Community
@Anonymous wrote:....
I have a code that labels the ex grade and the pipe ie.
The code ask the user to pick the base elevation, the profile, the label start and end location, and to mirror or move the label. It is all working but for some reason the elevations are not correct, can someone tell me what is wrong?
....
I didn't look too deeply, and don't fully understand everything that it's doing, but I have a guess....
You are setting the UCS to World at the beginning, with this:
(command "ucs" "save" "PPL")
(command "ucs" "")
to be reset back at the end. But then later you have this line:
(setq startpt (trans (car (vl-remove-if '(lambda (x) (< (car a) (car b))) grcoords)) 0 1))
which appears to be translating a point [which I think is used in calculating what's coming out wrong] from the World Coordinate System to the current Coordinate System. But the current CS is the WCS, so the (trans) function won't change anything.
I suspect that maybe you should not set the Coordinate System to WCS.
Kent Cooper, AIA 
@Anonymous wrote:....
(setq grcoords (vl-remove-if 'not (mapcar '(lambda (x)(if (= 10 (car x))(cdr x)))(entget grad))))
because If i go threw the code step by step and check the values the coordinate list that this returns is not on the exgrade line. ....
For me, that returns a correct list of WCS XY coordinates when applied to a Polyline whose entity name is stored in the 'grad' variable. Could it be not setting the right Polyline into that variable?
Kent Cooper, AIA
@Anonymous wrote:No it is getting the correct entity and it returns a list but if you plot the coords manually by drawing a line it does not correspond to a point on the line, I am guessing it should. Please open a new dwg file and draw a polyline on the layer C-ROAD-PROF, then run the code below. If you plot the coords are they on the line you drew? Maybe they are not soposto be but seems like they should??????
....
It returns the correct list of coordinates for me. Are you sure you're in the WCS?
Kent Cooper, AIA
@Anonymous wrote:Can anyone help me? Kent1Cooper you still there? Any ideas?
Yes, I'm still here. If listing the coordinates gives you different values than when you drew the Polyline, I don't have any other ideas. If drawing another Polyline with those coordinates gives you vertices in different places, the only other thing I can think of is that maybe you have running Object Snap modes that are finding locations to snap to on other objects.
Kent Cooper, AIA
@Anonymous wrote:Please see attachment.
Can you look at the file and tell me if you get the same coords as i do when?
All i know is that this is not working so can you tell me how to write the code in a different way so that it does work?
I may look deeper later [time to go home], and I haven't run the code on that yet. But one thing I notice is that the white Lines all have their endpoints on whole-number values in both X & Y directions [set Snap to 1 and you'll see that the cursor hits the endpoints, or turn on the Grid], but the red Polyline doesn't have its vertices positioned that way. Something seems to be rounding things off, but only the first two are at the nearest integer-coordinates location, so I don't know what's going on. Could it be a limit on number of significant figures? Does it do the same if you move the red one close to 0,0 instead of having the coordinate values so high?
Kent Cooper, AIA
@Anonymous wrote:I have now stepped threw the code and i think the problem comes into play here
(foreach p ptlist....
(setq dp (vlax-curve-getclosestpointto grad p)
pp (vlax-curve-getclosestpointto prof p))
....
I do not understand what this line of the code does
(polar p (* pi 0.75) 2.47)
(polar p (/ pi 4) 2.47))
or the values it returns, I looked at the help but i still dont understand what it doesSeams like using (vlax-curve-getclosestpointto grad p) could be a problem because it only gets a point closest to a curve, and in this case there are no curves
Keeping in mind that AutoLISP handles all angles in radians, look again at the AutoLISP Reference about (polar). It's just finding two points that are 2.47 units from p in the 135- and 45-degree directions.
The term "curve" doesn't need to mean things that are curved by the general definition -- in (vlax-curve...) functions it applies also to straight things. Anything with linearity [Line, Polyline, Arc, Circle, Ellipse, Spline, etc.] qualifies as a "curve," but not things without it, like Text.
If I'm reading the code right, I agree that the problem is most likely with the (vlax-curve-getClosestPointTo) function, but not because the Polylines don't count as "curves." I think the points represented by 'p' in the 'ptlist' are horizontally aligned in relation to the 'tline' which I think is along the base of the graph. The closest point on a sloped Polyline [either the grad or prof variable will usually, if not always, be sloped at varying angles] to a given point will not be directly above that given point, but will be the base of a perpendicular from the point to the Polyline, which perpendicular will usually not be vertical. I suspect you want the point that is directly vertically above the 'p' location on the Polyline. That could be found by drawing a temporary vertical Line through the p location, and using the (vla-intersectwith) function to find where that vertical intersects the Polyline. That function is not documented in the AutoLISP Reference, but do a Search for that here, and you'll find lots of examples.
Kent Cooper, AIA
@Anonymous wrote:Yes you are correct the tline is at the base of the graph, and yes i want location that is vertical to the p location, so i will look into the function and see if i can figure out how to draw a temp line.
I was thinking that maybe a way to do this would be to get the list intersetions along the grad and prof variable and the vertical grid lines, but i dont know how to select the lines at angel 90 on the layer C-ROAD-PROF-GRID-MAJR.
....
Finding those Lines would be easy -- just add another (ssget) after the first two here that are already in your code:
(setq grad (ssget "W" P1 P2 (list (cons 0 "lwpolyline")(cons 8 "C-ROAD-PROF"))));selects the gradeline
(setq prof (ssget "W" P1 P2 (list (cons 0 "lwpolyline")(cons 8 "C-SSWR-FORC"))));selects the forcemain line
(setq verts (ssget "W" P1 P2 '((0 . "LINE") (8 . "C-ROAD-PROF-GRID-MAJR"))));selects the major grid lines
But that would get you all of them, including the horizontal ones [there's no data on the angle that you can filter by]. The collection could be pared down to just the vertical ones easily enough, with a check on whether the X components of their endpoints are equal. This approach would also eliminate the need to ask the User for the first and last label locations, and [I think] the 'tline' and probably some other elements. I can't work that out further right now, but I may look at it again later.
Kent Cooper, AIA
![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")
