Jesse Cogswell

You would love my method filtering duplicates when building dynamic arrays. I use a FOR loop to go through the existing array with the code to add a new item to the array placed after the FOR loop and a LABEL pointer set at the very end. Inside the FOR loop, if a matching item is found, it uses a GOTO statement to bypass the code adding to the array. If it reaches the end of the loop without finding a match, it is allowed to get to the addition code. Here's the code in this plug-in that builds the array of object types using a ForEachObject(BuildTypeArray,(INSYMBOL & INVIEWPORT & (T=PLUGINOBJECT))) call: PROCEDURE BuildTypeArray(h:HANDLE); {ForEachObject callback that builds array of object types} LABEL 99; {Escape} VAR PIOName,locName:STRING; i:INTEGER; BSB:BOOLEAN; BEGIN PIOName:=GetName(GetRecord(h,NumRecords(h))); FOR i:=1 TO numPIOTypes DO BEGIN IF(PIOName = PIOTypes[i].name) THEN BEGIN PIOTypes[i].num:=PIOTypes[i].num + 1; GOTO 99; END; END; numPIOTypes:=numPIOTypes + 1; ALLOCATE PIOTypes[1..numPIOTypes]; PIOTypes[numPIOTypes].name:=PIOName; BSB:=GetLocalizedPluginName(PIOName,locName); IF(locName <> '') THEN PIOTypes[numPIOTypes].locName:=locName ELSE PIOTypes[numPIOTypes].locName:=PIOName; PIOTypes[numPIOTypes].num:=1; 99: {Escape} END;

All done. Here you go, lightly tested in VW2023 and VW2024: @michaelk I noticed in your screenshot that the two columns UUID and Parent UUID are visible. These are supposed to be hidden (have a column width of 0), did you expand the columns or is that how it presents on Mac? I'm asking because this plug-in will also use a hidden column to conceal the record name for each plug-in object so that it instead presents the localized name. If the Record Name column is visible when you run this, let me know and I'll change up how the programming works a little bit. To install this plug-in, follow the steps below: Download the attached Reset Plug-in Obj Classes.vsm file Open your Vectorworks User File in a file explorer / Finder window Easiest way to do this to go to your Vectorworks Preferences, select the User Folder tab, and click on the Explore (Windows) or Open in Finder (Mac) button Open the Plug-ins folder Place the downloaded Reset Plug-in Obj Classes.vsm file into the Plug-ins folder of your User Folder Restart Vectorworks Put the new plug-in into your workspace Go to Tools - Workspaces - Edit Current Workspace Select the Menus tab In the box on the left, find and expand the category JNC In the box on the right, find a menu to place the new command in, such as Edit or Modify or Tools Click and drag the Reset Plug-in Obj Classes command from the box on the left to the target menu in the box on the right Click OK I tested this and it seemed to work correctly. As a quick side note, as part of the process it will reset the objects after changing their class, so if you have objects that take a while to reset or a lot of objects will be affected by this, the operation may take a while to complete. Just a heads up. In my testing it was pretty fast, however. Reset Plug-in Obj Classes.vsm

