Re: [sage-release] Sage 8.9 released

2019-10-11 Thread Dima Pasechnik
On Fri, 11 Oct 2019, 16:04 Thierry,  wrote:

> On Fri, Oct 11, 2019 at 02:50:41PM -0500, Dima Pasechnik wrote:
> > I can't post there on ask.* from phone, I believe that installing
> > libncurses5-dev and libreadline-dev should allow the build to use system
> > readline.
>
> Unfortunately, what i understand from the post, the user does not have
> root acces, hence can not install packages.
>

asking the sysadmin to install two bog-standard OS packages should not be a
problem...


> > this is how we made it work on John's machines.
> > Naturally, the *.m4 files must not be removed,
>
> This was just a test to ensure that our version of readline will be
> installed.
>
> Ciao,
> Thierry
>
> > and Sage's ncurses package uninstalled with
> >
> > make ncurses-clean
> >
> > then
> >
> > ./configure
> >
> > make
> >
> > On Fri, 11 Oct 2019, 12:14 Thierry, 
> wrote:
> >
> > > Hi,
> > >
> > > there is also a readline-related issue on 8.9 there:
> > >
> > >
> https://ask.sagemath.org/question/48276/building-sage-89-from-source-fails/
> > >
> > > Ciao,
> > > Thierry
> > >
> > >
> > >
> > > On Wed, Oct 02, 2019 at 01:46:00PM -0500, Dima Pasechnik wrote:
> > > > On Wed, Oct 2, 2019 at 11:27 AM John Cremona  >
> > > wrote:
> > > > >
> > > > >
> > > > >
> > > > > On Wed, 2 Oct 2019 at 09:07, John Cremona 
> > > wrote:
> > > > >>
> > > > >>
> > > > >>
> > > > >> On Wed, 2 Oct 2019 at 02:22, Dima Pasechnik 
> > > wrote:
> > > > >>>
> > > > >>> something did not work as you installed libncurses5-dev,
> > > > >>> as you still appear to miss
> > > > >>> /usr/include/ncurses.h
> > > > >>> (according to the new log:
> > > > >>> configure:10315: checking for ncurses.h
> > > > >>> configure:10315: result: no )
> > > > >>>
> > > > >>> whereas
> > > https://packages.ubuntu.com/xenial/amd64/libncurses5-dev/filelist
> > > > >>> has this file listed...
> > > > >>>
> > > > >>> Could you check that
> > > > >>> /usr/include/ncurses.h
> > > > >>> is present on the box?
> > > > >>
> > > > >>
> > > > >> It was not, but I think that yesterday I was confused and had
> > > installed it on a different machine from the one I reported on (too
> many
> > > windows open).  I just checked by manually installing that package on
> all 6
> > > machines, which did do the install in 5 cases while on the 6th it was
> > > already there -- and that 6th machine was one for which the original
> build
> > > had worked OK!  (Yesterday I said that all 6 had failed the same way
> but in
> > > fact 5 had, the 6th was still building when I reported).
> > > > >>
> > > > >> For completeness I am attaching he latest config.log.
> > > > >>
> > > > >> I think that Samuel might have hit on something I did wrong.  My
> bash
> > > history shows
> > > > >>
> > > > >>  1989  tar zxf 8.9.tar.gz
> > > > >>  1990  cd sage-8.9
> > > > >>  1991  make configure
> > > > >>  1992  make
> > > > >>
> > > > >> with no ./configure step.  I must have thought that "make
> configure"
> > > was instead of that...
> > > > >>
> > > > >> I'll restart the 5 builds after doing ./configure (after
> > > ncurses-clean and readline-clean), and report back.
> > > > >
> > > > >
> > > > > All went successfully.   My only remaining question is: is I had
> done
> > > ./configure aafter make configure and before make, would the build have
> > > worked (without inctalling those ubuntu packages which had somehow
> never
> > > been needed for previous builds)?
> > > > >
> > > > "make configure" is only needed if Sage's main configure script is
> > > updated.
> > > > Just look at the target "configure" in the main Makefile
> > > >
> > > > I guess there could be a bug introduced in commit
> > > > 584735d9abe44b079df2566f7f73b6a69aaea0cf
> > > > as I mentioned above in the thread.
> > > > Feel free to open a ticket, saying that Sage's 8.9 readline builds
> > > > underlinked on ubuntu xenial
> > > >
> > > > > John
> > > > >
> > > > >>>
> > > > >>>
> > > > >>> On Wed, Oct 2, 2019 at 3:20 AM John Cremona <
> john.crem...@gmail.com>
> > > wrote:
> > > > >>> >
> > > > >>> >
> > > > >>> >
> > > > >>> > On Tue, 1 Oct 2019 at 18:20, Dima Pasechnik  >
> > > wrote:
> > > > >>> >>
> > > > >>> >> Hi John,
> > > > >>> >>
> > > > >>> >> Do you have
> > > > >>> >> libncurses5-dev
> > > > >>> >>  installed?
> > > > >>> >
> > > > >>> >
> > > > >>> > No only the package without -dev.
> > > > >>> >
> > > > >>> >>
> > > > >>> >> I guess not, as in your config.log I see
> > > > >>> >>
> > > > >>> >> conftest.cpp:56:21: fatal error: ncurses.h: No such file or
> > > directory
> > > > >>> >>
> > > > >>> >> and this header is provided by this package.
> > > > >>> >> I'd say it's a bug in Ubunty xenial that libreadline-dev does
> not
> > > > >>> >> depend on this package!
> > > > >>> >>
> > > > >>> >> Could you try installing it, then run
> > > > >>> >>
> > > > >>> >> make ncurses-clean
> > > > >>> >> make readline-clean
> > > > >>> >> ./configure
> > > > >>> >
> > > > >>> >
> > > > >>> > I did that.
> > > > >>> >
> > > > >>> >>
> > > > 

Re: [sage-release] Sage 8.9 released

2019-10-11 Thread Thierry
On Fri, Oct 11, 2019 at 02:50:41PM -0500, Dima Pasechnik wrote:
> I can't post there on ask.* from phone, I believe that installing
> libncurses5-dev and libreadline-dev should allow the build to use system
> readline.

Unfortunately, what i understand from the post, the user does not have
root acces, hence can not install packages.

> this is how we made it work on John's machines.
> Naturally, the *.m4 files must not be removed, 

This was just a test to ensure that our version of readline will be
installed.

Ciao,
Thierry

> and Sage's ncurses package uninstalled with
> 
> make ncurses-clean
> 
> then
> 
> ./configure
> 
> make
> 
> On Fri, 11 Oct 2019, 12:14 Thierry,  wrote:
> 
> > Hi,
> >
> > there is also a readline-related issue on 8.9 there:
> >
> > https://ask.sagemath.org/question/48276/building-sage-89-from-source-fails/
> >
> > Ciao,
> > Thierry
> >
> >
> >
> > On Wed, Oct 02, 2019 at 01:46:00PM -0500, Dima Pasechnik wrote:
> > > On Wed, Oct 2, 2019 at 11:27 AM John Cremona 
> > wrote:
> > > >
> > > >
> > > >
> > > > On Wed, 2 Oct 2019 at 09:07, John Cremona 
> > wrote:
> > > >>
> > > >>
> > > >>
> > > >> On Wed, 2 Oct 2019 at 02:22, Dima Pasechnik 
> > wrote:
> > > >>>
> > > >>> something did not work as you installed libncurses5-dev,
> > > >>> as you still appear to miss
> > > >>> /usr/include/ncurses.h
> > > >>> (according to the new log:
> > > >>> configure:10315: checking for ncurses.h
> > > >>> configure:10315: result: no )
> > > >>>
> > > >>> whereas
> > https://packages.ubuntu.com/xenial/amd64/libncurses5-dev/filelist
> > > >>> has this file listed...
> > > >>>
> > > >>> Could you check that
> > > >>> /usr/include/ncurses.h
> > > >>> is present on the box?
> > > >>
> > > >>
> > > >> It was not, but I think that yesterday I was confused and had
> > installed it on a different machine from the one I reported on (too many
> > windows open).  I just checked by manually installing that package on all 6
> > machines, which did do the install in 5 cases while on the 6th it was
> > already there -- and that 6th machine was one for which the original build
> > had worked OK!  (Yesterday I said that all 6 had failed the same way but in
> > fact 5 had, the 6th was still building when I reported).
> > > >>
> > > >> For completeness I am attaching he latest config.log.
> > > >>
> > > >> I think that Samuel might have hit on something I did wrong.  My bash
> > history shows
> > > >>
> > > >>  1989  tar zxf 8.9.tar.gz
> > > >>  1990  cd sage-8.9
> > > >>  1991  make configure
> > > >>  1992  make
> > > >>
> > > >> with no ./configure step.  I must have thought that "make configure"
> > was instead of that...
> > > >>
> > > >> I'll restart the 5 builds after doing ./configure (after
> > ncurses-clean and readline-clean), and report back.
> > > >
> > > >
> > > > All went successfully.   My only remaining question is: is I had done
> > ./configure aafter make configure and before make, would the build have
> > worked (without inctalling those ubuntu packages which had somehow never
> > been needed for previous builds)?
> > > >
> > > "make configure" is only needed if Sage's main configure script is
> > updated.
> > > Just look at the target "configure" in the main Makefile
> > >
> > > I guess there could be a bug introduced in commit
> > > 584735d9abe44b079df2566f7f73b6a69aaea0cf
> > > as I mentioned above in the thread.
> > > Feel free to open a ticket, saying that Sage's 8.9 readline builds
> > > underlinked on ubuntu xenial
> > >
> > > > John
> > > >
> > > >>>
> > > >>>
> > > >>> On Wed, Oct 2, 2019 at 3:20 AM John Cremona 
> > wrote:
> > > >>> >
> > > >>> >
> > > >>> >
> > > >>> > On Tue, 1 Oct 2019 at 18:20, Dima Pasechnik 
> > wrote:
> > > >>> >>
> > > >>> >> Hi John,
> > > >>> >>
> > > >>> >> Do you have
> > > >>> >> libncurses5-dev
> > > >>> >>  installed?
> > > >>> >
> > > >>> >
> > > >>> > No only the package without -dev.
> > > >>> >
> > > >>> >>
> > > >>> >> I guess not, as in your config.log I see
> > > >>> >>
> > > >>> >> conftest.cpp:56:21: fatal error: ncurses.h: No such file or
> > directory
> > > >>> >>
> > > >>> >> and this header is provided by this package.
> > > >>> >> I'd say it's a bug in Ubunty xenial that libreadline-dev does not
> > > >>> >> depend on this package!
> > > >>> >>
> > > >>> >> Could you try installing it, then run
> > > >>> >>
> > > >>> >> make ncurses-clean
> > > >>> >> make readline-clean
> > > >>> >> ./configure
> > > >>> >
> > > >>> >
> > > >>> > I did that.
> > > >>> >
> > > >>> >>
> > > >>> >>
> > > >>> >> and check whether system-wide readline is picked up?
> > > >>> >
> > > >>> >
> > > >>> > Now config.log includes these lines
> > > >>> >
> > > >>> > configure:13881: === checking whether to install the readline SPKG
> > ===
> > > >>> > configure:13925: checking Installing ncurses?
> > > >>> > configure:13928: result: Yes. Install readline as well.
> > > >>> > configure:14176: === checking whether to install the pari SPKG ===
> > > >>> > configure:14222: 

Re: [sage-release] Sage 8.9 released

2019-10-11 Thread Dima Pasechnik
I can't post there on ask.* from phone, I believe that installing
libncurses5-dev and libreadline-dev should allow the build to use system
readline.

this is how we made it work on John's machines.
Naturally, the *.m4 files must not be removed, and Sage's ncurses package
uninstalled with

make ncurses-clean

then

./configure

make

On Fri, 11 Oct 2019, 12:14 Thierry,  wrote:

> Hi,
>
> there is also a readline-related issue on 8.9 there:
>
> https://ask.sagemath.org/question/48276/building-sage-89-from-source-fails/
>
> Ciao,
> Thierry
>
>
>
> On Wed, Oct 02, 2019 at 01:46:00PM -0500, Dima Pasechnik wrote:
> > On Wed, Oct 2, 2019 at 11:27 AM John Cremona 
> wrote:
> > >
> > >
> > >
> > > On Wed, 2 Oct 2019 at 09:07, John Cremona 
> wrote:
> > >>
> > >>
> > >>
> > >> On Wed, 2 Oct 2019 at 02:22, Dima Pasechnik 
> wrote:
> > >>>
> > >>> something did not work as you installed libncurses5-dev,
> > >>> as you still appear to miss
> > >>> /usr/include/ncurses.h
> > >>> (according to the new log:
> > >>> configure:10315: checking for ncurses.h
> > >>> configure:10315: result: no )
> > >>>
> > >>> whereas
> https://packages.ubuntu.com/xenial/amd64/libncurses5-dev/filelist
> > >>> has this file listed...
> > >>>
> > >>> Could you check that
> > >>> /usr/include/ncurses.h
> > >>> is present on the box?
> > >>
> > >>
> > >> It was not, but I think that yesterday I was confused and had
> installed it on a different machine from the one I reported on (too many
> windows open).  I just checked by manually installing that package on all 6
> machines, which did do the install in 5 cases while on the 6th it was
> already there -- and that 6th machine was one for which the original build
> had worked OK!  (Yesterday I said that all 6 had failed the same way but in
> fact 5 had, the 6th was still building when I reported).
> > >>
> > >> For completeness I am attaching he latest config.log.
> > >>
> > >> I think that Samuel might have hit on something I did wrong.  My bash
> history shows
> > >>
> > >>  1989  tar zxf 8.9.tar.gz
> > >>  1990  cd sage-8.9
> > >>  1991  make configure
> > >>  1992  make
> > >>
> > >> with no ./configure step.  I must have thought that "make configure"
> was instead of that...
> > >>
> > >> I'll restart the 5 builds after doing ./configure (after
> ncurses-clean and readline-clean), and report back.
> > >
> > >
> > > All went successfully.   My only remaining question is: is I had done
> ./configure aafter make configure and before make, would the build have
> worked (without inctalling those ubuntu packages which had somehow never
> been needed for previous builds)?
> > >
> > "make configure" is only needed if Sage's main configure script is
> updated.
> > Just look at the target "configure" in the main Makefile
> >
> > I guess there could be a bug introduced in commit
> > 584735d9abe44b079df2566f7f73b6a69aaea0cf
> > as I mentioned above in the thread.
> > Feel free to open a ticket, saying that Sage's 8.9 readline builds
> > underlinked on ubuntu xenial
> >
> > > John
> > >
> > >>>
> > >>>
> > >>> On Wed, Oct 2, 2019 at 3:20 AM John Cremona 
> wrote:
> > >>> >
> > >>> >
> > >>> >
> > >>> > On Tue, 1 Oct 2019 at 18:20, Dima Pasechnik 
> wrote:
> > >>> >>
> > >>> >> Hi John,
> > >>> >>
> > >>> >> Do you have
> > >>> >> libncurses5-dev
> > >>> >>  installed?
> > >>> >
> > >>> >
> > >>> > No only the package without -dev.
> > >>> >
> > >>> >>
> > >>> >> I guess not, as in your config.log I see
> > >>> >>
> > >>> >> conftest.cpp:56:21: fatal error: ncurses.h: No such file or
> directory
> > >>> >>
> > >>> >> and this header is provided by this package.
> > >>> >> I'd say it's a bug in Ubunty xenial that libreadline-dev does not
> > >>> >> depend on this package!
> > >>> >>
> > >>> >> Could you try installing it, then run
> > >>> >>
> > >>> >> make ncurses-clean
> > >>> >> make readline-clean
> > >>> >> ./configure
> > >>> >
> > >>> >
> > >>> > I did that.
> > >>> >
> > >>> >>
> > >>> >>
> > >>> >> and check whether system-wide readline is picked up?
> > >>> >
> > >>> >
> > >>> > Now config.log includes these lines
> > >>> >
> > >>> > configure:13881: === checking whether to install the readline SPKG
> ===
> > >>> > configure:13925: checking Installing ncurses?
> > >>> > configure:13928: result: Yes. Install readline as well.
> > >>> > configure:14176: === checking whether to install the pari SPKG ===
> > >>> > configure:14222: checking installing gmp/mpir or readline?
> > >>> > configure:14224: result: yes; install pari as well
> > >>> > configure:14565: result: using Sage's pari SPKG
> > >>> > configure:14587: === checking whether to install the pari_nftables
> SPKG ===
> > >>> > configure:14631: checking installing pari?
> > >>> > configure:14634: result: yes; install pari_nftables as well
> > >>> > configure:14675: === checking whether to install the pkgconf SPKG
> ===
> > >>> > configure:14718: checking for pkg-config >= 0.29
> > >>> > configure:14794: result: /usr/bin/pkg-config
> > >>> > 

Re: [sage-release] Sage 8.9 released

2019-10-11 Thread Dima Pasechnik
On Fri, 11 Oct 2019, 12:36 Thierry,  wrote:

> Thanks for your answer, please could you cc me if you open a ticket ?
>

I will - might take a bit of time, as I am travelling from IMA at UMN back
to UK today.

>
> Ciao,
> Thierry
>
>
> On Fri, Oct 11, 2019 at 12:27:28PM -0500, Dima Pasechnik wrote:
> > yuck, this is a symptome of underlinking in readline.
> > In their infinite wisdom readline's devs refuse to fix this old bug;
> > it needs to be linked with libtinfo or other suchlike library, and it
> > keeps resufacings in various forms...
> >
> > I think I've made a patch to sage's readline for this, but it
> > sometimes breaks for reasons I need to look at.
> >
> > As to external readline, I guess our spkg-configure.m4 will lead one
> > to a broken build if the system's readline is underlinked,
> > cause our setup takes no account of this...
> >
> > So apparently ubuntu 16.04 has an underlinked readline.
> >
> >
> >
> > On Fri, Oct 11, 2019 at 12:14 PM Thierry
> >  wrote:
> > >
> > > Hi,
> > >
> > > there is also a readline-related issue on 8.9 there:
> > >
> > >
> https://ask.sagemath.org/question/48276/building-sage-89-from-source-fails/
> > >
> > > Ciao,
> > > Thierry
> > >
> > >
> > >
> > > On Wed, Oct 02, 2019 at 01:46:00PM -0500, Dima Pasechnik wrote:
> > > > On Wed, Oct 2, 2019 at 11:27 AM John Cremona 
> wrote:
> > > > >
> > > > >
> > > > >
> > > > > On Wed, 2 Oct 2019 at 09:07, John Cremona 
> wrote:
> > > > >>
> > > > >>
> > > > >>
> > > > >> On Wed, 2 Oct 2019 at 02:22, Dima Pasechnik 
> wrote:
> > > > >>>
> > > > >>> something did not work as you installed libncurses5-dev,
> > > > >>> as you still appear to miss
> > > > >>> /usr/include/ncurses.h
> > > > >>> (according to the new log:
> > > > >>> configure:10315: checking for ncurses.h
> > > > >>> configure:10315: result: no )
> > > > >>>
> > > > >>> whereas
> https://packages.ubuntu.com/xenial/amd64/libncurses5-dev/filelist
> > > > >>> has this file listed...
> > > > >>>
> > > > >>> Could you check that
> > > > >>> /usr/include/ncurses.h
> > > > >>> is present on the box?
> > > > >>
> > > > >>
> > > > >> It was not, but I think that yesterday I was confused and had
> installed it on a different machine from the one I reported on (too many
> windows open).  I just checked by manually installing that package on all 6
> machines, which did do the install in 5 cases while on the 6th it was
> already there -- and that 6th machine was one for which the original build
> had worked OK!  (Yesterday I said that all 6 had failed the same way but in
> fact 5 had, the 6th was still building when I reported).
> > > > >>
> > > > >> For completeness I am attaching he latest config.log.
> > > > >>
> > > > >> I think that Samuel might have hit on something I did wrong.  My
> bash history shows
> > > > >>
> > > > >>  1989  tar zxf 8.9.tar.gz
> > > > >>  1990  cd sage-8.9
> > > > >>  1991  make configure
> > > > >>  1992  make
> > > > >>
> > > > >> with no ./configure step.  I must have thought that "make
> configure" was instead of that...
> > > > >>
> > > > >> I'll restart the 5 builds after doing ./configure (after
> ncurses-clean and readline-clean), and report back.
> > > > >
> > > > >
> > > > > All went successfully.   My only remaining question is: is I had
> done ./configure aafter make configure and before make, would the build
> have worked (without inctalling those ubuntu packages which had somehow
> never been needed for previous builds)?
> > > > >
> > > > "make configure" is only needed if Sage's main configure script is
> updated.
> > > > Just look at the target "configure" in the main Makefile
> > > >
> > > > I guess there could be a bug introduced in commit
> > > > 584735d9abe44b079df2566f7f73b6a69aaea0cf
> > > > as I mentioned above in the thread.
> > > > Feel free to open a ticket, saying that Sage's 8.9 readline builds
> > > > underlinked on ubuntu xenial
> > > >
> > > > > John
> > > > >
> > > > >>>
> > > > >>>
> > > > >>> On Wed, Oct 2, 2019 at 3:20 AM John Cremona <
> john.crem...@gmail.com> wrote:
> > > > >>> >
> > > > >>> >
> > > > >>> >
> > > > >>> > On Tue, 1 Oct 2019 at 18:20, Dima Pasechnik 
> wrote:
> > > > >>> >>
> > > > >>> >> Hi John,
> > > > >>> >>
> > > > >>> >> Do you have
> > > > >>> >> libncurses5-dev
> > > > >>> >>  installed?
> > > > >>> >
> > > > >>> >
> > > > >>> > No only the package without -dev.
> > > > >>> >
> > > > >>> >>
> > > > >>> >> I guess not, as in your config.log I see
> > > > >>> >>
> > > > >>> >> conftest.cpp:56:21: fatal error: ncurses.h: No such file or
> directory
> > > > >>> >>
> > > > >>> >> and this header is provided by this package.
> > > > >>> >> I'd say it's a bug in Ubunty xenial that libreadline-dev does
> not
> > > > >>> >> depend on this package!
> > > > >>> >>
> > > > >>> >> Could you try installing it, then run
> > > > >>> >>
> > > > >>> >> make ncurses-clean
> > > > >>> >> make readline-clean
> > > > >>> >> ./configure
> > > > >>> >
> > > > >>> >
> > > > >>> > I did that.
> > > > >>> >
> > 

Re: [sage-release] Sage 8.9 released

2019-10-11 Thread Thierry
Thanks for your answer, please could you cc me if you open a ticket ?

Ciao,
Thierry


On Fri, Oct 11, 2019 at 12:27:28PM -0500, Dima Pasechnik wrote:
> yuck, this is a symptome of underlinking in readline.
> In their infinite wisdom readline's devs refuse to fix this old bug;
> it needs to be linked with libtinfo or other suchlike library, and it
> keeps resufacings in various forms...
> 
> I think I've made a patch to sage's readline for this, but it
> sometimes breaks for reasons I need to look at.
> 
> As to external readline, I guess our spkg-configure.m4 will lead one
> to a broken build if the system's readline is underlinked,
> cause our setup takes no account of this...
> 
> So apparently ubuntu 16.04 has an underlinked readline.
> 
> 
> 
> On Fri, Oct 11, 2019 at 12:14 PM Thierry
>  wrote:
> >
> > Hi,
> >
> > there is also a readline-related issue on 8.9 there:
> >
> > https://ask.sagemath.org/question/48276/building-sage-89-from-source-fails/
> >
> > Ciao,
> > Thierry
> >
> >
> >
> > On Wed, Oct 02, 2019 at 01:46:00PM -0500, Dima Pasechnik wrote:
> > > On Wed, Oct 2, 2019 at 11:27 AM John Cremona  
> > > wrote:
> > > >
> > > >
> > > >
> > > > On Wed, 2 Oct 2019 at 09:07, John Cremona  
> > > > wrote:
> > > >>
> > > >>
> > > >>
> > > >> On Wed, 2 Oct 2019 at 02:22, Dima Pasechnik  wrote:
> > > >>>
> > > >>> something did not work as you installed libncurses5-dev,
> > > >>> as you still appear to miss
> > > >>> /usr/include/ncurses.h
> > > >>> (according to the new log:
> > > >>> configure:10315: checking for ncurses.h
> > > >>> configure:10315: result: no )
> > > >>>
> > > >>> whereas 
> > > >>> https://packages.ubuntu.com/xenial/amd64/libncurses5-dev/filelist
> > > >>> has this file listed...
> > > >>>
> > > >>> Could you check that
> > > >>> /usr/include/ncurses.h
> > > >>> is present on the box?
> > > >>
> > > >>
> > > >> It was not, but I think that yesterday I was confused and had 
> > > >> installed it on a different machine from the one I reported on (too 
> > > >> many windows open).  I just checked by manually installing that 
> > > >> package on all 6 machines, which did do the install in 5 cases while 
> > > >> on the 6th it was already there -- and that 6th machine was one for 
> > > >> which the original build had worked OK!  (Yesterday I said that all 6 
> > > >> had failed the same way but in fact 5 had, the 6th was still building 
> > > >> when I reported).
> > > >>
> > > >> For completeness I am attaching he latest config.log.
> > > >>
> > > >> I think that Samuel might have hit on something I did wrong.  My bash 
> > > >> history shows
> > > >>
> > > >>  1989  tar zxf 8.9.tar.gz
> > > >>  1990  cd sage-8.9
> > > >>  1991  make configure
> > > >>  1992  make
> > > >>
> > > >> with no ./configure step.  I must have thought that "make configure" 
> > > >> was instead of that...
> > > >>
> > > >> I'll restart the 5 builds after doing ./configure (after ncurses-clean 
> > > >> and readline-clean), and report back.
> > > >
> > > >
> > > > All went successfully.   My only remaining question is: is I had done 
> > > > ./configure aafter make configure and before make, would the build have 
> > > > worked (without inctalling those ubuntu packages which had somehow 
> > > > never been needed for previous builds)?
> > > >
> > > "make configure" is only needed if Sage's main configure script is 
> > > updated.
> > > Just look at the target "configure" in the main Makefile
> > >
> > > I guess there could be a bug introduced in commit
> > > 584735d9abe44b079df2566f7f73b6a69aaea0cf
> > > as I mentioned above in the thread.
> > > Feel free to open a ticket, saying that Sage's 8.9 readline builds
> > > underlinked on ubuntu xenial
> > >
> > > > John
> > > >
> > > >>>
> > > >>>
> > > >>> On Wed, Oct 2, 2019 at 3:20 AM John Cremona  
> > > >>> wrote:
> > > >>> >
> > > >>> >
> > > >>> >
> > > >>> > On Tue, 1 Oct 2019 at 18:20, Dima Pasechnik  
> > > >>> > wrote:
> > > >>> >>
> > > >>> >> Hi John,
> > > >>> >>
> > > >>> >> Do you have
> > > >>> >> libncurses5-dev
> > > >>> >>  installed?
> > > >>> >
> > > >>> >
> > > >>> > No only the package without -dev.
> > > >>> >
> > > >>> >>
> > > >>> >> I guess not, as in your config.log I see
> > > >>> >>
> > > >>> >> conftest.cpp:56:21: fatal error: ncurses.h: No such file or 
> > > >>> >> directory
> > > >>> >>
> > > >>> >> and this header is provided by this package.
> > > >>> >> I'd say it's a bug in Ubunty xenial that libreadline-dev does not
> > > >>> >> depend on this package!
> > > >>> >>
> > > >>> >> Could you try installing it, then run
> > > >>> >>
> > > >>> >> make ncurses-clean
> > > >>> >> make readline-clean
> > > >>> >> ./configure
> > > >>> >
> > > >>> >
> > > >>> > I did that.
> > > >>> >
> > > >>> >>
> > > >>> >>
> > > >>> >> and check whether system-wide readline is picked up?
> > > >>> >
> > > >>> >
> > > >>> > Now config.log includes these lines
> > > >>> >
> > > >>> > configure:13881: === checking whether to install the readline SPKG 

Re: [sage-release] Sage 8.9 released

2019-10-11 Thread Dima Pasechnik
yuck, this is a symptome of underlinking in readline.
In their infinite wisdom readline's devs refuse to fix this old bug;
it needs to be linked with libtinfo or other suchlike library, and it
keeps resufacings in various forms...

