gh-103384: Generalize the regex pattern `BaseConfigurator.INDEX_PATTERN` to allow...

Co-authored-by: Vinay Sajip <vinay_sajip@yahoo.co.uk>
Co-authored-by: Alex Waygood <Alex.Waygood@Gmail.com>
Co-authored-by: Erlend E. Aasland <erlend@python.org>

diff --git a/Doc/library/logging.config.rst b/Doc/library/logging.config.rst
index 448978f43b6d1305f704444bafc0dafaf7bc8ff4..53fbd073c2670245581b4603075c255ab2f30135 100644 (file)
--- a/Doc/library/logging.config.rst
+++ b/Doc/library/logging.config.rst
@@ -685,7 +685,8 @@ resolve to ``'dev_team@domain.tld'`` and the string
 ``'support_team@domain.tld'``. The ``subject`` value could be accessed
 using either ``'cfg://handlers.email.subject'`` or, equivalently,
 ``'cfg://handlers.email[subject]'``.  The latter form only needs to be
-used if the key contains spaces or non-alphanumeric characters.  If an
+used if the key contains spaces or non-alphanumeric characters. Please note
+that the characters ``[`` and ``]`` are not allowed in the keys. If an
 index value consists only of decimal digits, access will be attempted
 using the corresponding integer value, falling back to the string
 value if needed.

diff --git a/Lib/logging/config.py b/Lib/logging/config.py
index a68281d3e359fd5649cf3b948b35e4f7c31998b0..41283f4d62726704eb576afeca8bd0b3d42ed266 100644 (file)
--- a/Lib/logging/config.py
+++ b/Lib/logging/config.py
@@ -378,7 +378,7 @@ class BaseConfigurator(object):
 
     WORD_PATTERN = re.compile(r'^\s*(\w+)\s*')
     DOT_PATTERN = re.compile(r'^\.\s*(\w+)\s*')
-    INDEX_PATTERN = re.compile(r'^\[\s*(\w+)\s*\]\s*')
+    INDEX_PATTERN = re.compile(r'^\[([^\[\]]*)\]\s*')
     DIGIT_PATTERN = re.compile(r'^\d+$')
 
