[issue15826] Increased test coverage of test_glob.py
New submission from Alessandro Moura: Hi, Here is a patch to increase test coverage of test_glob.py. I added tests where the glob patterns are of type bytes. The coverage (according to coverage.py) is now complete, except for what are clearly artefacts probably caused by the fact that glob is loaded before coverage measurements start (global statements are reported not covered but local ones are). -- components: Tests files: glob-cov-incr.patch keywords: patch messages: 169496 nosy: eng793 priority: normal severity: normal status: open title: Increased test coverage of test_glob.py type: enhancement versions: Python 3.3 Added file: http://bugs.python.org/file27067/glob-cov-incr.patch ___ Python tracker <http://bugs.python.org/issue15826> ___ ___ Python-bugs-list mailing list Unsubscribe: http://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
[issue15826] Increased test coverage of test_glob.py
Alessandro Moura added the comment: Agreed, here is the patch where this is done. -- Added file: http://bugs.python.org/file27078/test_glob.patch ___ Python tracker <http://bugs.python.org/issue15826> ___ ___ Python-bugs-list mailing list Unsubscribe: http://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
[issue15837] Added test to test_random.py testing Random.shuffle
New submission from Alessandro Moura: Random.shuffle does not have a test in test_random.py; the attached patch adds this test. In addition, I rewrote the documentation string for Random.shuffle, which apparently did not reflect recent changes in the code and was inconsistent with the definition of the method. This change is also part of this patch; I was not sure if this merited a separate issue, so I just included this here. On a related matter: in Random.shuffle there is a third optional argument which looks very odd to me: def shuffle(self, x, random=None, int=int): Besides being confusing to a user typing help(shuffle), what the "int" argument does in shuffle is to convert a float to an integer. But one could pass any one-argument function in principle, and it would be then very hard to predict what shuffle would do... it would not "shuffle" any more in the traditional sense - not with a uniform probability distribution. In summary, I don't see how that argument could be useful, although the people who wrote the library must have had something in mind... if so it would be a good idea to document it. -- components: Tests messages: 169603 nosy: eng793 priority: normal severity: normal status: open title: Added test to test_random.py testing Random.shuffle type: enhancement versions: Python 3.3 ___ Python tracker <http://bugs.python.org/issue15837> ___ ___ Python-bugs-list mailing list Unsubscribe: http://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
[issue15837] Added test to test_random.py testing Random.shuffle
Alessandro Moura added the comment: Sorry, here it is the patch. -- keywords: +patch Added file: http://bugs.python.org/file27082/random.patch ___ Python tracker <http://bugs.python.org/issue15837> ___ ___ Python-bugs-list mailing list Unsubscribe: http://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
[issue15826] Increased test coverage of test_glob.py
Alessandro Moura added the comment: test_bytes_glob_directory_with_trailing_slash already has a counterpart in the main class, but it is not easily handled by overriding self.glob and self.norm. Since it is only a single test, I just decided to override test_glob_directory_with_trailing_slash from the main class. But looking at this now I realize that I actually created another method with a different name by mistake; so the attached patch changes that. -- Added file: http://bugs.python.org/file27083/test_glob.patch ___ Python tracker <http://bugs.python.org/issue15826> ___ ___ Python-bugs-list mailing list Unsubscribe: http://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
[issue15826] Increased test coverage of test_glob.py
Alessandro Moura added the comment: Im OK with the patch changes. ''' /home/rdmurray/python/p33/Lib/os.py:263: BytesWarning: Comparison between bytes and string if tail == curdir: # xxx/newdir/. exists if xxx/newdir exists /home/rdmurray/python/p33/Lib/glob.py:63: BytesWarning: Comparison between bytes and string if basename == '': ''' Yes, I do see these warnings running with -b. I will look into them and try to fix them. -- ___ Python tracker <http://bugs.python.org/issue15826> ___ ___ Python-bugs-list mailing list Unsubscribe: http://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
