[lldb-dev] Moving pexpect and unittest2 to lldb/third_party

Thu Oct 22 12:56:48 PDT 2015

This is going in right now.  As it is a fairly large change, it wouldn't
surprise me if someone encounters an issue.  I tested this everywhere I can
and it seems fine, so please let me know if anyone encounters anything.

On Thu, Oct 22, 2015 at 11:39 AM Todd Fiala <todd.fiala at gmail.com> wrote:

> Yeah I think the biggest thing I wanted to check there was that there
> wasn't any unittest2 behavior present in that cut of unittest2 that didn't
> make it into the revamped version brought into the python distributions
> when they upgraded unittest.  Then it's just a big rename exercise on
> replacing unittest2 with unittest (again, after making sure that my
> expectations here are correct on this being valid).
>
> On Thu, Oct 22, 2015 at 11:33 AM, Zachary Turner <zturner at google.com>
> wrote:
>
>> Cool!  I probably won't delete it from the repo entirely, because that
>> entails mucking with the command line options of dotest to remove any
>> related options, and any initialization code in dotest.py or TestBase
>> subclasses related to unittest2.  For now I'll just delete the imports from
>> each individual test and remove the if __name__ == "main" blocks.  After
>> that though it should be a fairly straightforward follow-up to remove it
>> entirely if anyone wants to.
>>
>> On Thu, Oct 22, 2015 at 11:30 AM Todd Fiala <todd.fiala at gmail.com> wrote:
>>
>>> (I was eventually going to do this at some point after I verified it was
>>> indeed true).  It should just be called unittest in a stock distribution.
>>>
>>> On Thu, Oct 22, 2015 at 11:29 AM, Todd Fiala <todd.fiala at gmail.com>
>>> wrote:
>>>
>>>> We could also then remove unittest2 from inclusion in the lldb repo.
>>>>
>>>>
>>>> On Thu, Oct 22, 2015 at 11:28 AM, Todd Fiala <todd.fiala at gmail.com>
>>>> wrote:
>>>>
>>>>> I'd be okay with that.
>>>>>
>>>>> The unittest2 stuff looks like it was a vestige of being incorporated
>>>>> before unittest2 was stock (unitest) on Python 2.[6,7]?.  Everyone should
>>>>> have a unitest included that is effectively what we use as unittest2.
>>>>>
>>>>> -Todd
>>>>>
>>>>> On Thu, Oct 22, 2015 at 10:05 AM, Zachary Turner via lldb-dev <
>>>>> lldb-dev at lists.llvm.org> wrote:
>>>>>
>>>>>> I plan to put this in today.  Greg, should I just go ahead and delete
>>>>>> all the unittest2 stuff entirely?  TBH I'm all for anything that reduces
>>>>>> the complexity of the test suite.  It's got a couple hundred options that
>>>>>> nobody uses, seems like we should start whittling away at stuff that
>>>>>> doesn't get any use.
>>>>>>
>>>>>> If you prefer I leave it in that's less work for me since I have that
>>>>>> patch ready to go, but TBH I'd rather remove it if that's ok.
>>>>>>
>>>>>> On Thu, Oct 22, 2015 at 9:50 AM Zachary Turner <zturner at google.com>
>>>>>> wrote:
>>>>>>
>>>>>>> You can get pretty much the same effect though by just running
>>>>>>> dotest and passing it the folder that the .py file is in.   Then it only
>>>>>>> runs tests in that folder.  You can specify the filename too if you want to
>>>>>>> limit it to one name.  Sure, it's a few keystrokes less to just type
>>>>>>> TestMultithreaded.py or something, but given the extra complexity and the
>>>>>>> fact that it's running a totally different codepath, I wonder if the
>>>>>>> maintenance burden is worth it (I'm guessing no, since apparently it
>>>>>>> doesn't work well enough right now for anyone to use it)
>>>>>>>
>>>>>>> On Thu, Oct 22, 2015 at 9:47 AM Greg Clayton <gclayton at apple.com>
>>>>>>> wrote:
>>>>>>>
>>>>>>>> I believe it would import lldb correctly. I don't tend to run the
>>>>>>>> tests individually, but if it did work, I would use it more.
>>>>>>>>
>>>>>>>> > On Oct 22, 2015, at 9:26 AM, Zachary Turner via lldb-dev <
>>>>>>>> lldb-dev at lists.llvm.org> wrote:
>>>>>>>> >
>>>>>>>> > Todd, Greg, can you guys confirm this is true?   The import lldb
>>>>>>>> would succeed if someone had their PYTHONPATH set up just right, but if
