From 52d0df5f989903d46e6de4878e1d6a0e87a2c803 Mon Sep 17 00:00:00 2001 From: Eugenio Lacuesta Date: Wed, 12 May 2021 13:08:08 -0300 Subject: [PATCH 1/5] CaseInsensitiveDict (deprecate CaselessDict) --- scrapy/http/headers.py | 13 ++++--- scrapy/pipelines/files.py | 4 +- scrapy/utils/datatypes.py | 46 +++++++++++++++++++++++ tests/test_utils_datatypes.py | 71 +++++++++++++++++++++++++---------- 4 files changed, 107 insertions(+), 27 deletions(-) diff --git a/scrapy/http/headers.py b/scrapy/http/headers.py index 1a2b99b0a..dfbcf8361 100644 --- a/scrapy/http/headers.py +++ b/scrapy/http/headers.py @@ -1,5 +1,6 @@ from w3lib.http import headers_dict_to_raw -from scrapy.utils.datatypes import CaselessDict + +from scrapy.utils.datatypes import CaseInsensitiveDict, CaselessDict from scrapy.utils.python import to_unicode @@ -76,13 +77,13 @@ class Headers(CaselessDict): return headers_dict_to_raw(self) def to_unicode_dict(self): - """ Return headers as a CaselessDict with unicode keys + """ Return headers as a CaseInsensitiveDict with unicode keys and unicode values. Multiple values are joined with ','. """ - return CaselessDict( - (to_unicode(key, encoding=self.encoding), - to_unicode(b','.join(value), encoding=self.encoding)) - for key, value in self.items()) + return CaseInsensitiveDict( + (to_unicode(key, encoding=self.encoding), to_unicode(b','.join(value), encoding=self.encoding)) + for key, value in self.items() + ) def __copy__(self): return self.__class__(self) diff --git a/scrapy/pipelines/files.py b/scrapy/pipelines/files.py index 13ecd4e6c..2f1a25dfc 100644 --- a/scrapy/pipelines/files.py +++ b/scrapy/pipelines/files.py @@ -23,7 +23,7 @@ from scrapy.http import Request from scrapy.pipelines.media import MediaPipeline from scrapy.settings import Settings from scrapy.utils.boto import is_botocore_available -from scrapy.utils.datatypes import CaselessDict +from scrapy.utils.datatypes import CaseInsensitiveDict from scrapy.utils.ftp import ftp_store_file from scrapy.utils.log import failure_to_exc_info from scrapy.utils.misc import md5sum @@ -143,7 +143,7 @@ class S3FilesStore: """ Convert headers to botocore keyword agruments. """ # This is required while we need to support both boto and botocore. - mapping = CaselessDict({ + mapping = CaseInsensitiveDict({ 'Content-Type': 'ContentType', 'Cache-Control': 'CacheControl', 'Content-Disposition': 'ContentDisposition', diff --git a/scrapy/utils/datatypes.py b/scrapy/utils/datatypes.py index e31284a7f..ca6089e0f 100644 --- a/scrapy/utils/datatypes.py +++ b/scrapy/utils/datatypes.py @@ -6,14 +6,30 @@ This module must not depend on any module outside the Standard Library. """ import collections +import warnings import weakref from collections.abc import Mapping +from typing import Any, AnyStr + +from scrapy.exceptions import ScrapyDeprecationWarning class CaselessDict(dict): __slots__ = () + def __new__(cls, *args, **kwargs): + from scrapy.http.headers import Headers + + if issubclass(cls, CaselessDict) and not issubclass(cls, Headers): + warnings.warn( + "scrapy.utils.datatypes.CaselessDict is deprecated," + " please use scrapy.utils.datatypes.CaseInsensitiveDict instead", + category=ScrapyDeprecationWarning, + stacklevel=2, + ) + return super().__new__(cls, *args, **kwargs) + def __init__(self, seq=None): super().__init__() if seq: @@ -63,6 +79,36 @@ class CaselessDict(dict): return dict.pop(self, self.normkey(key), *args) +class CaseInsensitiveDict(collections.UserDict): + """A dict-like structure that accepts strings or bytes as keys and allows case-insensitive lookups. + + It also allows overriding key and value normalization by defining custom `normkey` and `normvalue` methods. + """ + + def __getitem__(self, key: AnyStr) -> Any: + return super().