[sage-release] Re: Sage 9.1.beta9 released

On Wednesday, April 1, 2020 at 3:02:49 PM UTC-7, Matthias Köppe wrote: > > Another variation to try is to use " . .homebrew-build-env " > before the build, which sets some environment variables so that homebrew's > "keg-only packages" are found. > Among other things, this affects libpng via zl

[sage-release] Re: Sage 9.1.beta9 released

Another variation to try is to use " . .homebrew-build-env " before the build, which sets some environment variables so that homebrew's "keg-only packages" are found. Among other things, this affects libpng via zlib. On Wednesday, April 1, 2020 at 2:05:31 PM UTC-4, John H Palmieri wrote: > > >

[sage-release] Re: Sage 9.1.beta9 released

On Wednesday, April 1, 2020 at 11:05:31 AM UTC-7, John H Palmieri wrote: > > > > On Tuesday, March 31, 2020 at 7:15:10 PM UTC-7, John H Palmieri wrote: >> >> >> >> On Tuesday, March 31, 2020 at 6:21:42 PM UTC-7, Matthias Köppe wrote: >>> >>> On Tuesday, March 31, 2020 at 8:02:48 PM UTC-4, John H

[sage-release] Re: Sage 9.1.beta9 released

On Tuesday, March 31, 2020 at 7:15:10 PM UTC-7, John H Palmieri wrote: > > > > On Tuesday, March 31, 2020 at 6:21:42 PM UTC-7, Matthias Köppe wrote: >> >> On Tuesday, March 31, 2020 at 8:02:48 PM UTC-4, John H Palmieri wrote: >>> >>> Here's the strange part: if I add the package libpng, then Sage

Re: [sage-release] Re: Sage 9.1.beta9 released

> Am 31.03.2020 um 23:25 schrieb Matthias Köppe : > > That is, the file `local/bin/sage-starts` is not found. This is with CentOS > 7.7. > > > This script was moved in https://trac.sagemath.org/ticket/23769 > It looks like the patchbot needs to be upda

Re: [sage-release] Re: Sage 9.1.beta9 released

> Am 31.03.2020 um 23:30 schrieb Matthias Köppe : > Looks like this new doctest (from https://trac.sagemath.org/ticket/29287) > needs some os.path.abspath in it. Please open a ticket for this. Ok, I have opened https://trac.sagemath.org/ticket/29446 for

[sage-release] Re: Sage 9.1.beta9 released

On Tuesday, March 31, 2020 at 6:21:42 PM UTC-7, Matthias Köppe wrote: > > On Tuesday, March 31, 2020 at 8:02:48 PM UTC-4, John H Palmieri wrote: >> >> Here's the strange part: if I add the package libpng, then Sage doesn't >> build. With these packages, it builds and passes tests (output from "b

[sage-release] Re: Sage 9.1.beta9 released

On Tuesday, March 31, 2020 at 8:02:48 PM UTC-4, John H Palmieri wrote: > > Here's the strange part: if I add the package libpng, then Sage doesn't > build. With these packages, it builds and passes tests (output from "brew > list --versions"): > > autoconf 2.69 > automake 1.16.2 > boost 1.72.0_1

[sage-release] Re: Sage 9.1.beta9 released

On Tuesday, March 31, 2020 at 8:02:48 PM UTC-4, John H Palmieri wrote: > > [...] > kiwisolver fails to build. For some reason, something in the build process > thinks that "-sdk" is a good flag to add to "g++": > > g++ -sdk macosx clang -bundle -undefined dynamic_lookup > -L/Users/palmieri/D

[sage-release] Re: Sage 9.1.beta9 released

Some good news and some weird news: in my attempts to debug this, I uninstalled a bunch of my homebrew packages, and I got Sage to build and pass tests. Yay. Here's the strange part: if I add the package libpng, then Sage doesn't build. With these packages, it builds and passes tests (output fr

[sage-release] Re: Sage 9.1.beta9 released

On Tuesday, March 31, 2020 at 1:40:19 PM UTC-4, Steven Trogdon wrote: > > I rebuilt Sage using the Sage-provided python instead my system python and > doctested Sage. The failure in doctest/test.py disappears. However, the > failures in rings/integer.pyx still persist. I suspect a system componen

