Add and implement all phase functions to tabular backends

[JonWel]

Needed for ExternalMedia.

[JonWel]

Maybe adding the line phases calc_phase(void){return _phase;}; to the TabularBackends.h would solve it.
Is there an ideal place to add this? (I was thinking of line 775, but an other one may be good too)?

Others similar functions may be added this way:

phases calc_phase(void){return _phase;};
void calc_specify_phase(phases phase){ specify_phase(phase); }
void calc_unspecify_phase(){ unspecify_phase(); }

Should I add them?

[ibell]

Currently phase is not used/implemented for tabular backends, is it strictly needed?

I'm unavailable until Tuesday of next week, you are on your own until them. I think I should be able to merge pull requests though.

[JonWel]

It is used to know when to apply twophase_derivsmoothing_xend or rho_smoothing_xend.

(adding this line made the code run but the answer is always 7, thus never two phases 6, I should have check that).

[ibell]

Maybe there is a bug in the phase determination for tabular backends
(probably there is actually!). Can you come up with a minimal example that
demonstrates the problem? It would be good if you could check all the
domains in the plot in
http://www.coolprop.org/coolprop/HighLevelAPI.html#phasesi-function

On Tue, Aug 4, 2015 at 12:43 AM, JonWel notifications@github.com wrote:

It's used to know when to apply twophase_derivsmoothing_xend or
rho_smoothing_xend.

(adding this line made the code run but the answer is always 7, thus
never two phases 6).

—
Reply to this email directly or view it on GitHub
#763 (comment).

[JonWel]

Trying to reproduce the plot from http://www.coolprop.org/coolprop/HighLevelAPI.html#phasesi-function only gives 7.0 (one phase) on all the tested domain (T = 280 to 700, P = 1e4 to 3e7, giving a matrix of 10000x10000).
So I guess the _phase variable is set to 7.0 somewhere and as there is no implementation of phase, it never change.

[ibell]

Yeah, that isn't all that surprising. I haven't implemented the phase
functions yet, and I will do so in the next week hopefully.

On Mon, Aug 24, 2015 at 4:42 AM, JonWel notifications@github.com wrote:

Trying to reproduce the plot from
http://www.coolprop.org/coolprop/HighLevelAPI.html#phasesi-function only
gives 7.0 (one phase) on the tested domain (T = 280 to 700, P = 1e4 to
3e7, giving a matrix of 10000x10000)

—
Reply to this email directly or view it on GitHub
#763 (comment).

[JonWel]

Will the specify_phase function be added to the tabular backends when the phase will be implemented? (The all in the title would make me say yes).