[R] Do not show a "message d'avis" that qbeta reports

[Marc Girondot]

Philippe has convinced me that the translation "Message d'avis" was the 
best based on the constraints in the way R displays messages. I agree 
that Avertissement was much better but it was too long.

Then the solution is to force the console messages to be in English when 
there is something strange:

 > warning("Essai")
Message d'avis :
Essai
 > Sys.setenv(LANG = "en")
 > warning("Essai")
Warning message:
Essai

Thanks a lot to the translation team for their great work.

Marc

Le 21/10/2021 à 07:38, SciViews a écrit :
> Martin Maechler wrote:
>> Currently, the French team is just one person
>
> This is not quite true. The people that contributed to the translation in French are listed here: https://github.com/phgrosjean/rfrench <https://github.com/phgrosjean/rfrench>. As you can see, there is a total of nine people among which two have actively contributed to the update of the translations for R 4.1. The page https://developer.r-project.org/TranslationTeams.html <https://developer.r-project.org/TranslationTeams.html> only exhibits the name of the contact person for each team.
>
> Regarding the translation of “Warning” into “Message d’avis” raised by Marc, I gave him an extended explanation offline, that I summarise here. “Warning” was finally translated into “avis” after considering “avertissement”, which is closer to the meaning in the context. However, we got problems with “avertissement” because some sentences became longer than in English. This produced sometimes weird display of the whole message, given that there are (was?) hard breaks in the middle of some of them.
>
> Finally, it is not “warning” that is translated to “message d’avis”, but it is “warning message” in the original sentences. For instance, for R 4.1.0 I, base/po/R-fr.po, I got line 1660: “Summary of (a total of %d) warning messages:” I don’t think that the translation proposed by Marc is good in that context. It would give "Résumé (d’un total de %d) attentions:” Indeed, it is currently Résumé (d’un total de %d) messages d’avis:”
>
> All the best,
>
> Philippe Grosjean
>
>> On 20 Oct 2021, at 22:11, Martin Maechler <maechler using stat.math.ethz.ch> wrote:
>>
>>>>>>> Duncan Murdoch
>>>>>>>     on Wed, 20 Oct 2021 07:17:30 -0400 writes:
>>> Translations are done by the translation teams, listed
>>> here:
>>> <https://developer.r-project.org/TranslationTeams.html>.
>>> If you'd like to offer help, you could contact someone on
>>> that list.
>>> Duncan Murdoch
>> Indeed.   Currently, the French team is just one person, and
>> maybe they will be happy to share the workload  or to get
>> another pair of eyes to look at it...
>>
>> Martin Maechler
>>
>>
>>> On 20/10/2021 6:47 a.m., Marc Girondot via R-help wrote:
>>>> Thanks ! It works.
>>>>
>>>> So "Warnings" has been translated in French by "Message
>>>> d'avis :" !  > warning("Essai") Message d'avis : Essai
>>>>
>>>> It is not the best translation...  I would prefer:
>>>> "Attention:"
>>>>
>>>> Marc
>>>>
>>>> Le 20/10/2021 à 12:28, Enrico Schumann a écrit :
>>>>> On Wed, 20 Oct 2021, Marc Girondot via R-help writes:
>>>>>
>>>>>> Dear R-helpers
>>>>>>
>>>>>> Do you know how to not show a "message d'avis" that
>>>>>> qbeta reports. suppressMessages and capture.output are
>>>>>> not working.
>>>>>>
>>>>>> (PS I Know that qbeta with shape2 being 1e-7 is
>>>>>> "strange"... this is obtained during an optimization
>>>>>> process. Just I don't want see the messages).
>>>>>>
>>>>>> Thanks
>>>>>>
>>>>>> Marc Girondot
>>>>>>
>>>>>> q <- qbeta(p=c(0.025, 0.975), shape1 = 3.3108797,
>>>>>> shape2 = 0.0000001) Message d'avis : Dans qbeta(p =
>>>>>> c(0.025, 0.975), shape1 = 3.3108797, shape2 = 1e-07) :
>>>>>>    qbeta(a, *) =: x0 with |pbeta(x0,*) - alpha| =
>>>>>> 0.024997 is not accurate
>>>>>>
>>>>>> suppressMessages(q <- qbeta(p=c(0.025, 0.975), shape1 =
>>>>>> 3.3108797, shape2 = 0.0000001)) Message d'avis : Dans
>>>>>> qbeta(p = c(0.025, 0.975), shape1 = 3.3108797, shape2 =
>>>>>> 1e-07) :   qbeta(a, *) =: x0 with |pbeta(x0,*) - alpha|
>>>>>> = 0.024997 is not accurate
>>>>>>
>>>>>> capture.output(q <- qbeta(p=c(0.025, 0.975), shape1 =
>>>>>> 3.3108797, shape2 = 0.0000001), type = "message")
>>>>>> character(0) Message d'avis : Dans qbeta(p = c(0.025,
>>>>>> 0.975), shape1 = 3.3108797, shape2 = 1e-07) :
>>>>>> qbeta(a, *) =: x0 with |pbeta(x0,*) - alpha| = 0.024997
>>>>>> is not accurate
>>>>>>
>>>>>> capture.output(q <- qbeta(p=c(0.025, 0.975), shape1 =
>>>>>> 3.3108797, shape2 = 0.0000001), type = "output")
>>>>>> character(0) Message d'avis : Dans qbeta(p = c(0.025,
>>>>>> 0.975), shape1 = 3.3108797, shape2 = 1e-07) :
>>>>>> qbeta(a, *) =: x0 with |pbeta(x0,*) - alpha| = 0.024997
>>>>>> is not accurate
>>>>>>
>>>>> Try 'suppressWarnings'.
>>>>>
>>>>>
>>>> ______________________________________________
>>>> R-help using r-project.org mailing list -- To UNSUBSCRIBE and
>>>> more, see https://stat.ethz.ch/mailman/listinfo/r-help
>>>> PLEASE do read the posting guide
>>>> http://www.R-project.org/posting-guide.html and provide
>>>> commented, minimal, self-contained, reproducible code.
>>>>
>>> ______________________________________________
>>> R-help using r-project.org mailing list -- To UNSUBSCRIBE and
>>> more, see https://stat.ethz.ch/mailman/listinfo/r-help
>>> PLEASE do read the posting guide
>>> http://www.R-project.org/posting-guide.html and provide
>>> commented, minimal, self-contained, reproducible code.
>> ______________________________________________
>> R-help using r-project.org mailing list -- To UNSUBSCRIBE and more, see
>> https://stat.ethz.ch/mailman/listinfo/r-help
>> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
>> and provide commented, minimal, self-contained, reproducible code.
>
> 	[[alternative HTML version deleted]]
>
> ______________________________________________
> R-help using r-project.org mailing list -- To UNSUBSCRIBE and more, see
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.