     value_converters = {

diff --git a/Lib/test/test_logging.py b/Lib/test/test_logging.py
index f26846f9663e5cec9accb561fa5b8e93fa13c8c4..c2e8ff5d463607244b491aa4d6524b37f3113db2 100644 (file)
--- a/Lib/test/test_logging.py
+++ b/Lib/test/test_logging.py
@@ -3662,7 +3662,28 @@ class ConfigDictTest(BaseTest):
         d = {
             'atuple': (1, 2, 3),
             'alist': ['a', 'b', 'c'],
-            'adict': {'d': 'e', 'f': 3 },
+            'adict': {
+                'd': 'e', 'f': 3 ,
+                'alpha numeric 1 with spaces' : 5,
+                'aplha numeric 1 %( - © ©ß¯' : 9,
+                'alpha numeric ] 1 with spaces' : 15,
+                'aplha ]] numeric 1 %( - © ©ß¯]' : 19,
+                ' aplha [ numeric 1 %( - © ©ß¯] ' : 11,
+                ' aplha ' : 32,
+                '' : 10,
+                'nest4' : {
+                    'd': 'e', 'f': 3 ,
+                    'alpha numeric 1 with spaces' : 5,
+                    'aplha numeric 1 %( - © ©ß¯' : 9,
+                    '' : 10,
+                    'somelist' :  ('g', ('h', 'i'), 'j'),
+                    'somedict' : {
+                        'a' : 1,
+                        'a with 1 and space' : 3,
+                        'a with ( and space' : 4,
+                    }
+                }
+            },
             'nest1': ('g', ('h', 'i'), 'j'),
             'nest2': ['k', ['l', 'm'], 'n'],
             'nest3': ['o', 'cfg://alist', 'p'],
@@ -3674,11 +3695,36 @@ class ConfigDictTest(BaseTest):
         self.assertEqual(bc.convert('cfg://nest2[1][1]'), 'm')
         self.assertEqual(bc.convert('cfg://adict.d'), 'e')
         self.assertEqual(bc.convert('cfg://adict[f]'), 3)
+        self.assertEqual(bc.convert('cfg://adict[alpha numeric 1 with spaces]'), 5)
+        self.assertEqual(bc.convert('cfg://adict[aplha numeric 1 %( - © ©ß¯]'), 9)
+        self.assertEqual(bc.convert('cfg://adict[]'), 10)
+        self.assertEqual(bc.convert('cfg://adict.nest4.d'), 'e')
+        self.assertEqual(bc.convert('cfg://adict.nest4[d]'), 'e')
+        self.assertEqual(bc.convert('cfg://adict[nest4].d'), 'e')
+        self.assertEqual(bc.convert('cfg://adict[nest4][f]'), 3)
+        self.assertEqual(bc.convert('cfg://adict[nest4][alpha numeric 1 with spaces]'), 5)
+        self.assertEqual(bc.convert('cfg://adict[nest4][aplha numeric 1 %( - © ©ß¯]'), 9)
+        self.assertEqual(bc.convert('cfg://adict[nest4][]'), 10)
+        self.assertEqual(bc.convert('cfg://adict[nest4][somelist][0]'), 'g')
+        self.assertEqual(bc.convert('cfg://adict[nest4][somelist][1][0]'), 'h')
+        self.assertEqual(bc.convert('cfg://adict[nest4][somelist][1][1]'), 'i')
+        self.assertEqual(bc.convert('cfg://adict[nest4][somelist][2]'), 'j')
+        self.assertEqual(bc.convert('cfg://adict[nest4].somedict.a'), 1)
+        self.assertEqual(bc.convert('cfg://adict[nest4].somedict[a]'), 1)
+        self.assertEqual(bc.convert('cfg://adict[nest4].somedict[a with 1 and space]'), 3)
+        self.assertEqual(bc.convert('cfg://adict[nest4].somedict[a with ( and space]'), 4)
+        self.assertEqual(bc.convert('cfg://adict.nest4.somelist[1][1]'), 'i')
+        self.assertEqual(bc.convert('cfg://adict.nest4.somelist[2]'), 'j')
+        self.assertEqual(bc.convert('cfg://adict.nest4.somedict.a'), 1)
+        self.assertEqual(bc.convert('cfg://adict.nest4.somedict[a]'), 1)
         v = bc.convert('cfg://nest3')
         self.assertEqual(v.pop(1), ['a', 'b', 'c'])
         self.assertRaises(KeyError, bc.convert, 'cfg://nosuch')
         self.assertRaises(ValueError, bc.convert, 'cfg://!')
         self.assertRaises(KeyError, bc.convert, 'cfg://adict[2]')
+        self.assertRaises(KeyError, bc.convert, 'cfg://adict[alpha numeric ] 1 with spaces]')
+        self.assertRaises(ValueError, bc.convert, 'cfg://adict[ aplha ]] numeric 1 %( - © ©ß¯] ]')
+        self.assertRaises(ValueError, bc.convert, 'cfg://adict[ aplha [ numeric 1 %( - © ©ß¯] ]')
 
     def test_namedtuple(self):
         # see bpo-39142

diff --git a/Misc/NEWS.d/next/Library/2023-04-09-05-30-41.gh-issue-103384.zAV7iB.rst b/Misc/NEWS.d/next/Library/2023-04-09-05-30-41.gh-issue-103384.zAV7iB.rst
new file mode 100644 (file)
index 0000000..3e9d3cf
--- /dev/null
+++ b/Misc/NEWS.d/next/Library/2023-04-09-05-30-41.gh-issue-103384.zAV7iB.rst
@@ -0,0 +1 @@
+Generalize the regex pattern ``BaseConfigurator.INDEX_PATTERN`` to allow spaces and non-alphanumeric characters in keys.