TMP: some fixes

numpy · bsipocz · May 25, 2023 · May 25, 2023 · May 26, 2023 · May 26, 2023
commit a4b478925e96436312932c24cc9dd07c77d3001a
diff --git a/numpy/__init__.py b/numpy/__init__.py
@@ -29,8 +29,7 @@
 
 Use the built-in ``help`` function to view a function's docstring::
 
-  >>> help(np.sort)
-  ... # doctest: +SKIP
+  >>> help(np.sort)  # doctest: +IGNORE_OUTPUT
 
 For some objects, ``np.info(obj)`` may provide additional help.  This is
 particularly true if you see the line "Help on ufunc object:" at the top
@@ -169,7 +168,7 @@
         where, zeros, zeros_like
     )
 
-    # NOTE: It's still under discussion whether these aliases 
+    # NOTE: It's still under discussion whether these aliases
     # should be removed.
     for ta in ["float96", "float128", "complex192", "complex256"]:
         try:
@@ -184,20 +183,20 @@
         histogram, histogram_bin_edges, histogramdd
     )
     from .lib._nanfunctions_impl import (
-        nanargmax, nanargmin, nancumprod, nancumsum, nanmax, nanmean, 
+        nanargmax, nanargmin, nancumprod, nancumsum, nanmax, nanmean,
         nanmedian, nanmin, nanpercentile, nanprod, nanquantile, nanstd,
         nansum, nanvar
     )
     from .lib._function_base_impl import (
-        select, piecewise, trim_zeros, copy, iterable, percentile, diff, 
+        select, piecewise, trim_zeros, copy, iterable, percentile, diff,
         gradient, angle, unwrap, sort_complex, flip, rot90, extract, place,
         vectorize, asarray_chkfinite, average, bincount, digitize, cov,
         corrcoef, median, sinc, hamming, hanning, bartlett, blackman,
         kaiser, trapz, i0, meshgrid, delete, insert, append, interp, quantile
     )
     from .lib._twodim_base_impl import (
-        diag, diagflat, eye, fliplr, flipud, tri, triu, tril, vander, 
-        histogram2d, mask_indices, tril_indices, tril_indices_from, 
+        diag, diagflat, eye, fliplr, flipud, tri, triu, tril, vander,
+        histogram2d, mask_indices, tril_indices, tril_indices_from,
         triu_indices, triu_indices_from
     )
     from .lib._shape_base_impl import (
@@ -206,7 +205,7 @@
         take_along_axis, tile, vsplit
     )
     from .lib._type_check_impl import (
-        iscomplexobj, isrealobj, imag, iscomplex, isreal, nan_to_num, real, 
+        iscomplexobj, isrealobj, imag, iscomplex, isreal, nan_to_num, real,
         real_if_close, typename, mintypecode, common_type
     )
     from .lib._arraysetops_impl import (
@@ -231,20 +230,21 @@
     )
     from .lib._index_tricks_impl import (
         diag_indices_from, diag_indices, fill_diagonal, ndindex, ndenumerate,
-        ix_, c_, r_, s_, ogrid, mgrid, unravel_index, ravel_multi_index, 
+        ix_, c_, r_, s_, ogrid, mgrid, unravel_index, ravel_multi_index,
         index_exp
     )
     from . import matrixlib as _mat
     from .matrixlib import (
         asmatrix, bmat, matrix
     )
 
+
     # public submodules are imported lazily, therefore are accessible from
     # __getattr__. Note that `distutils` (deprecated) and `array_api`
     # (experimental label) are not added here, because `from numpy import *`
     # must not raise any warnings - that's too disruptive.
     __numpy_submodules__ = {
-        "linalg", "fft", "dtypes", "random", "polynomial", "ma", 
+        "linalg", "fft", "dtypes", "random", "polynomial", "ma",
         "exceptions", "lib", "ctypeslib", "testing", "typing",
         "f2py", "test", "rec", "char", "core"
     }
@@ -388,7 +388,7 @@ def __getattr__(attr):
 
