upgrading R in sage ?

37 views
Skip to first unread message

Emmanuel Charpentier

unread,
Jan 24, 2013, 11:38:08 AM1/24/13
to sage-s...@googlegroups.com
Since a couple of versions ago, the package installation mechanism in R has changed, and some packages are no longer installable. Case in point : rjags, which needs code, whose installation fro CRAN packages will fail due to different "lazy loading" management.

Reading the SPKG file of the standard spkg for R makes me think that installing a newer R version could be as simple as creating a newer branch, installing the newer R source in that branch and recompiling. Is that correct ?

I tried to search relevant trac tickets, but this is not easy : "R" is not a good target for trac searching...

More general note : tghe R "community" is somewhat adamantine about running an up-to-date R : perusing the r-help mailing list, one can often see questions on older R version, almost always getting the standard answer "upgrade to <current version>  R." having at least an *optional* r-latest spkg might be useful...

Any advice ?

Volker Braun

unread,
Jan 24, 2013, 11:48:07 AM1/24/13
to sage-s...@googlegroups.com
Search for "r-project". We are trying to add that keyword to all R-related tickets. As you say, just searching for "R" is pointless ;-)

Jeroen Demeyer

unread,
Jan 24, 2013, 1:55:53 PM1/24/13
to sage-s...@googlegroups.com
Go for it! Create a Trac ticket and post a new spkg.

Emmanuel Charpentier

unread,
Jan 24, 2013, 3:01:33 PM1/24/13
to sage-s...@googlegroups.com
[ Snip... ]
> Any advice ? 
Go for it!  Create a Trac ticket and post a new spkg.

I will do that. But not immediately : RealLife(TM) has its exigences...

                                                                  Emmanuel Charpentier

Emmanuel Charpentier

unread,
Jan 24, 2013, 6:07:56 PM1/24/13
to sage-s...@googlegroups.com
[ Later ] almost done :
dropping the new source code was a breeze. I had to delete the install_parallel patch, which does something already in the new source. Installation was uneventful, and the new code do fix my gripe : rjags and coda can be installed now, and seem functional.

However, I have two ptestlong failures, (both expectable in hindsight...) :

sage -t  --long -force_lib devel/sage/sage/interfaces/r.py
**********************************************************************
File "/home/sage/sage-5.6/devel/sage-main/sage/interfaces/r.py", line 555:
    sage: r.version()
Expected:
    ((2, 14, 0), 'R version 2.14.0 (2011-10-31)')
Got:
    ((2, 15, 2), 'R version 2.15.2 (2012-10-26)')
**********************************************************************
File "/home/sage/sage-5.6/devel/sage-main/sage/interfaces/r.py", line 2049:
    sage: r.version()
Expected:
    ((2, 14, 0), 'R version 2.14.0 (2011-10-31)')
Got:
    ((2, 15, 2), 'R version 2.15.2 (2012-10-26)')
**********************************************************************
2 items had failures:
   1 of   4 in __main__.example_16
   1 of   4 in __main__.example_80
***Test Failed*** 2 failures.

I do not (yet) know where to find these doctests and how to present the relevant patches. Any pointers ? My view of the sage development process is not yet absolutely clear...

                                                                  Emmanuel Charpentier
 
 

Emmanuel Charpentier

unread,
Jan 24, 2013, 7:06:23 PM1/24/13
to sage-s...@googlegroups.com
Go for it!  Create a Trac ticket and post a new spkg. 

Done. That's ticket #14008 now. Still incomplete (lacking the patches for doctests, which will be done probably tomorrow).

Dima Pasechnik

unread,
Jan 24, 2013, 9:21:13 PM1/24/13
to sage-s...@googlegroups.com
On 2013-01-24, Emmanuel Charpentier <emanuel.c...@gmail.com> wrote:
> ------=_Part_98_13353589.1359068876230
> Content-Type: text/plain; charset=ISO-8859-1
> Content-Transfer-Encoding: quoted-printable
>
> Le jeudi 24 janvier 2013 21:01:33 UTC+1, Emmanuel Charpentier a =E9crit :
>
>> [ Snip... ]
>> > Any advice ?=20
>>
>>> Go for it! Create a Trac ticket and post a new spkg.=20
>>>
>>
>> I will do that. But not immediately : RealLife(TM) has its exigences...
>>
>> [ Later ] almost done :
> dropping the new source code was a breeze. I had to delete the=20
> install_parallel patch, which does something already in the new source.=20
> Installation was uneventful, and the new code do fix my gripe : rjags and=
>=20
> coda can be installed now, and seem functional.
>
> However, I have two ptestlong failures, (both expectable in hindsight...) :
>
these are trivial to fix: create a patch for
$SAGE_ROOT/devel/sage/sage/interfaces/r.py
and upload it on the ticket

(no need to create a new branch or anything, as this is a very harmless
patch, which shouldn't break anything of importance in your Sage
installation)

