GiD forum

Hi all,
Recently I needed information about the points (end-lines) defining the boundary of surfaces. I
tried to access them in Tcl through the command:

.central.s info list_entities Surfaces $surfnum

getting something like:

“NURBSURFACE
Num: 1 HigherEntity: 1 conditions: 2 material: 5
LAYER: Layer0
Meshing Info: (num=2,NOE=1) Elemtype=3 IsStructured=1 size=0 Meshing=Yes
NumLines: 4
Line: 4 Orientation: DIFF1ST
Line: 3 Orientation: DIFF1ST
Line: 2 Orientation: DIFF1ST
Line: 1 Orientation: DIFF1ST
Number of Control Points= 2 2 Degree= 1 1
Point 1,1 coords:-0.050000,-0.050000,0.000000
Point 1,2 coords:-0.050000,1.050000,0.000000
Point 2,1 coords:1.050000,-0.050000,0.000000
Point 2,2 coords:1.050000,1.050000,0.000000
Number of knots in U= 4
knot 1 value=0.000000
knot 2 value=0.000000
knot 3 value=1.000000
knot 4 value=1.000000
Number of knots in V= 4
knot 1 value=0.000000
knot 2 value=0.000000
knot 3 value=1.000000
knot 4 value=1.000000
Non rational
IsTrimmed: 1
Center: 0.500000 0.500000 0.000000
Normal: 0.000000 0.000000 -1.000000
END NURBSURFACE“


I noticed that the so called points are not always the end-points of the lines surrounding a
surface. Instead some points (normally the majority) are just outside the surface corners.
In the example above the real surface is a square:
Point 1,1 coords:0.0,0.0,0.0
Point 1,2 coords:0.0,1.0,0.0
Point 2,1 coords:1.0,0.0,0.0
Point 2,2 coords:1.0,1.0,0.0

I know nothing about surface modelling, but would like to ask if it is fine.
In any case, what would be the way of knowing the exact location of the points defining the
corners (end-lines) of a surface?
Thanks for any suggestion




Andrea Calaon

Ashurst Lodge, Ashurst
Southampton, Hampshire
SO40 7AA, UK
Tel. office: +44 23 80 293 223
Fax office: +44 23 80 292 853
Mobile: +44 7766 488 225
E-mail: andrea at beasy.com

The points "apparently" out of the surface, are because it's a trimmed
surface: it's defined by a base surface shape (in this case the planar
square), and a closed set of trimming curves. Then only is considerated the
surface domain "inside" this curves.
Note: you can use Geometry-Edit-Edit nurb surface... to show/edit the
control points of this base surface.

You need to list the information of the lines of this surface, then get the
start/end point number, and list this point to obtain the desired
coordinates information.

Regards
Enrique Escolano

----- Original Message -----
From: "Andrea Calaon" andrea at beasy.com
To: gidlist at gatxan.cimne.upc.es
Sent: Thursday, September 16, 2004 11:20 AM
Subject: [GiDlist] Surface definition Points


Hi all,
Recently I needed information about the points (end-lines) defining the
boundary of surfaces. I
tried to access them in Tcl through the command:

.central.s info list_entities Surfaces $surfnum

getting something like:

"NURBSURFACE
Num: 1 HigherEntity: 1 conditions: 2 material: 5
LAYER: Layer0
Meshing Info: (num=2,NOE=1) Elemtype=3 IsStructured=1 size=0 Meshing=Yes
NumLines: 4
Line: 4 Orientation: DIFF1ST
Line: 3 Orientation: DIFF1ST
Line: 2 Orientation: DIFF1ST
Line: 1 Orientation: DIFF1ST
Number of Control Points= 2 2 Degree= 1 1
Point 1,1 coords:-0.050000,-0.050000,0.000000
Point 1,2 coords:-0.050000,1.050000,0.000000
Point 2,1 coords:1.050000,-0.050000,0.000000
Point 2,2 coords:1.050000,1.050000,0.000000
Number of knots in U= 4
knot 1 value=0.000000
knot 2 value=0.000000
knot 3 value=1.000000
knot 4 value=1.000000
Number of knots in V= 4
knot 1 value=0.000000
knot 2 value=0.000000
knot 3 value=1.000000
knot 4 value=1.000000
Non rational
IsTrimmed: 1
Center: 0.500000 0.500000 0.000000
Normal: 0.000000 0.000000 -1.000000
END NURBSURFACE"


