From 24c10d2943c482c4d3ecc71d45df2d8c10fa5bb1 Mon Sep 17 00:00:00 2001 From: Pablo Galindo Salgado Date: Wed, 24 Nov 2021 22:21:23 +0000 Subject: [PATCH] bpo-45727: Only trigger the 'did you forgot a comma' error suggestion if inside parentheses (GH-29757) --- Grammar/python.gram | 3 ++- Lib/test/test_exceptions.py | 4 ++-- Lib/test/test_fstring.py | 2 +- .../2021-11-24-18-24-49.bpo-45727._xVbbo.rst | 3 +++ Parser/parser.c | 2 +- Parser/pegen.c | 4 +++- Parser/pegen.h | 1 + Parser/pegen_errors.c | 4 ++-- 8 files changed, 15 insertions(+), 8 deletions(-) create mode 100644 Misc/NEWS.d/next/Core and Builtins/2021-11-24-18-24-49.bpo-45727._xVbbo.rst diff --git a/Grammar/python.gram b/Grammar/python.gram index 60fbc400cf2..2c696a6a085 100644 --- a/Grammar/python.gram +++ b/Grammar/python.gram @@ -1084,7 +1084,8 @@ invalid_expression: # !(NAME STRING) is not matched so we don't show this error with some invalid string prefixes like: kf"dsfsdf" # Soft keywords need to also be ignored because they can be parsed as NAME NAME | !(NAME STRING | SOFT_KEYWORD) a=disjunction b=expression_without_invalid { - _PyPegen_check_legacy_stmt(p, a) ? NULL : RAISE_SYNTAX_ERROR_KNOWN_RANGE(a, b, "invalid syntax. Perhaps you forgot a comma?") } + _PyPegen_check_legacy_stmt(p, a) ? NULL : p->tokens[p->mark-1]->level == 0 ? NULL : + RAISE_SYNTAX_ERROR_KNOWN_RANGE(a, b, "invalid syntax. Perhaps you forgot a comma?") } | a=disjunction 'if' b=disjunction !('else'|':') { RAISE_SYNTAX_ERROR_KNOWN_RANGE(a, b, "expected 'else' after 'if' expression") } invalid_named_expression: diff --git a/Lib/test/test_exceptions.py b/Lib/test/test_exceptions.py index eee178c8ca2..7f087d085a8 100644 --- a/Lib/test/test_exceptions.py +++ b/Lib/test/test_exceptions.py @@ -226,14 +226,14 @@ class ExceptionTests(unittest.TestCase): check(b'Python = "\xcf\xb3\xf2\xee\xed" +', 1, 18) check('x = "a', 1, 5) check('lambda x: x = 2', 1, 1) - check('f{a + b + c}', 1, 1) + check('f{a + b + c}', 1, 2) check('[file for str(file) in []\n])', 1, 11) check('a = « hello » « world »', 1, 5) check('[\nfile\nfor str(file)\nin\n[]\n]', 3, 5) check('[file for\n str(file) in []]', 2, 2) check("ages = {'Alice'=22, 'Bob'=23}", 1, 16) check('match ...:\n case {**rest, "key": value}:\n ...', 2, 19) - check("a b c d e f", 1, 1) + check("[a b c d e f]", 1, 2) # Errors thrown by compile.c check('class foo:return 1', 1, 11) diff --git a/Lib/test/test_fstring.py b/Lib/test/test_fstring.py index 14a4c678fc9..bd1ca943c7c 100644 --- a/Lib/test/test_fstring.py +++ b/Lib/test/test_fstring.py @@ -944,7 +944,7 @@ x = ( "Bf''", "BF''",] double_quote_cases = [case.replace("'", '"') for case in single_quote_cases] - self.assertAllRaise(SyntaxError, 'unexpected EOF while parsing', + self.assertAllRaise(SyntaxError, 'invalid syntax', single_quote_cases + double_quote_cases) def test_leading_trailing_spaces(self): diff --git a/Misc/NEWS.d/next/Core and Builtins/2021-11-24-18-24-49.bpo-45727._xVbbo.rst