus)):
             token = next(self._lexer)
             next_factor = self._parse_factor()
@@ -98,7 +105,7 @@ def _parse_term(self) -> AstNode[QuantityT] | None:
             elif isinstance(token, _token.Minus):
                 factor = _ast.Sub(left=factor, right=next_factor)
 
-            token = self._lexer.peek()
+            token = self._peek_next_token()
 
         return factor
 
@@ -113,7 +120,7 @@ def _parse_factor(self) -> AstNode[QuantityT] | None:
         if unary is None:
             return None
 
-        token: _token.Token | None = self._lexer.peek()
+        token: _token.Token | None = self._peek_next_token()
         while token is not None and isinstance(token, (_token.Mul, _token.Div)):
             token = next(self._lexer)
             next_unary = self._parse_unary()
@@ -129,7 +136,7 @@ def _parse_factor(self) -> AstNode[QuantityT] | None:
             elif isinstance(token, _token.Div):
                 unary = _ast.Div(left=unary, right=next_unary)
 
-            token = self._lexer.peek()
+            token = self._peek_next_token()
 
         return unary
 
@@ -139,7 +146,7 @@ def _parse_unary(self) -> AstNode[QuantityT] | None:
         A unary is any expression that does not contain any binary
         operators outside of parentheses.
         """
-        token: _token.Token | None = self._lexer.peek()
+        token: _token.Token | None = self._peek_next_token()
         if token is not None and isinstance(token, _token.Minus):
             token = next(self._lexer)
             primary: AstNode[QuantityT] | None = self._parse_primary()
@@ -167,7 +174,7 @@ def _parse_bracketed(self) -> AstNode[QuantityT] | None:
                 message="Expected expression",
             )
 
-        token: _token.Token | None = self._lexer.peek()
+        token: _token.Token | None = self._peek_next_token()
         if token is None or not isinstance(token, _token.CloseParen):
             raise FormulaSyntaxError(
                 formula=self._formula,
@@ -194,7 +201,7 @@ def _parse_function_call(self) -> AstNode[QuantityT] | None:
 
         params: list[AstNode[QuantityT]] = []
 
-        token: _token.Token | None = self._lexer.peek()
+        token: _token.Token | None = self._peek_next_token()
         if token is None or not isinstance(token, _token.OpenParen):
             raise FormulaSyntaxError(
                 formula=self._formula,
@@ -213,7 +220,7 @@ def _parse_function_call(self) -> AstNode[QuantityT] | None:
                 )
             params.append(param)
 
-            token = self._lexer.peek()
+            token = self._peek_next_token()
             if token is None:
                 raise FormulaSyntaxError(
                     formula=self._formula,
@@ -245,7 +252,7 @@ def _parse_primary(self) -> AstNode[QuantityT] | None:
         - A function call
         - A bracketed expression
         """
-        token: _token.Token | None = self._lexer.peek()
+        token: _token.Token | None = self._peek_next_token()
         if token is None:
             return None
 
@@ -286,7 +293,7 @@ def parse(self) -> Formula[QuantityT]:
                 message="Empty formula",
             )
         # There should not be any tokens left
-        token = self._lexer.peek()
+        token = self._peek_next_token()
         if token is not None:
             raise FormulaSyntaxError(
                 formula=self._formula,