I noticed that the so called points are not always the end-points of the
lines surrounding a
surface. Instead some points (normally the majority) are just outside the
surface corners.
In the example above the real surface is a square:
Point 1,1 coords:0.0,0.0,0.0
Point 1,2 coords:0.0,1.0,0.0
Point 2,1 coords:1.0,0.0,0.0
Point 2,2 coords:1.0,1.0,0.0

I know nothing about surface modelling, but would like to ask if it is
fine.
In any case, what would be the way of knowing the exact location of the
points defining the
corners (end-lines) of a surface?
Thanks for any suggestion




Andrea Calaon

Ashurst Lodge, Ashurst
Southampton, Hampshire
SO40 7AA, UK
Tel. office: +44 23 80 293 223
Fax office: +44 23 80 292 853
Mobile: +44 7766 488 225
E-mail: andrea at beasy.com

_______________________________________________
GiDlist mailing list
GiDlist at gid.cimne.upc.es
http://gid.cimne.upc.es/mailman/listinfo/gidlist

Thanks Enrique!
I'll loop over the lines end extract the start/end points.
Regards

On 16 Sep 2004 at 19:35, Enrique Escolano wrote:

The points "apparently" out of the surface, are because it's a trimmed
surface: it's defined by a base surface shape (in this case the planar
square), and a closed set of trimming curves. Then only is considerated the
surface domain "inside" this curves.
Note: you can use Geometry-Edit-Edit nurb surface... to show/edit the
control points of this base surface.

You need to list the information of the lines of this surface, then get the
start/end point number, and list this point to obtain the desired
coordinates information.

Regards
Enrique Escolano



Andrea Calaon

Ashurst Lodge, Ashurst
Southampton, Hampshire
SO40 7AA, UK
Tel. office: +44 23 80 293 223
Fax office: +44 23 80 292 853
Mobile: +44 7766 488 225
E-mail: andrea at beasy.com

Hello,

It looks like nelem in version 7.4.9b does not include linear elements. I
could not find any posts about that. I went back to 7.2, in which nelem
does include the linear elements.

In brief, I have a model with:
Num. of linear elements=5
Num. of Triangle elements=832
Num. of Quadrilateral elements=72
Num. of nodes=2020

In version 7.2, nelem = 909.
In version 7.4.9b, nelem = 904.

Was the behavior of nelem changed on purpose?

Philippe

An HTML attachment was scrubbed...
URL: http://listas.cimne.upc.edu/pipermail/gidlist/attachments/20040923/f810be4f/attachment.htm

Hello Philippe.

*nelem has not changed on purpose.
I tried a small test and it worked ok.

Can you send your bas file, and and a small sample.
(can send directly to me: escolano at cimne.upc.es)

Regards
Enrique Escolano

----- Original Message -----
From: "Philippe Bouchilloux" pb at magsoft-flux.com
To: "GiDList" gidlist at gatxan.cimne.upc.es
Sent: Thursday, September 23, 2004 1:36 AM
Subject: {SPAM?} [GiDlist] nelem - Behavior changed?


Hello,

It looks like nelem in version 7.4.9b does not include linear elements. I
could not find any posts about that. I went back to 7.2, in which nelem
does include the linear elements.

In brief, I have a model with:
Num. of linear elements=5
Num. of Triangle elements=832
Num. of Quadrilateral elements=72
Num. of nodes=2020

In version 7.2, nelem = 909.
In version 7.4.9b, nelem = 904.

Was the behavior of nelem changed on purpose?

Philippe

_______________________________________________
GiDlist mailing list
GiDlist at gid.cimne.upc.es
http://gid.cimne.upc.es/mailman/listinfo/gidlist

The functions to get system information are very platform dependent.

We can provide directly to you some procedure to manage your problemtype register with the "GiD plus style".

Enrique Escolano

----- Original Message -----
From: De silva mahinda
To: gidlist at gatxan.cimne.upc.es
Sent: Thursday, September 23, 2004 4:03 AM
Subject: [GiDlist] Wants to read system information and show them as in Registry Module in GiD


Dear Sir,
We are using GiD as our pre and post processor. We made security, and its dialog box contain only lable to enter password. But we want to read some more information like in GiD and show them to user. At the moment what we are doing we read them inside main processor XFINAS and save them in another file and ask user to send that file.

if you can explain how to show the system information in Security dialog box as in your Registry Module , hope it will be easier for user as they can see it once they use. What we want to read at the moment is hard dive serial number and computer name

With regards,
De Silva




------------------------------------------------------------------------------
Help STOP spam with the new MSN 8 and get 2 months FREE* _______________________________________________ GiDlist mailing list GiDlist at gid.cimne.upc.es http://gid.cimne.upc.es/mailman/listinfo/gidlist
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://listas.cimne.upc.edu/pipermail/gidlist/attachments/20040929/898ac5f3/attachment.htm

Hello Enrique,

After further analysis, this is what I find: the nelem command is modified
by the remove command (for instance, remove(Linear) or remove(Triangle)).
Unfortunately, even the set elems(All) command does not restore the
situation to normal (i.e. nelem returns the total number of elements minus
the number of elements removed).

