From Joshua at bitbucket.org  Mon May  6 17:10:19 2019
From: Joshua at bitbucket.org (Joshua at bitbucket.org)
Date: Mon, 06 May 2019 21:10:19 +0000 (UTC)
Subject: [pypy-issue] Issue #3012: io.IncrementalNewlineDecoder.newlines
 behaves differently than CPython's (pypy/pypy)
Message-ID: <20190506211018.7981.83543@app-137.ash1.bb-inf.net>

New issue 3012: io.IncrementalNewlineDecoder.newlines behaves differently than CPython's
https://bitbucket.org/pypy/pypy/issues/3012/ioincrementalnewlinedecodernewlines

Joshua Oreman:

When io.IncrementalNewlineDecoder in translate mode is fed input one character at a time, it does correctly translate ?\\r\\n? sequences to ?\\n?, but does not record ?\\r\\n? in its newlines attribute \(which I understand is intended to represent the newline sequences that were encountered in the input\). Reproducer:

```
import io
import codecs

info = codecs.lookup("utf-8")
inner_decoder = info.incrementaldecoder("strict")
outer_decoder = io.IncrementalNewlineDecoder(inner_decoder, True)
msg = b"The quick brown fox jumps over the lazy dog.\r\n\n\r\r\n\n"
decoded = ""
for ch in msg:
    decoded += outer_decoder.decode(bytes([ch]))
decoded += outer_decoder.decode(b"", final=True)
assert decoded == "The quick brown fox jumps over the lazy dog.\n\n\n\n\n"
assert set(outer_decoder.newlines) == {"\r", "\n", "\r\n"}
```

This passes on CPython but fails on PyPy-3.6 v7.1.1: on the last line, outer\_decoder.newlines contains ?\\r? and ?\\n? but not ?\\r\\n?.

There is no issue in non-translating mode; the newlines property does wind up containing all three sequences in that case.



From Milton at bitbucket.org  Mon May  6 21:56:18 2019
From: Milton at bitbucket.org (Milton at bitbucket.org)
Date: Tue, 07 May 2019 01:56:18 +0000 (UTC)
Subject: [pypy-issue] Issue #3013: os.rename crashes rpython when you pass
 in a byte string filename (pypy/pypy)
Message-ID: <20190507015618.21113.93232@app-137.ash1.bb-inf.net>

New issue 3013: os.rename crashes rpython when you pass in a byte string filename
https://bitbucket.org/pypy/pypy/issues/3013/osrename-crashes-rpython-when-you-pass-in

Milton Thomas:

To reproduce:

```
import os
os.rename( bytes("a", "utf8"), bytes("b", "utf8"))
```

result:

```
RPython traceback:
  File "implement_8.c", line 402, in BuiltinActivation_UwS_ObjSpace_W_Root_W_Root_kwo_3
  File "pypy_module_posix.c", line 18574, in dispatch__star_0_2
Fatal RPython error: AssertionError
```

Discovered this because a library in the wild opened a file in read-binary mode and pulled the file path out as bytes.

Works in CPython 3.7.3.

OS is Windows 10, PyPy version is PyPy3.6 7.1.1.



From Jason at bitbucket.org  Tue May 14 18:32:47 2019
From: Jason at bitbucket.org (Jason at bitbucket.org)
Date: Tue, 14 May 2019 22:32:47 +0000 (UTC)
Subject: [pypy-issue] Issue #3014: JIT Issue inlining `struct.unpack('<HH',
 bytearray(...))` produces incorrect results (pypy/pypy)
Message-ID: <20190514223247.10503.18955@celery-worker-112.ash1.bb-inf.net>

New issue 3014: JIT Issue inlining `struct.unpack('<HH', bytearray(...))` produces incorrect results
https://bitbucket.org/pypy/pypy/issues/3014/jit-issue-inlining-structunpack-hh

Jason Madden:

Versions:

* Python 2.7.13 \(8cdda8b8cdb8, Apr 14 2019, 14:06:58\)  
  \[PyPy 7.1.1 with GCC 4.2.1 Compatible Apple LLVM 10.0.0 \(clang-1000.11.45.5\)\] \(macOS 10.14\)
* Python 3.6.1 \(784b254d6699, Apr 14 2019, 10:22:55\)  
  \[PyPy 7.1.1-beta0 with GCC 4.2.1 Compatible Apple LLVM 10.0.0 \(clang-1000.11.45.5\)\]
* Python 2.7.13 \(8cdda8b8cdb8, Apr 14 2019, 14:06:58\)  
  \[PyPy 7.1.1 with GCC 4.2.1 Compatible Apple LLVM 10.0.0 \(clang-1000.11.45.5\)\]

?

I?ve been working on getting RelStorage to run on PyPy with \(pure-Python\) mysql-connector-python at [https://github.com/zodb/relstorage/issues/228](https://github.com/zodb/relstorage/issues/228#issuecomment-492423284), but was running into a very strange, non-sensical error only on PyPy partway through the test sequence.