__getitem__(self.normkey(key)) + + def __setitem__(self, key: AnyStr, value: Any) -> None: + super().__setitem__(self.normkey(key), self.normvalue(value)) + + def __delitem__(self, key: AnyStr) -> None: + super().__delitem__(self.normkey(key)) + + def __contains__(self, key: AnyStr) -> bool: # type: ignore[override] + return super().__contains__(self.normkey(key)) + + def normkey(self, key: AnyStr) -> AnyStr: + """Method to normalize dictionary key access""" + return key.lower() + + def normvalue(self, value: Any) -> Any: + """Method to normalize values prior to be set""" + return value + + def __repr__(self) -> str: + return f"<{self.__class__.__name__}: {super().__repr__()}>" + + class LocalCache(collections.OrderedDict): """Dictionary with a finite number of keys. diff --git a/tests/test_utils_datatypes.py b/tests/test_utils_datatypes.py index e4bccf30e..c033cd537 100644 --- a/tests/test_utils_datatypes.py +++ b/tests/test_utils_datatypes.py @@ -1,26 +1,34 @@ import copy import unittest +import warnings from collections.abc import Mapping, MutableMapping +from scrapy.exceptions import ScrapyDeprecationWarning from scrapy.http import Request -from scrapy.utils.datatypes import CaselessDict, LocalCache, LocalWeakReferencedCache, SequenceExclude +from scrapy.utils.datatypes import ( + CaseInsensitiveDict, + CaselessDict, + LocalCache, + LocalWeakReferencedCache, + SequenceExclude, +) from scrapy.utils.python import garbage_collect __doctests__ = ['scrapy.utils.datatypes'] -class CaselessDictTest(unittest.TestCase): +class CaseInsensitiveDictMixin: def test_init_dict(self): seq = {'red': 1, 'black': 3} - d = CaselessDict(seq) + d = self.dict_class(seq) self.assertEqual(d['red'], 1) self.assertEqual(d['black'], 3) def test_init_pair_sequence(self): seq = (('red', 1), ('black', 3)) - d = CaselessDict(seq) + d = self.dict_class(seq) self.assertEqual(d['red'], 1) self.assertEqual(d['black'], 3) @@ -39,7 +47,7 @@ class CaselessDictTest(unittest.TestCase): return len(self._d) seq = MyMapping(red=1, black=3) - d = CaselessDict(seq) + d = self.dict_class(seq) self.assertEqual(d['red'], 1) self.assertEqual(d['black'], 3) @@ -64,12 +72,12 @@ class CaselessDictTest(unittest.TestCase): return len(self._d) seq = MyMutableMapping(red=1, black=3) - d = CaselessDict(seq) + d = self.dict_class(seq) self.assertEqual(d['red'], 1) self.assertEqual(d['black'], 3) def test_caseless(self): - d = CaselessDict() + d = self.dict_class() d['key_Lower'] = 1 self.assertEqual(d['KEy_loWer'], 1) self.assertEqual(d.get('KEy_loWer'), 1) @@ -79,19 +87,19 @@ class CaselessDictTest(unittest.TestCase): self.assertEqual(d.get('key_Lower'), 3) def test_delete(self): - d = CaselessDict({'key_lower': 1}) + d = self.dict_class({'key_lower': 1}) del d['key_LOWER'] self.assertRaises(KeyError, d.__getitem__, 'key_LOWER') self.assertRaises(KeyError, d.__getitem__, 'key_lower') def test_getdefault(self): - d = CaselessDict() + d = self.dict_class() self.assertEqual(d.get('c', 5), 5) d['c'] = 10 self.assertEqual(d.get('c', 5), 10) def test_setdefault(self): - d = CaselessDict({'a': 1, 'b': 2}) + d = self.dict_class({'a': 1, 'b': 2}) r = d.setdefault('A', 5) self.assertEqual(r, 1) @@ -104,15 +112,15 @@ class CaselessDictTest(unittest.TestCase): def test_fromkeys(self): keys = ('a', 'b') - d = CaselessDict.fromkeys(keys) + d = self.dict_class.fromkeys(keys) self.assertEqual(d['A'], None) self.assertEqual(d['B'], None) - d = CaselessDict.fromkeys(keys, 1) + d = self.dict_class.fromkeys(keys, 1) self.assertEqual(d['A'], 1) self.assertEqual(d['B'], 1) - instance = CaselessDict() + instance = self.dict_class() d = instance.fromkeys(keys) self.assertEqual(d['A'], None) self.assertEqual(d['B'], None) @@ -122,18 +130,19 @@ class CaselessDictTest(unittest.TestCase): self.assertEqual(d['B'], 