I'll send you a test case to your personal email address right away.

Philippe


*nelem has not changed on purpose.
I tried a small test and it worked ok.

Hello Philippe.
As you related, it exists some problem using *nelem (equivalent to *nelem(all)) after use *remove elems (nelem is not updated)

Until this bug is corrected (for the next beta 7.5.0b, maybe available the next week), can use a loop to count the number or elements instead to use *nelem.
for example:

*set elems(All)
*remove elems(Linear)
NELEM REMOVE LINEAR = *nelem
*set elems(All)
NELEM ALL ELEMENTS = *nelem ERROR!!
*set var n=0
*loop elems
*set var n=operation(n+1)
*end elems
NELEM ALL ELEMENTS = *n OK!!

In other hand, if only want count the number of elements of some type, instead to use *set/add/remove, can add a parameter to *nelem,
for example: *nelem(linear) *nelem(triangle) *nelem(quadrilateral) *nelem(all)

Regards
Enrique Escolano

----- Original Message -----
From: "Philippe Bouchilloux" pb at magsoft-flux.com
To: gidlist at gatxan.cimne.upc.es
Sent: Monday, October 04, 2004 8:58 PM
Subject: Re: [GiDlist] nelem - Behavior changed?


Hello Enrique,

After further analysis, this is what I find: the nelem command is modified
by the remove command (for instance, remove(Linear) or remove(Triangle)).
Unfortunately, even the set elems(All) command does not restore the
situation to normal (i.e. nelem returns the total number of elements minus
the number of elements removed).

I'll send you a test case to your personal email address right away.

Philippe


*nelem has not changed on purpose.
I tried a small test and it worked ok.

_______________________________________________
GiDlist mailing list
GiDlist at gid.cimne.upc.es
http://gid.cimne.upc.es/mailman/listinfo/gidlist

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://listas.cimne.upc.edu/pipermail/gidlist/attachments/20041011/c8e80b7c/attachment.htm

Thanks Enrique.

Philippe

At 10/11/2004 02:20 PM, you wrote:
Hello Philippe.
As you related, it exists some problem using *nelem (equivalent to
*nelem(all)) after use *remove elems (nelem is not updated)

Until this bug is corrected (for the next beta 7.5.0b, maybe available the
next week), can use a loop to count the number or elements instead to use
*nelem.
for example:

*set elems(All)
*remove elems(Linear)
NELEM REMOVE LINEAR = *nelem
*set elems(All)
NELEM ALL ELEMENTS = *nelem ERROR!!
*set var n=0
*loop elems
*set var n=operation(n+1)
*end elems
NELEM ALL ELEMENTS = *n OK!!

In other hand, if only want count the number of elements of some type,
instead to use *set/add/remove, can add a parameter to *nelem,
for example: *nelem(linear) *nelem(triangle) *nelem(quadrilateral) *nelem(all)

Regards
Enrique Escolano

----- Original Message -----
From: "Philippe Bouchilloux" mailto:pb at magsoft-flux.compb at magsoft-flux.com
To: mailto:gidlist at gatxan.cimne.upc.esgidlist at gatxan.cimne.upc.es
Sent: Monday, October 04, 2004 8:58 PM
Subject: Re: [GiDlist] nelem - Behavior changed?

Hello Enrique,

After further analysis, this is what I find: the nelem command is modified
by the remove command (for instance, remove(Linear) or remove(Triangle)).
Unfortunately, even the set elems(All) command does not restore the
situation to normal (i.e. nelem returns the total number of elements minus
the number of elements removed).

I'll send you a test case to your personal email address right away.

Philippe


*nelem has not changed on purpose.
I tried a small test and it worked ok.

_______________________________________________
GiDlist mailing list
mailto:GiDlist at gid.cimne.upc.esGiDlist at gid.cimne.upc.es
http://gid.cimne.upc.es/mailman/listinfo/gidlist


___________________________________________

Philippe Bouchilloux, Ph.D.
Magsoft Corporation
20 Prospect St.
Suite 311
Ballston Spa, NY 12020
Tel: (518) 884-0505
Fax: (518) 884-8688
http://www.magsoft-flux.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://listas.cimne.upc.edu/pipermail/gidlist/attachments/20041012/7067446d/attachment.htm