I think I've made a patch to sage's readline for this, but it
sometimes breaks for reasons I need to look at.

As to external readline, I guess our spkg-configure.m4 will lead one
to a broken build if the system's readline is underlinked,
cause our setup takes no account of this...

So apparently ubuntu 16.04 has an underlinked readline.



On Fri, Oct 11, 2019 at 12:14 PM Thierry
 wrote:
>
> Hi,
>
> there is also a readline-related issue on 8.9 there:
>
> https://ask.sagemath.org/question/48276/building-sage-89-from-source-fails/
>
> Ciao,
> Thierry
>
>
>
> On Wed, Oct 02, 2019 at 01:46:00PM -0500, Dima Pasechnik wrote:
> > On Wed, Oct 2, 2019 at 11:27 AM John Cremona  wrote:
> > >
> > >
> > >
> > > On Wed, 2 Oct 2019 at 09:07, John Cremona  wrote:
> > >>
> > >>
> > >>
> > >> On Wed, 2 Oct 2019 at 02:22, Dima Pasechnik  wrote:
> > >>>
> > >>> something did not work as you installed libncurses5-dev,
> > >>> as you still appear to miss
> > >>> /usr/include/ncurses.h
> > >>> (according to the new log:
> > >>> configure:10315: checking for ncurses.h
> > >>> configure:10315: result: no )
> > >>>
> > >>> whereas 
> > >>> https://packages.ubuntu.com/xenial/amd64/libncurses5-dev/filelist
> > >>> has this file listed...
> > >>>
> > >>> Could you check that
> > >>> /usr/include/ncurses.h
> > >>> is present on the box?
> > >>
> > >>
> > >> It was not, but I think that yesterday I was confused and had installed 
> > >> it on a different machine from the one I reported on (too many windows 
> > >> open).  I just checked by manually installing that package on all 6 
> > >> machines, which did do the install in 5 cases while on the 6th it was 
> > >> already there -- and that 6th machine was one for which the original 
> > >> build had worked OK!  (Yesterday I said that all 6 had failed the same 
> > >> way but in fact 5 had, the 6th was still building when I reported).
> > >>
> > >> For completeness I am attaching he latest config.log.
> > >>
> > >> I think that Samuel might have hit on something I did wrong.  My bash 
> > >> history shows
> > >>
> > >>  1989  tar zxf 8.9.tar.gz
> > >>  1990  cd sage-8.9
> > >>  1991  make configure
> > >>  1992  make
> > >>
> > >> with no ./configure step.  I must have thought that "make configure" was 
> > >> instead of that...
> > >>
> > >> I'll restart the 5 builds after doing ./configure (after ncurses-clean 
> > >> and readline-clean), and report back.
> > >
> > >
> > > All went successfully.   My only remaining question is: is I had done 
> > > ./configure aafter make configure and before make, would the build have 
> > > worked (without inctalling those ubuntu packages which had somehow never 
> > > been needed for previous builds)?
> > >
> > "make configure" is only needed if Sage's main configure script is updated.
> > Just look at the target "configure" in the main Makefile
> >
> > I guess there could be a bug introduced in commit
> > 584735d9abe44b079df2566f7f73b6a69aaea0cf
> > as I mentioned above in the thread.
> > Feel free to open a ticket, saying that Sage's 8.9 readline builds
> > underlinked on ubuntu xenial
> >
> > > John
> > >
> > >>>
> > >>>
> > >>> On Wed, Oct 2, 2019 at 3:20 AM John Cremona  
> > >>> wrote:
> > >>> >
> > >>> >
> > >>> >
> > >>> > On Tue, 1 Oct 2019 at 18:20, Dima Pasechnik  wrote:
> > >>> >>
> > >>> >> Hi John,
> > >>> >>
> > >>> >> Do you have
> > >>> >> libncurses5-dev
> > >>> >>  installed?
> > >>> >
> > >>> >
> > >>> > No only the package without -dev.
> > >>> >
> > >>> >>
> > >>> >> I guess not, as in your config.log I see
> > >>> >>
> > >>> >> conftest.cpp:56:21: fatal error: ncurses.h: No such file or directory
> > >>> >>
> > >>> >> and this header is provided by this package.
> > >>> >> I'd say it's a bug in Ubunty xenial that libreadline-dev does not
> > >>> >> depend on this package!
> > >>> >>
> > >>> >> Could you try installing it, then run
> > >>> >>
> > >>> >> make ncurses-clean
> > >>> >> make readline-clean
> > >>> >> ./configure
> > >>> >
> > >>> >
> > >>> > I did that.
> > >>> >
> > >>> >>
> > >>> >>
> > >>> >> and check whether system-wide readline is picked up?
> > >>> >
> > >>> >
> > >>> > Now config.log includes these lines
> > >>> >
> > >>> > configure:13881: === checking whether to install the readline SPKG ===
> > >>> > configure:13925: checking Installing ncurses?
> > >>> > configure:13928: result: Yes. Install readline as well.
> > >>> > configure:14176: === checking whether to install the pari SPKG ===
> > >>> > configure:14222: checking installing gmp/mpir or readline?
> > >>> > configure:14224: result: yes; install pari as well
> > >>> > configure:14565: result: using Sage's pari SPKG
> > >>> > configure:14587: === checking whether to install the pari_nftables 
> > >>> > SPKG ===
> > >>> 

Re: [sage-release] Sage 8.9 released

2019-10-11 Thread Thierry
Hi,

there is also a readline-related issue on 8.9 there:

https://ask.sagemath.org/question/48276/building-sage-89-from-source-fails/

Ciao,
Thierry



On Wed, Oct 02, 2019 at 01:46:00PM -0500, Dima Pasechnik wrote:
> On Wed, Oct 2, 2019 at 11:27 AM John Cremona  wrote:
> >
> >
> >
> > On Wed, 2 Oct 2019 at 09:07, John Cremona  wrote:
> >>
> >>
> >>
> >> On Wed, 2 Oct 2019 at 02:22, Dima Pasechnik  wrote:
> >>>
> >>> something did not work as you installed libncurses5-dev,
> >>> as you still appear to miss
> >>> /usr/include/ncurses.h
> >>> (according to the new log:
> >>> configure:10315: checking for ncurses.h
> >>> configure:10315: result: no )
> >>>
> >>> whereas https://packages.ubuntu.com/xenial/amd64/libncurses5-dev/filelist
> >>> has this file listed...
> >>>
> >>> Could you check that
> >>> /usr/include/ncurses.h
> >>> is present on the box?
> >>
> >>
> >> It was not, but I think that yesterday I was confused and had installed it 
> >> on a different machine from the one I reported on (too many windows open). 
> >>  I just checked by manually installing that package on all 6 machines, 
> >> which did do the install in 5 cases while on the 6th it was already there 
> >> -- and that 6th machine was one for which the original build had worked 
> >> OK!  (Yesterday I said that all 6 had failed the same way but in fact 5 
> >> had, the 6th was still building when I reported).
> >>
> >> For completeness I am attaching he latest config.log.
> >>
> >> I think that Samuel might have hit on something I did wrong.  My bash 
> >> history shows
> >>
> >>  1989  tar zxf 8.9.tar.gz
> >>  1990  cd sage-8.9
> >>  1991  make configure
> >>  1992  make
> >>
> >> with no ./configure step.  I must have thought that "make configure" was 
> >> instead of that...
> >>
> >> I'll restart the 5 builds after doing ./configure (after ncurses-clean and 
> >> readline-clean), and report back.
> >
> >
> > All went successfully.   My only remaining question is: is I had done 
> > ./configure aafter make configure and before make, would the build have 
> > worked (without inctalling those ubuntu packages which had somehow never 
> > been needed for previous builds)?
> >
> "make configure" is only needed if Sage's main configure script is updated.
> Just look at the target "configure" in the main Makefile
> 
> I guess there could be a bug introduced in commit
> 584735d9abe44b079df2566f7f73b6a69aaea0cf
> as I mentioned above in the thread.
> Feel free to open a ticket, saying that Sage's 8.9 readline builds
> underlinked on ubuntu xenial
> 
> > John
> >
> >>>
> >>>
> >>> On Wed, Oct 2, 2019 at 3:20 AM John Cremona  
> >>> wrote:
> >>> >
> >>> >
> >>> >
> >>> > On Tue, 1 Oct 2019 at 18:20, Dima Pasechnik  wrote:
> >>> >>
> >>> >> Hi John,
> >>> >>
> >>> >> Do you have
> >>> >> libncurses5-dev
> >>> >>  installed?
> >>> >
> >>> >
> >>> > No only the package without -dev.
> >>> >
> >>> >>
> >>> >> I guess not, as in your config.log I see
> >>> >>
> >>> >> conftest.cpp:56:21: fatal error: ncurses.h: No such file or directory
> >>> >>
> >>> >> and this header is provided by this package.
> >>> >> I'd say it's a bug in Ubunty xenial that libreadline-dev does not
> >>> >> depend on this package!
> >>> >>
> >>> >> Could you try installing it, then run
> >>> >>
> >>> >> make ncurses-clean
> >>> >> make readline-clean
> >>> >> ./configure
> >>> >
> >>> >
> >>> > I did that.
> >>> >
> >>> >>
> >>> >>
> >>> >> and check whether system-wide readline is picked up?
> >>> >
> >>> >
> >>> > Now config.log includes these lines
> >>> >
> >>> > configure:13881: === checking whether to install the readline SPKG ===
> >>> > configure:13925: checking Installing ncurses?
> >>> > configure:13928: result: Yes. Install readline as well.
> >>> > configure:14176: === checking whether to install the pari SPKG ===
> >>> > configure:14222: checking installing gmp/mpir or readline?
> >>> > configure:14224: result: yes; install pari as well
> >>> > configure:14565: result: using Sage's pari SPKG
> >>> > configure:14587: === checking whether to install the pari_nftables SPKG 
> >>> > ===
> >>> > configure:14631: checking installing pari?
> >>> > configure:14634: result: yes; install pari_nftables as well
> >>> > configure:14675: === checking whether to install the pkgconf SPKG ===
> >>> > configure:14718: checking for pkg-config >= 0.29
> >>> > configure:14794: result: /usr/bin/pkg-config
> >>> > configure:14807: result: using pkg-config from the system
> >>> > configure:14841: === checking whether to install the eclib SPKG ===
> >>> > configure:14886: checking installing ntl or pari?
> >>> > configure:14889: result: yes; install eclib as well
> >>> > configure:14974: === checking whether to install the 
> >>> > perl_term_readline_gnu SPKG ===
> >>> > configure:15066: checking for perl module Term::ReadLine
> >>> > configure:15076: result: ok
> >>> > configure:15084: checking Term::ReadLine module...
> >>> > configure:15088: result: non-GNU
> >>> >
> >>> > which 

Re: [sage-release] Sage 8.9 released

2019-10-07 Thread E. Madison Bray
On Mon, Oct 7, 2019 at 6:00 PM E. Madison Bray  wrote:
>
> On Mon, Oct 7, 2019 at 5:57 PM E. Madison Bray  wrote:
> >
> > When building from scratch on Cygwin I am getting some errors related
> > to OMP-related symbols being missing when linking a Sage module that
> > uses fflas-ffpack; specifically sage.libs.linbox_flint_interface:
> >
> > [sagelib-8.9] g++ -shared -Wl,--enable-auto-image-base
> > -L/opt/sagemath-8.9/local/lib -Wl,-rpath,/opt/sagemath-8.9/local/lib
> > -L/opt/sagemath-8.9/local/lib -Wl,-rpath,/opt/sagemath-8.9/local/lib
> > build/temp.cygwin-3.0.7-x86_64-2.7/build/cythonized/sage/libs/linbox/linbox_flint_interface.o
> > -L/opt/sagemath-8.9/local/lib
> > -L/opt/sagemath-8.9/local/lib/python2.7/config
> > -L/opt/sagemath-8.9/local/lib -llinbox -lntl -liml -lfflas -lffpack
> > -lgivaro -lblas -lflint -lmpfr -lgmp -lgmpxx -lstdc++ -lpython2.7 -o
> > build/lib.cygwin-3.0.7-x86_64-2.7/sage/libs/linbox/linbox_flint_interface.dll
> > [sagelib-8.9] 
> > build/temp.cygwin-3.0.7-x86_64-2.7/build/cythonized/sage/libs/linbox/linbox_flint_interface.o:
> > In function `FFPACK::rns_double::init(unsigned long, unsigned long,
> > double*, unsigned long, Givaro::Integer const*, unsigned long,
> > unsigned long, bool) const [clone ._omp_fn.0]':
> > [sagelib-8.9] 
> > /opt/sagemath-8.9/local/include/fflas-ffpack/field/rns-double.inl:95:
> > undefined reference to `GOMP_loop_ull_runtime_start'
> >
> > it goes on like that for several related errors.  I've never seen this
> > one before.  This is with gcc 7.4.0.
> >
> > This was doing a build from scratch for a release, in a clean Cygwin
> > install, so I'm not sure where the difference is.  One thing I can
> > see, comparing to a previous build log of my development sage from one
> > of the 8.9 release candidates (with fflas-ffpack 2.4.3) that that
> > module was compiled *without* -fopenmp, whereas on my clean build it
> > is passing -fopenmp and I'm not sure where the difference is coming
> > from.
>
> Ahah, I can also see that in my development build, configuring
> fflas-ffpack output:
>
> checking for OpenMP... no
>
> whereas on my release build it has
>
> checking for OpenMP... yes
>
>
> For now I don't need or care about OpenMP support, as I have not
> tested that at all on Windows.  I just need to figure out why that
> flag is being forcibly set...

