SJ-PT-23-100: Fix inconsistencies in error messages between C and Python implementations

4 files changed, 12 insertions, 7 deletions

diff --git a/CHANGES.txt b/CHANGES.txt
index b70200f..270b22e 100644
--- a/CHANGES.txt
+++ b/CHANGES.txt
@@ -6,6 +6,8 @@ Version 3.19.0 released 2023-04-XX
   an exception in Python 2.x; was probably unreachable (SJ-PT-23-02)
 * Backport the integer string length limitation from Python 3.11 to
   limit quadratic number parsing (SJ-PT-23-03)
+* Fix inconsistencies with error messages between the C and Python
+  implementations (SJ-PT-23-100)
 
 Version 3.18.4 released 2023-03-14
 
diff --git a/simplejson/_speedups.c b/simplejson/_speedups.c
index 43c589b..e08f4c4 100644
--- a/simplejson/_speedups.c
+++ b/simplejson/_speedups.c
@@ -1886,7 +1886,7 @@ _match_number_str(PyScannerObject *s, PyObject *pystr, Py_ssize_t start, Py_ssiz
     /* read a sign if it's there, make sure it's not the end of the string */
     if (str[idx] == '-') {
         if (idx >= end_idx) {
-            raise_errmsg(ERR_EXPECTING_VALUE, pystr, idx);
+            raise_errmsg(ERR_EXPECTING_VALUE, pystr, start);
             return NULL;
         }
         idx++;
@@ -1903,7 +1903,7 @@ _match_number_str(PyScannerObject *s, PyObject *pystr, Py_ssize_t start, Py_ssiz
     }
     /* no integer digits, error */
     else {
-        raise_errmsg(ERR_EXPECTING_VALUE, pystr, idx);
+        raise_errmsg(ERR_EXPECTING_VALUE, pystr, start);
         return NULL;
     }
 
@@ -1995,7 +1995,7 @@ _match_number_unicode(PyScannerObject *s, PyObject *pystr, Py_ssize_t start, Py_
     /* read a sign if it's there, make sure it's not the end of the string */
     if (PyUnicode_READ(kind, str, idx) == '-') {
         if (idx >= end_idx) {
-            raise_errmsg(ERR_EXPECTING_VALUE, pystr, idx);
+            raise_errmsg(ERR_EXPECTING_VALUE, pystr, start);
             return NULL;
         }
         idx++;
@@ -2015,7 +2015,7 @@ _match_number_unicode(PyScannerObject *s, PyObject *pystr, Py_ssize_t start, Py_
     }
     else {
         /* no integer digits, error */
-        raise_errmsg(ERR_EXPECTING_VALUE, pystr, idx);
+        raise_errmsg(ERR_EXPECTING_VALUE, pystr, start);
         return NULL;
     }
 
diff --git a/simplejson/decoder.py b/simplejson/decoder.py
index c28e445..bbde82c 100644
--- a/simplejson/decoder.py
+++ b/simplejson/decoder.py
@@ -93,6 +93,7 @@ def py_scanstring(s, end, encoding=None, strict=True,
         if chunk is None:
             raise JSONDecodeError(
                 "Unterminated string starting at", s, begin)
+        prev_end = end
         end = chunk.end()
         content, terminator = chunk.groups()
         # Content is contains zero or more unescaped string characters
@@ -107,7 +108,7 @@ def py_scanstring(s, end, encoding=None, strict=True,
         elif terminator != '\\':
             if strict:
                 msg = "Invalid control character %r at"
-                raise JSONDecodeError(msg, s, end)
+                raise JSONDecodeError(msg, s, prev_end)
             else:
                 _append(terminator)
                 continue
@@ -178,7 +179,7 @@ def JSONObject(state, encoding, strict, scan_once, object_hook,
             return pairs, end + 1
         elif nextchar != '"':
             raise JSONDecodeError(
-                "Expecting property name enclosed in double quotes",
+                "Expecting property name enclosed in double quotes or '}'",
                 s, end)
     end += 1
     while True:
diff --git a/simplejson/tests/test_fail.py b/simplejson/tests/test_fail.py
index 788f3a5..5f9a8f6 100644
--- a/simplejson/tests/test_fail.py
+++ b/simplejson/tests/test_fail.py
@@ -145,7 +145,7 @@ class TestFail(TestCase):
             ('["spam', 'Unterminated string starting at', 1),
             ('["spam"', "Expecting ',' delimiter", 7),
             ('["spam",', 'Expecting value', 8),
-            ('{', 'Expecting property name enclosed in double quotes', 1),
+            ('{', "Expecting property name enclosed in double quotes or '}'", 1),
             ('{"', 'Unterminated string starting at', 1),
             ('{"spam', 'Unterminated string starting at', 1),
             ('{"spam"', "Expecting ':' delimiter", 7),
@@ -156,6 +156,8 @@ class TestFail(TestCase):
             ('"', 'Unterminated string starting at', 0),
             ('"spam', 'Unterminated string starting at', 0),
             ('[,', "Expecting value", 1),
+            ('--', 'Expecting value', 0),
+            ('"\x18d', "Invalid control character %r", 1),
         ]
         for data, msg, idx in test_cases:
             try: