blib2to3 can raise TokenError and IndentationError too (#2671)

[etc/vim.git] / src / black / parsing.py

diff --git a/src/black/parsing.py b/src/black/parsing.py
index b673027022f44b4a2ebd5a9b4b3bd44d0b163b3b..c101643fe11f2e5473de5cc0a39f578804aa4397 100644 (file)
--- a/src/black/parsing.py
+++ b/src/black/parsing.py
@@ -43,6 +43,11 @@ except ImportError:
         ast3 = ast27 = ast
 
 
+PY310_HINT: Final[
+    str
+] = "Consider using --target-version py310 to parse Python 3.10 code."
+
+
 class InvalidInput(ValueError):
     """Raised when input source code fails all parse attempts."""
 
@@ -96,7 +101,8 @@ def lib2to3_parse(src_txt: str, target_versions: Iterable[TargetVersion] = ()) -
     if not src_txt.endswith("\n"):
         src_txt += "\n"
 
-    for grammar in get_grammars(set(target_versions)):
+    grammars = get_grammars(set(target_versions))
+    for grammar in grammars:
         drv = driver.Driver(grammar)
         try:
             result = drv.parse_string(src_txt, True)
@@ -117,6 +123,12 @@ def lib2to3_parse(src_txt: str, target_versions: Iterable[TargetVersion] = ()) -
             exc = InvalidInput(f"Cannot parse: {lineno}:{column}: {te.args[0]}")
 
     else:
+        if pygram.python_grammar_soft_keywords not in grammars and matches_grammar(
+            src_txt, pygram.python_grammar_soft_keywords
+        ):
+            original_msg = exc.args[0]
+            msg = f"{original_msg}\n{PY310_HINT}"
+            raise InvalidInput(msg) from None
         raise exc from None
 
     if isinstance(result, Leaf):
@@ -124,6 +136,16 @@ def lib2to3_parse(src_txt: str, target_versions: Iterable[TargetVersion] = ()) -
     return result
 
 
+def matches_grammar(src_txt: str, grammar: Grammar) -> bool:
+    drv = driver.Driver(grammar)
+    try:
+        drv.parse_string(src_txt, True)
+    except (ParseError, TokenError, IndentationError):
+        return False
+    else:
+        return True
+
+
 def lib2to3_unparse(node: Node) -> str:
     """Given a lib2to3 node, return its string representation."""
     code = str(node)