R devel - Dec 2017 - R CMD check warning about compiler warning flags

On 25/12/2017 7:00 AM, I?aki ?car wrote:
> 2017-12-25 12:30 GMT+01:00 Duncan Murdoch <murdoch.duncan at
gmail.com>:
>> The one negative aspect of Winston's effort is caused by this
weakness. If
>> you tell me that something happened in revision 73909, I know it was
recent.
>> If you tell me that something appeared in commit 2e80059, it wastes my
time
>> looking up that commit and putting it in context.
> 
> The output from "git describe" is something similar to an svn
> revision: it describes the number of commits since the previous
> annotated tag. Something like
<tag_name>-<#commits>-g<commit_hash>,
> e.g., v3.5.0-56-1234567.
But we're not using git, we're just sending emails back and forth.  I 
don't need to run svn to get some useful information from the number 
73909.  Winston's web page at 
https://github.com/wch/r-source/commit/2e80059 does display the number 
73909 if you know where to look, but his email doesn't contain it.

I don't know about other svn users, but I'd be perfectly content to read
something like "This appears to have been added in rev 73909 (diff shown 
here: https://github.com/wch/r-source/commit/2e80059)."

Duncan Murdoch

Maybe I'm new, and forgive my ignorance, but maybe in the future (~ X years
from now) the R Project could be managed entirely from github, by doing
pull requests and only R Core having commit rights...

Would make the forking process also easier... And could be a good roadmap.

But we're not using git, we're just sending emails back and forth.  I
don't
> need to run svn to get some useful information from the number 73909.
> Winston's web page at https://github.com/wch/r-source/commit/2e80059
does
> display the number 73909 if you know where to look, but his email
doesn't
> contain it.
>
> I don't know about other svn users, but I'd be perfectly content to
read
> something like "This appears to have been added in rev 73909 (diff
shown
> here: https://github.com/wch/r-source/commit/2e80059)."
>
> Duncan Murdoch
	[[alternative HTML version deleted]]

On 26 December 2017 at 00:00, Juan Telleria <jtelleriar at gmail.com>
wrote:
> Maybe I'm new, and forgive my ignorance, but maybe in the future (~ X
years
> from now) the R Project could be managed entirely from github, by doing
I strongly disagree. Are you aware that github is a commercial
company, github inc. [1] ?
What about gitlab? or Microsoft's codeplex? There are other services
similar to github, why github?
What happens if github goes out of business?

R-project should be maintained in the academic network and under
auspices of universities.


 [*]  GitHub, Inc.
       https://en.wikipedia.org/wiki/GitHub