[issue15845] Fixing some byte-to-string conversion warnings
New submission from Alessandro Moura: This is related to issue 15826. When run with the -b option, some glob.py and os.py functions give warnings due to byte-to-string conversions: amoura@amoura-laptop:~/cpython$ ./python -b -c "import glob; glob.glob(b'cover*/glob.cover')" /home/amoura/cpython/Lib/glob.py:64: BytesWarning: Comparison between bytes and string if basename == '': amoura@amoura-laptop:~/cpython$ ./python -b -c "import os; os.makedirs(b'tst/making/dirs')" /home/amoura/cpython/Lib/os.py:266: BytesWarning: Comparison between bytes and string if tail == cdir: # xxx/newdir/. exists if xxx/newdir exists The attached patch fixes this. There is a rather more mysterious phenomenon with exceptions (which is triggered by test_exceptions for ImportException, but it happens for any Exception class): >>> e = Exception(b'aaa') [60596 refs] >>> e.args[0] b'aaa' [60601 refs] >>> str(e) __main__:1: BytesWarning: str() on a bytes instance "b'aaa'" [60615 refs] >>> e.args[0] b'aaa' [60615 refs] >>> str(e) "b'aaa'" [60615 refs] >>> e.args[0] b'aaa' [60615 refs] In other words, if a bytes argument is given to the exception, the first call to str triggers the warning, but further calls don't. Is this worth pursuing? -- components: Library (Lib) files: os_glob_bytes.patch keywords: patch messages: 169683 nosy: eng793 priority: normal severity: normal status: open title: Fixing some byte-to-string conversion warnings type: enhancement versions: Python 3.3 Added file: http://bugs.python.org/file27092/os_glob_bytes.patch ___ Python tracker <http://bugs.python.org/issue15845> ___ ___ Python-bugs-list mailing list Unsubscribe: http://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
[issue15841] Some StringIO and BytesIO methods can succeed after close
Alessandro Moura added the comment: This also happens for the writable() and seekable() methods. The problem is that those methods do not check whether the buffers have been closed in stringio.c. This is fixed in the attached patch for StringIO. BytesIO should be the same, but bytesio.c is structured differently, and I still have to understand the code. I will try to do this, and then add tests for this issue - which should go in one of the mixins of test_memoryio.py, I presume. -- keywords: +patch nosy: +eng793 Added file: http://bugs.python.org/file27094/stringio.c.patch ___ Python tracker <http://bugs.python.org/issue15841> ___ ___ Python-bugs-list mailing list Unsubscribe: http://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
[issue15841] Some StringIO and BytesIO methods can succeed after close
Alessandro Moura added the comment: Here is the patch fixing the issue in both StringIO and BytesIO. In both cases, the problem is that in their respective c files, these methods always returned true, without testing whether the file was closed. Is this a recent rewrite? I am surprised this did not bite someone earlier. I also added a testcase to test_memoryio.py covering this issue; and I added the docstring for these methods, which were missing. This patch compiles and passes the entire test suite in my system. -- Added file: http://bugs.python.org/file27096/memio.patch ___ Python tracker <http://bugs.python.org/issue15841> ___ ___ Python-bugs-list mailing list Unsubscribe: http://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
[issue15841] Some StringIO and BytesIO methods can succeed after close
Alessandro Moura added the comment: Sorry, I should have seen the related issue (or just read the docs :)). Here is the patch fixing these issues, implementing the behaviour stated in the docs (raising ValueError after close). The tests revealed that the _pyio module also needed fixing, and this is done in the patch as well. -- Added file: http://bugs.python.org/file27102/memoryio.patch ___ Python tracker <http://bugs.python.org/issue15841> ___ ___ Python-bugs-list mailing list Unsubscribe: http://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
[issue15837] Added test to test_random.py testing Random.shuffle
Alessandro Moura added the comment: Comparing the execution time with and without the int=int argument of this command: amoura@amoura-laptop:~/cpython$ time ./python -c "from random import shuffle; lst=list(range(100)); shuffle(lst); print (len(lst))" I get with int=int: real0m13.755s user0m13.777s sys 0m0.124s and without it: real0m13.876s user0m13.701s sys 0m0.116s So it makes no difference in practice. On the other hand, removing this has a chance of braking existing code, if someone somewhere actually uses the third argument for something - I can't image what, but still... -- ___ Python tracker <http://bugs.python.org/issue15837> ___ ___ Python-bugs-list mailing list Unsubscribe: http://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
