Community
@JamaL9722060 wrote:I’m attaching another example in which the hatches are erased as the lisp command “merge hatch” is applied
What might be the issue here?
....
Hatch entity data is extraordinarily complicated, so it's hard even to analyze where everything is coming from, much less how to put it together from several patterns into one, but one thing I noticed:
A Hatch pattern made by either selecting internal points or closed objects will have a DXF code 91 value, or a NumberOfLoops VLA property, equal to the number of boundaries. The long skinny through-the-middle Hatch in your sample drawing was made from 7 boundaries, and the remaining three were made from single boundaries. That suggests to me that the overall merged pattern should have a DXF 91/NumberOfLoops value of 7 + 1 + 1 + 1 = 10. But it looks like the code assigns a value equal to the number of patterns selected to merge, which would be 4 in this case.
When I drew two Hatch patterns, each involving a single defining boundary, MergeHatch worked, which suggested to me that maybe that DXF 91/NumberOfLoops element could be the issue. But when I tried merging the three Hatches in your drawing that have single defining boundaries, omitting the one with 7, they were all deleted just as when that 7-loop one was included, so apparently that is not all of the problem.
Kent Cooper, AIA 
Jamal,
I've done a code to merge hatches, it was made a few years ago, the problem is, I can't find it!!!
So, I wrote this code, on the fly, it will merge overlapping hatches, it is not a fast code, because it uses multiple commands, but works.
Try it, is minimally tested...
(defun c:test (/ ss zer ss1 hpb olde itm ent bou ss2)
(alert
"\n Select hatch to Merge.\n\nFirst one will be the Properties Source."
)
(if (setq ss (ssget ":L" '((0 . "HATCH"))))
(progn
(command "_.undo" "_BE")
(setq olde (getvar 'cmdecho)
hpb (getvar 'hpbound)
itm -1
ss1 (ssadd)
zer (entlast)
)
(ssadd zer ss1)
(setvar 'hpbound 0)
(setvar 'cmdecho 0)
(while
(setq ent (ssname ss (setq itm (1+ itm))))
(command ".-hatchedit" ent "b" "r" "y")
);; while
(while
(setq zer (entnext zer))
(ssadd zer ss1)
);; while
(ssdel (ssname ss1 0) ss1)
(command "_.union" ss1 "")
(setq bou (entlast))
(command "_.hatch" "solid" "l" "")
(setq ss2 (ssget "_L"))
(command "_.copy"
(ssname ss 0)
""
"0,0,0"
"0,0,0"
"_.erase"
"p"
""
"_.matchprop"
"L"
ss2
""
"_.draworder"
ss2
""
"b"
"_.erase"
"L"
bou
ss
""
)
);; progn
);; if
(command "_.undo" "_E")
(setvar 'hpbound hpb)
(setvar 'cmdecho olde)
(princ)
);; test hope that helps
Henrique
Jamal,
as I wrote in my previous post, "will merge overlapping hatches"..
In your example the seven hatches are forming two distinct zones, and both zones are coincident only in one point, so it is impossible to define an common area to the two hatch zones, would be require at least two common points, or an overlap area, for the resulting boundary be only one.
Henrique
as I wrote in my previous post, "will merge overlapping hatches"..
In your example the seven hatches are forming two distinct zones, and both zones are coincident only in one point, so it is impossible to define an common area to the two hatch zones, would be require at least two common points, or an overlap area, for the resulting boundary be only one.
Henrique
Hi Jamal,
I understand your idea, but it would not be easy to create a tool that would merge hatches with a gap between them, mainly because to ensure that the final boundary would be desired, it would be almost impossible,
Jamal, in the example you attached, it would be possible to create a code to find a vertex coincident to both boundaries, and the two nearest points from the both boundaries, and then generate a new boundary, but your example is not the general case, many times, hatch boundaries may have very different shapes, and in that case it would be very difficult to create a tool that was reliable for all situations...
See the example I have attached, I think that exemplifies what I tried to say.
Cheers
Henrique
Jamal,
as I said:
"I never done anything to split an existing hatch, but let's see if I can think in a way to do it..."
This is how I can split an existing hatch, this code is not bullet-proof, is only a demo, and was made to try to demonstrate a way to get the hatch split...
Try it, is minimally tested...
(defun c:test (/ c e hpb itm olde p pts selp ss ss1 ss2 ss3)
(prompt "\nSelect the lines to split hatch: ")
(if (setq ss (ssget ":L" '((0 . "LINE,LWPOLYLINE,ARC"))))
(progn
(prompt "\nSelect the hatch to split: ")
(if (setq ss1 (ssget ":S:L" '((0 . "HATCH"))))
(progn
(command "_.undo" "_BE" "_ucs" "_W" "_plan" "")
(setq olde (getvar 'cmdecho)
hpb (getvar 'hpbound)
selp (getvar 'selectionpreview)
itm -1
fuzz 0.25)
(command ".-hatchedit" (ssname ss1 0) "_b" "_r" "_y"
"._extrude" "_L" "" "5")
(while
(setq e (ssname ss (setq itm (1+ itm))))
(command "._imprint" "_L" e "n" "")
(setq c (vlax-ename->vla-object e)
p (vlax-curve-getpointatdist
c
(/ (vlax-curve-getdistatparam c (vlax-curve-getendparam c))
2.
)
)
pts (cons p pts)
p (polar p (* pi 0.25) fuzz)
pts (cons p pts)
p (polar p (* pi 1.25) fuzz)
pts (cons p pts)
))
);; while
(command "._explode" "_L" "")
(setq ss2 (ssget "_P")
ss3 (ssget "_F" pts '((0 . "REGION"))))
(command "._erase" ss2 "_R" ss3 "")
(setq itm -1)
(while
(setq e (ssname ss3 (setq itm (1+ itm))))
(command "_.hatch" "solid" e ""
"_.matchprop" ss1 "_L" ""
"_.draworder" "_L" "" "_B")
);; while
(command "erase" ss1 ss3 "")
);; progn
);; if
);; progn
);; if
(setvar 'hpbound hpb)
(setvar 'selectionpreview selp)
(setvar 'cmdecho old)
(command "_ucs" "_P" "_zoom" "_P" "_.undo" "_E")
(princ)
);; testhope that helps
Henrique
![Autodesk Design & Make Report](https://forums.autodesk.com/t5/image/serverpage/image-id/1350519iCB55845EBFC43016/image-size/large?v=v2&px=999" "Autodesk Design & Make Report")
