Merge remote-tracking branch 'upstream/master' into deprecate-nonkeyword-args-replace

Author: MarcoGorelli

.github/workflows/database.yml

@@ -70,7 +70,7 @@ jobs:
     - uses: conda-incubator/setup-miniconda@v2
       with:
         activate-environment: pandas-dev
-        channel-priority: strict
+        channel-priority: flexible
         environment-file: ${{ matrix.ENV_FILE }}
         use-only-tar-bz2: true
 

.pre-commit-config.yaml

@@ -21,6 +21,7 @@ repos:
 -   repo: https://github.com/pre-commit/pre-commit-hooks
     rev: v3.4.0
     hooks:
+    -   id: debug-statements
     -   id: end-of-file-fixer
         exclude: \.txt$
     -   id: trailing-whitespace

asv_bench/benchmarks/strings.py

@@ -11,6 +11,19 @@
 from .pandas_vb_common import tm
 
 
+class Dtypes:
+    params = ["str", "string", "arrow_string"]
+    param_names = ["dtype"]
+
+    def setup(self, dtype):
+        from pandas.core.arrays.string_arrow import ArrowStringDtype  # noqa: F401
+
+        try:
+            self.s = Series(tm.makeStringIndex(10 ** 5), dtype=dtype)
+        except ImportError:
+            raise NotImplementedError
+
+
 class Construction:
 
     params = ["str", "string"]
@@ -49,18 +62,7 @@ def peakmem_cat_frame_construction(self, dtype):
         DataFrame(self.frame_cat_arr, dtype=dtype)
 
 
-class Methods:
-    params = ["str", "string", "arrow_string"]
-    param_names = ["dtype"]
-
-    def setup(self, dtype):
-        from pandas.core.arrays.string_arrow import ArrowStringDtype  # noqa: F401
-
-        try:
-            self.s = Series(tm.makeStringIndex(10 ** 5), dtype=dtype)
-        except ImportError:
-            raise NotImplementedError
-
+class Methods(Dtypes):
     def time_center(self, dtype):
         self.s.str.center(100)
 
@@ -83,6 +85,9 @@ def time_find(self, dtype):
     def time_rfind(self, dtype):
         self.s.str.rfind("[A-Z]+")
 
+    def time_fullmatch(self, dtype):
+        self.s.str.fullmatch("A")
+
     def time_get(self, dtype):
         self.s.str.get(0)
 
@@ -211,35 +216,26 @@ def time_cat(self, other_cols, sep, na_rep, na_frac):
         self.s.str.cat(others=self.others, sep=sep, na_rep=na_rep)
 
 
-class Contains:
+class Contains(Dtypes):
 
-    params = (["str", "string", "arrow_string"], [True, False])
+    params = (Dtypes.params, [True, False])
     param_names = ["dtype", "regex"]
 
     def setup(self, dtype, regex):
-        from pandas.core.arrays.string_arrow import ArrowStringDtype  # noqa: F401
-
-        try:
-            self.s = Series(tm.makeStringIndex(10 ** 5), dtype=dtype)
-        except ImportError:
-            raise NotImplementedError
+        super().setup(dtype)
 
     def time_contains(self, dtype, regex):
         self.s.str.contains("A", regex=regex)
 
 
-class Split:
+class Split(Dtypes):
 
-    params = (["str", "string", "arrow_string"], [True, False])
+    params = (Dtypes.params, [True, False])
     param_names = ["dtype", "expand"]
 
     def setup(self, dtype, expand):
-        from pandas.core.arrays.string_arrow import ArrowStringDtype  # noqa: F401
-
-        try:
-            self.s = Series(tm.makeStringIndex(10 ** 5), dtype=dtype).str.join("--")
-        except ImportError:
-            raise NotImplementedError
+        super().setup(dtype)
+        self.s = self.s.str.join("--")
 
     def time_split(self, dtype, expand):
         self.s.str.split("--", expand=expand)
@@ -248,17 +244,23 @@ def time_rsplit(self, dtype, expand):
         self.s.str.rsplit("--", expand=expand)
 
