2014-05-22 13:39 GMT-03:00 Thomas Spura tomspur@fedoraproject.org:

On Thu, May 22, 2014 at 6:35 PM, Paulo César Pereira de Andrade paulo.cesar.pereira.de.andrade@gmail.com wrote:

...

2014-05-22 12:32 GMT-03:00 Paul Howarth paul@city-fan.org:

...

On 17/05/14 17:39, Paulo César Pereira de Andrade wrote:

...

2014-05-16 21:23 GMT-03:00 Paul Howarth paul@city-fan.org:

...

On Fri, 16 May 2014 12:37:19 -0300 Paulo César Pereira de Andrade paulo.cesar.pereira.de.andrade@gmail.com wrote:

...

Hi sorry for adding the CC, but trying hard to get you attention :-)

2014-05-15 16:50 GMT-03:00 Paulo César Pereira de Andrade paulo.cesar.pereira.de.andrade@gmail.com: > > I am in the last steps on getting sagemath 6.2 working, so, > what should be done:

It would be better to coordinate the update, so...

> o Update to pari 2.7.0 -- optional

At least for sagemath it is ok to rebuild pari 2.7.0; I will use this patch:

https://abf.rosalinux.ru/openmandriva/sagemath/blob/master/sagemath-pari2.7.... and drop this:

http://pkgs.fedoraproject.org/cgit/sagemath.git/tree/sagemath-nopari2.6.patc...

Note that dist-git now has pari 2.7.1 but it's not built yet; that will change when the mass rebuild lands.

For sagemath I can apply the patches to support pari 2.7 later, but it probably would be better to build pari earlier. At least sagemath is already broken for almost a month in rawhide due to other dependencies.

Can you rebuild pari in the next few days please? This would also ensure that the mass rebuild would not let pass broken packages in case it builds packages that needs pari before pari itself.

I've now built pari 2.7.1 in Rawhide. I kept the old library around for now to avoid broken deps whilst packages are migrated to the new version.

Many thanks! Major issue now for sagemath is ipython, that I hope should be resolved and available soon.

I got now the unbundling exception approved and want to update python-tornado after the mass rebuild is over. Then I'll update ipython as ipthon requires a newer python-tornado.

Would also ipython 2.X be fine, or do you must have version 1.2.1? I was already working on the 2.X series before you were writing about version 1.2.1... :/

It starts and can start the notebook without problems, so I think it is safe. Only one warning:

---%<--- $ sage ┌────────────────────────────────────────────────────────────────────┐ │ Sage Version 6.2, Release Date: 2014-05-06 │ │ Type "notebook()" for the browser-based notebook interface. │ │ Type "help()" for help. │ └────────────────────────────────────────────────────────────────────┘ WARNING: Hook shutdown_hook is deprecated. Use the atexit module instead. sage: ---%<---

I have no reasons to think it could cause problems.

Greetings, Tom

Thanks, Paulo

2014-05-22 17:13 GMT-03:00 Thomas Spura tomspur@fedoraproject.org:

On Thu, May 22, 2014 at 10:10 PM, Paulo César Pereira de Andrade paulo.cesar.pereira.de.andrade@gmail.com wrote:

...

2014-05-22 13:39 GMT-03:00 Thomas Spura tomspur@fedoraproject.org:

...

On Thu, May 22, 2014 at 6:35 PM, Paulo César Pereira de Andrade paulo.cesar.pereira.de.andrade@gmail.com wrote:

...

2014-05-22 12:32 GMT-03:00 Paul Howarth paul@city-fan.org:

...

On 17/05/14 17:39, Paulo César Pereira de Andrade wrote:

...