         if attr in __former_attrs__:
             raise AttributeError(__former_attrs__[attr])
-        
+
         if attr in __expired_attributes__:
             raise AttributeError(
                 f"`np.{attr}` was removed in the NumPy 2.0 release. "
@@ -411,7 +411,7 @@ def __dir__():
             globals().keys() | __numpy_submodules__
         )
         public_symbols -= {
-            "matrixlib", "matlib", "tests", "conftest", "version", 
+            "matrixlib", "matlib", "tests", "conftest", "version",
             "compat", "distutils", "array_api"
         }
         return list(public_symbols)
@@ -480,7 +480,7 @@ def _mac_os_check():
     def hugepage_setup():
         """
         We usually use madvise hugepages support, but on some old kernels it
-        is slow and thus better avoided. Specifically kernel version 4.6 
+        is slow and thus better avoided. Specifically kernel version 4.6
         had a bug fix which probably fixed this:
         https://github.com/torvalds/linux/commit/7cf91a98e607c2f935dbcc177d70011e95b8faff
         """
@@ -489,7 +489,7 @@ def hugepage_setup():
             # If there is an issue with parsing the kernel version,
             # set use_hugepage to 0. Usage of LooseVersion will handle
             # the kernel version parsing better, but avoided since it
-            # will increase the import time. 
+            # will increase the import time.
             # See: #16679 for related discussion.
             try:
                 use_hugepage = 1

diff --git a/numpy/_core/arrayprint.py b/numpy/_core/arrayprint.py
@@ -61,6 +61,7 @@
     # str/False on the way in/out.
     'legacy': sys.maxsize}
 
+
 def _make_options_dict(precision=None, threshold=None, edgeitems=None,
                        linewidth=None, suppress=None, nanstr=None, infstr=None,
                        sign=None, formatter=None, floatmode=None, legacy=None):
@@ -241,7 +242,7 @@ def set_printoptions(precision=None, threshold=None, edgeitems=None,
 
     >>> np.set_printoptions(precision=4)
     >>> np.array([1.123456789])
-    [1.1235]
+    array([1.1235])
 
     Long arrays can be summarised:
 

diff --git a/numpy/_core/fromnumeric.py b/numpy/_core/fromnumeric.py
@@ -775,7 +775,7 @@ def partition(a, kth, axis=-1, kind='introselect', order=None):
     >>> a = np.array([7, 1, 7, 7, 1, 5, 7, 2, 3, 2, 6, 2, 3, 0])
     >>> p = np.partition(a, 4)
     >>> p
-    array([0, 1, 2, 1, 2, 5, 2, 3, 3, 6, 7, 7, 7, 7]) # may vary
+    array([0, 1, 2, 1, 2, 5, 2, 3, 3, 6, 7, 7, 7, 7])
 
     ``p[4]`` is 2;  all elements in ``p[:4]`` are less than or equal
     to ``p[4]``, and all elements in ``p[5:]`` are greater than or
@@ -873,13 +873,13 @@ def argpartition(a, kth, axis=-1, kind='introselect', order=None):
 
     >>> x = np.array([3, 4, 2, 1])
     >>> x[np.argpartition(x, 3)]
-    array([2, 1, 3, 4]) # may vary
+    array([2, 1, 3, 4])
     >>> x[np.argpartition(x, (1, 3))]
-    array([1, 2, 3, 4]) # may vary
+    array([1, 2, 3, 4])
 
     >>> x = [3, 4, 2, 1]
     >>> np.array(x)[np.argpartition(x, 3)]
-    array([2, 1, 3, 4]) # may vary
+    array([2, 1, 3, 4])
 
     Multi-dimensional array:
 
@@ -1034,14 +1034,14 @@ def sort(a, axis=-1, kind=None, order=None):
     >>> values = [('Arthur', 1.8, 41), ('Lancelot', 1.9, 38),
     ...           ('Galahad', 1.7, 38)]
     >>> a = np.array(values, dtype=dtype)       # create a structured array
-    >>> np.sort(a, order='height')                        # doctest: +SKIP
+    >>> np.sort(a, order='height')                        # doctest: +IGNORE_OUTPUT
     array([('Galahad', 1.7, 38), ('Arthur', 1.8, 41),
            ('Lancelot', 1.8999999999999999, 38)],
           dtype=[('name', '|S10'), ('height', '<f8'), ('age', '<i4')])
 
     Sort by age, then height if ages are equal:
 
-    >>> np.sort(a, order=['age', 'height'])               # doctest: +SKIP
+    >>> np.sort(a, order=['age', 'height'])               # doctest: +IGNORE_OUTPUT
     array([('Galahad', 1.7, 38), ('Lancelot', 1.8999999999999999, 38),
            ('Arthur', 1.8, 41)],
           dtype=[('name', '|S10'), ('height', '<f8'), ('age', '<i4')])
@@ -1585,8 +1585,8 @@ def squeeze(a, axis=None):
     >>> x = np.array([[1234]])
     >>> x.shape
     (1, 1)
-    >>> np.squeeze(x)
-    array(1234)  # 0d array
+    >>> np.squeeze(x)  # results a 0d array
+    array(1234)
     >>> np.squeeze(x).shape
     ()
     >>> np.squeeze(x)[()]
@@ -2547,7 +2547,7 @@ def all(a, axis=None, out=None, keepdims=np._NoValue, *, where=np._NoValue):
     >>> o=np.array(False)
     >>> z=np.all([-1, 4, 5], out=o)
     >>> id(z), id(o), z
-    (28293632, 28293632, array(True)) # may vary
+    (28293632, 28293632, array(True))
 