> sage -t --long -force_lib devel/sage/sage/interfaces/r.py
> **********************************************************************
> File "/home/sage/sage-5.6/devel/sage-main/sage/interfaces/r.py", line 555:
> sage: r.version()
> Expected:
> ((2, 14, 0), 'R version 2.14.0 (2011-10-31)')
> Got:
> ((2, 15, 2), 'R version 2.15.2 (2012-10-26)')
> **********************************************************************
> File "/home/sage/sage-5.6/devel/sage-main/sage/interfaces/r.py", line 2049:
> sage: r.version()
> Expected:
> ((2, 14, 0), 'R version 2.14.0 (2011-10-31)')
> Got:
> ((2, 15, 2), 'R version 2.15.2 (2012-10-26)')
> **********************************************************************
> 2 items had failures:
> 1 of 4 in __main__.example_16
> 1 of 4 in __main__.example_80
> ***Test Failed*** 2 failures.
>
> I do not (yet) know where to find these doctests and how to present the=20
> relevant patches. Any pointers ? My view of the sage development process is=
>=20
> not yet absolutely clear...
>
> Emmanuel=
>=20
>> Charpentier
>>
>=20
>=20
>
> --=20
> You received this message because you are subscribed to the Google Groups "=
> sage-support" group.
> To post to this group, send email to sage-s...@googlegroups.com.
> To unsubscribe from this group, send email to sage-support+unsubscribe@goog=
> legroups.com.
> Visit this group at http://groups.google.com/group/sage-support?hl=3Den.
>
>
>
> ------=_Part_98_13353589.1359068876230
> Content-Type: text/html; charset=ISO-8859-1
> Content-Transfer-Encoding: quoted-printable
>
> Le jeudi 24 janvier 2013 21:01:33 UTC+1, Emmanuel Charpentier a =E9crit&nbs=
> p;:<br><blockquote class=3D"gmail_quote" style=3D"margin: 0;margin-left: 0.=
> 8ex;border-left: 1px #ccc solid;padding-left: 1ex;">[ Snip... ]<div>&gt; An=
> y advice ?&nbsp;<br><blockquote class=3D"gmail_quote" style=3D"margin:0;mar=
> gin-left:0.8ex;border-left:1px #ccc solid;padding-left:1ex">Go for it! &nbs=
> p;Create a Trac ticket and post a new spkg.
><br></blockquote><div><br></div><div>I will do that. But not immediately : =
> RealLife(TM) has its exigences...</div></div><div><br></div></blockquote><d=
> iv>[ Later ] almost done :</div><div>dropping the new source code was a bre=
> eze. I had to delete the install_parallel patch, which does something alrea=
> dy in the new source. Installation was uneventful, and the new code do fix =
> my gripe : rjags and coda can be installed now, and seem functional.</div><=
> div><br></div><div>However, I have two ptestlong failures, (both expectable=
> in hindsight...) :</div><div><div><br></div><div>sage -t &nbsp;--long -for=
> ce_lib devel/sage/sage/interfaces/r.py</div><div>**************************=
> ********************************************</div><div>File "/home/sage/sag=
> e-5.6/devel/sage-main/sage/interfaces/r.py", line 555:</div><div>&nbsp; &nb=
> sp; sage: r.version()</div><div>Expected:</div><div>&nbsp; &nbsp; ((2, 14, =
> 0), 'R version 2.14.0 (2011-10-31)')</div><div>Got:</div><div>&nbsp; &nbsp;=
> ((2, 15, 2), 'R version 2.15.2 (2012-10-26)')</div><div>******************=
> ****************************************************</div><div>File "/home/=
> sage/sage-5.6/devel/sage-main/sage/interfaces/r.py", line 2049:</div><div>&=
> nbsp; &nbsp; sage: r.version()</div><div>Expected:</div><div>&nbsp; &nbsp; =
> ((2, 14, 0), 'R version 2.14.0 (2011-10-31)')</div><div>Got:</div><div>&nbs=
> p; &nbsp; ((2, 15, 2), 'R version 2.15.2 (2012-10-26)')</div><div>*********=
> *************************************************************</div><div>2 i=
> tems had failures:</div><div>&nbsp; &nbsp;1 of &nbsp; 4 in __main__.example=
> _16</div><div>&nbsp; &nbsp;1 of &nbsp; 4 in __main__.example_80</div><div>*=
> **Test Failed*** 2 failures.</div></div><div><br></div><div>I do not (yet) =
> know where to find these doctests and how to present the relevant patches. =
> Any pointers ? My view of the sage development process is not yet absolutel=
> y clear...</div><div><br></div><blockquote class=3D"gmail_quote" style=3D"m=
> argin: 0;margin-left: 0.8ex;border-left: 1px #ccc solid;padding-left: 1ex;"=
>><div></div><div>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &n=
> bsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; =
> &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp=
> ; &nbsp; &nbsp; &nbsp; Emmanuel Charpentier</div></blockquote><div>&nbsp;</=
> div><div>&nbsp;</div>
>
><p></p>
>
> -- <br />
> You received this message because you are subscribed to the Google Groups &=
> quot;sage-support&quot; group.<br />
> To post to this group, send email to sage-s...@googlegroups.com.<br />
> To unsubscribe from this group, send email to sage-support+unsubscribe@goog=
> legroups.com.<br />
> Visit this group at <a href=3D"http://groups.google.com/group/sage-support?=
> hl=3Den">http://groups.google.com/group/sage-support?hl=3Den</a>.<br />
> &nbsp;<br />
> &nbsp;<br />
>
> ------=_Part_98_13353589.1359068876230--
>

Emmanuel Charpentier

unread,
Jan 25, 2013, 9:11:55 AM1/25/13
to sage-s...@googlegroups.com
Patch uploaded, new spkg available here. "make ptestlong" tells me that "all test passed".  I think that trac#14008 is ready for a review...

Thank you to all people who have guided my first steps in the sage contribution labyrinth.

                                                                                   Emmanuel Charpentier

Reply all
Reply to author
Forward
0 new messages