[sage-release] Re: Sage 9.1.beta9 released

On Tuesday, March 31, 2020 at 4:18:08 PM UTC-4, Markus Wageringel wrote: > > Since this beta, I get the following permanent failure: > > sage -t --long src/sage/env.py > ** > File "src/sage/env.py", line 15, in sage.env > Failed ex

[sage-release] Re: Sage 9.1.beta9 released

On Tuesday, March 31, 2020 at 4:18:08 PM UTC-4, Markus Wageringel wrote: > > my patchbot failed multiple times with the following possibly related > error: > > ... > Sage build/upgrade complete! > make[1]: Verzeichnis „/amd/compute/sagebot/sage“ wird verlassen > sh: /amd/compute

[sage-release] Re: Sage 9.1.beta9 released

Since this beta, I get the following permanent failure: sage -t --long src/sage/env.py ** File "src/sage/env.py", line 15, in sage.env Failed example: out == repr((SAGE_ROOT, SAGE_LOCAL))

[sage-release] Re: Sage 9.1.beta9 released

I also see the error in rings/integer.pyx again now. This is https://trac.sagemath.org/ticket/28106. The doctests are limited to 3300 MB of memory be default, but this test exceeds it. The limit can be manually increased using the `--memlimit` option. Am Dienstag, 31. März 2020 19:40:19 UTC+2

[sage-release] Re: Sage 9.1.beta9 released

Testing with many optional and external packages, I get Using --optional=4ti2,cbc,ccache,cryptominisat,dot2tex,e_antic,external,glucose,latte_int,lidia,lrslib,memlimit,normaliz,notedown,openssl,pandoc_attributes,pycosat,pynormaliz,rst2ipynb,sage,sage_numerical_backends_coin,sage_numerical_backen

[sage-release] Re: Sage 9.1.beta9 released

I'm replying to my previous post. I rebuilt Sage using the Sage-provided python instead my system python and doctested Sage. The failure in doctest/test.py disappears. However, the failures in rings/integer.pyx still persist. I suspect a system component is the culprit. Any ideas which one? I h

Re: [sage-release] Re: Sage 9.1.beta9 released

On Monday, March 30, 2020 at 3:03:14 PM UTC-4, vdelecroix wrote: > > > Le 29/03/2020 à 21:50, Matthias Köppe a écrit : > > > > Let me remark that this is the first beta that tries to use the system > > python3 in a venv (#27824) instead of building our own copy of python3. > > Tests of building

Re: [sage-release] Re: Sage 9.1.beta9 released

On Tue, Mar 31, 2020 at 9:35 PM Steven Trogdon wrote: > > A feature or a bug? 'make distclean && make' always uses the system python to > build Sage. Is there some way to force building and using the Sage-supplied > python? sure, you can use ./configure -h to see how to have packages built Name

[sage-release] Re: Sage 9.1.beta9 released

A feature or a bug? 'make distclean && make' always uses the system python to build Sage. Is there some way to force building and using the Sage-supplied python? -- You received this message because you are subscribed to the Google Groups "sage-release" group. To unsubscribe from this group an

Re: [sage-release] Re: Sage 9.1.beta9 released

On Tue, Mar 31, 2020 at 7:46 PM Simon King wrote: > > Hi all, > > On 2020-03-31, Matthias Köppe wrote: > >> > I don't know, but I found these instructions: > >> > https://launchpad.net/~deadsnakes/+archive/ubuntu/ppa > >> > that also mention what packages should be installed > >> > >> Excellent!

[sage-release] Re: Sage 9.1.beta9 released

Hi all, On 2020-03-31, Matthias Köppe wrote: >> > I don't know, but I found these instructions: >> > https://launchpad.net/~deadsnakes/+archive/ubuntu/ppa >> > that also mention what packages should be installed >> >> Excellent! After installing them and doing ./configure, I see this line: >>

[sage-release] Re: Sage 9.1.beta9 released

On Monday, March 30, 2020 at 8:59:06 PM UTC-4, Simon King wrote: > > On 2020-03-31, Matthias Köppe > wrote: > > On Monday, March 30, 2020 at 8:30:34 PM UTC-4, Simon King wrote: > >> > >> I installed python3 after "sudo add-apt-repository ppa:deadsnakes/ppa". > >> Has that been a mistake? > >>