     """
     return _wrapreduction(a, np.logical_and, 'all', axis, None, out,
@@ -3088,8 +3088,8 @@ def prod(a, axis=None, dtype=None, out=None, keepdims=np._NoValue,
     raised on overflow.  That means that, on a 32-bit platform:
 
     >>> x = np.array([536870910, 536870910, 536870910, 536870910])
-    >>> np.prod(x)
-    16 # may vary
+    >>> np.prod(x)  # doctest: +SKIP
+    16
 
     The product of an empty array is the neutral element 1:
 
@@ -3520,7 +3520,7 @@ def mean(a, axis=None, dtype=None, out=None, keepdims=np._NoValue, *,
     Computing the mean in float64 is more accurate:
 
     >>> np.mean(a, dtype=np.float64)
-    0.55000000074505806 # may vary
+    0.55000000074505806
 
     Specifying a where argument:
 
@@ -3676,7 +3676,7 @@ def std(a, axis=None, dtype=None, out=None, ddof=0, keepdims=np._NoValue, *,
     --------
     >>> a = np.array([[1, 2], [3, 4]])
     >>> np.std(a)
-    1.1180339887498949 # may vary
+    1.1180339887498949
     >>> np.std(a, axis=0)
     array([1.,  1.])
     >>> np.std(a, axis=1)
@@ -3693,13 +3693,13 @@ def std(a, axis=None, dtype=None, out=None, ddof=0, keepdims=np._NoValue, *,
     Computing the standard deviation in float64 is more accurate:
 
     >>> np.std(a, dtype=np.float64)
-    0.44999999925494177 # may vary
+    0.44999999925494177
 
     Specifying a where argument:
 
     >>> a = np.array([[14, 8, 11, 10], [7, 9, 10, 11], [10, 15, 5, 10]])
     >>> np.std(a)
-    2.614064523559687 # may vary
+    2.614064523559687
     >>> np.std(a, where=[[True], [True], [False]])
     2.0
 
@@ -3881,15 +3881,15 @@ def var(a, axis=None, dtype=None, out=None, ddof=0, keepdims=np._NoValue, *,
     Computing the variance in float64 is more accurate:
 
     >>> np.var(a, dtype=np.float64)
-    0.20249999932944759 # may vary
+    0.20249999932944759
     >>> ((1-0.55)**2 + (0.1-0.55)**2)/2
     0.2025
 
     Specifying a where argument:
 
     >>> a = np.array([[14, 8, 11, 10], [7, 9, 10, 11], [10, 15, 5, 10]])
     >>> np.var(a)
-    6.833333333333333 # may vary
+    6.833333333333333
     >>> np.var(a, where=[[True], [True], [False]])
     4.0
    

diff --git a/numpy/lib/_arraysetops_impl.py b/numpy/lib/_arraysetops_impl.py
@@ -266,13 +266,13 @@ def unique(ar, return_index=False, return_inverse=False,
     array([1, 2, 3, 4, 6])
     >>> counts
     array([1, 3, 1, 1, 1])
-    >>> np.repeat(values, counts)
-    array([1, 2, 2, 2, 3, 4, 6])    # original order not preserved
+    >>> np.repeat(values, counts)  # original order not preserved
+    array([1, 2, 2, 2, 3, 4, 6])
 
     """
     ar = np.asanyarray(ar)
     if axis is None:
-        ret = _unique1d(ar, return_index, return_inverse, return_counts, 
+        ret = _unique1d(ar, return_index, return_inverse, return_counts,
                         equal_nan=equal_nan)
         return _unpack_tuple(ret)
 
@@ -871,11 +871,11 @@ def _in1d(ar1, ar2, assume_unique=False, invert=False, *, kind=None):
         # However, here we set the requirement that by default
         # the intermediate array can only be 6x
         # the combined memory allocation of the original
-        # arrays. See discussion on 
+        # arrays. See discussion on
         # https://github.com/numpy/numpy/pull/12065.
 
         if (
-            range_safe_from_overflow and 
+            range_safe_from_overflow and
             (below_memory_constraint or kind == 'table')
         ):
 

diff --git a/numpy/lib/_datasource.py b/numpy/lib/_datasource.py
@@ -41,6 +41,8 @@
 
 _open = open
 
+__doctest_skip__ = ['DataSource']
+
 
 def _check_mode(mode, encoding, newline):
     """Check mode and that encoding and newline are compatible.