Topic: ssget filter groupcode sequence?

[roy_043]

Can the (ssget) function filter for a groupcode sequence?

I am working on an application that currently stores two values (width and height) as 1040 groupcodes values in xdata. Using the code below will select all entities with either the height OR the width set to 1000.0. And I would like to select only those entities that have the width AND height set to 1000.0.

I am now thinking of combining the height and width in a groupcode 1010 (point) value, but would still be interested to know if there is any way to make (ssget) respect a groupcode sequence.

Code - Auto/Visual Lisp: [Select]

1. (sslength
2.   (ssget
3.     "_X"
4.     (list
5.       (list
6.         -3
7.         (list
8.           "MyApp"
9.           '(1002 . "{")
10.           (cons 1000 "WIDTH")
11.           (cons 1040 1000.0)
12.           '(1002 . "}")
13.           '(1002 . "{")
14.           (cons 1000 "HEIGHT")
15.           (cons 1040 1000.0)
16.           '(1002 . "}")
17.         )
18.       )
19.     )
20.   )
21. )

[ronjonp]

I don't think you can filter beyond the appname.

[Lee Mac]

I don't think you can filter beyond the appname.

Ditto, from the VLIDE Help:

Using the ssget filter-list, you can select all entities containing extended data for a particular application.

[roy_043]

I guess this feature is unique to Bricscad:

SR32664 - LISP/SDS/BRX: ssget("X" filter) failed to select entities using XData group code specifications, example: (ssget "X" (list (list -3 (list "APPID" (cons 1005 "323")))))

I definitely like it!

But is it generally speaking possible to filter for groupcode sequences? Let's say you want to select rectangles (lwpolylines) with at least two filleted corners. Can this be done with (ssget)?

[Lee Mac]

But is it generally speaking possible to filter for groupcode sequences? Let's say you want to select rectangles (lwpolylines) with at least two filleted corners. Can this be done with (ssget)?

After a very brief test, it would seem to appear that this is indeed possible.

To test, I created two simple LWPolylines:

Code - Auto/Visual Lisp: [Select]

1. (defun c:entmakeit ( )
2.     (entmake
3.        '(
4.             (0 . "LWPOLYLINE")
5.             (100 . "AcDbEntity")
6.             (100 . "AcDbPolyline")
7.             (90 . 3)
8.             (70 . 1)
9.             (10 0.0 0.0)
10.             (10 1.0 0.0)
11.             (10 1.0 1.0)
12.         )
13.     )
14.     (entmake
15.        '(
16.             (0 . "LWPOLYLINE")
17.             (100 . "AcDbEntity")
18.             (100 . "AcDbPolyline")
19.             (90 . 3)
20.             (70 . 1)
21.             (10 0.0 0.0)
22.             (10 1.0 0.0)
23.             (10 1.0 2.0)
24.         )
25.     )
26. )

Then used the following ssget filter to make a selection:

Code - Auto/Visual Lisp: [Select]

1. (defun c:ssgetit ( )
2.     (sssetfirst nil (ssget "_X" '((0 . "LWPOLYLINE") (10 1.0 0.0) (10 1.0 1.0))))
3. )

The result was as expected with only the first LWPolyline being selected by the filter, though, this simple test obviously isn't conclusive to encompass the claim for all DXF group codes and values.

[CAB]

This works for > one filleted corner, not sure how to test for > 1 corner.

Code: [Select]

(setq ss (ssget (list '(0 . "LWPOLYLINE") '(-4 . "/=") '(42 . 0.0))))

[roy_043]

@ Lee:
Your test is not conclusive because point (1.0 1.0) does not occur in the second lwpolyline.

If I use this version of (c:entmakeit) both entities are selected with your filter:

Code - Auto/Visual Lisp: [Select]

1. (defun c:entmakeit ( )
2.   (entmake
3.     '(
4.       (0 . "LWPOLYLINE")
5.       (100 . "AcDbEntity")
6.       (100 . "AcDbPolyline")
7.       (90 . 3)
8.       (70 . 1)
9.       (10 0.0 0.0)
10.       (10 1.0 0.0)
11.       (10 1.0 1.0)
12.     )
13.   )
14.   (entmake
15.     '(
16.       (0 . "LWPOLYLINE")
17.       (100 . "AcDbEntity")
18.       (100 . "AcDbPolyline")
19.       (90 . 5) ; Changed to 5.
20.       (70 . 1)
21.       (10 0.0 0.0)
22.       (10 1.0 0.0)
23.       (10 1.0 2.0)
24.       (10 1.0 1.0) ; Added point.
25.       (10 0.0 1.0) ; Added point.
26.     )
27.   )
28. )

[roy_043]

This works for > one filleted corner, not sure how to test for > 1 corner.

Code: [Select]

(setq ss (ssget (list '(0 . "LWPOLYLINE") '(-4 . "/=") '(42 . 0.0))))

This will select all lwpolylines without any straight segments. So filleted rectangles will not be selected.

This will select all rectangles with at least one filleted corner:

Code: [Select]

(setq ss (ssget (list '(0 . "LWPOLYLINE") '(-4 . "<OR") (cons 42 (- (sqrt 2) 1)) (cons 42 (- 1 (sqrt 2))) '(-4 . "OR>"))))

[Lee Mac]

@ Lee: Your test is not conclusive because point (1.0 1.0) does not occur in the second lwpolyline.

:self-faceplam: You're quite right roy, my test is meaningless  :ugly:

[roy_043]

Code: [Select]

(setq ss (ssget (list '(0 . "LWPOLYLINE") '(-4 . "/=") '(42 . 0.0))))

On reconsidering this filter I am having doubts if Bricscad's behaviour is correct. As explained before Bricscad will only select lwpolylines without any straight segments. It seems not logical that an lwpolyline with a curved segment and a straight segment does not pass the filter...?

[CAB]

Roy that filter will select any pline with a bulge and does not fit your requirements.
I think it can only be done by simulating the ssget with code.
Or post process the selection set to eliminate unwanted items.