b/Misc/NEWS.d/next/Core and Builtins/2021-11-24-18-24-49.bpo-45727._xVbbo.rst new file mode 100644 index 00000000000..d4b149ddccf --- /dev/null +++ b/Misc/NEWS.d/next/Core and Builtins/2021-11-24-18-24-49.bpo-45727._xVbbo.rst @@ -0,0 +1,3 @@ +Refine the custom syntax error that suggests that a comma may be missing to +trigger only when the expressions are detected between parentheses or +brackets. Patch by Pablo Galindo diff --git a/Parser/parser.c b/Parser/parser.c index b3aa35989ed..1cf6e356ad4 100644 --- a/Parser/parser.c +++ b/Parser/parser.c @@ -18298,7 +18298,7 @@ invalid_expression_rule(Parser *p) ) { D(fprintf(stderr, "%*c+ invalid_expression[%d-%d]: %s succeeded!\n", p->level, ' ', _mark, p->mark, "!(NAME STRING | SOFT_KEYWORD) disjunction expression_without_invalid")); - _res = _PyPegen_check_legacy_stmt ( p , a ) ? NULL : RAISE_SYNTAX_ERROR_KNOWN_RANGE ( a , b , "invalid syntax. Perhaps you forgot a comma?" ); + _res = _PyPegen_check_legacy_stmt ( p , a ) ? NULL : p -> tokens [p -> mark - 1] -> level == 0 ? NULL : RAISE_SYNTAX_ERROR_KNOWN_RANGE ( a , b , "invalid syntax. Perhaps you forgot a comma?" ); if (_res == NULL && PyErr_Occurred()) { p->error_indicator = 1; D(p->level--); diff --git a/Parser/pegen.c b/Parser/pegen.c index 4f51c63c443..ede281ac89c 100644 --- a/Parser/pegen.c +++ b/Parser/pegen.c @@ -170,6 +170,8 @@ initialize_token(Parser *p, Token *token, const char *start, const char *end, in return -1; } + token->level = p->tok->level; + const char *line_start = token_type == STRING ? p->tok->multi_line_start : p->tok->line_start; int lineno = token_type == STRING ? p->tok->first_lineno : p->tok->lineno; int end_lineno = p->tok->lineno; @@ -946,4 +948,4 @@ _PyPegen_run_parser_from_string(const char *str, int start_rule, PyObject *filen error: _PyTokenizer_Free(tok); return result; -} \ No newline at end of file +} diff --git a/Parser/pegen.h b/Parser/pegen.h index e5e712ab26b..78e75d7060c 100644 --- a/Parser/pegen.h +++ b/Parser/pegen.h @@ -35,6 +35,7 @@ typedef struct _memo { typedef struct { int type; PyObject *bytes; + int level; int lineno, col_offset, end_lineno, end_col_offset; Memo *memo; } Token; diff --git a/Parser/pegen_errors.c b/Parser/pegen_errors.c index 694184a03b0..93057d151db 100644 --- a/Parser/pegen_errors.c +++ b/Parser/pegen_errors.c @@ -399,7 +399,7 @@ _Pypegen_set_syntax_error(Parser* p, Token* last_token) { RAISE_SYNTAX_ERROR("error at start before reading any input"); } // Parser encountered EOF (End of File) unexpectedtly - if (p->tok->done == E_EOF) { + if (last_token->type == ERRORTOKEN && p->tok->done == E_EOF) { if (p->tok->level) { raise_unclosed_parentheses_error(p); } else { @@ -422,4 +422,4 @@ _Pypegen_set_syntax_error(Parser* p, Token* last_token) { // _PyPegen_tokenize_full_source_to_check_for_errors will override the existing // generic SyntaxError we just raised if errors are found. _PyPegen_tokenize_full_source_to_check_for_errors(p); -} \ No newline at end of file +}