@michaelk I think I told you when we met up last spring that I use GOTO statements often in my code as quick escape sequences since Pascal doesn't have a more traditional BREAK command (happy to be wrong about this, of course). It's used often in this particular script since the script has to traverse every single object in every single Symbol, Group, and Wall searching for objects of the target class. It uses the standard WHILE loop using FInGroup and NextObj, but if it finds an object using the class, it doesn't need to keep searching that object, so I use a GOTO statement to exit the WHILE loop. It doesn't save a huge amount of time or computational power, but if you run the command on a huge drawing with complicated symbols, it will be noticeable. However, that won't throw the error you're seeing since I'm not using a FOR or CASE structure in that example. In this script, I am using a GOTO in each. For the FOR loop, I'm using it to return the object type constant of a given parent object type name. I have an array of strings that is indexed by the object type constant selectors so that I can populate the Object Type and Parent Type columns of the dialog box. When you click on the Go to Selected button, it feeds the string found in the Parent Type column into a function that will then spit out the object type constant used in a CASE statement to determine how to get to the chosen object. That function is really simple and looks like this: FUNCTION GetObjectTypeIndex(objName:STRING) : INTEGER; {Searches object types for matching string and returns Object Number} LABEL 99; {Escape Sequence} VAR i:INTEGER; BEGIN IF(objName=sMiscResourceManager) THEN BEGIN i:=999; GOTO 99; END; FOR i:=1 TO kNumTypes DO BEGIN IF(objName=objTypes[i]) THEN GOTO 99; END; 99: GetObjectTypeIndex:=i; END; If it gets fed in "Resource Manager" as the parent, it will return 999 since the Resource Manager doesn't have an object type. I don't know for the life of me why I chose to use the GOTO to skip the FOR loop since I could have just put the FOR loop into the ELSE of the IF, but I did and it works, so I'm sticking by it. In the FOR loop, once it finds the match it doesn't need to complete the loop so instead it uses GOTO 99 to exit out. As for using a GOTO in a CASE structure, I wrote in a way to rebuild the list browser anytime you change target class or any of the Search options. The dialog handler is a big event CASE structure, and I have a LABEL in the part of the SetupDialogC event that adds and fills out the list browser. Anytime the user changes the class or the Search option, the list browser data is wiped out and a GOTO statements returns to that LABEL and regenerates the list browser. It's the best way I've found to be able to quickly regen the list browser without copying and pasting the generation code in each event that would require a rebuild. As penance for @grant_PD for hijacking their thread with nonsense, I'll get to writing up that command now.

I deal with this a lot doing venue drawings with fairly complex class structures. Often these venues are in educational institutions where the students may not have a firm grasp of classes vs. layers, so I do as much as I can to "idiot-proof" the drawings before handing them over. This is pretty easy to do with symbols, since you can set the Assign to Class attribute in the Symbol Options, but that doesn't work with plug-in objects. I think it would be a grand idea for there to be a way to setup default class mapping per PIO, but it would most certainly have to be drawing specific since you might not have the classes in your current drawing. Though I could also see a world in which you could set up the default class attributes and have Vectorworks build the class for you if it doesn't currently exist in the drawing. This would have to be a something that VW adds in the long term, but I wouldn't hold your breath for it. In the short term, it wouldn't be too much work to script up a dialog box that would populate a list of PIO types found in the drawing and let you map a class to them. Then, the script would scour the drawing and change all existing PIOs to their mapped class. I could write such a plug-in pretty quickly, so if I wrap up the drawing I'm currently working on, I might be able to put something together for you later this afternoon.

I think you are missing one set of parentheses in your SelectObj criteria. It should be (INSYMBOL & ((C <> 'NonPlot') & ((T = LOCUS) | (T = LOCUS3D)))). Also, a more efficient way to achieve what you are doing would be through a ForEachObject call. It would make the script one line since there's no need for variables. ForEachObject(DelObject,(INSYMBOL & ((C <> 'NonPlot') & ((T = LOCUS) | (T = LOCUS3D)))));

It's the only reason I still use VW2019 at this point. I do all of my development in 2019 so that the plug-ins will work in any version between 2019 and now. The language hasn't changed much at all, mostly new object variables and a handful of more convenient functions, but it's pretty rare when I come across something that must be in a newer version.

@jonp001 You really have to keep expanding until you get to an actual Vectorworks file. This is how deep I have to go until I get to a place I can import resources from: If you expand the folders until you get to the end and there isn't a VW file, then you may have a larger problem. I am just on Spotlight and don't have the Landmark package, so I can only check Plants based on the premium libraries, but this might be the issue you are having.

For menu commands, you can call another VS menu command by using DoMenuTextByName('Your Command Name',0); I assume it would be the same with Python. In this instance, I think the way I would go about doing this is to make the PDF export the main script and call the label creation script with vs.DoMenuTextByName. This should allow the label creation script to fully complete before exporting the PDF. If you put the label creation script into the Category Don't Put In Workspace, it won't populate in the Workspace Editor, so you can make sure someone doesn't try to use that script by mistake.

