[3.12] gh-104825: Remove implicit newline in the line attribute in tokens emitted in the tokenize module (GH-104846). (#104850)

(cherry picked from commit c8cf9b42eb2bfbd4c3e708ec28d32430248a1d7a)

1 files changed, 4 insertions, 0 deletions

diff --git a/Python/Python-tokenize.c b/Python/Python-tokenize.c
index f7e32d3..0023e30 100644
--- a/Python/Python-tokenize.c
+++ b/Python/Python-tokenize.c
@@ -123,6 +123,8 @@ _tokenizer_error(struct tok_state *tok)
     int result = 0;
 
     Py_ssize_t size = tok->inp - tok->buf;
+    assert(tok->buf[size-1] == '\n');
+    size -= 1; // Remove the newline character from the end of the line
     error_line = PyUnicode_DecodeUTF8(tok->buf, size, "replace");
     if (!error_line) {
         result = -1;
@@ -193,6 +195,8 @@ tokenizeriter_next(tokenizeriterobject *it)
     }
 
     Py_ssize_t size = it->tok->inp - it->tok->buf;
+    assert(it->tok->buf[size-1] == '\n');
+    size -= 1; // Remove the newline character from the end of the line
     PyObject *line = PyUnicode_DecodeUTF8(it->tok->buf, size, "replace");
     if (line == NULL) {
         Py_DECREF(str);