Backport recent fixes to 5.7.5 release with Python 3.8 and later comp…

…atibility (#340)

* Add CI jobs for Python 3.8, 3.9, and 3.10

* Adjust ruff configuration for Python 3.8

* Tweaks for Python 3.8 compatibility

* [dist] Update classifiers and requires-python for Python 3.8 support

* [dist] prepare 5.7.5 release

.github/workflows/default.yml

@@ -15,6 +15,9 @@ jobs:
         python-version: [
           '3.12',
           '3.11',
+          '3.10',
+          '3.9',
+          '3.8',
         ]
 
     steps:
@@ -33,7 +36,7 @@ jobs:
         python -m pip install --upgrade pip
         pip install --upgrade --requirement requirements-test.txt
     - name: lint
-      #      if: ${{ matrix.python-version == '3.12' }}
+      if: ${{ matrix.python-version == '3.12' }}
       run: |
         make lint
     - name: pytest
@@ -43,6 +46,7 @@ jobs:
       run: |
         make examples
     - name: documentation
+      if: ${{ matrix.python-version == '3.12' }}
       run: |
         make documentation
     - name: distributions

pyproject.toml

@@ -10,7 +10,7 @@ authors = [
 ]
 description = "TatSu takes a grammar in a variation of EBNF as input, and outputs a memoizing PEG/Packrat parser in Python."
 readme = "README.rst"
-requires-python = ">=3.11"
+requires-python = ">=3.8"
 keywords = []
 license = {file = "LICENSE.TXT"}
 classifiers = [
@@ -21,6 +21,9 @@ classifiers = [
     "Intended Audience :: Science/Research",
     "Environment :: Console",
     "Operating System :: OS Independent",
+    "Programming Language :: Python :: 3.8",
+    "Programming Language :: Python :: 3.9",
+    "Programming Language :: Python :: 3.10",
     "Programming Language :: Python :: 3.11",
     "Programming Language :: Python :: 3.12",
     "Topic :: Software Development :: Code Generators",

ruff.toml

@@ -44,7 +44,7 @@ lint.ignore = [
 ]
 exclude = []
 
-target-version = "py312"
+target-version = "py38"
 
 [lint.per-file-ignores]
 

tatsu/_version.py

@@ -1 +1 @@
-__version__ = '5.12.2b1'
+__version__ = '5.7.5'

tatsu/bootstrap.py

@@ -11,6 +11,8 @@
 
 # ruff: noqa: C405, COM812, I001, F401, PLR1702, PLC2801, SIM117
 
+from __future__ import annotations
+
 import sys
 from pathlib import Path
 

tatsu/collections/orderedset.py

@@ -1,14 +1,13 @@
 # NOTE: from https://github.com/LuminosoInsight/ordered-set/blob/master/ordered_set.py
+from __future__ import annotations
+
 import itertools
 from collections.abc import (
-    Iterable,
     Iterator,
     Mapping,
     MutableSequence,
-    MutableSet,
-    Sequence,
 )
-from typing import Any, TypeVar
+from typing import Any, Iterable, MutableSet, Sequence, TypeVar
 
 T = TypeVar('T')
 
@@ -29,7 +28,7 @@ def __getitem__(self, i):
             self._list_cache = list(self._map.keys())
         return self._list_cache[i]
 
-    def copy(self) -> 'OrderedSet[T]':
+    def copy(self) -> OrderedSet[T]:
         return self.__class__(self)
 
     def __getstate__(self):
@@ -78,25 +77,25 @@ def __repr__(self) -> str:
     def __eq__(self, other: Any) -> bool:
         return all(item in other for item in self)
 
-    def union(self, *other: Iterable[T]) -> 'OrderedSet[T]':
+    def union(self, *other: Iterable[T]) -> OrderedSet[T]:
         # do not split `str`
         outer = tuple(
-            [o] if not isinstance(o, set | Mapping | MutableSequence) else o
+            [o] if not isinstance(o, (set, Mapping, MutableSequence)) else o
             for o in other
         )
         inner = itertools.chain([self], *outer)
         items = itertools.chain.from_iterable(inner)
         return type(self)(itertools.chain(items))
 
