circular import when doing "import pandas" on pandas‑0.24.2‑cp38‑cp38m‑win_amd64.whl

[stonebig]

Maybe it's the Python-3.8.0.a4 windows dll change ?

just a remark, as it's all alpha. using cgohlke first wheel on Python-3.8.0a4, I have a strange message

Python 3.8.0a4 (v3.8.0a4:c1004b8546, May  7 2019, 12:46:11) [MSC v.1916 32 bit (Intel)] on win32
Type "copyright", "credits" or "license()" for more information.
>>> import pandas
Traceback (most recent call last):
  File "<pyshell#0>", line 1, in <module>
    import pandas
  File "C:\WinP\bd38\bu\WPy32-3800a4\python-3.8.0a4\lib\site-packages\pandas\__init__.py", line 49, in <module>
    from pandas.io.api import *
  File "C:\WinP\bd38\bu\WPy32-3800a4\python-3.8.0a4\lib\site-packages\pandas\io\api.py", line 8, in <module>
    from pandas.io.excel import ExcelFile, ExcelWriter, read_excel
  File "C:\WinP\bd38\bu\WPy32-3800a4\python-3.8.0a4\lib\site-packages\pandas\io\excel.py", line 34, in <module>
    from pandas.io.parsers import TextParser
  File "C:\WinP\bd38\bu\WPy32-3800a4\python-3.8.0a4\lib\site-packages\pandas\io\parsers.py", line 19, in <module>
    import pandas._libs.parsers as parsers
  File "pandas\_libs\parsers.pyx", line 1418, in init pandas._libs.parsers
AttributeError: partially initialized module 'pandas' has no attribute '_NA_VALUES' (most likely due to a circular import)
>>> 

[jreback]

we have no support / testing for 3.8 at all

[cgohlke]

This is more likely due to Cython-3.0a0 (git master from yesterday)

[gfyoung]

Feel free to continue the conversation, but due to the comment above, closing this.

[stonebig]

problem solved with the new pandas wheel.

[gfyoung]

@stonebig : That's great to hear!

[stonebig]

Other remark, still on this alpha stage., pandas‑0.24.2‑cp38‑cp38m‑win_amd64.whl of cgohlke now fails on this:

#Pandas 
import pandas as pd
import numpy as np

idx = pd.date_range('2000', '2005', freq='d', closed='left')
datas = pd.DataFrame({'Color':  [ 'green' if x> 1 else 'red' for x in np.random.randn(len(idx))], 
         'Measure': np.random.randn(len(idx)), 'Year': idx.year},
          index=idx.date)
datas.query('Measure > 0').groupby(['Color','Year']).size().unstack()

gives

---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
<ipython-input-25-99840eba54d9> in <module>
----> 1 datas.query('Measure > 0').groupby(['Color','Year']).size().unstack()

C:\WinP\bd38\bu\WPy64-3800a4\python-3.8.0a4.amd64\lib\site-packages\pandas\core\frame.py in query(self, expr, inplace, **kwargs)
   3086         kwargs['level'] = kwargs.pop('level', 0) + 1
   3087         kwargs['target'] = None
-> 3088         res = self.eval(expr, **kwargs)
   3089 
   3090         try:

C:\WinP\bd38\bu\WPy64-3800a4\python-3.8.0a4.amd64\lib\site-packages\pandas\core\frame.py in eval(self, expr, inplace, **kwargs)
   3201             kwargs['target'] = self
   3202         kwargs['resolvers'] = kwargs.get('resolvers', ()) + tuple(resolvers)
-> 3203         return _eval(expr, inplace=inplace, **kwargs)
   3204 
   3205     def select_dtypes(self, include=None, exclude=None):

C:\WinP\bd38\bu\WPy64-3800a4\python-3.8.0a4.amd64\lib\site-packages\pandas\core\computation\eval.py in eval(expr, parser, engine, truediv, local_dict, global_dict, resolvers, level, target, inplace)
    291                             target=target)
    292 
--> 293         parsed_expr = Expr(expr, engine=engine, parser=parser, env=env,
    294                            truediv=truediv)
    295 

C:\WinP\bd38\bu\WPy64-3800a4\python-3.8.0a4.amd64\lib\site-packages\pandas\core\computation\expr.py in __init__(self, expr, engine, parser, env, truediv, level)
    747         self.env.scope['truediv'] = truediv
    748         self._visitor = _parsers[parser](self.env, self.engine, self.parser)
--> 749         self.terms = self.parse()
    750 
    751     @property

C:\WinP\bd38\bu\WPy64-3800a4\python-3.8.0a4.amd64\lib\site-packages\pandas\core\computation\expr.py in parse(self)
    764     def parse(self):
    765         """Parse an expression"""
--> 766         return self._visitor.visit(self.expr)
    767 
    768     @property