I had some work I felt like procrastinating on this afternoon so I wrote you a plug-in that will take care of this for you. You will select a symbol and enter the number of symbols you want to insert. You can then select either a fixed height or set a range, same for rotation. There is a checkbox to group all inserted symbols (might be handy if you're generating a lot of them). Here's a video of it working in VW2019: 2023-12-01 15-51-28.mp4 For the example I created a basic symbol that was 6"x3", 1/8" thick. The marquee selection was 20'x20' with a height between 0'-0" and 10'-0". The plug-in was written in VW2019 and should work in any version from 2019 onward. I did test it in 2024 and it worked as expected. To install the plug-in, follow the steps below: Download the attached Randomize Symbol Insertion.vsm file Get access to your Vectorworks User Folder. The easiest way to do this is to: Open your Vectorworks Preferences Go to the User Folder tab Click on the Explore (Windows) / Open in Finder (Mac) button Insert the downloaded .vsm file into the Plug-ins folder inside the User Folder. Restart Vectorworks Insert the menu command into your workspace Go to Tools - Workspaces - Edit Current Workspace Click on the Menus tab In the box on the left, find and expand the JNC category In the box on the right, find a menu to put the command in, such as Tools or Edit Click and drag the Randomize Symbol Insertion command from the box on the left to the target menu on the right Click OK Let me know if anything weird happens. I tested it with 3D, 2D, and Hybrid symbols. When you use 2D symbols, it will place them on the 3D plane, so you want the symbols left on the Layer Plane, it's best to use the Fixed Height option set to 0. One thing to keep in mind, is that the marquee boundary box and entered heights will only limit the symbol's insertion point, not the geometry of the symbol itself. So if you are using this with large symbols, consider that when setting your insertion area. Randomize Symbol Insertion.vsm

How timely, this topic just came up a couple of hours ago. This thread should give you some ideas about where these objects might be and how to deal with them:

@zoomer I am 100% with you. I would love to have Vectorworks change the name of the Screen Plane when editing symbols to "2D Component" and allow its use in the OIP without having Legacy Features engaged. Or even better, an OIP option for planar options that only appear inside the symbol edit container, similar to the detail checkboxes. It seemed that my conversations with Steve would lead things in that direction, but it's been three years and nothing has changed in this regard. Have a look at my surprised face.

@zoomer Deprecating the Screen Plane was actually a huge bummer in my workflow. Doing lighting, I'm often near the end of the pipeline, so I get a "finished" model of the scenic or exhibit design. Unfortunately, most of the designers I've worked with don't use hybrid symbols, so I'll get a file full of 3D Symbols and occasionally (shudder) just Groups of objects. Converting the groups over to symbols is easy enough, just create the 2D geometry and Create Symbol, but editing 2D geometry into 3D Symbols can be a bit tricky. I do not like the Generate 2D from 3D Component command because you lose class information and it hard sets the lineweights, so I rarely use it. I developed a workflow of going into the 3D component edit and using Extract Face to quickly build geometry, then selecting all of it and sending it to the Screen Plane. Ta-da! Now you have a hybrid symbol. Video example below: JNC Screen Plane Demo.mp4 This process was made more difficult in 2022 when they deprecated the Screen Plane, as I would have to enable to Legacy Features, move the objects, then disable the Legacy Features anytime I wanted to use this process. After I complained on the forum when 2022 first came out, Steve Johnson from VW emailed me asking for a demo of my workflow. I recorded the video above, and we started talking about adding a plane option when inside the 3D edit component for something like "2D Symbol Component", but it never happened. So now I have a script in my workspace that will set the plane of any selected object to the Screen Plane that I can run when creating hybrid components.

Importing resources from another drawing shouldn't re-engage the 2D Legacy Features, since it's going to assume that you will be putting things on the Layer Plane or a 3D working plane. Even copying and pasting groups containing screen plane objects will result in all group objects being on the current working plane. I just did a quick test trying to force the features on by importing and copying and pasting as wasn't able to do it. You should be in the clear.

