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

[pablogsal]

https://github.com/python/cpython/commit/3d2ed8991f9f0f4bbefe4c6f5c8bbbb92259bac6
commit: 3d2ed8991f9f0f4bbefe4c6f5c8bbbb92259bac6
branch: 3.12
author: Pablo Galindo Salgado <Pablogsal at gmail.com>
committer: pablogsal <Pablogsal at gmail.com>
date: 2023-05-24T10:40:51Z
summary:

[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)

files:
A Misc/NEWS.d/next/Core and Builtins/2023-05-24-09-59-56.gh-issue-104825.mQesie.rst
M Lib/test/test_tabnanny.py
M Lib/test/test_tokenize.py
M Python/Python-tokenize.c

diff --git a/Lib/test/test_tabnanny.py b/Lib/test/test_tabnanny.py
index aa700118f735..cc122cafc798 100644
--- a/Lib/test/test_tabnanny.py
+++ b/Lib/test/test_tabnanny.py
@@ -222,7 +222,7 @@ def test_when_nannynag_error_verbose(self):
         """
         with TemporaryPyFile(SOURCE_CODES["nannynag_errored"]) as file_path:
             out = f"{file_path!r}: *** Line 3: trouble in tab city! ***\n"
-            out += "offending line: '\\tprint(\"world\")\\n'\n"
+            out += "offending line: '\\tprint(\"world\")'\n"
             out += "inconsistent use of tabs and spaces in indentation\n"
 
             tabnanny.verbose = 1
@@ -231,7 +231,7 @@ def test_when_nannynag_error_verbose(self):
     def test_when_nannynag_error(self):
         """A python source code file eligible for raising `tabnanny.NannyNag`."""
         with TemporaryPyFile(SOURCE_CODES["nannynag_errored"]) as file_path:
-            out = f"{file_path} 3 '\\tprint(\"world\")\\n'\n"
+            out = f"{file_path} 3 '\\tprint(\"world\")'\n"
             self.verify_tabnanny_check(file_path, out=out)
 
     def test_when_no_file(self):
@@ -341,7 +341,7 @@ def test_verbose_mode(self):
         """Should display more error information if verbose mode is on."""
         with TemporaryPyFile(SOURCE_CODES["nannynag_errored"]) as path:
             stdout = textwrap.dedent(
-                "offending line: '\\tprint(\"world\")\\n'"
+                "offending line: '\\tprint(\"world\")'"
             ).strip()
             self.validate_cmd("-v", path, stdout=stdout, partial=True)
 
@@ -349,6 +349,6 @@ def test_double_verbose_mode(self):
         """Should display detailed error information if double verbose is on."""
         with TemporaryPyFile(SOURCE_CODES["nannynag_errored"]) as path:
             stdout = textwrap.dedent(
-                "offending line: '\\tprint(\"world\")\\n'"
+                "offending line: '\\tprint(\"world\")'"
             ).strip()
             self.validate_cmd("-vv", path, stdout=stdout, partial=True)
diff --git a/Lib/test/test_tokenize.py b/Lib/test/test_tokenize.py
index 8e7ab3d4b7b5..fd9c919ce6a0 100644
--- a/Lib/test/test_tokenize.py
+++ b/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")
 
diff --git a/Misc/NEWS.d/next/Core and Builtins/2023-05-24-09-59-56.gh-issue-104825.mQesie.rst b/Misc/NEWS.d/next/Core and Builtins/2023-05-24-09-59-56.gh-issue-104825.mQesie.rst
new file mode 100644
index 000000000000..caf5d3527085
--- /dev/null
+++ b/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
diff --git a/Python/Python-tokenize.c b/Python/Python-tokenize.c
index f7e32d3af9a9..0023e303b96e 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);