It appears we used to configure fflas-ffpack with --disable-openmp,
however this ticket removed that:
https://trac.sagemath.org/ticket/27444 without bumping the package
patch level.
It also didn't remove the comment about disabling OpenMP support, even
though it no longer disables it.  I'm not sure it make sense to make
that the default; perhaps there should be a global config flag for
whether or not Sage and its dependencies should be built with OpenMP
support.



> > On Mon, Sep 30, 2019 at 1:19 AM Volker Braun  wrote:
> > >
> > > The "master" git branch has been updated to Sage-8.9. 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
> > >
> > > There was no change over 8.9.rc1
> > >
> > > --
> > > You received this message because you are subscribed to the Google Groups 
> > > "sage-release" group.
> > > To unsubscribe from this group and stop receiving emails from it, send an 
> > > email to sage-release+unsubscr...@googlegroups.com.
> > > To view this discussion on the web visit 
> > > https://groups.google.com/d/msgid/sage-release/ea48d604-eff9-4b8c-ab44-a0929a7e99f9%40googlegroups.com.

-- 
You received this message because you are subscribed to the Google Groups 
"sage-release" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-release+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-release/CAOTD34aDCyYgi6Q6aD1BixY0g-cf1G_1J8sZcbu58YmwcuHsYw%40mail.gmail.com.


Re: [sage-release] Sage 8.9 released

2019-10-07 Thread E. Madison Bray
On Mon, Oct 7, 2019 at 5:57 PM E. Madison Bray  wrote:
>
> When building from scratch on Cygwin I am getting some errors related
> to OMP-related symbols being missing when linking a Sage module that
> uses fflas-ffpack; specifically sage.libs.linbox_flint_interface:
>
> [sagelib-8.9] g++ -shared -Wl,--enable-auto-image-base
> -L/opt/sagemath-8.9/local/lib -Wl,-rpath,/opt/sagemath-8.9/local/lib
> -L/opt/sagemath-8.9/local/lib -Wl,-rpath,/opt/sagemath-8.9/local/lib
> build/temp.cygwin-3.0.7-x86_64-2.7/build/cythonized/sage/libs/linbox/linbox_flint_interface.o
> -L/opt/sagemath-8.9/local/lib
> -L/opt/sagemath-8.9/local/lib/python2.7/config
> -L/opt/sagemath-8.9/local/lib -llinbox -lntl -liml -lfflas -lffpack
> -lgivaro -lblas -lflint -lmpfr -lgmp -lgmpxx -lstdc++ -lpython2.7 -o
> build/lib.cygwin-3.0.7-x86_64-2.7/sage/libs/linbox/linbox_flint_interface.dll
> [sagelib-8.9] 
> build/temp.cygwin-3.0.7-x86_64-2.7/build/cythonized/sage/libs/linbox/linbox_flint_interface.o:
> In function `FFPACK::rns_double::init(unsigned long, unsigned long,
> double*, unsigned long, Givaro::Integer const*, unsigned long,
> unsigned long, bool) const [clone ._omp_fn.0]':
> [sagelib-8.9] 
> /opt/sagemath-8.9/local/include/fflas-ffpack/field/rns-double.inl:95:
> undefined reference to `GOMP_loop_ull_runtime_start'
>
> it goes on like that for several related errors.  I've never seen this
> one before.  This is with gcc 7.4.0.
>
> This was doing a build from scratch for a release, in a clean Cygwin
> install, so I'm not sure where the difference is.  One thing I can
> see, comparing to a previous build log of my development sage from one
> of the 8.9 release candidates (with fflas-ffpack 2.4.3) that that
> module was compiled *without* -fopenmp, whereas on my clean build it
> is passing -fopenmp and I'm not sure where the difference is coming
> from.

Ahah, I can also see that in my development build, configuring
fflas-ffpack output:

checking for OpenMP... no

whereas on my release build it has

checking for OpenMP... yes


For now I don't need or care about OpenMP support, as I have not
tested that at all on Windows.  I just need to figure out why that
flag is being forcibly set...


> On Mon, Sep 30, 2019 at 1:19 AM Volker Braun  wrote:
> >
> > The "master" git branch has been updated to Sage-8.9. 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
> >
> > There was no change over 8.9.rc1
> >
> > --
> > You received this message because you are subscribed to the Google Groups 
> > "sage-release" group.
> > To unsubscribe from this group and stop receiving emails from it, send an 
> > email to sage-release+unsubscr...@googlegroups.com.
> > To view this discussion on the web visit 
> > https://groups.google.com/d/msgid/sage-release/ea48d604-eff9-4b8c-ab44-a0929a7e99f9%40googlegroups.com.

-- 
You received this message because you are subscribed to the Google Groups 
"sage-release" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-release+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-release/CAOTD34YKKrE2eudAi2BgkMj%2BekzTB5gPrws5jhK6%3DntfN%2Bh37g%40mail.gmail.com.


Re: [sage-release] Sage 8.9 released

2019-10-07 Thread E. Madison Bray
When building from scratch on Cygwin I am getting some errors related
to OMP-related symbols being missing when linking a Sage module that
uses fflas-ffpack; specifically sage.libs.linbox_flint_interface:

[sagelib-8.9] g++ -shared -Wl,--enable-auto-image-base
-L/opt/sagemath-8.9/local/lib -Wl,-rpath,/opt/sagemath-8.9/local/lib
-L/opt/sagemath-8.9/local/lib -Wl,-rpath,/opt/sagemath-8.9/local/lib
build/temp.cygwin-3.0.7-x86_64-2.7/build/cythonized/sage/libs/linbox/linbox_flint_interface.o
-L/opt/sagemath-8.9/local/lib
-L/opt/sagemath-8.9/local/lib/python2.7/config
-L/opt/sagemath-8.9/local/lib -llinbox -lntl -liml -lfflas -lffpack
-lgivaro -lblas -lflint -lmpfr -lgmp -lgmpxx -lstdc++ -lpython2.7 -o
build/lib.cygwin-3.0.7-x86_64-2.7/sage/libs/linbox/linbox_flint_interface.dll
[sagelib-8.9] 
build/temp.cygwin-3.0.7-x86_64-2.7/build/cythonized/sage/libs/linbox/linbox_flint_interface.o:
In function `FFPACK::rns_double::init(unsigned long, unsigned long,
double*, unsigned long, Givaro::Integer const*, unsigned long,
unsigned long, bool) const [clone ._omp_fn.0]':
[sagelib-8.9] 
/opt/sagemath-8.9/local/include/fflas-ffpack/field/rns-double.inl:95:
undefined reference to `GOMP_loop_ull_runtime_start'

it goes on like that for several related errors.  I've never seen this
one before.  This is with gcc 7.4.0.

This was doing a build from scratch for a release, in a clean Cygwin
install, so I'm not sure where the difference is.  One thing I can
see, comparing to a previous build log of my development sage from one
of the 8.9 release candidates (with fflas-ffpack 2.4.3) that that
module was compiled *without* -fopenmp, whereas on my clean build it
is passing -fopenmp and I'm not sure where the difference is coming
from.

On Mon, Sep 30, 2019 at 1:19 AM Volker Braun  wrote:
>
> The "master" git branch has been updated to Sage-8.9. 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
>
> There was no change over 8.9.rc1
>
> --
> You received this message because you are subscribed to the Google Groups 
> "sage-release" group.
> To unsubscribe from this group and stop receiving emails from it, send an 
> email to sage-release+unsubscr...@googlegroups.com.
> To view this discussion on the web visit 
> https://groups.google.com/d/msgid/sage-release/ea48d604-eff9-4b8c-ab44-a0929a7e99f9%40googlegroups.com.

-- 
You received this message because you are subscribed to the Google Groups 
"sage-release" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-release+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-release/CAOTD34aMMVm9rUt057Vew3ON6VBcWQSva5zm3Hx%3D-StOeubSbw%40mail.gmail.com.


Re: [sage-release] Sage 8.9 released

2019-10-02 Thread Dima Pasechnik
On Wed, Oct 2, 2019 at 11:27 AM John Cremona  wrote:
>
>
>
> On Wed, 2 Oct 2019 at 09:07, John Cremona  wrote:
>>
>>
>>
>> On Wed, 2 Oct 2019 at 02:22, Dima Pasechnik  wrote:
>>>
>>> something did not work as you installed libncurses5-dev,
>>> as you still appear to miss
>>> /usr/include/ncurses.h
>>> (according to the new log:
>>> configure:10315: checking for ncurses.h
>>> configure:10315: result: no )
>>>
>>> whereas https://packages.ubuntu.com/xenial/amd64/libncurses5-dev/filelist
>>> has this file listed...
>>>
>>> Could you check that
>>> /usr/include/ncurses.h
>>> is present on the box?
>>
>>
>> It was not, but I think that yesterday I was confused and had installed it 
>> on a different machine from the one I reported on (too many windows open).  
>> I just checked by manually installing that package on all 6 machines, which 
>> did do the install in 5 cases while on the 6th it was already there -- and 
>> that 6th machine was one for which the original build had worked OK!  
>> (Yesterday I said that all 6 had failed the same way but in fact 5 had, the 
>> 6th was still building when I reported).
>>
>> For completeness I am attaching he latest config.log.
>>
>> I think that Samuel might have hit on something I did wrong.  My bash 
>> history shows
>>
>>  1989  tar zxf 8.9.tar.gz
>>  1990  cd sage-8.9
>>  1991  make configure
>>  1992  make
>>
>> with no ./configure step.  I must have thought that "make configure" was 
>> instead of that...
>>
>> I'll restart the 5 builds after doing ./configure (after ncurses-clean and 
>> readline-clean), and report back.
>
>
> All went successfully.   My only remaining question is: is I had done 
> ./configure aafter make configure and before make, would the build have 
> worked (without inctalling those ubuntu packages which had somehow never been 
> needed for previous builds)?
>
"make configure" is only needed if Sage's main configure script is updated.
Just look at the target "configure" in the main Makefile

I guess there could be a bug introduced in commit
584735d9abe44b079df2566f7f73b6a69aaea0cf
as I mentioned above in the thread.
Feel free to open a ticket, saying that Sage's 8.9 readline builds
underlinked on ubuntu xenial