2014-05-16 21:23 GMT-03:00 Paul Howarth paul@city-fan.org: > > On Fri, 16 May 2014 12:37:19 -0300 > Paulo César Pereira de Andrade > paulo.cesar.pereira.de.andrade@gmail.com wrote: > >> Hi sorry for adding the CC, but trying hard to get you attention :-) >> >> 2014-05-15 16:50 GMT-03:00 Paulo César Pereira de Andrade >> paulo.cesar.pereira.de.andrade@gmail.com: >>> >>> I am in the last steps on getting sagemath 6.2 working, so, >>> what should be done: >> >> >> It would be better to coordinate the update, so... >> >>> o Update to pari 2.7.0 -- optional >> >> >> At least for sagemath it is ok to rebuild pari 2.7.0; I will use >> this patch: >> >> https://abf.rosalinux.ru/openmandriva/sagemath/blob/master/sagemath-pari2.7.... >> and drop this: >> >> http://pkgs.fedoraproject.org/cgit/sagemath.git/tree/sagemath-nopari2.6.patc... > > > Note that dist-git now has pari 2.7.1 but it's not built yet; that will > change when the mass rebuild lands.

For sagemath I can apply the patches to support pari 2.7 later, but it probably would be better to build pari earlier. At least sagemath is already broken for almost a month in rawhide due to other dependencies.

Can you rebuild pari in the next few days please? This would also ensure that the mass rebuild would not let pass broken packages in case it builds packages that needs pari before pari itself.

I've now built pari 2.7.1 in Rawhide. I kept the old library around for now to avoid broken deps whilst packages are migrated to the new version.

Many thanks! Major issue now for sagemath is ipython, that I hope should be resolved and available soon.

I got now the unbundling exception approved and want to update python-tornado after the mass rebuild is over. Then I'll update ipython as ipthon requires a newer python-tornado.

Would also ipython 2.X be fine, or do you must have version 1.2.1? I was already working on the 2.X series before you were writing about version 1.2.1... :/

It starts and can start the notebook without problems, so I think it is safe. Only one warning:

---%<--- $ sage ┌────────────────────────────────────────────────────────────────────┐ │ Sage Version 6.2, Release Date: 2014-05-06 │ │ Type "notebook()" for the browser-based notebook interface. │ │ Type "help()" for help. │ └────────────────────────────────────────────────────────────────────┘ WARNING: Hook shutdown_hook is deprecated. Use the atexit module instead. sage: ---%<---

I have no reasons to think it could cause problems.

Great!

I'll then now prepare the pyhton-tornado update for next week and next is ipython. Should I let you know, when I have a scratch build of ipython before updating it in rawhide, so you can do the final testings?

Usually, if it is going to fail it fails at start (all I got was the warning), so I do not believe there will be any issues. But it would be ok to know about a scratch build and better yet when it is available in rawhide.

Greetings, Tom

Thanks, Paulo

2014-05-23 17:46 GMT-03:00 Thomas Spura tomspur@fedoraproject.org:

2014-05-22 22:23 GMT+02:00 Paulo César Pereira de Andrade paulo.cesar.pereira.de.andrade@gmail.com:

...

2014-05-22 17:13 GMT-03:00 Thomas Spura tomspur@fedoraproject.org:

...

On Thu, May 22, 2014 at 10:10 PM, Paulo César Pereira de Andrade paulo.cesar.pereira.de.andrade@gmail.com wrote:

...

2014-05-22 13:39 GMT-03:00 Thomas Spura tomspur@fedoraproject.org:

...

On Thu, May 22, 2014 at 6:35 PM, Paulo César Pereira de Andrade paulo.cesar.pereira.de.andrade@gmail.com wrote:

...