[sage-release] Re: Sage 9.1.beta9 released

Dear Matthias, On 2020-03-31, Matthias Köppe wrote: > On Monday, March 30, 2020 at 8:30:34 PM UTC-4, Simon King wrote: >> >> > Are you sure your /usr/bin/python comes from that distribution? As far >> as I >> > can see, xenial only has 3.5.1-3. >> >> I installed python3 after "sudo add-apt-rep

[sage-release] Re: Sage 9.1.beta9 released

On Monday, March 30, 2020 at 8:30:34 PM UTC-4, Simon King wrote: > > > Are you sure your /usr/bin/python comes from that distribution? As far > as I > > can see, xenial only has 3.5.1-3. > > I installed python3 after "sudo add-apt-repository ppa:deadsnakes/ppa". > Has that been a mistake? > >

[sage-release] Re: Sage 9.1.beta9 released

On 2020-03-30, Matthias Köppe wrote: > Hi Simon, > > On Monday, March 30, 2020 at 6:45:17 PM UTC-4, Simon King wrote: >> Concerning too old: >> $ uname -a >> Linux klap 4.15.0-91-generic #92~16.04.1-Ubuntu SMP Fri Feb 28 14:57:22 >> UTC 2020 x86_64 x86_64 x86_64 GNU/Linux >> $ cat /etc/issue

[sage-release] Re: Sage 9.1.beta9 released

On Monday, March 30, 2020 at 6:59:30 PM UTC-4, John H Palmieri wrote: > > I see the same in the log file for the Python 2 build. I wonder if this > has to do with upgrading Xcode, which I seem to have done 6 days ago. To be > clear, yesterday and today I have built 9.1.beta8 + Python 3 and 9.1.b

[sage-release] Re: Sage 9.1.beta9 released

On Monday, March 30, 2020 at 7:16:26 PM UTC-4, Matthias Köppe wrote: > > Hi Simon, > > On Monday, March 30, 2020 at 6:45:17 PM UTC-4, Simon King wrote: >> >> Trying what src/doc/en/installation/debian*.txt tells me: >> Note, selecting 'zlib1g-dev' instead of 'libz-dev' >> E: Unable to locate pack

[sage-release] Re: Sage 9.1.beta9 released

Hi Simon, On Monday, March 30, 2020 at 6:45:17 PM UTC-4, Simon King wrote: > > Trying what src/doc/en/installation/debian*.txt tells me: > Note, selecting 'zlib1g-dev' instead of 'libz-dev' > E: Unable to locate package libflint-arb-dev > E: Unable to locate package libplanarity-dev > E: Unabl

[sage-release] Re: Sage 9.1.beta9 released

I see the same in the log file for the Python 2 build. I wonder if this has to do with upgrading Xcode, which I seem to have done 6 days ago. To be clear, yesterday and today I have built 9.1.beta8 + Python 3 and 9.1.beta9 + Python 2, both successfully, despite this crash. It's 9.1.beta9 + Pyth

[sage-release] Re: Sage 9.1.beta9 released

Hi Matthias, On 2020-03-30, Matthias Köppe wrote: > This is outdated. We now generate an up-to-date version of this manual > section at boostrapping time. Use "cat src/doc/en/installation/debian*.txt" > to print it. > > Or re-run configure and take a look at the system package installation > h

[sage-release] Re: Sage 9.1.beta9 released

On Monday, March 30, 2020 at 5:54:19 PM UTC-4, Steven Trogdon wrote: > > *Gentoo [...]* > >> >> Please help with https://trac.sagemath.org/ticket/29105 so that builds on top of Gentoo can be properly tested. -- You received this message because you are subscribed to the Google Groups "sage-rel

[sage-release] Re: Sage 9.1.beta9 released

Hi Simon, On Monday, March 30, 2020 at 6:20:24 PM UTC-4, Simon King wrote: > > Let me add further disappointments: > > After making sure that python3.7.7 is installed on my laptop and after > installing the recommended system packages mentioned in > > http://doc.sagemath.org/html/en/installatio

