[Stackless] Pushing 2.7-slp branch to bitbucket
Anselm Kruis
a.kruis at science-computing.de
Wed Jan 15 23:54:56 CET 2014
Hi,
Bad news: I was unable to disable the end-of-line check.
what I did:
1. Pushed 4878b93d5294 to hg.python.org/stackless. That's the last good changeset.
2. Created a new head from 4878b93d5294 with the modified .hgeol: changeset fc87e2b7a52b.
3. Pushed fc87e2b7a52b to hg.python.org/stackless
4. Merged heads.
5. Tried to push 2.7-slp to hg.python.org/stackless. This failed again
kruis at VARDA /f/fg2/stackless/stackless
$ hg push -b 2.7-slp --new-branch ssh://hg@hg.python.org/stackless
pushing to ssh://hg@hg.python.org/stackless
running ssh hg at hg.python.org "hg -R stackless serve --stdio"
searching for changes
35 changesets found
remote: adding changesets
remote: adding manifests
remote: adding file changes
remote: added 35 changesets with 59 changes to 25 files
remote: error: pretxnchangegroup.eol hook failed: end-of-line check failed:
remote: Stackless/unittests/test_tracing.py in 92a8e005e75c should not have CRLF line endings
remote: Stackless/unittests/test_tracing.py in 5ffff9187628 should not have CRLF line endings
remote: Stackless/unittests/test_tracing.py in d5a022408fc2 should not have CRLF line endings
remote: Stackless/core/stackless_impl.h in d47f6ef396d5 should not have CRLF line endings
remote: transaction abort!
remote: rollback completed
remote: abort: end-of-line check failed:
remote: Stackless/unittests/test_tracing.py in 92a8e005e75c should not have CRLF line endings
remote: Stackless/unittests/test_tracing.py in 5ffff9187628 should not have CRLF line endings
remote: Stackless/unittests/test_tracing.py in d5a022408fc2 should not have CRLF line endings
remote: Stackless/core/stackless_impl.h in d47f6ef396d5 should not have CRLF line endings
6. Backed out the .hgeol change: I checked out fc87e2b7a52b and backed out
fc87e2b7a52b which created changeset 8b0f904e99be.
7. Pushed 8b0f904e99be to hg.python.org/stackless.
8. Merged heads an push to stackless-dev/stackless.
Perhaps we have the same problem on hg.python.org as described here:
https://mail.python.org/pipermail/python-dev/2011-March/110086.html
Perhaps we never nulled the server's checkout.
And now?
I propose to create a new head for 2.7-slp on 8b0f904e99be and reapply
all changes with fixed EOL on top of this head. Then we can close the
current head 24e0073. The same procedure would be required for 2.8-slp,
if we want to push it to hg.stackless.org. Create a new head based on
ff84d23 and reapply/remerge all later changes on it. Then close head
3fe1381.
I created a fork of stackless, that contains the proposed modification
of the history: https://bitbucket.org/akruis/stackless-fix-eol/
Cheers
Anselm
Am 15.01.2014 19:37, schrieb Anselm Kruis:
> Am 15.01.2014 18:58, schrieb Richard Tew:
>> I think the hook does a great job.
>
> So do I. But we need the hook on the "integration" repository. On a
> downstream mirror the hook is useless. :-(
>
> Cheers
> Anselm
>
>
> We've created this problem
>> ourselves, ideally we would have enabled the same hook on the
>> bitbucket repository when we moved there, and things would have been
>> golden.
>>
>> Forgetting to set up the eol stuff is so easy to do, that's why we
>> need the hook - and why I imagine Python.org has it.
>>
>> Cheers,
>> Richard.
>>
>> On 1/16/14, Christian Tismer <tismer at stackless.com> wrote:
>>> While my preference is to toss hg.python.org, because it gives us
>>> nothing but trouble,
>>> I propose to try this:
>>>
>>> - edit .hgeol
>>> - enter <name> = BIN explicitly for the four involved files at the end
>>> - ...(well, you wrote this, below)
>>> get the changes to the server
>>> - undo the .hgeol change
>>>
>>> This is boring stuff, this hook-thing really should not insist to
>>> check intermediate versions...
