[Debian-science-sagemath] GAP: issue related to compressed manual.six: PATCHES: reproducing issue

Ximin Luo infinity0 at debian.org
Fri Dec 16 23:05:00 UTC 2016


Jerome BENOIT:
> Hello Again,
> 
> On 14/12/16 23:19, Jerome BENOIT wrote:
>> Thanks for the patch. I am on my way to build the involved package
>> for the repository deb-sci-deb .
> 
>> Jerome
> 
>> On 14/12/16 19:46, Bill Allombert wrote:
>>> On Mon, Dec 12, 2016 at 10:19:18PM +0100, Bill Allombert wrote:
>>>> On Mon, Dec 12, 2016 at 06:06:37PM +0100, Bill Allombert wrote:
>>>>> On Sun, Dec 11, 2016 at 11:15:00PM +0000, Ximin Luo wrote:
>>>>>> Did you remove test.w before trying the tests? I think it is still
>>>>>> somehow autpgrp-related. Also the "corrupted" messages seem to be
>>>>>> separate from the actual failure of "no method found":
>>>>>
>>>>> I agree with you. But the "corrupted" messages are still a bug even
>>>>> if they do not affect Sage.
>>>>
>>>> I can reproduce this bug using a pristine gap installation without any
>>>> Debian stuff:
>>>>
>>>> rm -f workspace file file.gz
>>>> echo "abcdefgh" > file
>>>> gzip file
>>>> echo 'SaveWorkspace("workspace");' | bin/*/gap -q -b
>>>> echo 'ReadLine(InputTextFile("file"));' | bin/*/gap -q -b
>>>> echo 'ReadLine(InputTextFile("file"));' | bin/*/gap -q -b -L workspace
>>>>
>>>> I will report it upstream.
> 
>>> Upstream send me the attached patch which fix this problem.
>>> Please confirm this address the original issue.
> 
> I deposited a patched version of GAP at deb-sci-sage (4r8p6-1+sage23).
> 
> The test
> 
> ( cd sage; ./sage -t --long src/sage/interfaces/gap.py )
> 
> passes now. But, in a previous email, Ximin seems to speak about a work around.
> So, for plain confirmation, Ximin will gives the final answer.
> 

The work-around was just to uninstall gap-autpgrp (and the other packages with old manuals) when running sage and its tests.

Thankfully, I can confirm this patch works, even when gap-autpgrp is present.

$ sage -c 'print(gap.help("SymmetricGroup", pager=False)[:100])'
  
  50 Group Libraries
  
  When you start GAP, it already knows several groups. Currently GAP init

$ sudo aptitude install gap-autpgrp 
[..]
$ sage -c 'print(gap.help("SymmetricGroup", pager=False)[:100])'
Traceback (most recent call last):
  File "/usr/share/sage/bin/sage-eval", line 10, in <module>
    eval(compile(s,'<cmdline>','exec'))
  File "<cmdline>", line 1, in <module>
  File "/usr/lib/python2.7/dist-packages/sage/interfaces/gap.py", line 1334, in help
    line = Expect.eval(self, "? %s"%s)
  File "/usr/lib/python2.7/dist-packages/sage/interfaces/expect.py", line 1299, in eval
    for L in code.split('\n') if L != ''])
  File "/usr/lib/python2.7/dist-packages/sage/interfaces/gap.py", line 772, in _eval_line
    raise RuntimeError(message)
RuntimeError: Gap produced error output
Error, no method found! For debugging hints type ?Recovery from NoMethodFound
Error, no 1st choice method found for `+' on 2 arguments

   executing ? SymmetricGroup
# exit code 1

$ sudo aptitude install -t sid-sage gap gap-core gap-libs gap-online-help
$ sage -c 'print(gap.help("SymmetricGroup", pager=False)[:100])'
  
  50 Group Libraries
  
  When you start GAP, it already knows several groups. Currently GAP init

I also got similar results when running the gap.py doctests.

X

-- 
GPG: ed25519/56034877E1F87C35
GPG: rsa4096/1318EFAC5FBBDBCE
https://github.com/infinity0/pubkeys.git



More information about the Debian-science-sagemath mailing list