-    def __and__(self, other: Iterable[Iterable[T]]) -> 'OrderedSet[T]':
+    def __and__(self, other: Iterable[Iterable[T]]) -> OrderedSet[T]:
         return self.intersection(other)
 
-    def intersection(self, *other: Iterable[Iterable[T]]) -> 'OrderedSet[T]':
+    def intersection(self, *other: Iterable[Iterable[T]]) -> OrderedSet[T]:
         common = set.intersection(*other)  # type: ignore[var-annotated, arg-type]
         items = (item for item in self if item in common)
         return type(self)(items)
 
-    def difference(self, *other: Iterable[T]) -> 'OrderedSet[T]':
+    def difference(self, *other: Iterable[T]) -> OrderedSet[T]:
         other = set.union(*other)  # type: ignore[assignment, arg-type]
         items = (item for item in self if item not in other)
         return type(self)(items)
@@ -109,7 +108,7 @@ def issuperset(self, other: set[T]) -> bool:
             return False
         return all(item in self for item in other)
 
-    def symmetric_difference(self, other: set[T]) -> 'OrderedSet[T]':
+    def symmetric_difference(self, other: set[T]) -> OrderedSet[T]:
         cls = type(self)
         diff1 = cls(self).difference(other)
         diff2 = cls(other).difference(self)

tatsu/g2e/semantics.py

@@ -85,12 +85,12 @@ def syntactic_predicate(self, ast):
         return None
 
     def optional(self, ast):
-        if isinstance(ast, model.Group | model.Optional | model.Closure):
+        if isinstance(ast, (model.Group, model.Optional, model.Closure)):
             ast = ast.exp
         return model.Optional(ast)
 
     def closure(self, ast):
-        if isinstance(ast, model.Group | model.Optional):
+        if isinstance(ast, (model.Group, model.Optional)):
             ast = ast.exp
         return model.Closure(ast)
 

tatsu/grammars.py

@@ -865,7 +865,7 @@ def _nullable(self):
 
     @staticmethod
     def param_repr(p):
-        if isinstance(p, int | float) or (isinstance(p, str) and p.isalnum()):
+        if isinstance(p, (int, float)) or (isinstance(p, str) and p.isalnum()):
             return str(p)
         else:
             return repr(p)

tatsu/infos.py

@@ -3,6 +3,7 @@
 import copy
 import dataclasses
 import re
+import sys
 from collections.abc import Callable, Mapping
 from itertools import starmap
 from typing import Any, NamedTuple
@@ -11,6 +12,12 @@
 from .tokenizing import Tokenizer
 from .util.unicode_characters import C_DERIVE
 
+if sys.version_info < (3, 10):
+    import builtins
+
+    def zip(*iterables, strict=False):
+        return builtins.zip(*iterables)
+
 
 class UndefinedStr(str):
     pass
@@ -234,7 +241,7 @@ class RuleResult(NamedTuple):
     newstate: Any
 
 
-@dataclasses.dataclass(slots=True)
+@dataclasses.dataclass(**({'slots': True} if sys.version_info >= (3, 10) else {}))
 class ParseState:
     pos: int = 0
     ast: AST = dataclasses.field(default_factory=AST)

tatsu/mixins/indent.py

@@ -1,3 +1,5 @@
+from __future__ import annotations
+
 import io
 from contextlib import contextmanager
 

tatsu/ngcodegen/objectmodel.py

@@ -1,3 +1,5 @@
+from __future__ import annotations
+
 import builtins
 from collections import namedtuple
 

tatsu/ngcodegen/python.py

@@ -27,6 +27,8 @@
 
     # ruff: noqa: C405, COM812, I001, F401, PLR1702, PLC2801, SIM117
 
+    from __future__ import annotations
+
     import sys
     from pathlib import Path
 