In a nutshell,  the value of the `status_flag` in the OK result packet from the MySQL server is coming back with the `ServerFlag.MORE_RESULTS_EXIST` bit set, leading to the error. But that?s not actually the packet that the server is sending.

`protocol.py` has `parse_ok` that parses the `status_flag` from a packet \(a `bytearray`\) like this:

```python
    def parse_ok(self, packet):
        """Parse a MySQL OK-packet"""
        if not packet[4] == 0:
            raise errors.InterfaceError("Failed parsing OK packet (invalid).")

        ok_packet = {}
        ok_packet['orig_data'] = bytes(packet)
        try:
            ok_packet['field_count'] = struct_unpack('<xxxxB', packet[0:5])[0]
            (packet, ok_packet['affected_rows']) = utils.read_lc_int(packet[5:])
            (packet, ok_packet['insert_id']) = utils.read_lc_int(packet)
            (ok_packet['status_flag'], # This is the bad value
             ok_packet['warning_count']) = struct_unpack('<HH', packet[0:4]) 
            packet = packet[4:]
            if packet:
                (packet, ok_packet['info_msg']) = utils.read_lc_string(packet)
                ok_packet['info_msg'] = ok_packet['info_msg'].decode('utf-8')
        except ValueError:
            raise errors.InterfaceError("Failed parsing OK packet.")
        return ok_packet
```

\( `struct_unpack` is a small helper function for Python 2 that wraps a `bytearray` into a `buffer` and uses `struct.unpack_from`; I get the same results if I modify the code to use `struct.unpack` directly. Python 3 uses struct.unpack directly already.\)

I tweaked the error message to report the packet value, and to also unpack the data at that time. \(This is a cold branch of code that?s never been taken before\)

```python
 if self._have_next_result: # The status_flag was not what we wanted.
     from .catch23 import PY2, struct_unpack
     raise errors.InterfaceError(
         'Use cmd_query_iter for statements with multiple queries.'
         '\n%s\n%r\n%r\nUnpacks to:%r' % (
             query, result, raw_result, 
             struct_unpack('<HH', raw_result[0:4])))
```

```
InterfaceError: Use cmd_query_iter for statements with multiple queries.
    INSERT INTO temp_pack_visit (zoid, keep_tid)
    SELECT zoid, 0
    FROM current_object
    WHERE tid > 274639514178723874
{'field_count': 0, 'affected_rows': 0, 'insert_id': 0,
 'status_flag': 3480, 'warning_count': 0, 'info_msg': u'Records: 0  Duplicates: 0  Warnings: 0'}

bytearray(b'.\x00\x00\x01\x00\x00\x00\x02\x00\x00\x00&Records: 0  Duplicates: 0  Warnings: 0')
    
Unpacks to: (46, 256)
```

You can see that protocol.py's `parse_ok` method gets a `status_flag` of `3480`, while performing the same unpacking on the same data in this function gets a value of `46` \(which is what CPython gets too\).

\*\*Running with `--jit off` solves the problem and lets the tests complete successfully. Also, just running with `--jit inlining=0` solves the problem too.\*\*

How can I help with this? I imagine you?ll want a JIT log of some sort?



From dholth at bitbucket.org  Fri May 17 10:01:31 2019
From: dholth at bitbucket.org (dholth at bitbucket.org)
Date: Fri, 17 May 2019 14:01:31 +0000 (UTC)
Subject: [pypy-issue] Issue #3015: black crashes on pypy 7 with 'PosixPath'
 does not support the buffer interface (pypy/pypy)
Message-ID: <20190517140131.9750.9619@celery-worker-112.ash1.bb-inf.net>

New issue 3015: black crashes on pypy 7 with 'PosixPath' does not support the buffer interface
https://bitbucket.org/pypy/pypy/issues/3015/black-crashes-on-pypy-7-with-posixpath

Daniel Holth:

[https://github.com/python/black/issues/780](https://github.com/python/black/issues/780)

black complains when running under pypy 7 but seems to work anyway. The error is

`error: cannot format x.py: 'PosixPath' does not support the buffer interface`



From piotrjurkiewicz at bitbucket.org  Sat May 18 17:05:17 2019
From: piotrjurkiewicz at bitbucket.org (piotrjurkiewicz at bitbucket.org)
Date: Sat, 18 May 2019 21:05:17 +0000 (UTC)
Subject: [pypy-issue] Issue #3016: AttributeError: 'memoryview' object has
 no attribute 'obj' (pypy/pypy)
Message-ID: <20190518210517.38549.98481@celery-worker-108.ash1.bb-inf.net>

New issue 3016: AttributeError: 'memoryview' object has no attribute 'obj'
https://bitbucket.org/pypy/pypy/issues/3016/attributeerror-memoryview-object-has-no

Piotr Jurkiewicz:

Apparently there is no attribute `obj` in `memoryview`, which points to the underlying object of the `memoryview`.

```
Traceback (most recent call last):
  File "/usr/lib/pypy3/lib-python/3/runpy.py", line 193, in _run_module_as_main
    "__main__", mod_spec)
  File "/usr/lib/pypy3/lib-python/3/runpy.py", line 85, in _run_code
    exec(code, run_globals)
  File "/home/jurkiew/flow_stats/flow_stats/convert.py", line 50, in <module>
  File "/home/jurkiew/flow_stats/flow_stats/convert.py", line 46, in main
  File "/home/jurkiew/flow_stats/flow_stats/convert.py", line 26, in convert
    for flow in reader(file):
  File "/home/jurkiew/flow_stats/flow_stats/lib/io.py", line 158, in read_arrays
    obj = mv.obj
AttributeError: 'memoryview' object has no attribute 'obj'
```



From JanB42 at bitbucket.org  Wed May 22 08:54:45 2019
From: JanB42 at bitbucket.org (JanB42 at bitbucket.org)
Date: Wed, 22 May 2019 12:54:45 +0000 (UTC)
Subject: [pypy-issue] Issue #3017: PyPy3.6 v7.1.1 Datetime Module
 ValueError: Invalid argument (pypy/pypy)
Message-ID: <20190522125445.37571.46402@celery-worker-110.ash1.bb-inf.net>

New issue 3017: PyPy3.6 v7.1.1 Datetime Module ValueError: Invalid argument
https://bitbucket.org/pypy/pypy/issues/3017/pypy36-v711-datetime-module-valueerror

JanB:

I used PyPy3.5 v7.0 together with the openpyxl module and it worked fine. After switching to PyPy3.6 v7.1.1 I get the attached ValueError with the last three lines pointing to the Python3 Datetime module. I am rather unexperienced, is this an PyPy issue or does it have something to do with the openpyxl module?



From Jan at bitbucket.org  Sun May 26 00:06:44 2019
From: Jan at bitbucket.org (Jan at bitbucket.org)
Date: Sun, 26 May 2019 04:06:44 +0000 (UTC)
Subject: [pypy-issue] Issue #3018: numpy.array_equal behaves differently
 from python3 when comparing ctype types (pypy/pypy)
Message-ID: <20190526040644.1526.83438@app-137.ash1.bb-inf.net>

New issue 3018: numpy.array_equal behaves differently from python3 when comparing ctype types
https://bitbucket.org/pypy/pypy/issues/3018/numpyarray_equal-behaves-differently-from

Jan Vesely:

```python
Python 3.6.1 (784b254d669919c872a505b807db8462b6140973, Apr 21 2019, 14:47:54)
[PyPy 7.1.1-beta0 with GCC 8.2.0] on linux
Type "help", "copyright", "credits" or "license" for more information.
And now for something completely different: ``PyPy is vast, and contains
multitudes''
>>>> import ctypes
>>>> import numpy
>>>> a = ctypes.c_int(1)
>>>> numpy.array_equal(a, True)
False
>>>> numpy.array_equal(a, 1)
False
```

vs.

```
Python 3.6.5 (default, Feb 24 2019, 17:47:37) 
[GCC 8.2.0] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import numpy
>>> import ctypes
>>> a = ctypes.c_int(1)
>>> numpy.array_equal(a, 1)
True
>>> numpy.array_equal(a, True)
True
```

Both instances are running numpy-1.15.4



From Andrew at bitbucket.org  Sun May 26 14:13:26 2019
From: Andrew at bitbucket.org (Andrew at bitbucket.org)
Date: Sun, 26 May 2019 18:13:26 +0000 (UTC)
Subject: [pypy-issue] Issue #3020: Error Code Handling in Multiprocessing
 Tests (pypy/pypy)
Message-ID: <20190526181325.17663.71238@celery-worker-112.ash1.bb-inf.net>

New issue 3020: Error Code Handling in Multiprocessing Tests
https://bitbucket.org/pypy/pypy/issues/3020/error-code-handling-in-multiprocessing

Andrew Lawrence:

There is some behavior of error codes that I cannot explain in test\_multiprocessing\_spawn.py.

When running the tests lib-python\\3\\test\\test\_multiprocessing\_spawn.py WithManagerTestLock.test\_lock WithProcessesTestBarrier.test\_barrier on Windows the error code leaks from test\_lock into test\_barrier causing test\_barrier to fail.             

In CPython the error codes do not leak between the tests like this.

?

I have also noticed that doing p \_winapi.GetLastError\(\) from pdb inside WithProcessesTestBarrier.test\_barrier in CPython produces the error code from the previous test but print\(\_winapi.GetLastError\(\)\) in the test itself does not result in the error.

Also inserting the following code into \_test\_multiprocessing.py results in different behaviour on both CPython and windows when running test\_multiprocessing\_spawn.py WithManagerTestLock.test\_lock 

```
class ManagerMixin(object):
  def tearDownClass(cls):
        import ctypes
        ctypes.windll.kernel32.SetLastError(996)
        print(_winapi.GetLastError())
        print(ctypes.windll.kernel32.GetLastError())
```

The final print statement produces 0 in CPython and 996 in PyPy