Open embray opened 6 years ago
Setting new milestone based on a cursory review of ticket status, priority, and last modification date.
Setting a new milestone for this ticket based on a cursory review.
Open as of Sage 10.3.beta0:
cunningham_tables
(no DESTDIR)
frobby
(no DESTDIR -- fixed in #36839)gap3
(no DESTDIR)lie
(no DESTDIR)polymake
(no DESTDIR)polytopes_db_4d
(no DESTDIR)saclib
(no DESTDIR)
This is a meta-ticket for tickets that address updating other packages to use the helper functions in
sage-dist-helpers
where possible. There are a few advantages to doing this:The second bullet point is especially motivated right now by #22509. Since it requires updates to nearly every package in Sage anyways, it makes more sense to update most packages to use the helper functions, and then add
DESTDIR
support through the helper functions. In particular, the vast majority of Python and autotools packages can be givenDESTDIR
support en masse by this approach.24014 and #24018 -- Update Python packages that are installed with
pip
to usesdh_pip_install
24025 -- Update "trivial" autotools packages to use
sdh_configure
,sdh_make
, andsdh_install
. By "trivial" I mean packages with very simple build scripts that can be mechanically updated with simple replacements.25038 Update some standard packages that were excluded from #24025 due to conflicts with other existing tickets
#24692 - gdb (optional)#26021 - tides (optional)27039 Update some optional packages that use autotools in a standard way but were just excluded from #24025 due to not being standard packages
????? Same as previous, but for experimental packages:
25043 - python2/python3
????? - compilerwrapper (experimental)
????? - gambit (optional)
22626 - gap
26856 - gap_packages (optional)
27016 - gcc (optional)
????? - git (optional)
#25001 #27016 - gfortran (optional)????? - gmp (optional)
25045 - mpfr
25086 - nauty
25087 - ppl
25049 - r
????? - termcap (has a
configure
from a very old autoconf)????? - topcom (its existing spkg-install uses
install-strip
instead ofinstall
and one can see why: the debugging info for this library is quite large for some reason; need an alternative tosdh_make_install
that allows alternate install targets)configure
but they still usemake
in such a way that can be replaced withsdh_make
). It might also be possible to narrow down one or two other classes of packages that can be simplified with new helper functions.25035 - arb - has a hand-written
configure
script????? - cmake (optional)
29150 - coxeter3 (optional)
25035 - flint- has a hand-written
configure
script25042 - freetype - has a hand-written
configure
script????? - gap3 (experimental) - do we care?
25100 - gfan
25100 - lcalc
????? - lie (experimental)
29148 - mcqd (optional)
????? - modular_decomposition (experimental)
27511 - normaliz (optional)
25037 - ntl - has a hand-written
configure
script25052 - openblas
????? - openssl
25144 - palp
25143 - pari
????? - perl_term_readline_gnu (experimental)
27398 - plantri (optional)
????? - polymake (experimental)
????? - qepcad (experimental)
25100 - ratpoints
25100 - rubiks
????? - saclib (optional)
????? - scipoptsuite (experimental)
????? - sip (optional)
25100 - symmetrica
25099 - sympow
25048 - tachyon
25142 - zlib - has a hand-written
configure
script25085 - zn_poly -
configure
is just a wrapper for a custom Python script calledmakemakefile.py
pip install
does not suffice:25092 - cysignals
25051 - numpy
25051 - pillow
25051 - pip
????? - scons (optional)
25051 - setuptools
$SAGE_DESTDIR_LOCAL
.25147 - boost_cropped
25040 - combinatorial_designs
25040 - graphs
????? - jmol
25039 - mathjax
25040 - pari_galdata
25040 - pari_seadata_small
25040 - polytopes_db
25039 - thebe
25040 - threejs
????? - atlas (optional)
????? - conway_polynomials
????? - elliptic_curves
Component: build
Issue created by migration from https://trac.sagemath.org/ticket/24024