bpo-35766: Merge typed_ast back into CPython (GH-11645)

Author: gvanrossum

Doc/library/ast.rst

@@ -126,16 +126,33 @@ The abstract grammar is currently defined as follows:
 Apart from the node classes, the :mod:`ast` module defines these utility functions
 and classes for traversing abstract syntax trees:
 
-.. function:: parse(source, filename='<unknown>', mode='exec')
+.. function:: parse(source, filename='<unknown>', mode='exec', *, type_comments=False)
 
    Parse the source into an AST node.  Equivalent to ``compile(source,
    filename, mode, ast.PyCF_ONLY_AST)``.
 
+   If ``type_comments=True`` is given, the parser is modified to check
+   and return type comments as specified by :pep:`484` and :pep:`526`.
+   This is equivalent to adding :data:`ast.PyCF_TYPE_COMMENTS` to the
+   flags passed to :func:`compile()`.  This will report syntax errors
+   for misplaced type comments.  Without this flag, type comments will
+   be ignored, and the ``type_comment`` field on selected AST nodes
+   will always be ``None``.  In addition, the locations of ``# type:
+   ignore`` comments will be returned as the ``type_ignores``
+   attribute of :class:`Module` (otherwise it is always an empty list).
+
+   In addition, if ``mode`` is ``'func_type'``, the input syntax is
+   modified to correspond to :pep:`484` "signature type comments",
+   e.g. ``(str, int) -> List[str]``.
+
    .. warning::
       It is possible to crash the Python interpreter with a
       sufficiently large/complex string due to stack depth limitations
       in Python's AST compiler.
 
+   .. versionchanged:: 3.8
+      Added ``type_comments=True`` and ``mode='func_type'``.
+
 
 .. function:: literal_eval(node_or_string)
 

Doc/library/token-list.inc

@@ -69,6 +69,13 @@ the :mod:`tokenize` module.
    always be an ``ENCODING`` token.
 
 
+.. data:: TYPE_COMMENT
+
+   Token value indicating that a type comment was recognized.  Such
+   tokens are only produced when :func:`ast.parse()` is invoked with
+   ``type_comments=True``.
+
+
 .. versionchanged:: 3.5
    Added :data:`AWAIT` and :data:`ASYNC` tokens.
 
@@ -78,3 +85,6 @@ the :mod:`tokenize` module.
 .. versionchanged:: 3.7
    Removed :data:`AWAIT` and :data:`ASYNC` tokens. "async" and "await" are
    now tokenized as :data:`NAME` tokens.
+
+.. versionchanged:: 3.8
+   Added :data:`TYPE_COMMENT`.

Doc/library/token.rst

@@ -7,7 +7,9 @@
 #       single_input is a single interactive statement;
 #       file_input is a module or sequence of commands read from an input file;
 #       eval_input is the input for the eval() functions.
+#       func_type_input is a PEP 484 Python 2 function type comment
 # NB: compound_stmt in single_input is followed by extra NEWLINE!
+# NB: due to the way TYPE_COMMENT is tokenized it will always be followed by a NEWLINE
 single_input: NEWLINE | simple_stmt | compound_stmt NEWLINE
 file_input: (NEWLINE | stmt)* ENDMARKER
 eval_input: testlist NEWLINE* ENDMARKER
@@ -17,14 +19,14 @@ decorators: decorator+
 decorated: decorators (classdef | funcdef | async_funcdef)
 
 async_funcdef: 'async' funcdef
-funcdef: 'def' NAME parameters ['->' test] ':' suite
+funcdef: 'def' NAME parameters ['->' test] ':' [TYPE_COMMENT] func_body_suite
 