@@ -94,7 +96,7 @@ def walk_Grammar(self, grammar: grammars.Grammar):
 
     def walk_Rule(self, rule: grammars.Rule):
         def param_repr(p):
-            if isinstance(p, int | float):
+            if isinstance(p, (int, float)):
                 return str(p)
             else:
                 return repr(p.split('::')[0])

tatsu/objectmodel.py

@@ -112,7 +112,7 @@ def with_parent(node):
             return node
 
         def children_of(child):
-            if isinstance(child, weakref.ReferenceType | weakref.ProxyType):
+            if isinstance(child, (weakref.ReferenceType, weakref.ProxyType)):
                 return
             elif isinstance(child, Node):
                 yield with_parent(child)
@@ -121,7 +121,7 @@ def children_of(child):
                     if name.startswith('_'):
                         continue
                     yield from children_of(value)
-            elif isinstance(child, list | tuple):
+            elif isinstance(child, (list, tuple)):
                 yield from (
                     with_parent(c) for c in child if isinstance(c, Node)
                 )

tatsu/parser.py

@@ -1,3 +1,5 @@
+from __future__ import annotations
+
 import re
 from typing import Any
 

tatsu/synth.py

@@ -1,3 +1,5 @@
+from __future__ import annotations
+
 from collections.abc import Mapping
 from typing import Any
 

tatsu/util/_common.py

@@ -186,7 +186,7 @@ def decode_match(match):
 
 def isiter(value):
     return isinstance(value, Iterable) and not isinstance(
-        value, str | bytes | bytearray,
+        value, (str, bytes, bytearray),
     )
 
 
@@ -241,19 +241,19 @@ def timestamp():
 
 
 def asjson(obj, seen=None):  # noqa: PLR0911, PLR0912
-    if obj is None or isinstance(obj, int | float | str | bool):
+    if obj is None or isinstance(obj, (int, float, str, bool)):
         return obj
 
     if seen is None:
         seen = set()
     elif id(obj) in seen:
         return f'{type(obj).__name__}@{id(obj)}'
 
-    if isinstance(obj, Mapping | AsJSONMixin) or isiter(obj):
+    if isinstance(obj, (Mapping, AsJSONMixin)) or isiter(obj):
         seen.add(id(obj))
 
     try:
-        if isinstance(obj, weakref.ReferenceType | weakref.ProxyType):
+        if isinstance(obj, (weakref.ReferenceType, weakref.ProxyType)):
             return f'{obj.__class__.__name__}@0x{hex(id(obj)).upper()[2:]}'
         elif hasattr(obj, '__json__'):
             return obj.__json__(seen=seen)
@@ -304,7 +304,7 @@ def plainjson(obj):
             for name, value in obj.items()
             if name not in {'__class__', 'parseinfo'}
         }
-    elif isinstance(obj, weakref.ReferenceType | weakref.ProxyType):
+    elif isinstance(obj, (weakref.ReferenceType, weakref.ProxyType)):
         return '@ref'
     elif isinstance(obj, str) and obj.startswith('@'):
         return '@ref'

tatsu/walkers.py

@@ -25,7 +25,7 @@ def __init__(self):
         )._walker_cache  # pylint: disable=no-member
 
     def walk(self, node: Node | list[Node], *args, **kwargs) -> Any:
-        if isinstance(node, list | tuple):
+        if isinstance(node, (list, tuple)):
             return [self.walk(n, *args, **kwargs) for n in node]
 
         if isinstance(node, Mapping):

test/grammar/semantics_test.py

@@ -100,7 +100,7 @@ def test_builder_basetype_codegen(self):
         self.assertTrue(hasattr(ast, 'a'))
         self.assertTrue(hasattr(ast, 'b'))
 
-        self.assertTrue(issubclass(D, A | B | C))
+        self.assertTrue(issubclass(D, (A, B, C)))
 
     def test_optional_attributes(self):
         grammar = r"""