> John
>
>>>
>>>
>>> On Wed, Oct 2, 2019 at 3:20 AM John Cremona  wrote:
>>> >
>>> >
>>> >
>>> > On Tue, 1 Oct 2019 at 18:20, Dima Pasechnik  wrote:
>>> >>
>>> >> Hi John,
>>> >>
>>> >> Do you have
>>> >> libncurses5-dev
>>> >>  installed?
>>> >
>>> >
>>> > No only the package without -dev.
>>> >
>>> >>
>>> >> I guess not, as in your config.log I see
>>> >>
>>> >> conftest.cpp:56:21: fatal error: ncurses.h: No such file or directory
>>> >>
>>> >> and this header is provided by this package.
>>> >> I'd say it's a bug in Ubunty xenial that libreadline-dev does not
>>> >> depend on this package!
>>> >>
>>> >> Could you try installing it, then run
>>> >>
>>> >> make ncurses-clean
>>> >> make readline-clean
>>> >> ./configure
>>> >
>>> >
>>> > I did that.
>>> >
>>> >>
>>> >>
>>> >> and check whether system-wide readline is picked up?
>>> >
>>> >
>>> > Now config.log includes these lines
>>> >
>>> > configure:13881: === checking whether to install the readline SPKG ===
>>> > configure:13925: checking Installing ncurses?
>>> > configure:13928: result: Yes. Install readline as well.
>>> > configure:14176: === checking whether to install the pari SPKG ===
>>> > configure:14222: checking installing gmp/mpir or readline?
>>> > configure:14224: result: yes; install pari as well
>>> > configure:14565: result: using Sage's pari SPKG
>>> > configure:14587: === checking whether to install the pari_nftables SPKG 
>>> > ===
>>> > configure:14631: checking installing pari?
>>> > configure:14634: result: yes; install pari_nftables as well
>>> > configure:14675: === checking whether to install the pkgconf SPKG ===
>>> > configure:14718: checking for pkg-config >= 0.29
>>> > configure:14794: result: /usr/bin/pkg-config
>>> > configure:14807: result: using pkg-config from the system
>>> > configure:14841: === checking whether to install the eclib SPKG ===
>>> > configure:14886: checking installing ntl or pari?
>>> > configure:14889: result: yes; install eclib as well
>>> > configure:14974: === checking whether to install the 
>>> > perl_term_readline_gnu SPKG ===
>>> > configure:15066: checking for perl module Term::ReadLine
>>> > configure:15076: result: ok
>>> > configure:15084: checking Term::ReadLine module...
>>> > configure:15088: result: non-GNU
>>> >
>>> > which does not look right.  Complete config.log isattached.
>>> >
>>> >
>>> >>
>>> >>
>>> >> If yes, this should cure these machines...
>>> >>
>>> >> Dima
>>> >>
>>> >> On Tue, Oct 1, 2019 at 9:01 PM John Cremona  
>>> >> wrote:
>>> >> >
>>> >> >
>>> >> >
>>> >> > On Tue, 1 Oct 2019 at 13:42, Dima Pasechnik  wrote:
>>> >> >>
>>> >> >> On Tue, Oct 1, 2019 at 7:43 PM John Cremona  
>>> >> >> wrote:
>>> >> >> >
>>> >> >> >
>>> >> >> >
>>> >> >> > On Tue, 1 Oct 2019 at 12:08, Dima Pasechnik  
>>> >> 

Re: [sage-release] Sage 8.9 released

2019-10-02 Thread John Cremona
On Wed, 2 Oct 2019 at 09:07, John Cremona  wrote:

>
>
> On Wed, 2 Oct 2019 at 02:22, Dima Pasechnik  wrote:
>
>> something did not work as you installed libncurses5-dev,
>> as you still appear to miss
>> /usr/include/ncurses.h
>> (according to the new log:
>> configure:10315: checking for ncurses.h
>> configure:10315: result: no )
>>
>> whereas https://packages.ubuntu.com/xenial/amd64/libncurses5-dev/filelist
>> has this file listed...
>>
>> Could you check that
>> /usr/include/ncurses.h
>> is present on the box?
>>
>
> It was not, but I think that yesterday I was confused and had installed it
> on a different machine from the one I reported on (too many windows open).
> I just checked by manually installing that package on all 6 machines, which
> did do the install in 5 cases while on the 6th it was already there -- and
> that 6th machine was one for which the original build had worked OK!
> (Yesterday I said that all 6 had failed the same way but in fact 5 had, the
> 6th was still building when I reported).
>
> For completeness I am attaching he latest config.log.
>
> I think that Samuel might have hit on something I did wrong.  My bash
> history shows
>
>  1989  tar zxf 8.9.tar.gz
>  1990  cd sage-8.9
>  1991  make configure
>  1992  make
>
> with no ./configure step.  I must have thought that "make configure" was
> instead of that...
>
> I'll restart the 5 builds after doing ./configure (after ncurses-clean and
> readline-clean), and report back.
>
>>
All went successfully.   My only remaining question is: is I had done
./configure aafter make configure and before make, would the build have
worked (without inctalling those ubuntu packages which had somehow never
been needed for previous builds)?

John


>
>> On Wed, Oct 2, 2019 at 3:20 AM John Cremona 
>> wrote:
>> >
>> >
>> >
>> > On Tue, 1 Oct 2019 at 18:20, Dima Pasechnik  wrote:
>> >>
>> >> Hi John,
>> >>
>> >> Do you have
>> >> libncurses5-dev
>> >>  installed?
>> >
>> >
>> > No only the package without -dev.
>> >
>> >>
>> >> I guess not, as in your config.log I see
>> >>
>> >> conftest.cpp:56:21: fatal error: ncurses.h: No such file or directory
>> >>
>> >> and this header is provided by this package.
>> >> I'd say it's a bug in Ubunty xenial that libreadline-dev does not
>> >> depend on this package!
>> >>
>> >> Could you try installing it, then run
>> >>
>> >> make ncurses-clean
>> >> make readline-clean
>> >> ./configure
>> >
>> >
>> > I did that.
>> >
>> >>
>> >>
>> >> and check whether system-wide readline is picked up?
>> >
>> >
>> > Now config.log includes these lines
>> >
>> > configure:13881: === checking whether to install the readline SPKG ===
>> > configure:13925: checking Installing ncurses?
>> > configure:13928: result: Yes. Install readline as well.
>> > configure:14176: === checking whether to install the pari SPKG ===
>> > configure:14222: checking installing gmp/mpir or readline?
>> > configure:14224: result: yes; install pari as well
>> > configure:14565: result: using Sage's pari SPKG
>> > configure:14587: === checking whether to install the pari_nftables SPKG
>> ===
>> > configure:14631: checking installing pari?
>> > configure:14634: result: yes; install pari_nftables as well
>> > configure:14675: === checking whether to install the pkgconf SPKG ===
>> > configure:14718: checking for pkg-config >= 0.29
>> > configure:14794: result: /usr/bin/pkg-config
>> > configure:14807: result: using pkg-config from the system
>> > configure:14841: === checking whether to install the eclib SPKG ===
>> > configure:14886: checking installing ntl or pari?
>> > configure:14889: result: yes; install eclib as well
>> > configure:14974: === checking whether to install the
>> perl_term_readline_gnu SPKG ===
>> > configure:15066: checking for perl module Term::ReadLine
>> > configure:15076: result: ok
>> > configure:15084: checking Term::ReadLine module...
>> > configure:15088: result: non-GNU
>> >
>> > which does not look right.  Complete config.log isattached.
>> >
>> >
>> >>
>> >>
>> >> If yes, this should cure these machines...
>> >>
>> >> Dima
>> >>
>> >> On Tue, Oct 1, 2019 at 9:01 PM John Cremona 
>> wrote:
>> >> >
>> >> >
>> >> >
>> >> > On Tue, 1 Oct 2019 at 13:42, Dima Pasechnik 
>> wrote:
>> >> >>
>> >> >> On Tue, Oct 1, 2019 at 7:43 PM John Cremona 
>> wrote:
>> >> >> >
>> >> >> >
>> >> >> >
>> >> >> > On Tue, 1 Oct 2019 at 12:08, Dima Pasechnik 
>> wrote:
>> >> >> >>
>> >> >> >> On Tue, Oct 1, 2019 at 6:59 PM Dima Pasechnik 
>> wrote:
>> >> >> >> >
>> >> >> >> > It seems that these machines don't have readline development
>> files
>> >> >> >> > properly installed.
>> >> >> >> >
>> >> >> >> > Could you attach config.log ?
>> >> >> >> >
>> >> >> >> I mean the "main" config.log, not the ones from package builds.
>> >> >> >
>> >> >> >
>> >> >> > Here it is.  I looks as if some version of readline was
>> downloaded and installed.
>> >> >> no, it says that Sage's readline will be built, because according
>> to the test,
>> >> >> there is 

Re: [sage-release] Sage 8.9 released

2019-10-01 Thread Dima Pasechnik
something did not work as you installed libncurses5-dev,
as you still appear to miss
/usr/include/ncurses.h
(according to the new log:
configure:10315: checking for ncurses.h
configure:10315: result: no )

whereas https://packages.ubuntu.com/xenial/amd64/libncurses5-dev/filelist
has this file listed...

Could you check that
/usr/include/ncurses.h
is present on the box?

On Wed, Oct 2, 2019 at 3:20 AM John Cremona  wrote:
>
>
>
> On Tue, 1 Oct 2019 at 18:20, Dima Pasechnik  wrote:
>>
>> Hi John,
>>
>> Do you have
>> libncurses5-dev
>>  installed?
>
>
> No only the package without -dev.
>
>>
>> I guess not, as in your config.log I see
>>
>> conftest.cpp:56:21: fatal error: ncurses.h: No such file or directory
>>
>> and this header is provided by this package.
>> I'd say it's a bug in Ubunty xenial that libreadline-dev does not
>> depend on this package!
>>
>> Could you try installing it, then run
>>
>> make ncurses-clean
>> make readline-clean
>> ./configure
>
>
> I did that.
>
>>
>>
>> and check whether system-wide readline is picked up?
>
>
> Now config.log includes these lines
>
> configure:13881: === checking whether to install the readline SPKG ===
> configure:13925: checking Installing ncurses?
> configure:13928: result: Yes. Install readline as well.
> configure:14176: === checking whether to install the pari SPKG ===
> configure:14222: checking installing gmp/mpir or readline?
> configure:14224: result: yes; install pari as well
> configure:14565: result: using Sage's pari SPKG
> configure:14587: === checking whether to install the pari_nftables SPKG ===
> configure:14631: checking installing pari?
> configure:14634: result: yes; install pari_nftables as well
> configure:14675: === checking whether to install the pkgconf SPKG ===
> configure:14718: checking for pkg-config >= 0.29
> configure:14794: result: /usr/bin/pkg-config
> configure:14807: result: using pkg-config from the system
> configure:14841: === checking whether to install the eclib SPKG ===
> configure:14886: checking installing ntl or pari?
> configure:14889: result: yes; install eclib as well
> configure:14974: === checking whether to install the perl_term_readline_gnu 
> SPKG ===
> configure:15066: checking for perl module Term::ReadLine
> configure:15076: result: ok
> configure:15084: checking Term::ReadLine module...
> configure:15088: result: non-GNU
>
> which does not look right.  Complete config.log isattached.
>
>
>>
>>
>> If yes, this should cure these machines...
>>
>> Dima
>>
>> On Tue, Oct 1, 2019 at 9:01 PM John Cremona  wrote:
>> >
>> >
>> >
>> > On Tue, 1 Oct 2019 at 13:42, Dima Pasechnik  wrote:
>> >>
>> >> On Tue, Oct 1, 2019 at 7:43 PM John Cremona  
>> >> wrote:
>> >> >
>> >> >
>> >> >
>> >> > On Tue, 1 Oct 2019 at 12:08, Dima Pasechnik  wrote:
>> >> >>
>> >> >> On Tue, Oct 1, 2019 at 6:59 PM Dima Pasechnik  
>> >> >> wrote:
>> >> >> >
>> >> >> > It seems that these machines don't have readline development files
>> >> >> > properly installed.
>> >> >> >
>> >> >> > Could you attach config.log ?
>> >> >> >
>> >> >> I mean the "main" config.log, not the ones from package builds.
>> >> >
>> >> >
>> >> > Here it is.  I looks as if some version of readline was downloaded and 
>> >> > installed.
>> >> no, it says that Sage's readline will be built, because according to the 
>> >> test,
>> >> there is no ncurses (or tinfo) installed.
>> >>
>> >> So I am puzzled that you say that the machine has libreadline-dev 
>> >> installed.
>> >> Could you post the contents of this package?
>> >
>> >
>> > $ apt show libreadline-dev
>> > Package: libreadline-dev
>> > Version: 6.3-8ubuntu2
>> > Priority: optional
>> > Section: libdevel
>> > Source: readline6
>> > Origin: Ubuntu
>> > Maintainer: Ubuntu Developers 
>> > Original-Maintainer: Matthias Klose 
>> > Bugs: https://bugs.launchpad.net/ubuntu/+filebug
>> > Installed-Size: 6,144 B
>> > Depends: libreadline6-dev (= 6.3-8ubuntu2)
>> > Conflicts: libreadline-gplv2-dev
>> > Supported: 9m
>> > Download-Size: 992 B
>> > APT-Manual-Installed: yes
>> > APT-Sources: http://gb.archive.ubuntu.com/ubuntu xenial/main amd64 Packages
>> > Description: GNU readline and history libraries, development files
>> >  The GNU readline library aids in the consistency of user interface
>> >  across discrete programs that need to provide a command line
>> >  interface.
>> >  .
>> >  The GNU history library provides a consistent user interface for
>> >  recalling lines of previously typed input.
>> >  .
>> >  This package is a dependency package depending on libreadline6-dev.
>> >
>> >>
>> >> As well, I'd like to look at the output of
>> >>
>> >> $ ldd `find /usr -name libreadline.so`
>> >>
>> >> - which should look more or less like:
>> >>
>> >> linux-vdso.so.1 (0x7ffc2a2ce000)
>> >> libtinfo.so.6 => /lib/x86_64-linux-gnu/libtinfo.so.6 (0x79942a162000)
>> >> libc.so.6 => /lib/x86_64-linux-gnu/libc.so.6 (0x799429fa1000)
>> >> /lib64/ld-linux-x86-64.so.2 (0x79942a3f3000)
>> >
>> >
>> > It finds 

Re: [sage-release] Sage 8.9 released

2019-10-01 Thread Samuel Lelièvre
Le mar. 1 oct. 2019 à 12:38, John Cremona:
>
> For a change instead of updating the git repository to build 8.9 I downloaded
> the file 8.9.tar.gz from the Releases section on github.  After unpacking and
> "make configure" I then did "make".

Shouldn't there be an extra step between `make configure` and `make`?

$ make configure
$ ./configure
$ make

-- 
You received this message because you are subscribed to the Google Groups 
"sage-release" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-release+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-release/CAEcArF17gdxyYposoiau0TYOKbZCZOL36bW-KwgNcny7XW7HrQ%40mail.gmail.com.


