From 005fc4b5e0fea000d25dd217d6845271db21986f Mon Sep 17 00:00:00 2001 From: Pablo Galindo Salgado Date: Tue, 5 Jul 2022 17:39:21 +0100 Subject: [PATCH] [3.10] gh-94360: Fix a tokenizer crash when reading encoded files with syntax errors from stdin (GH-94386) MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit * gh-94360: Fix a tokenizer crash when reading encoded files with syntax errors from stdin Signed-off-by: Pablo Galindo * nitty nit Co-authored-by: Łukasz Langa . (cherry picked from commit 36fcde61ba48c4e918830691ecf4092e4e3b9b99) Co-authored-by: Pablo Galindo Salgado --- .../2022-06-28-14-20-36.gh-issue-94360.DiEnen.rst | 2 ++ Parser/pegen.c | 8 ++++---- Parser/tokenizer.c | 10 +++++++++- 3 files changed, 15 insertions(+), 5 deletions(-) create mode 100644 Misc/NEWS.d/next/Core and Builtins/2022-06-28-14-20-36.gh-issue-94360.DiEnen.rst diff --git a/Misc/NEWS.d/next/Core and Builtins/2022-06-28-14-20-36.gh-issue-94360.DiEnen.rst b/Misc/NEWS.d/next/Core and Builtins/2022-06-28-14-20-36.gh-issue-94360.DiEnen.rst new file mode 100644 index 00000000000000..0a74ba38b0ac43 --- /dev/null +++ b/Misc/NEWS.d/next/Core and Builtins/2022-06-28-14-20-36.gh-issue-94360.DiEnen.rst @@ -0,0 +1,2 @@ +Fixed a tokenizer crash when reading encoded files with syntax errors from +``stdin`` with non utf-8 encoded text. Patch by Pablo Galindo diff --git a/Parser/pegen.c b/Parser/pegen.c index c04824315c7a36..414059d12363cf 100644 --- a/Parser/pegen.c +++ b/Parser/pegen.c @@ -446,12 +446,12 @@ get_error_line(Parser *p, Py_ssize_t lineno) Py_ssize_t relative_lineno = p->starting_lineno ? lineno - p->starting_lineno + 1 : lineno; for (int i = 0; i < relative_lineno - 1; i++) { - char *new_line = strchr(cur_line, '\n') + 1; - assert(new_line != NULL && new_line <= buf_end); - if (new_line == NULL || new_line > buf_end) { + char *new_line = strchr(cur_line, '\n'); + assert(new_line != NULL && new_line + 1 < buf_end); + if (new_line == NULL || new_line + 1 > buf_end) { break; } - cur_line = new_line; + cur_line = new_line + 1; } char *next_newline; diff --git a/Parser/tokenizer.c b/Parser/tokenizer.c index 267ccec33fd770..579474c43354ff 100644 --- a/Parser/tokenizer.c +++ b/Parser/tokenizer.c @@ -328,6 +328,10 @@ tok_concatenate_interactive_new_line(struct tok_state *tok, const char *line) { Py_ssize_t current_size = tok->interactive_src_end - tok->interactive_src_start; Py_ssize_t line_size = strlen(line); + char last_char = line[line_size > 0 ? line_size - 1 : line_size]; + if (last_char != '\n') { + line_size += 1; + } char* new_str = tok->interactive_src_start; new_str = PyMem_Realloc(new_str, current_size + line_size + 1); @@ -341,7 +345,11 @@ tok_concatenate_interactive_new_line(struct tok_state *tok, const char *line) { return -1; } strcpy(new_str + current_size, line); - + if (last_char != '\n') { + /* Last line does not end in \n, fake one */ + new_str[current_size + line_size - 1] = '\n'; + new_str[current_size + line_size] = '\0'; + } tok->interactive_src_start = new_str; tok->interactive_src_end = new_str + current_size + line_size; return 0;