[sage-devel] Could some kind soul review #24969 ?

Trac#24969 upgrades R to the recently released 3.4.4. Having it in Sage 8.2 could be useful to Sage's R users (which are not legion, conceded...). As usual with R, Review on Mac and Windows are more than welcome... -- Emmanuel Charpentier -- You

[sage-devel] Re: Internet access during tests

On 2018-04-07, Thierry wrote: > I am sorry to insist, but what is wrong with adding > > # optional -- internet > > on doctests that might connect to the internet ? 1. As I was reminded by Jeroen, internet access is blocked during package installation.

Re: [sage-devel] Re: Internet access during tests

On Sat, Apr 07, 2018 at 10:10:09AM +, Simon King wrote: > > Hi Dima, > > On 2018-04-07, Dima Pasechnik wrote: > > You know, VW has vast experience with designing this sort of "tests" ;-) > > Yes, it reminds of that. Sorry. > > But what else could I do? In interactive

[sage-devel] Re: Internet access during tests

Hi Dima, On 2018-04-07, Dima Pasechnik wrote: > By the way, there are (optional) GAP packages included in Sage that fetch > things from > the internet, namely atlasrep. > > Atlasrep does caching of the data it downloads, though. So does p_group_cohomology. It ships a

[sage-devel] Re: Internet access during tests

Hi Dima, On 2018-04-07, Dima Pasechnik wrote: > You know, VW has vast experience with designing this sort of "tests" ;-) Yes, it reminds of that. Sorry. But what else could I do? In interactive sessions I do not see that problem (actually even when the interactive session

[sage-devel] Re: Internet access during tests

On Saturday, April 7, 2018 at 11:27:35 AM UTC+2, Simon King wrote: > > Hi Thierry, > > On 2018-04-07, Thierry wrote: > > What is wrong with adding the following to the tests that try to connect > > to the internet, whether they absolutely depend on it or not ? >

[sage-devel] Re: Internet access during tests

Hi Simon, On Saturday, April 7, 2018 at 10:07:43 AM UTC+2, Simon King wrote: > > On 2018-04-06, Thierry wrote: > >> 1. Is it really true that Sage currently does NOT block internet > >>access during doctests, i.e., has returned to the situation more > >>

[sage-devel] Re: Internet access during tests

Hi Thierry, On 2018-04-07, Thierry wrote: > What is wrong with adding the following to the tests that try to connect > to the internet, whether they absolutely depend on it or not ? > > # optional -- internet It would mean that these tests wouldn't be

Re: [sage-devel] Re: Internet access during tests

On Sat, Apr 07, 2018 at 08:05:21AM +, Simon King wrote: > Hi Thierry, > > On 2018-04-06, Thierry wrote: > >> 1. Is it really true that Sage currently does NOT block internet > >>access during doctests, i.e., has returned to the situation more > >>than

[sage-devel] Re: Internet access during tests

Thierry wrote: > sudo iptables -A OUTPUT -j ACCEPT -o lo > sudo iptables -A OUTPUT -j ACCEPT -o 127.0.0.1 > sudo iptables -A OUTPUT -j DROP > > But i guess this is not what you want. I don't know the details, but it looks like nowadays there are ways to do something a little more fine-grained:

[sage-devel] Re: Internet access during tests

Hi Thierry, On 2018-04-06, Thierry wrote: >> 1. Is it really true that Sage currently does NOT block internet >>access during doctests, i.e., has returned to the situation more >>than three years ago? > > I am not sure internet access was ever blocked.

[sage-devel] Re: Newest Xcode is causing doctest failures

Hi John, haven't you forgotten the update of command-line tools? could you post the outputs of your softwareupdate and sysctl -n machdep.cpu.brand_string (see example on https://trac.sagemath.org/ticket/25113) I cannot reproduce these errors. Dima On Saturday, March 31, 2018 at 9:46:41 PM