1) def test_contains(self): - d = CaselessDict() + d = self.dict_class() d['a'] = 1 assert 'a' in d + assert 'A' in d def test_pop(self): - d = CaselessDict() + d = self.dict_class() d['a'] = 1 self.assertEqual(d.pop('A'), 1) self.assertRaises(KeyError, d.pop, 'A') def test_normkey(self): - class MyDict(CaselessDict): + class MyDict(self.dict_class): def normkey(self, key): return key.title() @@ -142,7 +151,7 @@ class CaselessDictTest(unittest.TestCase): self.assertEqual(list(d.keys()), ['Key-One']) def test_normvalue(self): - class MyDict(CaselessDict): + class MyDict(self.dict_class): def normvalue(self, value): if value is not None: return value + 1 @@ -171,11 +180,35 @@ class CaselessDictTest(unittest.TestCase): self.assertEqual(d.get('key'), 2) def test_copy(self): - h1 = CaselessDict({'header1': 'value'}) + h1 = self.dict_class({'header1': 'value'}) h2 = copy.copy(h1) self.assertEqual(h1, h2) self.assertEqual(h1.get('header1'), h2.get('header1')) - assert isinstance(h2, CaselessDict) + assert isinstance(h2, self.dict_class) + + +class CaseInsensitiveDictTest(CaseInsensitiveDictMixin, unittest.TestCase): + dict_class = CaseInsensitiveDict + + def test_repr(self): + d = self.dict_class({"foo": "bar"}) + self.assertEqual(repr(d), "") + + +class CaselessDictTest(CaseInsensitiveDictMixin, unittest.TestCase): + dict_class = CaselessDict + + def test_deprecation_message(self): + with warnings.catch_warnings(record=True) as caught: + self.dict_class({"foo": "bar"}) + + self.assertEqual(len(caught), 1) + self.assertTrue(issubclass(caught[0].category, ScrapyDeprecationWarning)) + self.assertEqual( + "scrapy.utils.datatypes.CaselessDict is deprecated," + " please use scrapy.utils.datatypes.CaseInsensitiveDict instead", + str(caught[0].message), + ) class SequenceExcludeTest(unittest.TestCase): From bbeed6ae8fd9aed3651b104e4cc3e56495e1b1b9 Mon Sep 17 00:00:00 2001 From: Eugenio Lacuesta Date: Thu, 19 Aug 2021 14:09:30 -0300 Subject: [PATCH 2/5] CaseInsensitiveDict: preserve original keys (only lookups are key-insensitive) --- scrapy/utils/datatypes.py | 36 ++++++++++++++++++++++------------- tests/test_utils_datatypes.py | 13 +++++++++++-- 2 files changed, 34 insertions(+), 15 deletions(-) diff --git a/scrapy/utils/datatypes.py b/scrapy/utils/datatypes.py index ca6089e0f..1d56811f0 100644 --- a/scrapy/utils/datatypes.py +++ b/scrapy/utils/datatypes.py @@ -79,35 +79,45 @@ class CaselessDict(dict): return dict.pop(self, self.normkey(key), *args) -class CaseInsensitiveDict(collections.UserDict): +class CaseInsensitiveDict(collections.UserDict,): """A dict-like structure that accepts strings or bytes as keys and allows case-insensitive lookups. It also allows overriding key and value normalization by defining custom `normkey` and `normvalue` methods. """ + def __init__(self, *args, **kwargs) -> None: + self._keys: dict = {} + super().__init__(*args, **kwargs) + def __getitem__(self, key: AnyStr) -> Any: - return super().__getitem__(self.normkey(key)) + normalized_key = self.normkey(key) + return super().__getitem__(self._keys[normalized_key.lower()]) def __setitem__(self, key: AnyStr, value: Any) -> None: - super().__setitem__(self.normkey(key), self.normvalue(value)) + normalized_key = self.normkey(key) + if normalized_key.lower() in self._keys: + del self[self._keys[normalized_key.lower()]] + super().__setitem__(normalized_key, self.normvalue(value)) + self._keys[normalized_key.lower()] = normalized_key def __delitem__(self, key: AnyStr) -> None: - super().__delitem__(self.normkey(key)) + normalized_key = self.normkey(key) + stored_key = self._keys.pop(normalized_key.lower()) + super().__delitem__(stored_key) def __contains__(self, key: AnyStr) -> bool: # type: ignore[override] - return super().