Re: [sage-devel] Using hidden attributes in methods

2015-06-05 Thread Jeroen Demeyer
On 2015-06-04 18:23, Bruno Grenet wrote: Thus if I implement a method `number_of_terms` in the class `Polynomial_generic_sparse`, I am tempted to implement the third solution. Since this is a class where the attribute is defined, you don't need to mangle anything. Otherwise, I would say that

Re: [sage-devel] What is the parent() of a vector from a subspace?

2015-06-05 Thread Johan S. R. Nielsen
On Wednesday, June 3, 2015 at 11:02:42 PM UTC+2, David Roe wrote: For a vector space V over an inexact field, the statement that a vector v belongs to a particular subspace W cannot be checked by examining an approximation to v in V. For example, (2 + O(3^50), 4 + O(3^50)) may or may

[sage-devel] Online Bug report form: is it dead?

2015-06-05 Thread Nathann Cohen
Hello everybody, It is apparently possible to report bugs through an online form on Sage's website [1]. Is anybody reading those reports? If nobody tracks them and if nobody volunteers to do it from now on, I propose that we remove this link. By having users submit bug reports that nobody

[sage-devel] Re: Online Bug report form: is it dead?

2015-06-05 Thread Jakob Kroeker
where do the sagenb reports end? here? https://docs.google.com/spreadsheets/d/1GF_-hM-3DP0YcViieBWIhsign3JHegTlj32cs8ymPZE/pub Am Freitag, 5. Juni 2015 12:52:06 UTC+2 schrieb Nathann Cohen: Hello everybody, It is apparently possible to report bugs through an online form on Sage's

[sage-devel] Re: sage-6.7 build failed on Cygwin64 for python-2.7.8 No module named crypt

2015-06-05 Thread Jean-Pierre Flori
On Friday, June 5, 2015 at 2:49:40 PM UTC+2, Stefan Auracher wrote: the logfile ~/sage-6.7/logs/pkgs/python-2.7.8.log at the very end after make[3]: Leaving directory '/home/XX/sage-6.7/local/var/tmp/sage/build/python-2.7.8/src' failed with: Installing valgrind suppression

[sage-devel] sage-6.7 build failed on Cygwin64 for python-2.7.8 No module named crypt

2015-06-05 Thread Stefan Auracher
the logfile ~/sage-6.7/logs/pkgs/python-2.7.8.log at the very end after make[3]: Leaving directory '/home/XX/sage-6.7/local/var/tmp/sage/build/python-2.7.8/src' failed with: Installing valgrind suppression file... Testing importing of various modules... ctypes module imported OK math

Re: [sage-devel] Using hidden attributes in methods

2015-06-05 Thread Bruno Grenet
Thank you for your answers. I was actually aware of hidden methods being accessible from inside the class, but I somehow gave a bad example... I note the advices that 1. Having to do manual mangling probably reveals some bad structuration, and 2. Instead of do manual mangling, one can

[sage-devel] Re: Online Bug report form: is it dead?

2015-06-05 Thread kcrisman
Occasionally some of us go through the Google docs spreadsheet and try to see what is on them. Note that only older versions of the sage notebook should have this link there, I *thought* it was replaced with a link to ask.sagemath. The Report Bugs link at the top of www.sagemath.org goes to

[sage-devel] Re: need help posting to trac

2015-06-05 Thread kcrisman
I have just made a *lot* of revisions to the files in the sage/sandpiles/ directory and would like to post then to the trac system to be reviewed and included in a future version of sage. I have followed the instructions on the collaborative development with git-trac page (

Re: [sage-devel] Re: Extremely broad bug in Sage Integral Computations

2015-06-05 Thread kcrisman
Try it again with domain:complex Both x and -x are real if x is real. If setting domain:complex fixes something, then it is certainly poorly named. Well, I can't answer for Maxima internals :-) but I would agree that there are often very subtle side-effects in Maxima from things like

[sage-devel] Re: need help posting to trac

2015-06-05 Thread David Perkinson
Once I have successfully navigated this process, I would happily provide feedback on the documentation. (Thanks for your tolerance with my questions!) On Friday, June 5, 2015 at 7:15:39 PM UTC+1, kcrisman wrote: I have just made a *lot* of revisions to the files in the sage/sandpiles/

[sage-devel] need help posting to trac

2015-06-05 Thread David Perkinson
I have just made a *lot* of revisions to the files in the sage/sandpiles/ directory and would like to post then to the trac system to be reviewed and included in a future version of sage. I have followed the instructions on the collaborative development with git-trac page

[sage-devel] Re: sage-6.7 build failed on Cygwin64 for python-2.7.8 No module named crypt

2015-06-05 Thread kcrisman
I'm not an expert on compiling on Cygwin. What can I do? Possibly the already installed python 2.7.10 is a problem ? Follow http://trac.sagemath.org/wiki/Cygwin64Port :) On Cygwin64 libcrypt-devel is not installed by default; Is there any way we could get some of that information in

Re: [sage-devel] Re: Extremely broad bug in Sage Integral Computations

2015-06-05 Thread rjf
On Thursday, June 4, 2015 at 6:21:43 AM UTC-7, kcrisman wrote: 1. anyone who believes that sqrt(x^2) is |x| is mistaken. As Bill Page says, there are two values. Ah, the old 'function versus expression' debate. I really missed that. 2. any system that automatically produces |x|

[sage-devel] Re: need help posting to trac

2015-06-05 Thread Volker Braun
On Friday, June 5, 2015 at 8:15:39 PM UTC+2, kcrisman wrote: You presumably shouldn't have to clone, I *think* every Sage copy comes with a git repo at this point - is that correct, all? Yes. The most important git command is git status and tells you where you currently are. The preferred

[sage-devel] Re: need help posting to trac

2015-06-05 Thread kcrisman
Once I have successfully navigated this process, I would happily provide feedback on the documentation. (Thanks for your tolerance with my questions!) Opposite - thank you for your tolerance of the challenge to contribution imposed by the new workflow. (For quick things, you can simply