gh-104825: Remove implicit newline in the line attribute in tokens emitted in the tokenize module

Author: pablogsal

Lib/test/test_tokenize.py

@@ -103,7 +103,7 @@ def k(x):
             e.exception.msg,
             'unindent does not match any outer indentation level')
         self.assertEqual(e.exception.offset, 9)
-        self.assertEqual(e.exception.text, '  x += 5\n')
+        self.assertEqual(e.exception.text, '  x += 5')
 
     def test_int(self):
         # Ordinary integers and binary operators
@@ -1157,7 +1157,7 @@ def readline():
 
         # skip the initial encoding token and the end tokens
         tokens = list(_tokenize(readline(), encoding='utf-8'))[:-2]
-        expected_tokens = [TokenInfo(3, '"ЉЊЈЁЂ"', (1, 0), (1, 7), '"ЉЊЈЁЂ"\n')]
+        expected_tokens = [TokenInfo(3, '"ЉЊЈЁЂ"', (1, 0), (1, 7), '"ЉЊЈЁЂ"')]
         self.assertEqual(tokens, expected_tokens,
                          "bytes not decoded with encoding")
 

Lib/tokenize.py

@@ -518,8 +518,8 @@ def error(message, filename=None, location=None):
             if args.exact:
                 token_type = token.exact_type
             token_range = "%d,%d-%d,%d:" % (token.start + token.end)
-            print("%-20s%-15s%-15r" %
-                  (token_range, tok_name[token_type], token.string))
+            print("%-20s%-15s%-15r%-15r" %
+                  (token_range, tok_name[token_type], token.string, token.line))
     except IndentationError as err:
         line, column = err.args[1][1:3]
         error(err.args[0], filename, (line, column))

Misc/NEWS.d/next/Core and Builtins/2023-05-24-09-59-56.gh-issue-104825.mQesie.rst

@@ -0,0 +1,2 @@
+Tokens emitted by the :mod:`tokenize` module do not include an implicit
+``\n`` character in the ``line`` attribute anymore. Patch by Pablo Galindo

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);