>>>
>>> On 15/01/14 10:36, Anselm Kruis wrote:
>>>> Any progress so far? I found old discussions on python-dev:
>>>> https://mail.python.org/pipermail/python-dev/2011-March/110063.html
>>>> https://mail.python.org/pipermail/python-dev/2012-May/119214.html
>>>> https://mail.python.org/pipermail/python-committers/2011-May/001685.html
>>>>
>>>>
>>>> Perhaps it is possible to
>>>> - push a change to the the .hgeol file to hg.python.org
>>>> - merge this change back into the stackless-dev/stackless repo on
>>>> bitbucket
>>>> - push everything to hg.python.org
>>>> - revert the change to .hgeol
>>>> But this procedure is really ugly!
>>>>
>>>> If it is not possible to disable this hook, I volunteer to rewrite the
>>>> history. After all it were my commits, that caused all this trouble.
>>>> Fortunately stackless-dev/stackless is not yet public. Therefore it
>>>> shouldn't be to hard, to update all clones. To rewrite the history I
>>>> would create new heads just before my commits, reapply all changes
>>>> with correct line endings and finally close the old heads.
>>>>
>>>> Cheers
>>>> Anselm
>>>>
>>>>
>>>> Am 14.01.2014 12:44, schrieb Anselm Kruis:
>>>>> Now that our integration repository is on bitbucket, we should try to
>>>>> get the checks disabled for hg.python.org/stackless.
>>>>>
>>>>> Cheers
>>>>> Anselm
>>>>>
>>>>> Am 14.01.2014 05:29, schrieb Richard Tew:
>>>>>> Or someone can get the line ending thing disabled for the
>>>>>> hg.python.org repo on the server-side. Even if just for several days
>>>>>> for those revisions to be pushed. That's what we did when importing
>>>>>> the repository IIRC.
>>>>>>
>>>>>> Cheers,
>>>>>> Richard.
>>>>>>
>>>>>> On 1/14/14, Christian Tismer <tismer at stackless.com> wrote:
>>>>>>> I have checked some of these revisions.
>>>>>>> The reports are from versions which are not current.
>>>>>>>
>>>>>>> My hope was that the versions are only checked in the most
>>>>>>> recent revision, which was unfortunately not the case.
>>>>>>>
>>>>>>> I think hg.python.org is pushing the rules too far here.
>>>>>>> Anyway, if we want to check out changes to hg.python.org,
>>>>>>> I fear we have to re-write history.
>>>>>>>
>>>>>>> This seems to affect four files.
>>>>>>> We would need to:
>>>>>>>
>>>>>>> - undo all the changes from where they occurred, fix the files, and
>>>>>>> re-apply
>>>>>>> the changes. This is a lot of work.
>>>>>>>
>>>>>>> - this also means history re-writing.
>>>>>>>
>>>>>>> cheers - Chris
>>>>>>>
>>>>>>>
>>>>>>> On 13.01.14 06:38, Richard Tew wrote:
>>>>>>>> D:\VCS\HG\Python\stackless>hg push -b 2.7-slp --new-branch
>>>>>>>> ssh://hg@hg.python.org/stackless
>>>>>>>> *** failed to import extension xeol: No module named xeol
>>>>>>>> pushing to ssh://hg@hg.python.org/stackless
>>>>>>>> searching for changes
>>>>>>>> remote: adding changesets
>>>>>>>> remote: adding manifests
>>>>>>>> remote: adding file changes
>>>>>>>> remote: added 68 changesets with 159 changes to 62 files
>>>>>>>> remote: error: pretxnchangegroup.eol hook failed: end-of-line check
>>>>>>>> failed:
>>>>>>>> remote: Stackless/unittests/test_tracing.py in 92a8e005e75c
>>>>>>>> should
>>>>>>>> not have CRLF line endings
>>>>>>>> remote: Stackless/unittests/test_tracing.py in 5ffff9187628
>>>>>>>> should
>>>>>>>> not have CRLF line endings
>>>>>>>> remote: Stackless/unittests/test_tracing.py in d5a022408fc2
>>>>>>>> should
>>>>>>>> not have CRLF line endings
>>>>>>>> remote: Stackless/core/stackless_impl.h in d47f6ef396d5 should
>>>>>>>> not
>>>>>>>> have CRLF line endings
>>>>>>>> remote: transaction abort!
