CELL_DATA output for VTK, Gerris.

[s f1]

 Dear Basilisk users,

I have the attached simple Gerris run file (the run takes less than a second). (I am still using Gerris as I need contact angle bc.)

In the attaced file there is no two-phase flow, just a simple example from Gerris tutorial. 

My question:

I would like to output CELL_DATA for VTK visualisation.

By default, the VTK output is POINT_DATA at nodes of a cell. I tried several combinations of GfsOutputSimulation and found using "format = text" works but needs quite an effort to organising the data to match VTK cell structure data output. If I have a test case comprising of even 5000 cells that would be a daunting task.

Also, I see that the "format = gfs" writes the cell data which is exactly what I want.

Reason why I am doing?

Might sound funny!! I have few other data in VTK format that I post-process using Paraview. I realise that the color bar in gfsview brighter compared to that of paraview and this is causing issues in visualising the output. Therefore, I am trying to write VTK output and visualise in paraview.

Hope anyone can advise me,

Sai

[Stephane Popinet]

Hi Sai,

> (I am still using Gerris as I need contact angle bc.)

Basilisk can now do contact angles, also in 3D and also in parallel
(unlike Gerris). See:

http://basilisk.fr/src/test/sessile.c
http://basilisk.fr/src/test/sessile3D.c

cheers,

Stephane

[H LIU]

Hi Stephane,

I find that if the "two-phase.h" is included together with "contact.h", a warning will appear during the compilation: 

-------------------------------------------------------------------------------------------
$BASILISK/contact.h:12 warning : "interface_normal" redefined

12 | #define interface_normal(point, c) interface_normal (point, c)

-------------------------------------------------------------------------------------------

and the compilation can be completed and the program can run normally， but I wonder if redefinition of interface_normal influences the simulation result or not.

Thank you very much!

Hu LIU

在 2019年2月22日星期五 UTC+8下午3:27:50，Stephane Popinet写道：

[Venkata Sai Anvesh Sangadi]

Hi LIU and Stephane, 

>I find that if the "two-phase.h" is included together with "contact.h", a warning will appear during the compilation: 

>-------------------------------------------------------------------------------------------
>$BASILISK/contact.h:12 warning : "interface_normal" redefined

>12 | #define interface_normal(point, c) interface_normal (point, c)

>-------------------------------------------------------------------------------------------

I too am facing the same warning and I am concerned if redefinition of the normal will affect the results. Additionally, please suggest  a workaround to this. 
Thanks in Advance. 

Anvesh Sangadi 

IIT Hyderbad, India