[sage-release] Re: Sage 9.1.beta9 released

Hi! On 2020-03-30, Dima Pasechnik wrote: > On Tue, 31 Mar 2020, 03:03 Vincent Delecroix, <20100.delecr...@gmail.com> > wrote: >> Le 29/03/2020 à 21:50, Matthias Köppe a écrit : >> > >> > Let me remark that this is the first beta that tries to use the system >> > python3 in a venv (#27824) instead

[sage-release] Re: Sage 9.1.beta9 released

*Gentoo* Incremental build 9.1.beta8 -> 9.1.beta9 After make python3-clean && make I have the following failures: -- sage -t --long --warn-long 115.0 src/sage/doctest/test.py # 1 doctest failed sage -t --long --warn-long 115.0

[sage-release] Re: Sage 9.1.beta9 released

>From your config.log ... rather suspicious: configure:10189: g++ -std=gnu++11 -o conftest -g -O2 conftest.cpp -lgmp -lm >&5 configure:10189: $? = 0 configure:10206: result: -lgmp configure:10231: will use system package and not install SPKG mpir configure:10265: checking absolute name of co

Re: [sage-release] Re: Sage 9.1.beta9 released

I ran "brew upgrade" and it only upgraded a few packages: tox, cmake, boost, automake, and some port of emacs. I rebuilt Sage and it didn't help. (Just to be sure, I also built 9.1.beta8, and that one passes tests.) On Monday, March 30, 2020 at 9:30:26 AM UTC-7, Dima Pasechnik wrote: > > I vagu

Re: [sage-release] Re: Sage 9.1.beta9 released

On Tue, 31 Mar 2020, 03:03 Vincent Delecroix, <20100.delecr...@gmail.com> wrote: > > > Le 29/03/2020 à 21:50, Matthias Köppe a écrit : > > > > Let me remark that this is the first beta that tries to use the system > > python3 in a venv (#27824) instead of building our own copy of python3. > > Test

Re: [sage-release] Re: Sage 9.1.beta9 released

Le 29/03/2020 à 21:50, Matthias Köppe a écrit : Let me remark that this is the first beta that tries to use the system python3 in a venv (#27824) instead of building our own copy of python3. Tests of building from scratch or (after "make python3-clean") would be valuable. Very disappointing

Re: [sage-release] Re: Sage 9.1.beta9 released

On Monday, March 30, 2020 at 12:23:00 PM UTC-4, John H Palmieri wrote: > > I should also mention that I have built every beta from scratch for Python > 2 and 3 for quite a while, and these failures are new. So if the gcc in > homebrew is getting picked up, it's because of some new change in Sage.

Re: [sage-release] Re: Sage 9.1.beta9 released

I vaguely recall seeing similar errors on Homebrew, they were due to stale Homebrew. Could you try updating it? On Tue, Mar 31, 2020 at 12:23 AM John H Palmieri wrote: > > I should also mention that I have built every beta from scratch for Python 2 > and 3 for quite a while, and these failures a

Re: [sage-release] Re: Sage 9.1.beta9 released

I should also mention that I have built every beta from scratch for Python 2 and 3 for quite a while, and these failures are new. So if the gcc in homebrew is getting picked up, it's because of some new change in Sage. On Sunday, March 29, 2020 at 10:01:31 PM UTC-7, Dima Pasechnik wrote: > > Co

Re: [sage-release] Re: Sage 9.1.beta9 released

On Mon, Mar 30, 2020 at 10:14 PM Thierry wrote: > > On Mon, Mar 30, 2020 at 10:04:46PM +0800, Dima Pasechnik wrote: > > On Mon, 30 Mar 2020 at 21:47, Thierry > > wrote: > > > > > Hi, > > > > > > On Sun, Mar 29, 2020 at 12:50:18PM -0700, Matthias Köppe wrote: > > > > On Sunday, March 29, 2020 at 8

Re: [sage-release] Re: Sage 9.1.beta9 released

On Mon, Mar 30, 2020 at 10:04:46PM +0800, Dima Pasechnik wrote: > On Mon, 30 Mar 2020 at 21:47, Thierry > wrote: > > > Hi, > > > > On Sun, Mar 29, 2020 at 12:50:18PM -0700, Matthias Köppe wrote: > > > On Sunday, March 29, 2020 at 8:59:43 AM UTC-4, Volker Braun wrote: > > > > > > > > As always, yo