C:\WinP\bd38\bu\WPy64-3800a4\python-3.8.0a4.amd64\lib\site-packages\pandas\core\computation\expr.py in visit(self, node, **kwargs)
    329         method = 'visit_' + node.__class__.__name__
    330         visitor = getattr(self, method)
--> 331         return visitor(node, **kwargs)
    332 
    333     def visit_Module(self, node, **kwargs):

C:\WinP\bd38\bu\WPy64-3800a4\python-3.8.0a4.amd64\lib\site-packages\pandas\core\computation\expr.py in visit_Module(self, node, **kwargs)
    335             raise SyntaxError('only a single expression is allowed')
    336         expr = node.body[0]
--> 337         return self.visit(expr, **kwargs)
    338 
    339     def visit_Expr(self, node, **kwargs):

C:\WinP\bd38\bu\WPy64-3800a4\python-3.8.0a4.amd64\lib\site-packages\pandas\core\computation\expr.py in visit(self, node, **kwargs)
    329         method = 'visit_' + node.__class__.__name__
    330         visitor = getattr(self, method)
--> 331         return visitor(node, **kwargs)
    332 
    333     def visit_Module(self, node, **kwargs):

C:\WinP\bd38\bu\WPy64-3800a4\python-3.8.0a4.amd64\lib\site-packages\pandas\core\computation\expr.py in visit_Expr(self, node, **kwargs)
    338 
    339     def visit_Expr(self, node, **kwargs):
--> 340         return self.visit(node.value, **kwargs)
    341 
    342     def _rewrite_membership_op(self, node, left, right):

C:\WinP\bd38\bu\WPy64-3800a4\python-3.8.0a4.amd64\lib\site-packages\pandas\core\computation\expr.py in visit(self, node, **kwargs)
    329         method = 'visit_' + node.__class__.__name__
    330         visitor = getattr(self, method)
--> 331         return visitor(node, **kwargs)
    332 
    333     def visit_Module(self, node, **kwargs):

C:\WinP\bd38\bu\WPy64-3800a4\python-3.8.0a4.amd64\lib\site-packages\pandas\core\computation\expr.py in visit_Compare(self, node, **kwargs)
    664             op = self.translate_In(ops[0])
    665             binop = ast.BinOp(op=op, left=node.left, right=comps[0])
--> 666             return self.visit(binop)
    667 
    668         # recursive case: we have a chained comparison, a CMP b CMP c, etc.

C:\WinP\bd38\bu\WPy64-3800a4\python-3.8.0a4.amd64\lib\site-packages\pandas\core\computation\expr.py in visit(self, node, **kwargs)
    329         method = 'visit_' + node.__class__.__name__
    330         visitor = getattr(self, method)
--> 331         return visitor(node, **kwargs)
    332 
    333     def visit_Module(self, node, **kwargs):

C:\WinP\bd38\bu\WPy64-3800a4\python-3.8.0a4.amd64\lib\site-packages\pandas\core\computation\expr.py in visit_BinOp(self, node, **kwargs)
    433 
    434     def visit_BinOp(self, node, **kwargs):
--> 435         op, op_class, left, right = self._maybe_transform_eq_ne(node)
    436         left, right = self._maybe_downcast_constants(left, right)
    437         return self._maybe_evaluate_binop(op, op_class, left, right)

C:\WinP\bd38\bu\WPy64-3800a4\python-3.8.0a4.amd64\lib\site-packages\pandas\core\computation\expr.py in _maybe_transform_eq_ne(self, node, left, right)
    372             left = self.visit(node.left, side='left')
    373         if right is None:
--> 374             right = self.visit(node.right, side='right')
    375         op, op_class, left, right = self._rewrite_membership_op(node, left,
    376                                                                 right)

C:\WinP\bd38\bu\WPy64-3800a4\python-3.8.0a4.amd64\lib\site-packages\pandas\core\computation\expr.py in visit(self, node, **kwargs)
    328 
    329         method = 'visit_' + node.__class__.__name__
--> 330         visitor = getattr(self, method)
    331         return visitor(node, **kwargs)
    332 

AttributeError: 'PandasExprVisitor' object has no attribute 'visit_Constant'

[jorisvandenbossche]

Although it's that we don't have any official support of testing for Python 3.8, it is still nice that people already start testing it, so issues can be catched early (both on our side as on Python's side). So thanks for that @stonebig !

For the last error: I am not very familiar with the DataFrame.query code, but that seems like potentially due to a change in AST ?

[stonebig]

error still present with python-3.8.0b1