>>>>>>>> remote: rollback completed
>>>>>>>> remote: abort: end-of-line check failed:
>>>>>>>> remote: Stackless/unittests/test_tracing.py in 92a8e005e75c
>>>>>>>> should
>>>>>>>> not have CRLF line endings
>>>>>>>> remote: Stackless/unittests/test_tracing.py in 5ffff9187628
>>>>>>>> should
>>>>>>>> not have CRLF line endings
>>>>>>>> remote: Stackless/unittests/test_tracing.py in d5a022408fc2
>>>>>>>> should
>>>>>>>> not have CRLF line endings
>>>>>>>> remote: Stackless/core/stackless_impl.h in d47f6ef396d5 should
>>>>>>>> not
>>>>>>>> have CRLF line endings
>>>>>>>>
>>>>>>>> Hmm...
>>>>>>>>
>>>>>>>> Any ideas?
>>>>>>>>
>>>>>>>> Cheers,
>>>>>>>> Richard.
>>>>>>>>
>>>>>>>> _______________________________________________
>>>>>>>> Stackless mailing list
>>>>>>>> Stackless at stackless.com
>>>>>>>> http://www.stackless.com/mailman/listinfo/stackless
>>>>>>>
>>>>>>>
>>>>>>> --
>>>>>>> Christian Tismer :^) <mailto:tismer at stackless.com>
>>>>>>> Software Consulting : Have a break! Take a ride on
>>>>>>> Python's
>>>>>>> Karl-Liebknecht-Str. 121 : *Starship*
>>>>>>> http://starship.python.net/
>>>>>>> 14482 Potsdam : PGP key ->
>>>>>>> http://pgp.uni-mainz.de
>>>>>>> phone +49 173 24 18 776 fax +49 (30) 700143-0023
>>>>>>> PGP 0x57F3BF04 9064 F4E1 D754 C2FF 1619 305B C09C 5A3B 57F3
>>>>>>> BF04
>>>>>>> whom do you want to sponsor today?
>>>>>>> http://www.stackless.com/
>>>>>>>
>>>>>>>
>>>>>>> _______________________________________________
>>>>>>> Stackless mailing list
>>>>>>> Stackless at stackless.com
>>>>>>> http://www.stackless.com/mailman/listinfo/stackless
>>>>>>>
>>>>>>
>>>>>> _______________________________________________
>>>>>> Stackless mailing list
>>>>>> Stackless at stackless.com
>>>>>> http://www.stackless.com/mailman/listinfo/stackless
>>>>>>
>>>>>
>>>>
>>>
>>>
>>> --
>>> Christian Tismer :^) <mailto:tismer at stackless.com>
>>> Software Consulting : Have a break! Take a ride on Python's
>>> Karl-Liebknecht-Str. 121 : *Starship* http://starship.python.net/
>>> 14482 Potsdam : PGP key -> http://pgp.uni-mainz.de
>>> phone +49 173 24 18 776 fax +49 (30) 700143-0023
>>> PGP 0x57F3BF04 9064 F4E1 D754 C2FF 1619 305B C09C 5A3B 57F3 BF04
>>> whom do you want to sponsor today? http://www.stackless.com/
>>>
>>>
>>> _______________________________________________
>>> Stackless mailing list
>>> Stackless at stackless.com
>>> http://www.stackless.com/mailman/listinfo/stackless
>>>
>>
>> _______________________________________________
>> Stackless mailing list
>> Stackless at stackless.com
>> http://www.stackless.com/mailman/listinfo/stackless
>>
>
--
Dipl. Phys. Anselm Kruis science + computing ag
Senior Solution Architect Ingolstädter Str. 22
email A.Kruis at science-computing.de 80807 München, Germany
phone +49 89 356386 874 fax 737 www.science-computing.de
--
Vorstandsvorsitzender/Chairman of the board of management:
Gerd-Lothar Leonhart
Vorstand/Board of Management:
Dr. Bernd Finkbeiner, Michael Heinrichs,
Dr. Arno Steitz, Dr. Ingrid Zech
Vorsitzender des Aufsichtsrats/
Chairman of the Supervisory Board:
Philippe Miltin
Sitz/Registered Office: Tuebingen
Registergericht/Registration Court: Stuttgart
Registernummer/Commercial Register No.: HRB 382196
More information about the Stackless
mailing list