[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: guile-2.0 and debian
From: |
Thomas Morley |
Subject: |
Re: guile-2.0 and debian |
Date: |
Wed, 23 Nov 2016 10:33:10 +0100 |
2016-11-23 9:23 GMT+01:00 Antonio Ospite <address@hidden>:
> On Wed, 23 Nov 2016 00:25:03 +0100
> Thomas Morley <address@hidden> wrote:
>
> [...]
>> Hi Antonio,
>>
>> I figured to do a regtest-comparison between builds with guile 1.8.8
>> and guile 2.0.13:
>>
>> For that I had to get back guile 1.8.8 and did a build from current master,
>> then I did 'make test-baseline'.
>> Then I copied the entire folder 'lilypond-git/build/input' elsewhere.
>>
>> As second step I got guile 2.0.13 back
>> (Which is pretty tedious, because it's not in the distro, even not for
>> Ubuntu 16.10, if I'm not mistaken.)
>> Did a build with your _previous_ patches. (Your mail with the new
>> patch-set came in while it was running already.)
>> Copied 'lilypond-git/build/input' back into the new build.
>> And did 'make check'
>>
>> This is pretty tedious as well. Anyone with a better suggestion?
>>
>
> You could install debian stable in a virtual machine.
>
> Or for a more lightweight approach you can create a debian stable tree
> using debootstrap and run a shell from it in a container with
> systemd-nspawn, this is what I did for my quick tests with guile-1.8.
>
> The same goes for people wanting to try lilypond with guile-2.0.13, in
> that case a debian unstable container is to be used.
>
> I can elaborate more if there is interest.
I've already set up debian unstable in a VM, but somewhere I made a
mistake and didn't try to fix it up to now ...
>
>> I decided not to abort the running process. Even if the results may be
>> outdated they may give some hints for further TODOs.
>> It seems they are all instances of the same already known issue,
>> though. I attach them anyway, other readers of this thread may be
>> interested...
>>
>> Because index.html contains not the linked images, they are attached
>> as screenshots
>>
>
> Some browsers allow to save a complete web page, with all the linked
> images in a directory, at least firefox does.
My last mail wasn't distributed to the mailing-list, because of too
huge attachments. Can't say I'm really surprised.
I'll not try to circumvent it, it would only document an intermediate state.
>
> BTW the results are promising, with my latest patchset the UTF-8
> characters should be rendered fine. The images are not pixel perfect
> because when using guile-2.0 the floating point numbers in the
> postscript output are formatted slightly differently and that results
> in different positioning of the symbols on the score, but I haven't
> looked deeply into that.
The regtest-comparison with your recent patches are fine.
No issue visible!!
I'll attach the file.
The image for test-output-distance.ly will be missing, but that's our
test, if a regtest-comparison did happen at all, so not a problem.
After a quick glance over the other stuff I didn't notice problems.
Cheers,
Harmr
index.html
Description: Text document
- Re: guile-2.0 and debian, (continued)
- Re: guile-2.0 and debian, David Kastrup, 2016/11/26
- Re: guile-2.0 and debian, Thomas Morley, 2016/11/26
- Re: guile-2.0 and debian, Thomas Morley, 2016/11/26
- Re: guile-2.0 and debian, Thomas Morley, 2016/11/28
- Re: guile-2.0 and debian, Antonio Ospite, 2016/11/29
- Re: guile-2.0 and debian, Federico Bruni, 2016/11/24
- RE: guile-2.0 and debian, Andrew Bernard, 2016/11/24
- Re: guile-2.0 and debian,
Thomas Morley <=
- Re: guile-2.0 and debian, Thomas Morley, 2016/11/23
- Re: guile-2.0 and debian, Thomas Morley, 2016/11/24
- Re: guile-2.0 and debian, David Pirotte, 2016/11/23
- Re: guile-2.0 and debian, Thomas Morley, 2016/11/23
- Re: guile-2.0 and debian, Masamichi Hosoda, 2016/11/24
- Re: guile-2.0 and debian, Antonio Ospite, 2016/11/19