__contains__(self.normkey(key)) - - def normkey(self, key: AnyStr) -> AnyStr: - """Method to normalize dictionary key access""" - return key.lower() - - def normvalue(self, value: Any) -> Any: - """Method to normalize values prior to be set""" - return value + normalized_key = self.normkey(key) + return normalized_key.lower() in self._keys def __repr__(self) -> str: return f"<{self.__class__.__name__}: {super().__repr__()}>" + def normkey(self, key: AnyStr) -> AnyStr: + return key + + def normvalue(self, value: Any) -> Any: + return value + class LocalCache(collections.OrderedDict): """Dictionary with a finite number of keys. diff --git a/tests/test_utils_datatypes.py b/tests/test_utils_datatypes.py index c033cd537..5faaabe81 100644 --- a/tests/test_utils_datatypes.py +++ b/tests/test_utils_datatypes.py @@ -1,4 +1,5 @@ import copy +from typing import Iterator import unittest import warnings from collections.abc import Mapping, MutableMapping @@ -191,8 +192,16 @@ class CaseInsensitiveDictTest(CaseInsensitiveDictMixin, unittest.TestCase): dict_class = CaseInsensitiveDict def test_repr(self): - d = self.dict_class({"foo": "bar"}) - self.assertEqual(repr(d), "") + d1 = self.dict_class({"foo": "bar"}) + self.assertEqual(repr(d1), "") + d2 = self.dict_class({"AsDf": "QwErTy", "FoO": "bAr"}) + self.assertEqual(repr(d2), "") + + def test_iter(self): + d = self.dict_class({"AsDf": "QwErTy", "FoO": "bAr"}) + iterkeys = iter(d) + self.assertIsInstance(iterkeys, Iterator) + self.assertEqual(list(iterkeys), ["AsDf", "FoO"]) class CaselessDictTest(CaseInsensitiveDictMixin, unittest.TestCase): From 10ebf6384ed58253c237224d523e602b1f3c2224 Mon Sep 17 00:00:00 2001 From: Eugenio Lacuesta Date: Thu, 19 Aug 2021 14:12:55 -0300 Subject: [PATCH 3/5] Remove unnecessary comma --- scrapy/utils/datatypes.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/scrapy/utils/datatypes.py b/scrapy/utils/datatypes.py index 1d56811f0..6eeabe1ee 100644 --- a/scrapy/utils/datatypes.py +++ b/scrapy/utils/datatypes.py @@ -79,7 +79,7 @@ class CaselessDict(dict): return dict.pop(self, self.normkey(key), *args) -class CaseInsensitiveDict(collections.UserDict,): +class CaseInsensitiveDict(collections.UserDict): """A dict-like structure that accepts strings or bytes as keys and allows case-insensitive lookups. It also allows overriding key and value normalization by defining custom `normkey` and `normvalue` methods. From 1c031b8a8dd719e6011ee29889bc8181cdbc9a9b Mon Sep 17 00:00:00 2001 From: Eugenio Lacuesta Date: Thu, 12 May 2022 13:10:08 -0300 Subject: [PATCH 4/5] Underscore CaseInsensitiveDict normkey/normvalue --- scrapy/utils/datatypes.py | 19 +++++++++---------- tests/test_utils_datatypes.py | 10 +++++++--- 2 files changed, 16 insertions(+), 13 deletions(-) diff --git a/scrapy/utils/datatypes.py b/scrapy/utils/datatypes.py index f45e1c9b8..807a95504 100644 --- a/scrapy/utils/datatypes.py +++ b/scrapy/utils/datatypes.py @@ -80,9 +80,8 @@ class CaselessDict(dict): class CaseInsensitiveDict(collections.UserDict): - """A dict-like structure that accepts strings or bytes as keys and allows case-insensitive lookups. - - It also allows overriding key and value normalization by defining custom `normkey` and `normvalue` methods. + """A dict-like structure that accepts strings or bytes + as keys and allows case-insensitive lookups. """ def __init__(self, *args, **kwargs) -> None: @@ -90,32 +89,32 @@ class CaseInsensitiveDict(collections.UserDict): super().__init__(*args, **kwargs) def __getitem__(self, key: AnyStr) -> Any: - normalized_key = self.normkey(key) + normalized_key = self._normkey(key) return super().__getitem__(self._keys[normalized_key.lower()]) def __setitem__(self, key: AnyStr, value: Any) -> None: - normalized_key = self.normkey(key) + normalized_key = self._normkey(key) if normalized_key.lower() in self._keys: del self[self._keys[normalized_key.lower()]] - super().