Re: [sage-release] Sage 8.9 released

2019-10-01 Thread 'Justin C. Walker' via sage-release



> On Sep 29, 2019, at 16:19 , Volker Braun  wrote:
> 
> The "master" git branch has been updated to Sage-8.9. 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
> 
> There was no change over 8.9.rc1

Built w/o problems from a fresh clone of the develop branch, and all tests 
(‘ptestlong’) passed!

This is on three macOS systems:

10.11.6 (mid-2015 MBP, Quad-core Core i7)
10.13.6 (2017 iMac Pro, 18-core Xeon W)
10.14.6 (2017 MBP, Quad-core Core i7)

Odd that, on 10.13.6, 8.9.rc1 had a repeatable failure testing 
(src/sage/libs/singular/polynomial.pyx), but not with 8.9.  ?!?

Justin

--
Justin C. Walker, Curmudgeon at Large
Director
Institute for the Enhancement of the Director's Income
--
In mathematics you don't understand
   things. You just get used to them. 
   --John von Neumann
--

-- 
You received this message because you are subscribed to the Google Groups 
"sage-release" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-release+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-release/EFF03370-C2ED-41E8-B0C0-0C1A9A74121D%40mac.com.


Re: [sage-release] Sage 8.9 released

2019-10-01 Thread Isuru Fernando
conda packages are available now. We are now patch free except for
https://trac.sagemath.org/ticket/28533 on Debian based distributions.

There are a couple of warnings in werkzeug that was not present when I
tested 8.8. Other than that, few test failures from 8.8 remain the same.

Isuru.

On Tue, Oct 1, 2019 at 8:01 AM John Cremona  wrote:

>
>
> On Tue, 1 Oct 2019 at 13:42, Dima Pasechnik  wrote:
>
>> On Tue, Oct 1, 2019 at 7:43 PM John Cremona 
>> wrote:
>> >
>> >
>> >
>> > On Tue, 1 Oct 2019 at 12:08, Dima Pasechnik  wrote:
>> >>
>> >> On Tue, Oct 1, 2019 at 6:59 PM Dima Pasechnik 
>> wrote:
>> >> >
>> >> > It seems that these machines don't have readline development files
>> >> > properly installed.
>> >> >
>> >> > Could you attach config.log ?
>> >> >
>> >> I mean the "main" config.log, not the ones from package builds.
>> >
>> >
>> > Here it is.  I looks as if some version of readline was downloaded and
>> installed.
>> no, it says that Sage's readline will be built, because according to the
>> test,
>> there is no ncurses (or tinfo) installed.
>>
>> So I am puzzled that you say that the machine has libreadline-dev
>> installed.
>> Could you post the contents of this package?
>>
>
> $ apt show libreadline-dev
> Package: libreadline-dev
> Version: 6.3-8ubuntu2
> Priority: optional
> Section: libdevel
> Source: readline6
> Origin: Ubuntu
> Maintainer: Ubuntu Developers 
> Original-Maintainer: Matthias Klose 
> Bugs: https://bugs.launchpad.net/ubuntu/+filebug
> Installed-Size: 6,144 B
> Depends: libreadline6-dev (= 6.3-8ubuntu2)
> Conflicts: libreadline-gplv2-dev
> Supported: 9m
> Download-Size: 992 B
> APT-Manual-Installed: yes
> APT-Sources: http://gb.archive.ubuntu.com/ubuntu xenial/main amd64
> Packages
> Description: GNU readline and history libraries, development files
>  The GNU readline library aids in the consistency of user interface
>  across discrete programs that need to provide a command line
>  interface.
>  .
>  The GNU history library provides a consistent user interface for
>  recalling lines of previously typed input.
>  .
>  This package is a dependency package depending on libreadline6-dev.
>
>
>> As well, I'd like to look at the output of
>>
>> $ ldd `find /usr -name libreadline.so`
>>
>> - which should look more or less like:
>>
>> linux-vdso.so.1 (0x7ffc2a2ce000)
>> libtinfo.so.6 => /lib/x86_64-linux-gnu/libtinfo.so.6 (0x79942a162000)
>> libc.so.6 => /lib/x86_64-linux-gnu/libc.so.6 (0x799429fa1000)
>> /lib64/ld-linux-x86-64.so.2 (0x79942a3f3000)
>>
>
> It finds the system one as well as two sage installation ones.  The system
> one is like what you expected:
>
> /usr/local/sage/sage-8.8/local/lib/libreadline.so:
> linux-vdso.so.1 =>  (0x7ffd459fe000)
> libtinfo.so.6 => /usr/local/sage/sage-8.8/local/lib/libtinfo.so.6
> (0x7f9940e43000)
> libc.so.6 => /lib/x86_64-linux-gnu/libc.so.6 (0x7f9940a73000)
> /lib64/ld-linux-x86-64.so.2 (0x7f99412d3000)
> /usr/local/sage/sage-8.9/local/lib/libreadline.so:
> linux-vdso.so.1 =>  (0x7ffe9e4ce000)
> libc.so.6 => /lib/x86_64-linux-gnu/libc.so.6 (0x7f00db8e3000)
> /lib64/ld-linux-x86-64.so.2 (0x7f00dbf03000)
> /usr/lib/x86_64-linux-gnu/libreadline.so:
> linux-vdso.so.1 =>  (0x7ffebbdd6000)
> libtinfo.so.5 => /lib/x86_64-linux-gnu/libtinfo.so.5
> (0x7f2e72d1b000)
> libc.so.6 => /lib/x86_64-linux-gnu/libc.so.6 (0x7f2e7294b000)
> /lib64/ld-linux-x86-64.so.2 (0x7f2e73193000)
>
>
>>
>> I gather that the test we have for system's ncurses is too
>> restrictive, and doesn't work
>> even though it should... (So this is a potential Sage bug)
>>
>> To continue with this problem:
>> Indeed, readline gets built by Sage, but it's a bit broken, as I
>> gather it's not linked
>> against libtinfo:
>>
>> To check this, please post the output of
>>
>>
> $ ldd /usr/local/sage/sage-8.9/local/lib/libreadline.so.6
> linux-vdso.so.1 =>  (0x7ffce3bbe000)
> libc.so.6 => /lib/x86_64-linux-gnu/libc.so.6 (0x7fbee541b000)
> /lib64/ld-linux-x86-64.so.2 (0x7fbee5a3b000)
>
>
>
>>
>> I don't know why this doesn't work, it might be something went wrong in
>> --
>> commit 584735d9abe44b079df2566f7f73b6a69aaea0cf
>> Author: Dima Pasechnik 
>> Date:   Wed May 1 10:58:00 2019 +0100
>>
>> get termcap library name from the readline's configure
>> 
>>
>> (so this might be another Sage bug)
>>
>> 
>> Fixing at least one of these bugs might cure the systems you have, but
>> more info is needed...
>> What OS is that, precisely?
>>
>
> $ lsb_release -a
> No LSB modules are available.
> Distributor ID: Ubuntu
> Description:Ubuntu 16.04.6 LTS
> Release:16.04
> Codename:   xenial
>
> (Same for all 6 machines.  I have tried to be consistent by reporting
> answers to your questions all from the same one.)

Re: [sage-release] Sage 8.9 released

2019-10-01 Thread Dima Pasechnik
Hi John,

Do you have
libncurses5-dev
 installed?
I guess not, as in your config.log I see

conftest.cpp:56:21: fatal error: ncurses.h: No such file or directory

and this header is provided by this package.
I'd say it's a bug in Ubunty xenial that libreadline-dev does not
depend on this package!

Could you try installing it, then run

make ncurses-clean
make readline-clean
./configure

and check whether system-wide readline is picked up?

If yes, this should cure these machines...

Dima

On Tue, Oct 1, 2019 at 9:01 PM John Cremona  wrote:
>
>
>
> On Tue, 1 Oct 2019 at 13:42, Dima Pasechnik  wrote:
>>
>> On Tue, Oct 1, 2019 at 7:43 PM John Cremona  wrote:
>> >
>> >
>> >
>> > On Tue, 1 Oct 2019 at 12:08, Dima Pasechnik  wrote:
>> >>
>> >> On Tue, Oct 1, 2019 at 6:59 PM Dima Pasechnik  wrote:
>> >> >
>> >> > It seems that these machines don't have readline development files
>> >> > properly installed.
>> >> >
>> >> > Could you attach config.log ?
>> >> >
>> >> I mean the "main" config.log, not the ones from package builds.
>> >
>> >
>> > Here it is.  I looks as if some version of readline was downloaded and 
>> > installed.
>> no, it says that Sage's readline will be built, because according to the 
>> test,
>> there is no ncurses (or tinfo) installed.
>>
>> So I am puzzled that you say that the machine has libreadline-dev installed.
>> Could you post the contents of this package?
>
>
> $ apt show libreadline-dev
> Package: libreadline-dev
> Version: 6.3-8ubuntu2
> Priority: optional
> Section: libdevel
> Source: readline6
> Origin: Ubuntu
> Maintainer: Ubuntu Developers 
> Original-Maintainer: Matthias Klose 
> Bugs: https://bugs.launchpad.net/ubuntu/+filebug
> Installed-Size: 6,144 B
> Depends: libreadline6-dev (= 6.3-8ubuntu2)
> Conflicts: libreadline-gplv2-dev
> Supported: 9m
> Download-Size: 992 B
> APT-Manual-Installed: yes
> APT-Sources: http://gb.archive.ubuntu.com/ubuntu xenial/main amd64 Packages
> Description: GNU readline and history libraries, development files
>  The GNU readline library aids in the consistency of user interface
>  across discrete programs that need to provide a command line
>  interface.
>  .
>  The GNU history library provides a consistent user interface for
>  recalling lines of previously typed input.
>  .
>  This package is a dependency package depending on libreadline6-dev.
>
>>
>> As well, I'd like to look at the output of
>>
>> $ ldd `find /usr -name libreadline.so`
>>
>> - which should look more or less like:
>>
>> linux-vdso.so.1 (0x7ffc2a2ce000)
>> libtinfo.so.6 => /lib/x86_64-linux-gnu/libtinfo.so.6 (0x79942a162000)
>> libc.so.6 => /lib/x86_64-linux-gnu/libc.so.6 (0x799429fa1000)
>> /lib64/ld-linux-x86-64.so.2 (0x79942a3f3000)
>
>
> It finds the system one as well as two sage installation ones.  The system 
> one is like what you expected:
>
> /usr/local/sage/sage-8.8/local/lib/libreadline.so:
> linux-vdso.so.1 =>  (0x7ffd459fe000)
> libtinfo.so.6 => /usr/local/sage/sage-8.8/local/lib/libtinfo.so.6 
> (0x7f9940e43000)
> libc.so.6 => /lib/x86_64-linux-gnu/libc.so.6 (0x7f9940a73000)
> /lib64/ld-linux-x86-64.so.2 (0x7f99412d3000)
> /usr/local/sage/sage-8.9/local/lib/libreadline.so:
> linux-vdso.so.1 =>  (0x7ffe9e4ce000)
> libc.so.6 => /lib/x86_64-linux-gnu/libc.so.6 (0x7f00db8e3000)
> /lib64/ld-linux-x86-64.so.2 (0x7f00dbf03000)
> /usr/lib/x86_64-linux-gnu/libreadline.so:
> linux-vdso.so.1 =>  (0x7ffebbdd6000)
> libtinfo.so.5 => /lib/x86_64-linux-gnu/libtinfo.so.5 
> (0x7f2e72d1b000)
> libc.so.6 => /lib/x86_64-linux-gnu/libc.so.6 (0x7f2e7294b000)
> /lib64/ld-linux-x86-64.so.2 (0x7f2e73193000)
>
>>
>>
>> I gather that the test we have for system's ncurses is too
>> restrictive, and doesn't work
>> even though it should... (So this is a potential Sage bug)
>>
>> To continue with this problem:
>> Indeed, readline gets built by Sage, but it's a bit broken, as I
>> gather it's not linked
>> against libtinfo:
>>
>> To check this, please post the output of
>>
>
> $ ldd /usr/local/sage/sage-8.9/local/lib/libreadline.so.6
> linux-vdso.so.1 =>  (0x7ffce3bbe000)
> libc.so.6 => /lib/x86_64-linux-gnu/libc.so.6 (0x7fbee541b000)
> /lib64/ld-linux-x86-64.so.2 (0x7fbee5a3b000)
>
>
>>
>>
>> I don't know why this doesn't work, it might be something went wrong in
>> --
>> commit 584735d9abe44b079df2566f7f73b6a69aaea0cf
>> Author: Dima Pasechnik 
>> Date:   Wed May 1 10:58:00 2019 +0100
>>
>> get termcap library name from the readline's configure
>> 
>>
>> (so this might be another Sage bug)
>>
>> 
>> Fixing at least one of these bugs might cure the systems you have, but
>> more info is needed...
>> What OS is that, precisely?
>
>
> $ lsb_release -a
> No LSB modules are available.
> Distributor ID: Ubuntu
> 

Re: [sage-release] Sage 8.9 released

2019-10-01 Thread John Cremona
On Tue, 1 Oct 2019 at 13:42, Dima Pasechnik  wrote:

> On Tue, Oct 1, 2019 at 7:43 PM John Cremona 
> wrote:
> >
> >
> >
> > On Tue, 1 Oct 2019 at 12:08, Dima Pasechnik  wrote:
> >>
> >> On Tue, Oct 1, 2019 at 6:59 PM Dima Pasechnik 
> wrote:
> >> >
> >> > It seems that these machines don't have readline development files
> >> > properly installed.
> >> >
> >> > Could you attach config.log ?
> >> >
> >> I mean the "main" config.log, not the ones from package builds.
> >
> >
> > Here it is.  I looks as if some version of readline was downloaded and
> installed.
> no, it says that Sage's readline will be built, because according to the
> test,
> there is no ncurses (or tinfo) installed.
>
> So I am puzzled that you say that the machine has libreadline-dev
> installed.
> Could you post the contents of this package?
>

$ apt show libreadline-dev
Package: libreadline-dev
Version: 6.3-8ubuntu2
Priority: optional
Section: libdevel
Source: readline6
Origin: Ubuntu
Maintainer: Ubuntu Developers 
Original-Maintainer: Matthias Klose 
Bugs: https://bugs.launchpad.net/ubuntu/+filebug
Installed-Size: 6,144 B
Depends: libreadline6-dev (= 6.3-8ubuntu2)
Conflicts: libreadline-gplv2-dev
Supported: 9m
Download-Size: 992 B
APT-Manual-Installed: yes
APT-Sources: http://gb.archive.ubuntu.com/ubuntu xenial/main amd64 Packages
Description: GNU readline and history libraries, development files
 The GNU readline library aids in the consistency of user interface
 across discrete programs that need to provide a command line
 interface.
 .
 The GNU history library provides a consistent user interface for
 recalling lines of previously typed input.
 .
 This package is a dependency package depending on libreadline6-dev.


