Mirror
/
cpython
mirror of https://github.com/python/cpython
4
1
Fork 0
Code Issues Releases Wiki Activity

bpo-45738: Fix computation of error location for invalid continuation (GH-29550) 

characters in the parser
This commit is contained in:
Pablo Galindo Salgado 2021-11-14 01:06:41 +00:00 committed by GitHub
parent f8da00ef04
commit 25835c518a
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
4 changed files with 14 additions and 12 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
Lib/test/test_syntax.py
View File

@ -1505,7 +1505,13 @@ def func2():
def test_invalid_line_continuation_error_position(self): def test_invalid_line_continuation_error_position(self):
self._check_error(r"a = 3 \ 4", self._check_error(r"a = 3 \ 4",
"unexpected character after line continuation character", "unexpected character after line continuation character",
lineno=1, offset=9) lineno=1, offset=8)
self._check_error('1,\\#\n2',
"unexpected character after line continuation character",
lineno=1, offset=4)
self._check_error('\nfgdfgf\n1,\\#\n2\n',
"unexpected character after line continuation character",
lineno=3, offset=4)
def test_invalid_line_continuation_left_recursive(self): def test_invalid_line_continuation_left_recursive(self):
# Check bpo-42218: SyntaxErrors following left-recursive rules # Check bpo-42218: SyntaxErrors following left-recursive rules

2
Misc/NEWS.d/next/Core and Builtins/2021-11-14-00-14-45.bpo-45738.e0cgKd.rst Normal file
View File

@ -0,0 +1,2 @@
Fix computation of error location for invalid continuation characters in the
parser. Patch by Pablo Galindo.

15
Parser/pegen.c
View File

@ -351,14 +351,7 @@ tokenizer_error(Parser *p)
msg = "too many levels of indentation"; msg = "too many levels of indentation";
break; break;
case E_LINECONT: { case E_LINECONT: {
char* loc = strrchr(p->tok->buf, '\n'); col_offset = p->tok->cur - p->tok->buf - 1;
const char* last_char = p->tok->cur - 1;
if (loc != NULL && loc != last_char) {
col_offset = p->tok->cur - loc - 1;
p->tok->buf = loc;
} else {
col_offset = last_char - p->tok->buf - 1;
}
msg = "unexpected character after line continuation character"; msg = "unexpected character after line continuation character";
break; break;
} }
@ -366,7 +359,9 @@ tokenizer_error(Parser *p)
msg = "unknown parsing error"; msg = "unknown parsing error";
} }
RAISE_ERROR_KNOWN_LOCATION(p, errtype, p->tok->lineno, col_offset, p->tok->lineno, -1, msg); RAISE_ERROR_KNOWN_LOCATION(p, errtype, p->tok->lineno,
col_offset >= 0 ? col_offset : 0,
p->tok->lineno, -1, msg);
return -1; return -1;
} }
@ -497,7 +492,7 @@ _PyPegen_raise_error_known_location(Parser *p, PyObject *errtype,
does not physically exist */ does not physically exist */
assert(p->tok->fp == NULL || p->tok->fp == stdin || p->tok->done == E_EOF || !uses_utf8_codec); assert(p->tok->fp == NULL || p->tok->fp == stdin || p->tok->done == E_EOF || !uses_utf8_codec);
if (p->tok->lineno <= lineno) { if (p->tok->lineno <= lineno && p->tok->inp > p->tok->buf) {
Py_ssize_t size = p->tok->inp - p->tok->buf; Py_ssize_t size = p->tok->inp - p->tok->buf;
error_line = PyUnicode_DecodeUTF8(p->tok->buf, size, "replace"); error_line = PyUnicode_DecodeUTF8(p->tok->buf, size, "replace");
} }

1
Parser/tokenizer.c
View File

@ -1970,7 +1970,6 @@ tok_get(struct tok_state *tok, const char **p_start, const char **p_end)
c = tok_nextc(tok); c = tok_nextc(tok);
if (c != '\n') { if (c != '\n') {
tok->done = E_LINECONT; tok->done = E_LINECONT;
tok->cur = tok->inp;
return ERRORTOKEN; return ERRORTOKEN;
} }
c = tok_nextc(tok); c = tok_nextc(tok);