Opened 9 years ago

Closed 9 years ago

#11415 closed defect (fixed)

update pynac to 0.2.3

Reported by: burcin Owned by: burcin
Priority: major Milestone: sage-4.7.1
Component: packages: standard Keywords:
Cc: kcrisman Merged in: sage-4.7.1.alpha4
Authors: Richard Kreckel, Jens Vollinga, Burcin Erocal Reviewers: Karl-Dieter Crisman, François Bissey
Report Upstream: N/A Work issues:
Branch: Commit:
Dependencies: Stopgaps:

Description

Yet another pynac release, including fixes for #10964, #11352 and #9240.

http://sage.math.washington.edu/home/burcin/pynac/pynac-0.2.3.spkg

Attached patch fixes some doctest to remove extra braces printed when typesetting mul objects.

Attachments (1)

trac_11415-remove_extra_braces.patch (1.3 KB) - added by burcin 9 years ago.

Download all attachments as: .zip

Change History (13)

Changed 9 years ago by burcin

comment:1 Changed 9 years ago by kcrisman

  • Cc kcrisman added

comment:2 follow-up: Changed 9 years ago by kcrisman

  • Reviewers set to Karl-Dieter Crisman

I'm confused about

changeset:   197:3c39cd396893
user:        Burcin Erocal <burcin@erocal.org>
date:        Tue May 31 16:37:48 2011 +0200
summary:     Use "%lf" to print doubles instead of "%f".

Not that it seems to be affecting anything, though one would think it would print things differently...

Anyway, I don't see that in the Ginac commit list, but I don't see it mentioned in any of the relevant tickets either.

Otherwise, so far, so good. Now to look at the actual patches and tickets involved. #9240 I will probably not get to right away.

comment:3 Changed 9 years ago by kcrisman

Positive review on the patch in this ticket.

comment:4 Changed 9 years ago by kcrisman

  • Status changed from new to needs_review

Positive review modulo an answer to the question about why the change from && to & was necessary in the Pynac patch for #9240.

comment:5 in reply to: ↑ 2 ; follow-up: Changed 9 years ago by fbissey

Replying to kcrisman:

I'm confused about

changeset:   197:3c39cd396893
user:        Burcin Erocal <burcin@erocal.org>
date:        Tue May 31 16:37:48 2011 +0200
summary:     Use "%lf" to print doubles instead of "%f".

Not that it seems to be affecting anything, though one would think it would print things differently...

Hi Karl,

Burcin introduced this on my report because gcc complains. See here https://bitbucket.org/burcin/pynac/issue/2/correct-format-in-numericcpp. This is not about printing but about entering data if you look at the commit https://bitbucket.org/burcin/pynac/changeset/3c39cd396893 this is in scanf not printf. So printing is not affected.

comment:6 in reply to: ↑ 5 Changed 9 years ago by kcrisman

Burcin introduced this on my report because gcc complains. See here https://bitbucket.org/burcin/pynac/issue/2/correct-format-in-numericcpp. This is not about printing but about entering data if you look at the commit https://bitbucket.org/burcin/pynac/changeset/3c39cd396893 this is in scanf not printf. So printing is not affected.

Thanks, Francois! I have to say, David Kirkby would love this - someone fixing a warning and not just an actual error in an upstream package! If you watch Sage compile, it's amazing just how many warnings there are - small wonder we need gcc!

And since I know very little about C and its derivatives (enough to get in trouble, I guess), that explanation was actually very helpful.

Now I just need to know the answer about the bitwise and versus boolean and.

comment:7 follow-up: Changed 9 years ago by fbissey

I actually didn't report all the warnings. Just the easy to fix one. There's some of that as well:

/scratch/portage/sci-libs/pynac-0.2.2/work/pynac-0.2.2/src/ginac/numeric.cpp: In function 'PyObject* Integer(const long int&)':
/scratch/portage/sci-libs/pynac-0.2.2/work/pynac-0.2.2/src/ginac/numeric.cpp:173:56: warning: deprecated conversion from string constant to 'char*'
/scratch/portage/sci-libs/pynac-0.2.2/work/pynac-0.2.2/src/ginac/numeric.cpp: In function 'void GiNaC::coerce(GiNaC::Number_T&, GiNaC::Number_T&, const GiNaC::Number_T&, const GiNaC::Number_T&)':
/scratch/portage/sci-libs/pynac-0.2.2/work/pynac-0.2.2/src/ginac/numeric.cpp:262:67: warning: deprecated conversion from string constant to 'char*'

But you wouldn't believe how widespread this one is. Just look at your singular build logs for example when you have time. I may attempt your other question but in #9240 (he says looking at his C manual and looking at the code).

comment:8 Changed 9 years ago by kcrisman

Yes, I've seen lots of those as well.

comment:9 Changed 9 years ago by kcrisman

  • Reviewers changed from Karl-Dieter Crisman to Karl-Dieter Crisman, François Bissey
  • Status changed from needs_review to positive_review

Well, fbissey seems to be convinced in #9240 that the && to & is ok, and it doesn't break anything, and it's certainly in the right spirit (just what kind of and), so I think that suffices for the new Pynac. #9240 itself needs some clarification, at least for me.

comment:10 in reply to: ↑ 7 Changed 9 years ago by burcin

Replying to fbissey:

I actually didn't report all the warnings. Just the easy to fix one. There's some of that as well:

/scratch/portage/sci-libs/pynac-0.2.2/work/pynac-0.2.2/src/ginac/numeric.cpp: In function 'PyObject* Integer(const long int&)':
/scratch/portage/sci-libs/pynac-0.2.2/work/pynac-0.2.2/src/ginac/numeric.cpp:173:56: warning: deprecated conversion from string constant to 'char*'
/scratch/portage/sci-libs/pynac-0.2.2/work/pynac-0.2.2/src/ginac/numeric.cpp: In function 'void GiNaC::coerce(GiNaC::Number_T&, GiNaC::Number_T&, const GiNaC::Number_T&, const GiNaC::Number_T&)':
/scratch/portage/sci-libs/pynac-0.2.2/work/pynac-0.2.2/src/ginac/numeric.cpp:262:67: warning: deprecated conversion from string constant to 'char*'

But you wouldn't believe how widespread this one is. Just look at your singular build logs for example when you have time. I may attempt your other question but in #9240 (he says looking at his C manual and looking at the code).

I try to fix the warnings I know about. Thanks Francois for reporting the %lf issue.

Unfortunately the ones above are due to the python headers not declaring the char* arguments of some functions as const. I don't think it's a good idea to manually cast to a nonconstant type to suppress these warnings.

Many thanks for the reviews Karl-Dieter.

comment:11 Changed 9 years ago by jdemeyer

  • Component changed from symbolics to packages

comment:12 Changed 9 years ago by jdemeyer

  • Merged in set to sage-4.7.1.alpha4
  • Resolution set to fixed
  • Status changed from positive_review to closed
Note: See TracTickets for help on using tickets.