> As well, I'd like to look at the output of
>
> $ ldd `find /usr -name libreadline.so`
>
> - which should look more or less like:
>
> linux-vdso.so.1 (0x7ffc2a2ce000)
> libtinfo.so.6 => /lib/x86_64-linux-gnu/libtinfo.so.6 (0x79942a162000)
> libc.so.6 => /lib/x86_64-linux-gnu/libc.so.6 (0x799429fa1000)
> /lib64/ld-linux-x86-64.so.2 (0x79942a3f3000)
>

It finds the system one as well as two sage installation ones.  The system
one is like what you expected:

/usr/local/sage/sage-8.8/local/lib/libreadline.so:
linux-vdso.so.1 =>  (0x7ffd459fe000)
libtinfo.so.6 => /usr/local/sage/sage-8.8/local/lib/libtinfo.so.6
(0x7f9940e43000)
libc.so.6 => /lib/x86_64-linux-gnu/libc.so.6 (0x7f9940a73000)
/lib64/ld-linux-x86-64.so.2 (0x7f99412d3000)
/usr/local/sage/sage-8.9/local/lib/libreadline.so:
linux-vdso.so.1 =>  (0x7ffe9e4ce000)
libc.so.6 => /lib/x86_64-linux-gnu/libc.so.6 (0x7f00db8e3000)
/lib64/ld-linux-x86-64.so.2 (0x7f00dbf03000)
/usr/lib/x86_64-linux-gnu/libreadline.so:
linux-vdso.so.1 =>  (0x7ffebbdd6000)
libtinfo.so.5 => /lib/x86_64-linux-gnu/libtinfo.so.5
(0x7f2e72d1b000)
libc.so.6 => /lib/x86_64-linux-gnu/libc.so.6 (0x7f2e7294b000)
/lib64/ld-linux-x86-64.so.2 (0x7f2e73193000)


>
> I gather that the test we have for system's ncurses is too
> restrictive, and doesn't work
> even though it should... (So this is a potential Sage bug)
>
> To continue with this problem:
> Indeed, readline gets built by Sage, but it's a bit broken, as I
> gather it's not linked
> against libtinfo:
>
> To check this, please post the output of
>
>
$ ldd /usr/local/sage/sage-8.9/local/lib/libreadline.so.6
linux-vdso.so.1 =>  (0x7ffce3bbe000)
libc.so.6 => /lib/x86_64-linux-gnu/libc.so.6 (0x7fbee541b000)
/lib64/ld-linux-x86-64.so.2 (0x7fbee5a3b000)



>
> I don't know why this doesn't work, it might be something went wrong in
> --
> commit 584735d9abe44b079df2566f7f73b6a69aaea0cf
> Author: Dima Pasechnik 
> Date:   Wed May 1 10:58:00 2019 +0100
>
> get termcap library name from the readline's configure
> 
>
> (so this might be another Sage bug)
>
> 
> Fixing at least one of these bugs might cure the systems you have, but
> more info is needed...
> What OS is that, precisely?
>

$ lsb_release -a
No LSB modules are available.
Distributor ID: Ubuntu
Description:Ubuntu 16.04.6 LTS
Release:16.04
Codename:   xenial

(Same for all 6 machines.  I have tried to be consistent by reporting
answers to your questions all from the same one.)


>
> Sorry for trouble,
>

Not at all, thanks for your help.

On a more positive note, my python3 build on one of these machines was fine
after git pull trac develop.



> Dima
>
>
> And Python2 does not do a very good job gathering what extra libs are
> needed,
>  as its confugure output says
>
> checking how to link readline libs... -lreadline -ltermcap
>
> but then the error happens in the linking where one does not see
> -ltermcap, but
> rather -lcursesw
>
> (so this might be a Python2 bug)
>
>
>
> >
> > I can of course install more 

Re: [sage-release] Sage 8.9 released

2019-10-01 Thread Dima Pasechnik
On Tue, Oct 1, 2019 at 7:43 PM John Cremona  wrote:
>
>
>
> On Tue, 1 Oct 2019 at 12:08, Dima Pasechnik  wrote:
>>
>> On Tue, Oct 1, 2019 at 6:59 PM Dima Pasechnik  wrote:
>> >
>> > It seems that these machines don't have readline development files
>> > properly installed.
>> >
>> > Could you attach config.log ?
>> >
>> I mean the "main" config.log, not the ones from package builds.
>
>
> Here it is.  I looks as if some version of readline was downloaded and 
> installed.
no, it says that Sage's readline will be built, because according to the test,
there is no ncurses (or tinfo) installed.

So I am puzzled that you say that the machine has libreadline-dev installed.
Could you post the contents of this package?
As well, I'd like to look at the output of

$ ldd `find /usr -name libreadline.so`

- which should look more or less like:

linux-vdso.so.1 (0x7ffc2a2ce000)
libtinfo.so.6 => /lib/x86_64-linux-gnu/libtinfo.so.6 (0x79942a162000)
libc.so.6 => /lib/x86_64-linux-gnu/libc.so.6 (0x799429fa1000)
/lib64/ld-linux-x86-64.so.2 (0x79942a3f3000)

I gather that the test we have for system's ncurses is too
restrictive, and doesn't work
even though it should... (So this is a potential Sage bug)

To continue with this problem:
Indeed, readline gets built by Sage, but it's a bit broken, as I
gather it's not linked
against libtinfo:

To check this, please post the output of

ldd /usr/local/sage/sage-8.9/local/lib/libreadline.so.6

I don't know why this doesn't work, it might be something went wrong in
--
commit 584735d9abe44b079df2566f7f73b6a69aaea0cf
Author: Dima Pasechnik 
Date:   Wed May 1 10:58:00 2019 +0100

get termcap library name from the readline's configure


(so this might be another Sage bug)


Fixing at least one of these bugs might cure the systems you have, but
more info is needed...
What OS is that, precisely?

Sorry for trouble,
Dima


And Python2 does not do a very good job gathering what extra libs are needed,
 as its confugure output says

checking how to link readline libs... -lreadline -ltermcap

but then the error happens in the linking where one does not see -ltermcap, but
rather -lcursesw

(so this might be a Python2 bug)



>
> I can of course install more ubuntu packages on these machines (though I do 
> have libreadline-dev installed already) but that's not quite the point here.
>
> John
>
>>
>> >
>> > On Tue, Oct 1, 2019 at 6:38 PM John Cremona  wrote:
>> > >
>> > > For a change instead of updating the git repository to build 8.9 I 
>> > > downloaded the file 8.9.tar.gz from the Releases section on github.  
>> > > After unpacking and "make configure" I then did "make".  On 6 different 
>> > > machines, all running ubuntu, all the builds failed with
>> > >
>> > > Error building Sage.
>> > >
>> > > The following package(s) may have failed to build (not necessarily
>> > > during this run of 'make all-start'):
>> > >
>> > > * package: python2-2.7.15.p1
>> > >   log file: /usr/local/sage/sage-8.9/logs/pkgs/python2-2.7.15.p1.log
>> > >   build directory: 
>> > > /usr/local/sage/sage-8.9/local/var/tmp/sage/build/python2-2.7.15.p1
>> > >
>> > > One logfile is attached, they all fail the same easy: something about 
>> > > readline.
>> > >
>> > > The point here is not so much what I should do to get around this, but 
>> > > rather, what is wrong with "make configure; make" which led it to happen 
>> > > on machines where the prerequisites for Sage are certainly all installed 
>> > > since they all have working versions of 8.8 (and many previous, and in 
>> > > some cases 8.9 prereleases) built from the github repo.
>> > >
>> > > John
>> > >
>> > > On Tue, 1 Oct 2019 at 09:07, Samuel Lelièvre  
>> > > wrote:
>> > >>
>> > >> ## Debian
>> > >>
>> > >> On Debian GNU/Linux 10 (buster), upgraded from
>> > >> SageMath 8.9.rc0 via git and tested SageMath 8.9,
>> > >> both py2 and py3. In both cases:
>> > >> - `make`: ok
>> > >> - `make testlong`: all tests pass!
>> > >>
>> > >> ## Cygwin
>> > >>
>> > >> On Cygwin64, tested SageMath 8.9, both py2 and py3,
>> > >> with the following hardware and operating system:
>> > >>
>> > >> - laptop: HP ProBook 640 G1
>> > >> - processor: Intel Core i7-4610M, 3.00 GHz
>> > >> - memory: 8 GB
>> > >> - Cygwin64 version: CYGWIN_NT-6.1 3.0.7 (0.338/5/3)
>> > >> - Windows: Windows 7 Professional (64-bit), Service Pack 1
>> > >>
>> > >> ### Summary
>> > >>
>> > >> Here is a summary following this structure:
>> > >>   - [file]  # [failures or timeouts during make testlong]
>> > >> --> [result when tested individually]; [ticket references]
>> > >>
>> > >> - py2:
>> > >>   - src/sage/doctest/forker.py  # 4 doctests failed
>> > >> --> passes when tested individually
>> > >>   - src/sage/parallel/map_reduce.py  # Timed out after testing finished
>> > >> --> passes when tested individually; might be #27537
>> > >>
>> > >> - py3
>> > >>   - 

Re: [sage-release] Sage 8.9 released

2019-10-01 Thread Dima Pasechnik
On Tue, Oct 1, 2019 at 6:59 PM Dima Pasechnik  wrote:
>
> It seems that these machines don't have readline development files
> properly installed.
>
> Could you attach config.log ?
>
I mean the "main" config.log, not the ones from package builds.

>
> On Tue, Oct 1, 2019 at 6:38 PM John Cremona  wrote:
> >
> > For a change instead of updating the git repository to build 8.9 I 
> > downloaded the file 8.9.tar.gz from the Releases section on github.  After 
> > unpacking and "make configure" I then did "make".  On 6 different machines, 
> > all running ubuntu, all the builds failed with
> >
> > Error building Sage.
> >
> > The following package(s) may have failed to build (not necessarily
> > during this run of 'make all-start'):
> >
> > * package: python2-2.7.15.p1
> >   log file: /usr/local/sage/sage-8.9/logs/pkgs/python2-2.7.15.p1.log
> >   build directory: 
> > /usr/local/sage/sage-8.9/local/var/tmp/sage/build/python2-2.7.15.p1
> >
> > One logfile is attached, they all fail the same easy: something about 
> > readline.
> >
> > The point here is not so much what I should do to get around this, but 
> > rather, what is wrong with "make configure; make" which led it to happen on 
> > machines where the prerequisites for Sage are certainly all installed since 
> > they all have working versions of 8.8 (and many previous, and in some cases 
> > 8.9 prereleases) built from the github repo.
> >
> > John
> >
> > On Tue, 1 Oct 2019 at 09:07, Samuel Lelièvre  
> > wrote:
> >>
> >> ## Debian
> >>
> >> On Debian GNU/Linux 10 (buster), upgraded from
> >> SageMath 8.9.rc0 via git and tested SageMath 8.9,
> >> both py2 and py3. In both cases:
> >> - `make`: ok
> >> - `make testlong`: all tests pass!
> >>
> >> ## Cygwin
> >>
> >> On Cygwin64, tested SageMath 8.9, both py2 and py3,
> >> with the following hardware and operating system:
> >>
> >> - laptop: HP ProBook 640 G1
> >> - processor: Intel Core i7-4610M, 3.00 GHz
> >> - memory: 8 GB
> >> - Cygwin64 version: CYGWIN_NT-6.1 3.0.7 (0.338/5/3)
> >> - Windows: Windows 7 Professional (64-bit), Service Pack 1
> >>
> >> ### Summary
> >>
> >> Here is a summary following this structure:
> >>   - [file]  # [failures or timeouts during make testlong]
> >> --> [result when tested individually]; [ticket references]
> >>
> >> - py2:
> >>   - src/sage/doctest/forker.py  # 4 doctests failed
> >> --> passes when tested individually
> >>   - src/sage/parallel/map_reduce.py  # Timed out after testing finished
> >> --> passes when tested individually; might be #27537
> >>
> >> - py3
> >>   - src/sage/coding/linear_code.py  # Timed out
> >> --> passes when tested individually; might be #26119
> >>   - src/sage/libs/eclib/interface.py  # 1 doctest failed
> >> --> still fails when tested individually; see #28472 (positive review)
> >>   - src/sage/rings/finite_rings/finite_field_base.pyx  # 1 doctest failed
> >> --> passes when tested individually
> >>   - src/sage/rings/polynomial/polynomial_rational_flint.pyx  # 1 doctest 
> >> failed
> >> --> still fails when tested individually; this is #28334
> >>   - src/sage_setup/clean.py  # 1 doctest failed
> >> --> still fails when tested individually
> >>
> >> Read on for the details for each of py2 and py3.
> >>
> >> ### Python 2
> >>
> >> Upgraded from SageMath 8.9.rc0 via git.
> >> Ran `make`: success. Ran `make testlong` and individually
> >> retested files that failed or timed out, see below.
> >>
> >> $ make testlong
> >> ...
> >> --
> >> sage -t --long --warn-long 87.2 src/sage/doctest/forker.py  # 4 doctests 
> >> failed
> >> sage -t --long --warn-long 87.2 src/sage/parallel/map_reduce.py  #
> >> Timed out after testing finished
> >> --
> >>
> >> $ ./sage -t --long --warn-long 87.2 src/sage/doctest/forker.py
> >> Running doctests with ID 2019-10-01-08-19-19-2f160fbc.
> >> Git branch: develop
> >> Using --optional=build,dochtml,python2,sage
> >> Doctesting 1 file.
> >> sage -t --long --warn-long 87.2 src/sage/doctest/forker.py
> >> [449 tests, 34.21 s]
> >> --
> >> All tests passed!
> >> --
> >> Total time for all tests: 81.7 seconds
> >> cpu time: 5.5 seconds
> >> cumulative wall time: 34.2 seconds
> >>
> >> $ ./sage -t --long --warn-long 87.2 src/sage/parallel/map_reduce.py
> >> Running doctests with ID 2019-10-01-08-21-35-400c0707.
> >> Git branch: develop
> >> Using --optional=build,dochtml,python2,sage
> >> Doctesting 1 file.
> >> sage -t --long --warn-long 87.2 src/sage/parallel/map_reduce.py
> >> [294 tests, 33.43 s]
> >> --
> >> All tests passed!
> >> --
> >> Total time for all 