[issue15837] Added test to test_random.py testing Random.shuffle
Alessandro Moura added the comment: The int=int still makes no difference, but if the second argument is set to random.random, we get a big speedup, regardless of whether the third argument is there: without int=int: amoura@amoura-laptop:~/cpython$ time ./python -c "import random; lst=list(range(100)); random.shuffle(lst,random.random); print (len(lst))" 100 real0m7.082s user0m6.952s sys 0m0.116s With int=int: amoura@amoura-laptop:~/cpython$ time ./python -c "import random; lst=list(range(100)); random.shuffle(lst,random.random); print (len(lst))" 100 real0m7.281s user0m7.156s sys 0m0.100s Without second argument: amoura@amoura-laptop:~/cpython$ time ./python -c "import random; lst=list(range(100)); random.shuffle(lst); print (len(lst))" 100 real0m13.783s user0m13.609s sys 0m0.108s This could be because of the many tests of whether the 2nd argument is None in the loop. -- ___ Python tracker <http://bugs.python.org/issue15837> ___ ___ Python-bugs-list mailing list Unsubscribe: http://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
[issue15837] Added test to test_random.py testing Random.shuffle
Alessandro Moura added the comment: Yup. This is the result of simply eliminating the condition in the loop and just using the second argument (for the purposes of testing this only): amoura@amoura-laptop:~/cpython$ time ./python -c "import random; lst=list(range(100)); random.shuffle(lst,random.random); print (len(lst))" 100 real0m7.330s user0m7.148s sys 0m0.092s -- ___ Python tracker <http://bugs.python.org/issue15837> ___ ___ Python-bugs-list mailing list Unsubscribe: http://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
[issue15841] Some StringIO and BytesIO methods can succeed after close
Alessandro Moura added the comment: Thanks. Here is the amended patch with your suggestions implemented. -- Added file: http://bugs.python.org/file27117/memoryio.patch ___ Python tracker <http://bugs.python.org/issue15841> ___ ___ Python-bugs-list mailing list Unsubscribe: http://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
[issue15841] Some StringIO and BytesIO methods can succeed after close
Alessandro Moura added the comment: I just emailed my contributor's agreement. -- ___ Python tracker <http://bugs.python.org/issue15841> ___ ___ Python-bugs-list mailing list Unsubscribe: http://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
[issue15842] Some SocketIO methods can succeed after close()
Alessandro Moura added the comment: This patch fixes the problem, making those methods raise a ValueError exception after close. I also added one test case for this issue. -- keywords: +patch nosy: +eng793 Added file: http://bugs.python.org/file27128/socket.patch ___ Python tracker <http://bugs.python.org/issue15842> ___ ___ Python-bugs-list mailing list Unsubscribe: http://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
[issue15872] shutil.rmtree(..., ignore_errors=True) doesn't ignore all errors
Alessandro Moura added the comment: Yes, confirmed. When checking whether the provided path is a directory, rmtree does not check whether ignore_errors is set. According to the docstring, "If ignore_errors is set, errors are ignored". Presumably this means any error, in which case this is not the desired behaviour. The attached patch fixes this. -- keywords: +patch nosy: +eng793 Added file: http://bugs.python.org/file27138/shutil.patch ___ Python tracker <http://bugs.python.org/issue15872> ___ ___ Python-bugs-list mailing list Unsubscribe: http://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
[issue15872] shutil.rmtree(..., ignore_errors=True) doesn't ignore all errors
Alessandro Moura added the comment: Added test to patch. -- Added file: http://bugs.python.org/file27139/shutil.patch ___ Python tracker <http://bugs.python.org/issue15872> ___ ___ Python-bugs-list mailing list Unsubscribe: http://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
[issue15842] Some SocketIO methods can succeed after close()
Alessandro Moura added the comment: Fixed seekable(), and amended tests; see patch. -- Added file: http://bugs.python.org/file27149/socket.patch ___ Python tracker <http://bugs.python.org/issue15842> ___ ___ Python-bugs-list mailing list Unsubscribe: http://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