When you set your criteria to search, you can uncheck all of the Search Within checkboxes. Though they deprecated the Screen Plane, Vectorworks still uses it on hybrid objects to determine 2D vs 3D components. If you have a 2D or hybrid symbol, all of the 2D geometry will be on the screen plane by default, and this is set when the symbol is made. So, if you are building a hybrid symbol containing both 3D and planar geometry, when you create the symbol, Vectorworks will automatically assign any planar geometry on the Layer Plane (IE, not on the 3D Plane) to the Screen Plane to separate the 2D and 3D geometry for the hybrid symbol. Likewise for plug-in objects that have hybrid geometry, all of the 2D is technically on the Screen Plane. Viewport annotations (and anything placed on a Sheet Layer) are also automatically placed on the Screen Plane behind the scenes. What I'm getting at is that objects within all of those things currently on the Screen Plane must be on the Screen Plane and are not counted when assessing whether you can disable the Legacy 2D Features. One thing that you will want to make sure of is that you have all classes and layers visible and have both your Class and Layer View settings set to Show/Snap/Modify. The criteria box will count all eligible objects across the drawing, but the selection will only result in what you can currently select based on the current visibility and selection settings. This will also affect objects inside of groups, since the objects will be counted but not necessarily selected (kind of, they will be selected but you wouldn't see that until you enter the group edit container). There is a fairly simple script that you can run that will migrate all offending objects to the Layer Plane and thus allow you to disable Legacy 2D Features. Note that this script does not come with a warranty, and while it shouldn't bork anything in your drawing, I take no responsibility if it does. PROCEDURE SwitchToLayerPlane; PROCEDURE Execute(h:HANDLE); CONST kIsScreen = 1160; BEGIN SetObjectVariableBoolean(h,kIsScreen,FALSE); END; BEGIN ForEachObject(Execute,(PLA='Screen Plane')); END; Run(SwitchToLayerPlane); To run this script, go to your Resource Manager, right-click and select New Resource - Script. This will ask you to select or name a "Script Palette" (something like "Drawing Macros" will work just fine, don't select Saved Views if it shows up in the dialog box), then name the script whatever you want ("Migrate All to Layer Plane" would be a fine name). It will now open a notepad for your script. Make sure the Language is set to Vectorscript, and copy and paste the above code in. It should look like this: Click the OK button to close the notepad. This will generate a floating palette with your new script in it. Double click the script to run it. This should move anything on the screen plane over to the layer plane excluding the objects inside symbols, plug-in objects, viewport annotations, and on sheet layers. After running it, you should be able to disable the Legacy 2D Features. I have seen some weirdness in very old drawings where some objects are more or less "stuck" on the screen plane and it won't let you disable the features, but it should work for most up to date drawings.

I think we have truly lost this thread. Ultimately, this is a wishlist post asking for Rhino.inside integration in Vectorworks. There's any number of reasons users could want this beyond just NURBS modelling. Vectorworks has many instances of integrating "live links" with other programs such as Unreal Datasmith, C4D, Enscape, Vision, and Lightwright, so there's precedence for this and Rhino.inside looks to be another viable integration. If there is some computational or licensing barrier, it's up to the folks at Vectorworks to tell us. Currently, the only response from them in this thread is at the very top from @JuanP saying that they looked into it and determined that users weren't interested. This thread, aside from one user, proves otherwise. This thread should be much less questioning why someone would want this functionality and a show-and-tell, and should instead be a show of hands of people who want this functionality in Vectorworks. I've never touched Rhino and rarely need NURBS modelling, but I would love to see Vectorworks integrate more technology from other sources to give us more options while also stressing their developers and engineers less so that we can maybe get fully functional window and door tools.

Assuming that you are using a Viewport on a Sheet Layer, you should look into the Detail Level parameter in the Viewport's Object Info Palette. When you are editing a symbol, the Object Info Palette for each object will have a field called Display at Detail Level, with checkboxes for Low, Medium, and High. What you would need to do is to find the details that are lost at that size and uncheck the Low checkbox in the OIP or build simplified geometry and make sure that only Low is selected. Then, you'd set your Viewport to Low Detail. I can understand that this may be very tedious at this point, as you seem to have a finished model, so it would mean editing a good number of symbols. But, this can be a very valuable thing to incorporate in your workflow if you find yourself working with big and complicated models. There's even an option in Document Settings called Auto display detail levels for design layers that will set the viewed detail level of design layers based on the scale settings. EDIT: I see in your signature that you are using VW Fundamentals. I'm not sure if this functionality is in Fundamentals, as Fundamentals is very stripped down compared to the other Vectorworks modules.