2014-05-22 12:32 GMT-03:00 Paul Howarth paul@city-fan.org: > On 17/05/14 17:39, Paulo César Pereira de Andrade wrote: >> >> 2014-05-16 21:23 GMT-03:00 Paul Howarth paul@city-fan.org: >>> >>> On Fri, 16 May 2014 12:37:19 -0300 >>> Paulo César Pereira de Andrade >>> paulo.cesar.pereira.de.andrade@gmail.com wrote: >>> >>>> Hi sorry for adding the CC, but trying hard to get you attention :-) >>>> >>>> 2014-05-15 16:50 GMT-03:00 Paulo César Pereira de Andrade >>>> paulo.cesar.pereira.de.andrade@gmail.com: >>>>> >>>>> I am in the last steps on getting sagemath 6.2 working, so, >>>>> what should be done: >>>> >>>> >>>> It would be better to coordinate the update, so... >>>> >>>>> o Update to pari 2.7.0 -- optional >>>> >>>> >>>> At least for sagemath it is ok to rebuild pari 2.7.0; I will use >>>> this patch: >>>> >>>> https://abf.rosalinux.ru/openmandriva/sagemath/blob/master/sagemath-pari2.7.... >>>> and drop this: >>>> >>>> http://pkgs.fedoraproject.org/cgit/sagemath.git/tree/sagemath-nopari2.6.patc... >>> >>> >>> Note that dist-git now has pari 2.7.1 but it's not built yet; that will >>> change when the mass rebuild lands. >> >> >> For sagemath I can apply the patches to support pari 2.7 later, but >> it probably would be better to build pari earlier. At least sagemath >> is already broken for almost a month in rawhide due to other >> dependencies. >> >> Can you rebuild pari in the next few days please? This would also >> ensure that the mass rebuild would not let pass broken packages >> in case it builds packages that needs pari before pari itself. > > > I've now built pari 2.7.1 in Rawhide. I kept the old library around for now > to avoid broken deps whilst packages are migrated to the new version.

Many thanks! Major issue now for sagemath is ipython, that I hope should be resolved and available soon.

I got now the unbundling exception approved and want to update python-tornado after the mass rebuild is over. Then I'll update ipython as ipthon requires a newer python-tornado.

Would also ipython 2.X be fine, or do you must have version 1.2.1? I was already working on the 2.X series before you were writing about version 1.2.1... :/

It starts and can start the notebook without problems, so I think it is safe. Only one warning:

---%<--- $ sage ┌────────────────────────────────────────────────────────────────────┐ │ Sage Version 6.2, Release Date: 2014-05-06 │ │ Type "notebook()" for the browser-based notebook interface. │ │ Type "help()" for help. │ └────────────────────────────────────────────────────────────────────┘ WARNING: Hook shutdown_hook is deprecated. Use the atexit module instead. sage: ---%<---

I have no reasons to think it could cause problems.

Great!

I'll then now prepare the pyhton-tornado update for next week and next is ipython. Should I let you know, when I have a scratch build of ipython before updating it in rawhide, so you can do the final testings?

Usually, if it is going to fail it fails at start (all I got was the warning), so I do not believe there will be any issues. But it would be ok to know about a scratch build and better yet when it is available in rawhide.

I think the ipython package won't change much after this scratch build: http://koji.fedoraproject.org/koji/taskinfo?taskID=6883972

I'm waiting for feedback in the fpc ticket, if all bundled() provides are sane. So possibly by the end of the next week I can built it. Or beginning of June before the mass rebuilt.

I was expecting rebuilding the documentation would not load ipython deps, but it did, so it should be the only issue preventing sagemath update now:

https://kojipkgs.fedoraproject.org//work/tasks/3246/6903246/build.log

Greetings, Tom

Thanks, Paulo

2014-05-29 21:08 GMT-03:00 Thomas Spura tomspur@fedoraproject.org:

Hi Paulo,

2014-05-28 18:23 GMT+02:00 Paulo César Pereira de Andrade paulo.cesar.pereira.de.andrade@gmail.com:

[huge snip]

...

I was expecting rebuilding the documentation would not load ipython deps, but it did, so it should be the only issue preventing sagemath update now:

https://kojipkgs.fedoraproject.org//work/tasks/3246/6903246/build.log

The TerminalInteractiveShell is at the same place in version 1 and 2, so sagemath should at least in theory work with both versions.

I just built it in rawhide. Please give it a try: http://koji.fedoraproject.org/koji/taskinfo?taskID=6909927

Well, I already knew it would work :-) sagemath 6.2 is now available in rawhide. Unless there is a lot of people asking for it (and helping), I think working on a sagemath 6.2 update for f20 should not be done; need significant work and ipython backported would be one of the easiest tasks...

So, I should close a few sagemath bug reports as NEXTRELEASE.

Greetings, Tom

Thanks, Paulo