 
-class Dummies:
-    params = ["str", "string", "arrow_string"]
-    param_names = ["dtype"]
+class Extract(Dtypes):
 
-    def setup(self, dtype):
-        from pandas.core.arrays.string_arrow import ArrowStringDtype  # noqa: F401
+    params = (Dtypes.params, [True, False])
+    param_names = ["dtype", "expand"]
 
-        try:
-            self.s = Series(tm.makeStringIndex(10 ** 5), dtype=dtype).str.join("|")
-        except ImportError:
-            raise NotImplementedError
+    def setup(self, dtype, expand):
+        super().setup(dtype)
+
+    def time_extract_single_group(self, dtype, expand):
+        with warnings.catch_warnings(record=True):
+            self.s.str.extract("(\\w*)A", expand=expand)
+
+
+class Dummies(Dtypes):
+    def setup(self, dtype):
+        super().setup(dtype)
+        self.s = self.s.str.join("|")
 
     def time_get_dummies(self, dtype):
         self.s.str.get_dummies("|")
@@ -279,3 +281,9 @@ def setup(self):
     def time_vector_slice(self):
         # GH 2602
         self.s.str[:5]
+
+
+class Iter(Dtypes):
+    def time_iter(self, dtype):
+        for i in self.s:
+            pass

ci/deps/actions-37-db-min.yaml

@@ -31,7 +31,8 @@ dependencies:
   - openpyxl
   - pandas-gbq
   - google-cloud-bigquery>=1.27.2 # GH 36436
-  - pyarrow=0.17 # GH 38803
+  - protobuf>=3.12.4
+  - pyarrow=0.17.1 # GH 38803
   - pytables>=3.5.1
   - scipy
   - xarray=0.12.3

ci/deps/actions-37-db.yaml

@@ -31,7 +31,7 @@ dependencies:
   - pandas-gbq
   - google-cloud-bigquery>=1.27.2 # GH 36436
   - psycopg2
-  - pyarrow>=0.15.0
+  - pyarrow>=0.17.0
   - pymysql
   - pytables
   - python-snappy

ci/deps/actions-37-minimum_versions.yaml

@@ -23,7 +23,7 @@ dependencies:
   - pytables=3.5.1
   - python-dateutil=2.7.3
   - pytz=2017.3
-  - pyarrow=0.15
+  - pyarrow=0.17.0
   - scipy=1.2
   - xlrd=1.2.0
   - xlsxwriter=1.0.2

ci/deps/actions-37.yaml

@@ -18,7 +18,7 @@ dependencies:
   - numpy=1.19
   - python-dateutil
   - nomkl
-  - pyarrow=0.15.1
+  - pyarrow
   - pytz
   - s3fs>=0.4.0
   - moto>=1.3.14

ci/deps/azure-macos-37.yaml

@@ -1,6 +1,7 @@
 name: pandas-dev
 channels:
   - defaults
+  - conda-forge
 dependencies:
   - python=3.7.*
 
@@ -21,7 +22,7 @@ dependencies:
   - numexpr
   - numpy=1.17.3
   - openpyxl
-  - pyarrow=0.15.1
+  - pyarrow=0.17.0
   - pytables
   - python-dateutil==2.7.3
   - pytz

ci/deps/azure-windows-37.yaml

@@ -26,7 +26,7 @@ dependencies:
   - numexpr
   - numpy=1.17.*
   - openpyxl
-  - pyarrow=0.15
+  - pyarrow=0.17.0
   - pytables
   - python-dateutil
   - pytz

ci/deps/azure-windows-38.yaml

@@ -25,7 +25,7 @@ dependencies:
   - numpy=1.18.*
   - openpyxl
   - jinja2
-  - pyarrow>=0.15.0
+  - pyarrow>=0.17.0
   - pytables
   - python-dateutil
   - pytz