For those playing along at home or searching for this solution somewhere down the line, the test code above changed to use the Rpstr_GetValueStr and Rpstr_SetValueStr implementation that will survive encryption is below: PROCEDURE TestPythonEncrypt; VAR testString:DYNARRAY[] OF CHAR; BSB:BOOLEAN; BEGIN testString:='Vectorscript'; Rpstr_SetValueStr('testString',testString); AlrtDialog(Concat('Test: ',Rpstr_GetValueStr('testString',''))); PythonBeginContext; PythonExecute('import vs'); PythonExecute('vs.AlrtDialog("Python is executing")'); PythonExecute('vs.AlrtDialog(vs.Rpstr_GetValueStr("testString",""))'); PythonExecute('vs.Rpstr_SetValueStr("testString","Python")'); PythonExecute('vs.AlrtDialog(vs.Rpstr_GetValueStr("testString",""))'); PythonEndContext; AlrtDialog(Concat('Test: ',Rpstr_GetValueStr('testString',''))); BSB:=Rpstr_RemoveValue('testString'); END; Run(TestPythonEncrypt);

@JBenghiat I just tested this and you appear to have saved the day. Totally works. Thank you also to @Pat Stanford for the enlightening reading material. Answered some questions while raising several others.

I'm afraid that I already know the answer to this one, but I would most definitely love to be proven wrong. I have a Vectorscript plug-in that uses PythonExecute and vs.GetVSVar and vs.SetVSVar to get a file's Last Modified date using the pathlib Python library. This all works beautifully until I encrypt the plug-in, in which case it breaks and crashes Vectorworks. The issue seems to stem from vs.GetVSVar not working correctly and returning a "NoneType" object rather than a DYNARRAY[] OF CHAR of the supplied variable. I did some tests and it seems that vs.GetVSVar and vs.SetVSVar do not work within encrypted plug-ins. I'm guessing that the variable names are encrypted but not in the same way that strings are, so when you pass in the variable name in a string, it no longer matches with the actual variable name so it returns NIL. Here's my test code: PROCEDURE TestPythonEncrypt; VAR testString:DYNARRAY[] OF CHAR; BEGIN testString:='Vectorscript'; AlrtDialog(Concat('Test: ',testString)); PythonBeginContext; PythonExecute('import vs'); PythonExecute('vs.AlrtDialog("Python is executing")'); PythonExecute('vs.AlrtDialog(vs.GetVSVar("testString"))'); PythonExecute('vs.SetVSVar("testString","Python")'); PythonExecute('vs.AlrtDialog(vs.GetVSVar("testString"))'); PythonEndContext; AlrtDialog(Concat('Test: ',testString)); END; Run(TestPythonEncrypt); It should return: Test: Vectorscript Python is executing Vectorscript Python Test: Python After encryption, it instead returns: Test: Vectorscript Python is executing Test: Vectorscript This tells me that Python is working, but not the vs.GetVSVar or vs.SetVSVar functions. The whole point of the plug-in is lost without being able to pull a file's Last Modified date, and I'd rather not send out an unencrypted plug-in. Any ideas on a solution?

I posted about this back in 2022: Even though luminous area is included in the IES spec, Vectorworks ignores it and treats the custom light as more or less a point source. I'm guessing that this is a limitation of the Vectorworks rendering engine. For linear fixtures, I've cheated by having multiple custom lights per fixture, but it really doesn't work well for things like LED tape. It's unfortunate that using custom lights with distribution files is the only way to get photo-accurate beam/field falloff, but doesn't work correctly for linear fixtures.