>>>>>>>> really none of us care about it, I'm with Tamas in that I'd rather remove
>>>>>>>> it.
>>>>>>>> >
>>>>>>>> > On Thu, Oct 22, 2015 at 2:55 AM Tamas Berghammer <
>>>>>>>> tberghammer at google.com> wrote:
>>>>>>>> > Hi Zach,
>>>>>>>> >
>>>>>>>> > I think nobody is using the "if __name__ == '__main__'" block as
>>>>>>>> executing a test file directly isn't working at the moment (the "import
>>>>>>>> lldb" command fails). If you plan to change all test file then I would
>>>>>>>> prefer to remove the reference to unittest2 from them for simplicity if
>>>>>>>> nobody have an objection against it.
>>>>>>>> >
>>>>>>>> > Tamas
>>>>>>>> >
>>>>>>>> > On Wed, Oct 21, 2015 at 8:57 PM Zachary Turner via lldb-dev <
>>>>>>>> lldb-dev at lists.llvm.org> wrote:
>>>>>>>> > TL;DR - Nobody has to do anything, this is just a heads up that a
>>>>>>>> 400+ file CL is coming.
>>>>>>>> >
>>>>>>>> > IANAL, but I've been told by one that I need to move all third
>>>>>>>> party code used by LLDB to lldb/third_party.  Currently there is only one
>>>>>>>> thing there: the Python `six` module used for creating code that is
>>>>>>>> portable across Python 2 and Python 3.
>>>>>>>> >
>>>>>>>> > The only other 2 instances that I'm aware of are pexpect and
>>>>>>>> unittest2, which are under lldb/test.  I've got some patches locally which
>>>>>>>> move pexpect and unittest2 to lldb/third_party.  I'll hold off on checking
>>>>>>>> them in for a bit to give people a chance to see this message first,
>>>>>>>> because otherwise you might be surprised when you see a CL with 400 files
>>>>>>>> being checked in.
>>>>>>>> >
>>>>>>>> > Nobody will have to do anything after this CL goes in, and
>>>>>>>> everything should continue to work exactly as it currently does.
>>>>>>>> >
>>>>>>>> > The main reason for the churn is that pretty much every single
>>>>>>>> test in LLDB does something like this:
>>>>>>>> >
>>>>>>>> > import unittest2
>>>>>>>> >
>>>>>>>> > ...
>>>>>>>> >
>>>>>>>> > if __name__ == '__main__':
>>>>>>>> >     import atexit
>>>>>>>> >     lldb.SBDebugger.Initialize()
>>>>>>>> >     atexit.register(lambda: lldb.SBDebugger.Terminate())
>>>>>>>> >     unittest2.main()
>>>>>>>> >
>>>>>>>> > This worked when unittest2 was a subfolder of test, but not when
>>>>>>>> it's somewhere else.  Since LLDB's python code is not organized into a
>>>>>>>> standard python package and we treat the scripts like dotest etc as
>>>>>>>> standalone scripts, the way I've made this work is by introducing a module
>>>>>>>> called lldb_shared under test which, when you import it, fixes up sys.path
>>>>>>>> to correctly add all the right locations under lldb/third_party.
>>>>>>>> >
>>>>>>>> > So, every single test now needs a line at the top to import
>>>>>>>> lldb_shared.
>>>>>>>> >
>>>>>>>> > TBH I don't even know if we need this unittest2 stuff anymore
>>>>>>>> (does anyone even use it?)  but even if the answer is no, then that still
>>>>>>>> means changing every file to delete the import statement and the if
>>>>>>>> __name__ == '__main__': block.
>>>>>>>> >
>>>>>>>> > If there are no major concerns I plan to check this in by the end
>>>>>>>> of the day, or tomorrow.
>>>>>>>> > _______________________________________________
>>>>>>>> > lldb-dev mailing list
>>>>>>>> > lldb-dev at lists.llvm.org
>>>>>>>> > http://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-dev
>>>>>>>> > _______________________________________________
>>>>>>>> > lldb-dev mailing list
>>>>>>>> > lldb-dev at lists.llvm.org
>>>>>>>> > http://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-dev
>>>>>>>>
>>>>>>>>
>>>>>> _______________________________________________
>>>>>> lldb-dev mailing list
>>>>>> lldb-dev at lists.llvm.org
>>>>>> http://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-dev
>>>>>>
>>>>>>
>>>>>
>>>>>
>>>>> --
>>>>> -Todd
>>>>>
>>>>
>>>>
>>>>
>>>> --
>>>> -Todd
>>>>
>>>
>>>
>>>
>>> --
>>> -Todd
>>>
>>
>
>
> --
> -Todd
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/lldb-dev/attachments/20151022/ca17ba7a/attachment.html>