Overkill does not delete duplicates - attached

[babicdusko]

I'm trying to clean up a file with 400k objects with overkill but it does not work as intended. It does not recognize some objects that seem like exact duplicates to me. I extracted 2 such objects and attached here.

I've tried the following steps:

1. Run overkill with tolerance 0 and all boxes unchecked

2. Tolerance 1, boxes off

3. Tolerance 0, "ignore object property boxes" checked, "option" boxes unchecked

4-8. Tolerance 0, "ignore object property boxes" unchecked, "option" boxes checked one-by-one

The above steps do not work.

Steps that do work:

9. The checkbox combo that does delete duplicate lines but overall does a poor job because it breaks the polyline is "optimize segments within polylines" and "combine colinear objects that partially overlap".

10. Also, if I explode polylines, then run overkill, it does delete duplicates as it should.

But I want to keep polylines.

I can't figure out what's wrong with these objects.

[cadffm · Accepted solution]

Hi,

>>"I can't figure out what's wrong with these objects."

OVERKILL is not a simple visual check,

1. this function compares the object data!

2. Your pline starting at different coordinates

3. Your pline vertic are in opposite directions

Picture, I colored the segement 1,2,3,4 in the same colors

[jreidKVSUZ · Accepted solution]

I tested OVERKILL in all options and for some reason this does not work like you said in your drawing. But I did a search and found a LISP file that does, and it keeps the last line / lines drawn.

I copied all programing into a LISP file and called it MYOVERKILL like the creator called it.

Here is the link to the original creator back in 2017:
https://www.cadtutor.net/forum/topic/63371-delete-double-lines-controled/

Hope this works for you. If so, please click the “Solved” Button.

Thanks, JRR!

[babicdusko]

Thanks jreidKVSUZ but I didn't get what exactly are you referring to when you say "tested OVERKILL in all options and for some reason this does not work like you said in your drawing"? My issue was not draw order related.

[babicdusko]

Thanks Sebastian, you're amazing!

So there's no way to delete those "duplicates" without exploding polylines?

[cadffm]

---

@babicdusko  schrieb:

Thanks jreidKVSUZ but I didn't get what exactly are you referring to when you say "tested OVERKILL in all options and for some reason this does not work like you said in your drawing"? My issue was not draw order related.

---

Whatever, but you should test the tool from m.rebar (attached by jreidKVSUZ above)

Download this .lsp file

Open your .dwg file

run APPOAD, load this .lsp

run new command MYOVERKILL

>"So there's no way to delete those "duplicates" without exploding polylines?"

Try 3rd party tools, like myoverkill for example (there are more, out there in www)

[jreidKVSUZ]

What I mean is, by checking and unchecking the boxes in the Overkill command. 

None of them worked. But if you download the LISP file and load it up like CADFFM said to do, the LISP file removes the Plines/ Lines you don't want.

See options image below. Once again, I checked and unchecked them all and it did not work like your post said. But to answer your question on exploding. If you explode what is in this drawing, Overkill works great. But you lose the PLINEs. So download, install the MYOVERKILL.lsp file and all works great like you want it do. The LISP uses the Draworder to keep the latest of the duplicates. 

Hope This helps! If so, please click the “Solved” Button.

JRR!

[babicdusko]

Thanks jreidKVSUZ,

I tried your LISP and it does work on deleting questioned duplicates. What it does not give is any customization over the process (tolerance) and does not give any results (how many deleted objects at the end). I want to delete only exact duplicates (tolerance 0.000). Does myoverkill command has any tolerance?

This combo does delete duplicates but at the cost of braking the polyline.

[cadffm]

Hi,

>"exact duplicates (tolerance 0.000)"

exact, or 0.000, you have to decide - it's not the same..

1.23453 is not the same as 1.23451 or 1.234530157

>" Does myoverkill command has any tolerance?"

The current setting is 1e-4 (0.0001), you can controle the tolerance by using the internal lisp function directly:

Command MYOVERKILL fires (OVERKILL-MR (ssorder) 100 1e-4)

you can do the same with your own tolerance value, for example:

Command: (OVERKILL-MR (ssorder) 100 0.0000001)

[babicdusko]

Thanks for the reply Sebastian.

Gosh, overkill command raises many questions.

   >"exact, or 0.000, you have to decide - it's not the same.."

I've always thought that the number of significant zeros doesn't matter as long as all significant figures are zeros. That being said, I though that 0=0.0=0.000=0.000000. So, what's the precision when OVERKILL (ACAD default command) is set to 0.000000? Do you wanna say there's still tolerance? In other words is the tolerance of 0.000000 the same as 0.0000001?

Apart from that,

I was testing overkill options as I was not familiar with the command and found another issue in the above attached file - "combine colinear objects that partially overlaps" option requires significant tolerance increase in order to work. If I open a new drawing, draw 2 partially overlapped line and run the overkill with minimum tolerance (set to 0.000000), it will combine the lines. If I copy those lines and paste to overkill.dwg the same overkill settings do not work. I have to increase the tolerance to get the same result. What might be the reason?

Thanks

[babicdusko]

It seems that the file is too heavy for the command so it doesn't work.

Command: MYOVERKILL
; error: Exception occurred: 0xC0000005 (Access Violation)
; warning: unwind skipped on exception
; error: Exception occurred: 0xC0000005 (Access Violation)