__setitem__(normalized_key, self.normvalue(value)) + super().__setitem__(normalized_key, self._normvalue(value)) self._keys[normalized_key.lower()] = normalized_key def __delitem__(self, key: AnyStr) -> None: - normalized_key = self.normkey(key) + normalized_key = self._normkey(key) stored_key = self._keys.pop(normalized_key.lower()) super().__delitem__(stored_key) def __contains__(self, key: AnyStr) -> bool: # type: ignore[override] - normalized_key = self.normkey(key) + normalized_key = self._normkey(key) return normalized_key.lower() in self._keys def __repr__(self) -> str: return f"<{self.__class__.__name__}: {super().__repr__()}>" - def normkey(self, key: AnyStr) -> AnyStr: + def _normkey(self, key: AnyStr) -> AnyStr: return key - def normvalue(self, value: Any) -> Any: + def _normvalue(self, value: Any) -> Any: return value diff --git a/tests/test_utils_datatypes.py b/tests/test_utils_datatypes.py index 5faaabe81..0a724f237 100644 --- a/tests/test_utils_datatypes.py +++ b/tests/test_utils_datatypes.py @@ -1,8 +1,8 @@ import copy -from typing import Iterator import unittest import warnings from collections.abc import Mapping, MutableMapping +from typing import Iterator from scrapy.exceptions import ScrapyDeprecationWarning from scrapy.http import Request @@ -144,19 +144,23 @@ class CaseInsensitiveDictMixin: def test_normkey(self): class MyDict(self.dict_class): - def normkey(self, key): + def _normkey(self, key): return key.title() + normkey = _normkey # deprecated CaselessDict class + d = MyDict() d['key-one'] = 2 self.assertEqual(list(d.keys()), ['Key-One']) def test_normvalue(self): class MyDict(self.dict_class): - def normvalue(self, value): + def _normvalue(self, value): if value is not None: return value + 1 + normvalue = _normvalue # deprecated CaselessDict class + d = MyDict({'key': 1}) self.assertEqual(d['key'], 2) self.assertEqual(d.get('key'), 2) From 2c65066ad9e293630da2c594af06ad483abe800d Mon Sep 17 00:00:00 2001 From: Eugenio Lacuesta Date: Fri, 27 May 2022 19:56:42 -0300 Subject: [PATCH 5/5] Avoid exceptions on copy --- scrapy/utils/datatypes.py | 7 +++++-- tests/test_utils_datatypes.py | 8 +++++++- 2 files changed, 12 insertions(+), 3 deletions(-) diff --git a/scrapy/utils/datatypes.py b/scrapy/utils/datatypes.py index 807a95504..fd5ac3b08 100644 --- a/scrapy/utils/datatypes.py +++ b/scrapy/utils/datatypes.py @@ -94,8 +94,11 @@ class CaseInsensitiveDict(collections.UserDict): def __setitem__(self, key: AnyStr, value: Any) -> None: normalized_key = self._normkey(key) - if normalized_key.lower() in self._keys: - del self[self._keys[normalized_key.lower()]] + try: + lower_key = self._keys[normalized_key.lower()] + del self[lower_key] + except KeyError: + pass super().__setitem__(normalized_key, self._normvalue(value)) self._keys[normalized_key.lower()] = normalized_key diff --git a/tests/test_utils_datatypes.py b/tests/test_utils_datatypes.py index 0a724f237..36df9006f 100644 --- a/tests/test_utils_datatypes.py +++ b/tests/test_utils_datatypes.py @@ -187,9 +187,15 @@ class CaseInsensitiveDictMixin: def test_copy(self): h1 = self.dict_class({'header1': 'value'}) h2 = copy.copy(h1) + assert isinstance(h2, self.dict_class) self.assertEqual(h1, h2) self.assertEqual(h1.get('header1'), h2.get('header1')) - assert isinstance(h2, self.dict_class) + self.assertEqual(h1.get('header1'), h2.get('HEADER1')) + h3 = h1.copy() + assert isinstance(h3, self.dict_class) + self.assertEqual(h1, h3) + self.assertEqual(h1.get('header1'), h3.get('header1')) + self.assertEqual(h1.get('header1'), h3.get('HEADER1')) class CaseInsensitiveDictTest(CaseInsensitiveDictMixin, unittest.TestCase):