Re: [sympy] Status of old/new assumptions systems

2016-01-19 Thread Oscar Benjamin
On 18 January 2016 at 20:47, Aaron Meurer wrote: >>> For now, I would recommend using the old assumptions. However, you can >>> play with the new assumptions. The goal in making the new assumptions >>> read the old was to make this easier to do. >> >> But it also makes it a

Re: [sympy] Status of old/new assumptions systems

2016-01-19 Thread Aaron Meurer
On Tue, Jan 19, 2016 at 7:57 AM, Oscar Benjamin wrote: > On 18 January 2016 at 20:47, Aaron Meurer wrote: > >>> For now, I would recommend using the old assumptions. However, you can > >>> play with the new assumptions. The goal in making the new

Re: [sympy] Status of old/new assumptions systems

2016-01-18 Thread Oscar Benjamin
On 15 January 2016 at 15:06, Aaron Meurer wrote: > > In the released version (0.7.6.1), the systems are independent. In > master, the new assumptions (ask(), Q) read the old assumptions on > Symbols (is_positive, positive=True). > > For performance purposes, this was done in a

Re: [sympy] Status of old/new assumptions systems

2016-01-18 Thread Aaron Meurer
On Mon, Jan 18, 2016 at 6:32 AM, Oscar Benjamin wrote: > On 15 January 2016 at 15:06, Aaron Meurer wrote: >> >> In the released version (0.7.6.1), the systems are independent. In >> master, the new assumptions (ask(), Q) read the old assumptions on

Re: [sympy] Status of old/new assumptions systems

2016-01-15 Thread Aaron Meurer
In the released version (0.7.6.1), the systems are independent. In master, the new assumptions (ask(), Q) read the old assumptions on Symbols (is_positive, positive=True). For performance purposes, this was done in a way that prevents the new assumptions from noticing contradictions (it was added