The packages needing packages external to sage to be installed raise the question wether optional packages are allowed to have dependencies other then sage. To me it feels a bit weird if you first need to install something outside of sage before you can install the package, however this seems like a minor issue related to the patchbot failures.
On Friday, 15 September 2017 22:01:40 UTC+2, Maarten Derickx wrote: > > In my ongoing quest of trying to get fewer patchbot failures I started at > least with installing all optional packages. I now have every optional > package returned by:: > > sage: optional_packages() > > except the following: > > gmp - this package installs fine but one has to choose between mpir > and gmp > igraph -this needs libxml2 could by solved by apt get libxml2-dev > python_igraph - you wouldn't guess, but this depends on igraph > libtheora - a known problem #23732 > mpfrcx - added so recently (#11806) that the tar is not on the mirror > network yet > > Sage builds fine after this but the documentation suddenly does not build > anymore because it misses Graphviz. The total memory used by this install > is 15 GB (7 GB more then the 8.3 GB it had before installing the optional > packages). > > The doctest don't pass anymore, to be precise there are 48 failures, there > were 3 that I saw before and I know the trac tickets of. The other 45 can > be found at https://trac.sagemath.org/ticket/23871 if any of the files > affected look familiar and you know an open ticket for it please let me > know. I am now working on sorting out the 45 failures at #23871 . > > > > > On Friday, 15 September 2017 10:31:58 UTC+2, vdelecroix wrote: >> >> In case you did not notice, you are just preventing an important >> discussion concerning the management of optional packages to happen... >> >> On 15/09/2017 10:28, Jeroen Demeyer wrote: >> > On 2017-09-15 09:00, Jori Mäntysalo wrote: >> >> If so, a >> >> bug in Sage could -- at least in theory -- lead a compromise to the >> >> server. >> > >> > It's a *feature* of Sage (actually Python) that it can run arbitrary >> > code, not a bug. >> > >> > -- You received this message because you are subscribed to the Google Groups "sage-devel" group. To unsubscribe from this group and stop receiving emails from it, send an email to sage-devel+unsubscr...@googlegroups.com. To post to this group, send email to sage-devel@googlegroups.com. Visit this group at https://groups.google.com/group/sage-devel. For more options, visit https://groups.google.com/d/optout.