fix(types): address several type issues, add type check job

lark/lark.py

@@ -1,6 +1,7 @@
 from abc import ABC, abstractmethod
 import sys, os, pickle, hashlib
 import tempfile
+import types
 from typing import (
     TypeVar, Type, List, Dict, Iterator, Callable, Union, Optional, Sequence,
     Tuple, Iterable, IO, Any, TYPE_CHECKING, Collection
@@ -27,9 +28,10 @@ from .grammar import Rule
 
 import re
 try:
-    import regex  # type: ignore
+    import regex
+    _has_regex = True
 except ImportError:
-    regex = None
+    _has_regex = False
 
 
 ###{standalone
@@ -253,11 +255,12 @@ class Lark(Serialize):
 
     def __init__(self, grammar: 'Union[Grammar, str, IO[str]]', **options) -> None:
         self.options = LarkOptions(options)
+        re_module: types.ModuleType
 
         # Set regex or re module
         use_regex = self.options.regex
         if use_regex:
-            if regex:
+            if _has_regex:
                 re_module = regex
             else:
                 raise ImportError('`regex` module must be installed if calling `Lark(regex=True)`.')
@@ -267,7 +270,7 @@ class Lark(Serialize):
         # Some, but not all file-like objects have a 'name' attribute
         if self.options.source_path is None:
             try:
-                self.source_path = grammar.name
+                self.source_path = grammar.name  # type: ignore[union-attr]
             except AttributeError:
                 self.source_path = '<string>'
         else:
@@ -275,7 +278,7 @@ class Lark(Serialize):
 
         # Drain file-like objects to get their contents
         try:
-            read = grammar.read
+            read = grammar.read  # type: ignore[union-attr]
         except AttributeError:
             pass
         else:

lark/tree.py

@@ -5,6 +5,7 @@ from typing import List, Callable, Iterator, Union, Optional, Generic, TypeVar,
 
 if TYPE_CHECKING:
     from .lexer import TerminalDef, Token
+    import rich
     if sys.version_info >= (3, 8):
         from typing import Literal
     else:

lark/utils.py

@@ -109,9 +109,10 @@ class SerializeMemoizer(Serialize):
 
 
 try:
-    import regex  # type: ignore
+    import regex
+    _has_regex = True
 except ImportError:
-    regex = None
+    _has_regex = False
 
 if sys.version_info >= (3, 11):
     import re._parser as sre_parse
@@ -123,7 +124,7 @@ else:
 categ_pattern = re.compile(r'\\p{[A-Za-z_]+}')
 
 def get_regexp_width(expr):
-    if regex:
+    if _has_regex:
         # Since `sre_parse` cannot deal with Unicode categories of the form `\p{Mn}`, we replace these with
         # a simple letter, which makes no difference as we are only trying to get the possible lengths of the regex
         # match here below.
@@ -135,7 +136,7 @@ def get_regexp_width(expr):
     try:
         return [int(x) for x in sre_parse.parse(regexp_final).getwidth()]
     except sre_constants.error:
-        if not regex:
+        if not _has_regex:
             raise ValueError(expr)
         else:
             # sre_parse does not support the new features in regex. To not completely fail in that case,
@@ -223,15 +224,16 @@ def combine_alternatives(lists):
 
 try:
     import atomicwrites
+    _has_atomicwrites = True
 except ImportError:
-    atomicwrites = None  # type: ignore[assigment]
+    _has_atomicwrites = False
 
 class FS:
     exists = staticmethod(os.path.exists)
 
     @staticmethod
     def open(name, mode="r", **kwargs):
-        if atomicwrites and "w" in mode:
+        if _has_atomicwrites and "w" in mode:
             return atomicwrites.atomic_write(name, mode=mode, overwrite=True, **kwargs)
         else:
             return open(name, mode, **kwargs)

pyproject.toml

@@ -0,0 +1,18 @@
+[build-system]
+requires = ["setuptools>=42"]
+build-backend = "setuptools.build_meta"
+
+
+[tool.mypy]
+files = "lark"
+python_version = "3.6"
+show_error_codes = true
+enable_error_code = ["ignore-without-code"]
+exclude = [
+  "^lark/__pyinstaller",
+]
+
+# You can disable imports or control per-module/file settings here
+[[tool.mypy.overrides]]
+module = [ "js2py" ]
+ignore_missing_imports = true

tox.ini

@@ -1,16 +1,30 @@
 [tox]
-envlist = py36, py37, py38, py39, pypy, pypy3
-skip_missing_interpreters=true
+envlist = py36, py37, py38, py39, py310, pypy3, type
+skip_missing_interpreters = true
 
 [testenv]
 whitelist_externals = git
 deps =
     -rtest-requirements.txt
+passenv =
+    TERM
 
 # to always force recreation and avoid unexpected side effects
-recreate=True
+recreate = True
 
-commands=
+commands =
     git submodule sync -q
     git submodule update --init
     python -m tests {posargs}
+
+[testenv:type]
+description = run type check on code base
+skip_install = true
+recreate = false
+deps =
+    mypy==0.950
+    types-atomicwrites
+    types-regex
+    rich
+commands =
+    mypy