In my mind, there are two: There is some direct integration with fixture modes in Vectorworks, letting you bypass some of the current awkwardness with GDTF fixtures. The Fixture Mode parameter in the Lighting Device OIP will automatically be exported into the matching Vision profile. But this only works if there is a Vision profile (otherwise you have to wait for one) and that the Vision profile is correct (not always guaranteed). Vision is much cheaper than L8. But that's about it. L8 has far more features (better laser control, water features and pyro, animated models, better handling for reflection), and the final output is much more polished. I went in on a full Vision license in 2017 right after Vectorworks bought them because of the relative ease of getting a model from Vectorworks to Vision and with the hope that having Vectorworks behind them, we'd see some meaningful upgrades. Unfortunately, the software has mostly stayed the same with only minor changes from year to year (and some features removed because they never worked. The "Attachments" feature that would handle things like color scrollers and remote dimmers was entirely broken between 2017 and 2019, before being completely removed in 2020. Bad luck if you ever need to do scrollers or play with lens options in your VL500s, there's no way to do it). And now with the ubiquity of the MVR file format and GDTF fixture profiles, the one major advantage that Vision had is largely moot. I bought Vision primarily to do some moving light preprogramming for mostly theatre, dance, and opera and with both ETC and MA consoles having a built in solution where you have XYZ control of fixture focus, I barely ever use Vision anymore. If I were in the market for a previs software for concert and event work, I would most definitely look into L8, Capture, or Depence2 before considering Vision. Hell, I'd probably even look at WYSIWYG if you need something more budget conscious. Or you could look into Carbon for Unreal, which looks pretty promising and is very affordable.

I've seen this kind of thing when the Clip Cube is active, especially when you have textures with image transparency shaders. Is it happening in 3D views or in Top/Plan as well?

The quick and dirty way to do it would be to go to Tools - Custom Selection, select Execute Immediately then select All Objects as the criteria. There will be a counter at the bottom of the dialog box that shows the number of objects. You can check the Search Within Symbols, Plug-in Objects, and/or Viewport Annotations depending on what kind of count you are looking for. If you want something a little faster, you can create a script (Vectorscript format) with something like: AlrtDialog(Concat(Count(ALL),' objects in drawing')); for all objects on layers, but not going inside of symbols (each symbol would count as one object), plug-in objects, or viewport annotations. AlrtDialog(Concat(Count(INSYMBOL & (ALL)),' objects in drawing and symbols')); would count all objects on layers and inside symbols (each component of the symbol would be counted separately and in addition to the symbol, so a drawing consisting of 2 symbols, each with 2 rectangles would come to a count of 6 objects). For a full count of every single object, inside symbols, plug-ins, and annotations, the script would look like this: AlrtDialog(Concat(INSYMBOL & INOBJECT & INVIEWPORT & (ALL)),' total objects in drawing')); EDIT: I realize now that you weren't specific about what kind of objects you are trying to count. I usually use the Custom Selection method outlined above, just setting the criteria to match exactly what I'm trying to count and then reading the value at the bottom of the dialog box. Another method would be to make a Worksheet with a cell with the =COUNT() function and inserting the criteria within the parentheses.

Unfortunately, that password lock is a byproduct of the student version watermark and, as far as I have been able to find, not possible to bypass without getting the watermark removed. It's one of the reasons I shudder when I get scenic or exhibit drawings with the student watermark, as it's much harder to later incorporate my drawings into the larger PDF packets that I need to generate. One way to kind of get around it is to use the Print function and to print to PDF using the Microsoft or Adobe rather than doing the traditional Export PDF or Publish functions. You'll lose some functionality (no way to do PDF layers, automatic hyperlinks will be broken, may lose ability to import PDF to Illustrator and retain vectors, etc), and you'll have to do this separately for each sheet and merge them in a PDF later, but at least it won't be password protected.

This is most likely related to a setting found under File - Document Settings - Spotlight Preferences. In the Classes and Color tab, there are a couple of things to check out. There is a section to determine what class the fixture should be in, with options for: Currently active class Symbol Definition class Automatic assignment using a class set in this dialog or tracked by a Lighting Device Field My guess is that you have something set not to your liking here. There are also options to change the visual attributes of the Lighting Device based on the color field. This also might be causing you some grief. This dialog may be slightly different depending on which version of Vectorworks you are using, but will generally have the same options if you're on VW2019 or newer.