 parameters: '(' [typedargslist] ')'
-typedargslist: (tfpdef ['=' test] (',' tfpdef ['=' test])* [',' [
-        '*' [tfpdef] (',' tfpdef ['=' test])* [',' ['**' tfpdef [',']]]
-      | '**' tfpdef [',']]]
-  | '*' [tfpdef] (',' tfpdef ['=' test])* [',' ['**' tfpdef [',']]]
-  | '**' tfpdef [','])
+typedargslist: (tfpdef ['=' test] (',' [TYPE_COMMENT] tfpdef ['=' test])* (TYPE_COMMENT | [',' [TYPE_COMMENT] [
+        '*' [tfpdef] (',' [TYPE_COMMENT] tfpdef ['=' test])* (TYPE_COMMENT | [',' [TYPE_COMMENT] ['**' tfpdef [','] [TYPE_COMMENT]]])
+      | '**' tfpdef [','] [TYPE_COMMENT]]])
+  | '*' [tfpdef] (',' [TYPE_COMMENT] tfpdef ['=' test])* (TYPE_COMMENT | [',' [TYPE_COMMENT] ['**' tfpdef [','] [TYPE_COMMENT]]])
+  | '**' tfpdef [','] [TYPE_COMMENT])
 tfpdef: NAME [':' test]
 varargslist: (vfpdef ['=' test] (',' vfpdef ['=' test])* [',' [
         '*' [vfpdef] (',' vfpdef ['=' test])* [',' ['**' vfpdef [',']]]
@@ -39,7 +41,7 @@ simple_stmt: small_stmt (';' small_stmt)* [';'] NEWLINE
 small_stmt: (expr_stmt | del_stmt | pass_stmt | flow_stmt |
              import_stmt | global_stmt | nonlocal_stmt | assert_stmt)
 expr_stmt: testlist_star_expr (annassign | augassign (yield_expr|testlist) |
-                     ('=' (yield_expr|testlist_star_expr))*)
+                     [('=' (yield_expr|testlist_star_expr))+ [TYPE_COMMENT]] )
 annassign: ':' test ['=' (yield_expr|testlist)]
 testlist_star_expr: (test|star_expr) (',' (test|star_expr))* [',']
 augassign: ('+=' | '-=' | '*=' | '@=' | '/=' | '%=' | '&=' | '|=' | '^=' |
@@ -71,13 +73,13 @@ compound_stmt: if_stmt | while_stmt | for_stmt | try_stmt | with_stmt | funcdef
 async_stmt: 'async' (funcdef | with_stmt | for_stmt)
 if_stmt: 'if' namedexpr_test ':' suite ('elif' namedexpr_test ':' suite)* ['else' ':' suite]
 while_stmt: 'while' test ':' suite ['else' ':' suite]
-for_stmt: 'for' exprlist 'in' testlist ':' suite ['else' ':' suite]
+for_stmt: 'for' exprlist 'in' testlist ':' [TYPE_COMMENT] suite ['else' ':' suite]
 try_stmt: ('try' ':' suite
            ((except_clause ':' suite)+
             ['else' ':' suite]
             ['finally' ':' suite] |
            'finally' ':' suite))
-with_stmt: 'with' with_item (',' with_item)*  ':' suite
+with_stmt: 'with' with_item (',' with_item)*  ':' [TYPE_COMMENT] suite
 with_item: test ['as' expr]
 # NB compile.c makes sure that the default except clause is last
 except_clause: 'except' [test ['as' NAME]]
@@ -150,3 +152,14 @@ encoding_decl: NAME
 
 yield_expr: 'yield' [yield_arg]
 yield_arg: 'from' test | testlist_star_expr
+
+# the TYPE_COMMENT in suites is only parsed for funcdefs,
+# but can't go elsewhere due to ambiguity
+func_body_suite: simple_stmt | NEWLINE [TYPE_COMMENT NEWLINE] INDENT stmt+ DEDENT
+
+func_type_input: func_type NEWLINE* ENDMARKER
+func_type: '(' [typelist] ')' '->' test
+# typelist is a modified typedargslist (see above)
+typelist: (test (',' test)* [','
+       ['*' [test] (',' test)* [',' '**' test] | '**' test]]
+     |  '*' [test] (',' test)* [',' '**' test] | '**' test)

Grammar/Grammar

@@ -55,6 +55,8 @@ ELLIPSIS                '...'
 COLONEQUAL              ':='
 
 OP
+TYPE_IGNORE
+TYPE_COMMENT
 ERRORTOKEN
 
 # These aren't used by the C tokenizer but are needed for tokenize.py

Grammar/Tokens

@@ -22,6 +22,7 @@ PyAPI_FUNC(PyCodeObject *) PyNode_Compile(struct _node *, const char *);
 #define PyCF_DONT_IMPLY_DEDENT 0x0200
 #define PyCF_ONLY_AST 0x0400
 #define PyCF_IGNORE_COOKIE 0x0800
+#define PyCF_TYPE_COMMENTS 0x1000
 
 #ifndef Py_LIMITED_API
 typedef struct {
@@ -85,10 +86,10 @@ PyAPI_FUNC(int) _PyAST_Optimize(struct _mod *, PyArena *arena, int optimize);
 
 #endif /* !Py_LIMITED_API */
 
-/* These definitions must match corresponding definitions in graminit.h.
-   There's code in compile.c that checks that they are the same. */
+/* These definitions must match corresponding definitions in graminit.h. */
 #define Py_single_input 256
 #define Py_file_input 257
 #define Py_eval_input 258
+#define Py_func_type_input 345
 
 #endif /* !Py_COMPILE_H */