Re: [sage-release] Sage 8.9 released

2019-10-01 Thread Dima Pasechnik
It seems that these machines don't have readline development files
properly installed.

Could you attach config.log ?


On Tue, Oct 1, 2019 at 6:38 PM John Cremona  wrote:
>
> For a change instead of updating the git repository to build 8.9 I downloaded 
> the file 8.9.tar.gz from the Releases section on github.  After unpacking and 
> "make configure" I then did "make".  On 6 different machines, all running 
> ubuntu, all the builds failed with
>
> Error building Sage.
>
> The following package(s) may have failed to build (not necessarily
> during this run of 'make all-start'):
>
> * package: python2-2.7.15.p1
>   log file: /usr/local/sage/sage-8.9/logs/pkgs/python2-2.7.15.p1.log
>   build directory: 
> /usr/local/sage/sage-8.9/local/var/tmp/sage/build/python2-2.7.15.p1
>
> One logfile is attached, they all fail the same easy: something about 
> readline.
>
> The point here is not so much what I should do to get around this, but 
> rather, what is wrong with "make configure; make" which led it to happen on 
> machines where the prerequisites for Sage are certainly all installed since 
> they all have working versions of 8.8 (and many previous, and in some cases 
> 8.9 prereleases) built from the github repo.
>
> John
>
> On Tue, 1 Oct 2019 at 09:07, Samuel Lelièvre  
> wrote:
>>
>> ## Debian
>>
>> On Debian GNU/Linux 10 (buster), upgraded from
>> SageMath 8.9.rc0 via git and tested SageMath 8.9,
>> both py2 and py3. In both cases:
>> - `make`: ok
>> - `make testlong`: all tests pass!
>>
>> ## Cygwin
>>
>> On Cygwin64, tested SageMath 8.9, both py2 and py3,
>> with the following hardware and operating system:
>>
>> - laptop: HP ProBook 640 G1
>> - processor: Intel Core i7-4610M, 3.00 GHz
>> - memory: 8 GB
>> - Cygwin64 version: CYGWIN_NT-6.1 3.0.7 (0.338/5/3)
>> - Windows: Windows 7 Professional (64-bit), Service Pack 1
>>
>> ### Summary
>>
>> Here is a summary following this structure:
>>   - [file]  # [failures or timeouts during make testlong]
>> --> [result when tested individually]; [ticket references]
>>
>> - py2:
>>   - src/sage/doctest/forker.py  # 4 doctests failed
>> --> passes when tested individually
>>   - src/sage/parallel/map_reduce.py  # Timed out after testing finished
>> --> passes when tested individually; might be #27537
>>
>> - py3
>>   - src/sage/coding/linear_code.py  # Timed out
>> --> passes when tested individually; might be #26119
>>   - src/sage/libs/eclib/interface.py  # 1 doctest failed
>> --> still fails when tested individually; see #28472 (positive review)
>>   - src/sage/rings/finite_rings/finite_field_base.pyx  # 1 doctest failed
>> --> passes when tested individually
>>   - src/sage/rings/polynomial/polynomial_rational_flint.pyx  # 1 doctest 
>> failed
>> --> still fails when tested individually; this is #28334
>>   - src/sage_setup/clean.py  # 1 doctest failed
>> --> still fails when tested individually
>>
>> Read on for the details for each of py2 and py3.
>>
>> ### Python 2
>>
>> Upgraded from SageMath 8.9.rc0 via git.
>> Ran `make`: success. Ran `make testlong` and individually
>> retested files that failed or timed out, see below.
>>
>> $ make testlong
>> ...
>> --
>> sage -t --long --warn-long 87.2 src/sage/doctest/forker.py  # 4 doctests 
>> failed
>> sage -t --long --warn-long 87.2 src/sage/parallel/map_reduce.py  #
>> Timed out after testing finished
>> --
>>
>> $ ./sage -t --long --warn-long 87.2 src/sage/doctest/forker.py
>> Running doctests with ID 2019-10-01-08-19-19-2f160fbc.
>> Git branch: develop
>> Using --optional=build,dochtml,python2,sage
>> Doctesting 1 file.
>> sage -t --long --warn-long 87.2 src/sage/doctest/forker.py
>> [449 tests, 34.21 s]
>> --
>> All tests passed!
>> --
>> Total time for all tests: 81.7 seconds
>> cpu time: 5.5 seconds
>> cumulative wall time: 34.2 seconds
>>
>> $ ./sage -t --long --warn-long 87.2 src/sage/parallel/map_reduce.py
>> Running doctests with ID 2019-10-01-08-21-35-400c0707.
>> Git branch: develop
>> Using --optional=build,dochtml,python2,sage
>> Doctesting 1 file.
>> sage -t --long --warn-long 87.2 src/sage/parallel/map_reduce.py
>> [294 tests, 33.43 s]
>> --
>> All tests passed!
>> --
>> Total time for all tests: 34.0 seconds
>> cpu time: 5.2 seconds
>> cumulative wall time: 33.4 seconds
>>
>> ### Python 3
>>
>> Upgraded from SageMath 8.9.rc0 via git.
>> Ran `make`: success. Ran `make testlong` and individually
>> retested files that failed or timed out, see below.
>>
>> $ make testlong
>> ...
>> 

Re: [sage-release] Sage 8.9 released

2019-10-01 Thread Samuel Lelièvre
## Debian

On Debian GNU/Linux 10 (buster), upgraded from
SageMath 8.9.rc0 via git and tested SageMath 8.9,
both py2 and py3. In both cases:
- `make`: ok
- `make testlong`: all tests pass!

## Cygwin

On Cygwin64, tested SageMath 8.9, both py2 and py3,
with the following hardware and operating system:

- laptop: HP ProBook 640 G1
- processor: Intel Core i7-4610M, 3.00 GHz
- memory: 8 GB
- Cygwin64 version: CYGWIN_NT-6.1 3.0.7 (0.338/5/3)
- Windows: Windows 7 Professional (64-bit), Service Pack 1

### Summary

Here is a summary following this structure:
  - [file]  # [failures or timeouts during make testlong]
--> [result when tested individually]; [ticket references]

- py2:
  - src/sage/doctest/forker.py  # 4 doctests failed
--> passes when tested individually
  - src/sage/parallel/map_reduce.py  # Timed out after testing finished
--> passes when tested individually; might be #27537

- py3
  - src/sage/coding/linear_code.py  # Timed out
--> passes when tested individually; might be #26119
  - src/sage/libs/eclib/interface.py  # 1 doctest failed
--> still fails when tested individually; see #28472 (positive review)
  - src/sage/rings/finite_rings/finite_field_base.pyx  # 1 doctest failed
--> passes when tested individually
  - src/sage/rings/polynomial/polynomial_rational_flint.pyx  # 1 doctest failed
--> still fails when tested individually; this is #28334
  - src/sage_setup/clean.py  # 1 doctest failed
--> still fails when tested individually

Read on for the details for each of py2 and py3.

### Python 2

Upgraded from SageMath 8.9.rc0 via git.
Ran `make`: success. Ran `make testlong` and individually
retested files that failed or timed out, see below.

$ make testlong
...
--
sage -t --long --warn-long 87.2 src/sage/doctest/forker.py  # 4 doctests failed
sage -t --long --warn-long 87.2 src/sage/parallel/map_reduce.py  #
Timed out after testing finished
--

$ ./sage -t --long --warn-long 87.2 src/sage/doctest/forker.py
Running doctests with ID 2019-10-01-08-19-19-2f160fbc.
Git branch: develop
Using --optional=build,dochtml,python2,sage
Doctesting 1 file.
sage -t --long --warn-long 87.2 src/sage/doctest/forker.py
[449 tests, 34.21 s]
--
All tests passed!
--
Total time for all tests: 81.7 seconds
cpu time: 5.5 seconds
cumulative wall time: 34.2 seconds

$ ./sage -t --long --warn-long 87.2 src/sage/parallel/map_reduce.py
Running doctests with ID 2019-10-01-08-21-35-400c0707.
Git branch: develop
Using --optional=build,dochtml,python2,sage
Doctesting 1 file.
sage -t --long --warn-long 87.2 src/sage/parallel/map_reduce.py
[294 tests, 33.43 s]
--
All tests passed!
--
Total time for all tests: 34.0 seconds
cpu time: 5.2 seconds
cumulative wall time: 33.4 seconds

### Python 3

Upgraded from SageMath 8.9.rc0 via git.
Ran `make`: success. Ran `make testlong` and individually
retested files that failed or timed out, see below.

$ make testlong
...
--
sage -t --long --warn-long 87.2 src/sage/coding/linear_code.py  # Timed out
sage -t --long --warn-long 87.2 src/sage/libs/eclib/interface.py  # 1
doctest failed
sage -t --long --warn-long 87.2
src/sage/rings/finite_rings/finite_field_base.pyx  # 1 doctest failed
sage -t --long --warn-long 87.2
src/sage/rings/polynomial/polynomial_rational_flint.pyx  # 1 doctest
failed
sage -t --long --warn-long 87.2 src/sage_setup/clean.py  # 1 doctest failed
--

$ ./sage -t --long --warn-long 87.2 src/sage/coding/linear_code.py
Running doctests with ID 2019-10-01-07-58-10-19e3fc53.
Git branch: develop
Using --optional=build,dochtml,python2,sage
Doctesting 1 file.
sage -t --long --warn-long 87.2 src/sage/coding/linear_code.py
[749 tests, 28.33 s]
--
All tests passed!
--
Total time for all tests: 29.2 seconds
cpu time: 25.8 seconds
cumulative wall time: 28.3 seconds

$ ./sage -t --long --warn-long 87.2 src/sage/libs/eclib/interface.py
Running doctests with ID 2019-10-01-07-58-59-bd646d17.
Git branch: develop
Using --optional=build,dochtml,python2,sage
Doctesting 1 file.
sage -t --long --warn-long 87.2 src/sage/libs/eclib/interface.py
**
File "src/sage/libs/eclib/interface.py", line 597, in
sage.libs.eclib.interface.mwrank_EllipticCurve.saturate
Failed example:
E.saturation([Q1,Q2])

Re: [sage-release] Sage 8.9 released

2019-09-29 Thread Samuel Lelièvre
Le lun. 30 sept. 2019 à 01:19, Volker Braun:
>
> The "master" git branch has been updated to Sage-8.9. [...]
>
> There was no change over 8.9.rc1

Thanks for the release!

Will there be py2 and py3 binaries for SageMath 8.9?

Also, a problem some users faced with the macOS binary
for SageMath 8.8 was as follows: installing extra packages
with pip requires to first run `sage -i openssl` followed by
`sage -f python2`, but this would fail to build scipy 1.2.0
and leave users with a nonworking Sage. Matthias Koeppe
diagnosed the likely cause and opened an issue about it
on binary-pkg:
https://github.com/sagemath/binary-pkg/issues/19

-- 
You received this message because you are subscribed to the Google Groups 
"sage-release" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-release+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-release/CAEcArF2SAxi8_Om7dmm9Qfor0pxo0H%3D1faBYjGzY5A%2BUDWpnsQ%40mail.gmail.com.