Re: [sage-release] Re: Sage 9.1.beta9 released

On Mon, 30 Mar 2020 at 21:47, Thierry wrote: > Hi, > > On Sun, Mar 29, 2020 at 12:50:18PM -0700, Matthias Köppe wrote: > > On Sunday, March 29, 2020 at 8:59:43 AM UTC-4, Volker Braun wrote: > > > > > > As always, you can get the latest beta version from the "develop" git > > > branch. Alternative

Re: [sage-release] Re: Sage 9.1.beta9 released

Hi, On Sun, Mar 29, 2020 at 12:50:18PM -0700, Matthias Köppe wrote: > On Sunday, March 29, 2020 at 8:59:43 AM UTC-4, Volker Braun wrote: > > > > As always, you can get the latest beta version from the "develop" git > > branch. Alternatively, the self-contained source tarball is at > > http://www

Re: [sage-release] Re: Sage 9.1.beta9 released

On Monday, March 30, 2020 at 1:49:28 AM UTC-4, Dima Pasechnik wrote: > > On Mon, Mar 30, 2020 at 1:12 PM Matthias Köppe > wrote: > > > > By the way, current homebrew does not seem to have Python 3.7.x. > really? > https://github.com/Homebrew/homebrew-core/blob/master/Formula/python.rb > still

[sage-release] Re: Sage 9.1.beta9 released

Volker Braun wrote: > I don't see any major issues for 9.1 any more, so we should soon have > the first rc. If you have anything that you thing should go in then > now is the time to finish it up ;-) I'd really like #28517 to go in, because it fixes some serious issues reported by a user attempti

[sage-release] Re: Sage 9.1.beta9 released

On Debian testing running on core i7 + 16 GB RAM, upgrading 9.1.beta6 to 9.1.beta9 gets 13 permanent errors : -- sage -t --long src/sage/schemes/elliptic_curves/ell_number_field.py # 5 doctests failed sage -t --long src/sage/sc

Re: [sage-release] Re: Sage 9.1.beta9 released

On Mon, Mar 30, 2020 at 1:12 PM Matthias Köppe wrote: > > By the way, current homebrew does not seem to have Python 3.7.x. really? https://github.com/Homebrew/homebrew-core/blob/master/Formula/python.rb still shows 3.7.7 > I am not sure how to test with historical snapshots of the homebrew > di

[sage-release] Re: Sage 9.1.beta9 released

By the way, current homebrew does not seem to have Python 3.7.x. I am not sure how to test with historical snapshots of the homebrew distribution. On Monday, March 30, 2020 at 1:08:44 AM UTC-4, Matthias Köppe wrote: > > Please post config.log and output of "brew list --versions" > > On Monday, M

[sage-release] Re: Sage 9.1.beta9 released

Please post config.log and output of "brew list --versions" On Monday, March 30, 2020 at 12:36:25 AM UTC-4, John H Palmieri wrote: > > On OS X with a homebrew installation of Python 3.7.7 I get some failures, > mostly to do with pari, including messages like this: > > PARI/GP ERROR: > *** bug

Re: [sage-release] Re: Sage 9.1.beta9 released

Could it be that gcc in Homebrew gets picked to compile C/C++ code? On Mon, Mar 30, 2020 at 12:36 PM John H Palmieri wrote: > > On OS X with a homebrew installation of Python 3.7.7 I get some failures, > mostly to do with pari, including messages like this: > > PARI/GP ERROR: > *** bug in PA

[sage-release] Re: Sage 9.1.beta9 released

On OS X with a homebrew installation of Python 3.7.7 I get some failures, mostly to do with pari, including messages like this: PARI/GP ERROR: *** bug in PARI/GP (Segmentation Fault), please report. and this: fatal error: 'gmp.h' file not found #include "gmp.h" ^~~

[sage-release] Re: Sage 9.1.beta9 released

On Sunday, March 29, 2020 at 8:59:43 AM UTC-4, Volker Braun wrote: > > As always, you can get the latest beta version from the "develop" git > branch. Alternatively, the self-contained source tarball is at > http://www.sagemath.org/download-latest.html > > I don't see any major issues for 9.1 an