[Python-checkins] gh-93283: Improve error message for f-string with invalid conversion character (GH-93349)

[serhiy-storchaka]

https://github.com/python/cpython/commit/07df8d5b2c715c9aa770c47986d59e272bf049f0
commit: 07df8d5b2c715c9aa770c47986d59e272bf049f0
branch: main
author: Serhiy Storchaka <storchaka at gmail.com>
committer: serhiy-storchaka <storchaka at gmail.com>
date: 2022-05-31T20:38:29+03:00
summary:

gh-93283: Improve error message for f-string with invalid conversion character (GH-93349)

files:
A Misc/NEWS.d/next/Core and Builtins/2022-05-30-14-50-03.gh-issue-93283.XDO2ZQ.rst
M Lib/test/test_fstring.py
M Parser/string_parser.c

diff --git a/Lib/test/test_fstring.py b/Lib/test/test_fstring.py
index 0c3372f033551..e8bf420d699ed 100644
--- a/Lib/test/test_fstring.py
+++ b/Lib/test/test_fstring.py
@@ -590,7 +590,9 @@ def test_format_specifier_expressions(self):
         self.assertEqual(f'{-10:{"-"}#{1}0{"x"}}', '      -0xa')
         self.assertEqual(f'{10:#{3 != {4:5} and width}x}', '       0xa')
 
-        self.assertAllRaise(SyntaxError, "f-string: expecting '}'",
+        self.assertAllRaise(SyntaxError,
+                            """f-string: invalid conversion character 'r{"': """
+                            """expected 's', 'r', or 'a'""",
                             ["""f'{"s"!r{":10"}}'""",
 
                              # This looks like a nested format spec.
@@ -1012,19 +1014,28 @@ def test_conversions(self):
         # Not a conversion, but show that ! is allowed in a format spec.
         self.assertEqual(f'{3.14:!<10.10}', '3.14!!!!!!')
 
-        self.assertAllRaise(SyntaxError, 'f-string: invalid conversion character',
-                            ["f'{3!g}'",
-                             "f'{3!A}'",
-                             "f'{3!3}'",
-                             "f'{3!G}'",
-                             "f'{3!!}'",
+        self.assertAllRaise(SyntaxError, "f-string: expecting '}'",
+                            ["f'{3!'",
+                             "f'{3!s'",
+                             "f'{3!g'",
+                             ])
+
+        self.assertAllRaise(SyntaxError, 'f-string: missed conversion character',
+                            ["f'{3!}'",
+                             "f'{3!:'",
                              "f'{3!:}'",
-                             "f'{3! s}'",  # no space before conversion char
                              ])
 
-        self.assertAllRaise(SyntaxError, "f-string: expecting '}'",
-                            ["f'{x!s{y}}'",
-                             "f'{3!ss}'",
+        for conv in 'g', 'A', '3', 'G', '!', ' s', 's ', ' s ', 'ä', 'ɐ', 'ª':
+            self.assertAllRaise(SyntaxError,
+                                "f-string: invalid conversion character %r: "
+                                "expected 's', 'r', or 'a'" % conv,
+                                ["f'{3!" + conv + "}'"])
+
+        self.assertAllRaise(SyntaxError,
+                            "f-string: invalid conversion character 'ss': "
+                            "expected 's', 'r', or 'a'",
+                            ["f'{3!ss}'",
                              "f'{3!ss:}'",
                              "f'{3!ss:s}'",
                              ])
diff --git a/Misc/NEWS.d/next/Core and Builtins/2022-05-30-14-50-03.gh-issue-93283.XDO2ZQ.rst b/Misc/NEWS.d/next/Core and Builtins/2022-05-30-14-50-03.gh-issue-93283.XDO2ZQ.rst
new file mode 100644
index 0000000000000..550e86988b25a
--- /dev/null
+++ b/Misc/NEWS.d/next/Core and Builtins/2022-05-30-14-50-03.gh-issue-93283.XDO2ZQ.rst	
@@ -0,0 +1,2 @@
+Improve error message for invalid syntax of conversion character in f-string
+expressions.
diff --git a/Parser/string_parser.c b/Parser/string_parser.c
index 9c12d8ca101d0..c56ed20ad4cfc 100644
--- a/Parser/string_parser.c
+++ b/Parser/string_parser.c
@@ -767,27 +767,43 @@ fstring_find_expr(Parser *p, const char **str, const char *end, int raw, int rec
     /* Check for a conversion char, if present. */
     if (**str == '!') {
         *str += 1;
-        if (*str >= end) {
-            goto unexpected_end_of_string;
+        const char *conv_start = *str;
+        while (1) {
+            if (*str >= end) {
+                goto unexpected_end_of_string;
+            }
+            if (**str == '}' || **str == ':') {
+                break;
+            }
+            *str += 1;
+        }
+        if (*str == conv_start) {
+            RAISE_SYNTAX_ERROR(
+                      "f-string: missed conversion character");
+            goto error;
         }
 
-        conversion = (unsigned char)**str;
-        *str += 1;
-
+        conversion = (unsigned char)*conv_start;
         /* Validate the conversion. */
-        if (!(conversion == 's' || conversion == 'r' || conversion == 'a')) {
-            RAISE_SYNTAX_ERROR(
-                      "f-string: invalid conversion character: "
-                      "expected 's', 'r', or 'a'");
+        if ((*str != conv_start + 1) ||
+            !(conversion == 's' || conversion == 'r' || conversion == 'a'))
+        {
+            PyObject *conv_obj = PyUnicode_FromStringAndSize(conv_start,
+                                                             *str-conv_start);
+            if (conv_obj) {
+                RAISE_SYNTAX_ERROR(
+                        "f-string: invalid conversion character %R: "
+                        "expected 's', 'r', or 'a'",
+                        conv_obj);
+                Py_DECREF(conv_obj);
+            }
             goto error;
         }
 
     }
 
     /* Check for the format spec, if present. */
-    if (*str >= end) {
-        goto unexpected_end_of_string;
-    }
+    assert(*str < end);
     if (**str == ':') {
         *str += 1;
         if (*str >= end) {