Petri

Use 'Reshape' tool. It can be a bit fiddly to get an edge moved in the correct ange (ie. 'parallel'), but with correct constraits, doable. I don't think you will see a fundamental change in the situation any time soon, as the behaviour of rectangles is very deep in the fundamental layers of VW's graphics engine (in fact, in the operating system.)

quote: Originally posted by Katie: For your purpses Andy, the split tool can achieve what you want to do, but with an extra step. VW 11 has the extra step eliminated. Pardon me for saying this, but - first NNA takes some very useful functionality out and essentially ruins a tool - then we are offered at least partial reinstatement as a part of an expensive upgrade. Clever marketing...

Strange. When I copy & paste from Word, I get perfectly editable text.

With Workspace Editor, you can add a keyboard shortcut to the command. [ 06-06-2004, 07:19 PM: Message edited by: Petri ]

I don't understand this discussion. When I copy and paste text from Word, I even get fonts correctly so I'm not quite sure why one would want to do it as PICT - unless it is a table or very complex tab-formatting. And then, a text-only PICT is, AFAK, not a bitmap image: letters stay as letters (well, on the Mac at least) and have full resolution - and the file size does not get inflated. Am I missing something?

We just got the HP 130nr yesterday. To my delight (and almost surprise), the software is OK at least with OS X - there is also an OS 9 version, but I have not tested it yet. So, Kevin - you may need to reconsider your general opinion... (and yes, I know exactly how bad the 455CA is!) Whether one needs roll feed is a somewhat complex matter: if you almost always print on one medium OR if you have long print runs (either drawings or copies), roll feed is definitely better, but if every other print is different OR you don't print much, the benefits are less significant. Location of the printer also matters - if it is next to your workstation (as it would be without network connection!), sheet feed is not too bad.

There is a tiny chance that the points are in a sensible sequence. No harm done it you try this script: PROCEDURE JoinTheDots; VAR obHd : HANDLE; x1, y1, x2, y2 : REAL; BEGIN obHd := LACTLAYER; GETLOCPT(obHd, x1, y1); WHILE obHd <> NIL DO BEGIN obHd := PREVOBJ(obHd); GETLOCPT(obHd, x2, y2); MOVETO(x1, y1); LINETO(x2, y2); x1 := x2; y1 := y2; END; END; RUN(JoinTheDots);

I think the above only applies to parametric objects (PIOs) - tools and menu commands pretty much work as palette scripts. (And at least in VW 9, some ONLY work as palette scripts...) This is not perhaps an orthodox method, but it works for me as comes to PIOs: I use constants in place of parameters, with the same names they will eventually have. When I'm close, I create the PIO, but the script only consists of a single $INCLUDE, so I can edit the script with BBEdit. (You may need to clear the cache with the NNA script for the purpose; can't remember how it was/is distributed...) I've tried to use $DEBUG, but I guess I'm not quite clever enough to really utilise it, so I just have to do everything by trial and error... I don't want to sound patronising, but do try to use subroutines (procedures and functions) instead of spaghetti code. The examples are pretty good guidance in this respect. I wish I had realised this 10 years ago!

There may be a very simple solution: change the scale of the layer. Have you checked the measurements of the imported survey? The survey is very likely to be done in 1:1 scale, but with what units, that you have to ask. AutoCAD does not have scales at all in modelspace (and no explicit units anywhere); theoretically (and sometimes in practice) it is possible to draw in paperspace by scaling everything you draw as you would be on tracing paper, but this is not common practice.

Jaws PDF Creator is a workaround - and at USD/E 75, not too expensive. Of course it is a hassle to create the PDFs, but then again, you only really need this for drawings you issue and having a copy is not a bad idea. (Well, that's how I rationalised it...)

Damian, I'm not sure if I can follow your solution - but the thing is that VW cannot possibly know that some objects should be kept at a particular location in relation to page when you change scale. If you would scale in page setup, things would be different. If VW 11 does not fix this for you, remind me to give you a couple of relevant tools when we meet next time - they'll make it reasonably easy to keep all sheet layers at 1:1 scale. Cheers, Petri

Carmelhill, Your class tool works: I finally tested it. The layer tool should also work as it looks to be OK. The tool does not at present 'stay on' - you need another loop for the functionality you seem to want.

For those who happen to be working on Macs it would be quite possible to set up a FileMaker Pro database for selecting and placing plants. Only on the Mac, however, because then FileMaker and VW would have a common language - AppleScript. No, at this stage I won't volunteer my services - all in all it is not quite simple to do as a generic system that would suit everyone and paying customers for VW add-ons are hard to come by. The plant database of LandMark 9 is also in a FileMaker file - don't know what the situation is currently. Never looked at it as I have my own plant database system, but adding a couple of modules to it for project-specific sets and plant/symbol mapping plus the AppleScript part would be quite possible. So, you are in doubt? Well, I did write a FileMaker Pro solution for defining paint colours as VW 10+ gradients. Free from VectorDepot, if someone wants to explore the concept.

How about adding 'RUN' to both? No, I did not test, but the code seems to be OK in other respects. Or do you get syntax errors? However, eg. for just hiding classes the procedure could be simplified to work on a selected object (ie. the standard object-verb syntax) and shortened to HIDECLASS(GETCLASS(FSACTLAYER)); which is what I use when I sort out imported AutoCAD drawings with 300+ classes and don't need any fancy and time-consuming dialogs. (Yes, it is one line only, no procedure name, no 'Run'.) Anyway, on VectorDepot there are already several competing & complementing class management tool kits ( including some of mine which focus on the more esoteric CAD management aspects). I believe there are tools for the purpose of the above scripts.

Exploding should no longer be needed, however, turning off layers does not, AFAIK, make any difference unless you say WBLOCK. I don't know what ArchiCAD does in its AutoCAD export, but if it exports everything as blocks with attributes, 4-5 minutes is not a long time for import. Could ArchiCAD create XREFs? Highly unlikely, but you could try binding anyway. [ 05-17-2004, 05:44 AM: Message edited by: Petri ]

When you 'show database headers', you have buttons for sorting and summarising. Instructions are in the manual.

quote: Originally posted by abrodin@alaska.net: I successfully created a DTM using 3D loci at the triangulated points imported from autocad, 166 points. It doesn't have as much detail as I would like. Considering how all DTM programs work (=TIN, triangulated irregular nodes), that 166 is likely to be all the actual, surveyed data there is and the level of detail cannot be increased. Everything between the points is done by interpolation; if you compare your contours with AutoCAD contours, they are likely to be identical, take or leave a few centimetres.

Robert, 51000 points does not sound much to me so if the VW 10 DTM chokes on that, it is still useless. OK, it may be a lot for a single house site, but there are other types of projects... With filtering, I have spent endless hours in getting rid of intersecting polygons; when converting the 3D polys to 3D loci, the problem per se goes away, but of course produces wrong results. And boy, is it difficult to find these intersections, as there is a dialog for each and every one of them, reporting page-related coordinates! Why can one buy a $200 DTM add-on to AutoCAD, one which works with hundreds of thousands of points, can use attribute data to generate the model (the standard way of getting terrain info from GIS systems), does slope analysis correctly etc etc, but one has to pay significantly more for the inferior (mostly useless) DTM capability in VW?

Katie, you SHOULD tell how these style (class) names are modifid to be useful or point the person to the appropriate tech note or manual section. I'm sick and tired of giving the same instructions over and over again.

Pamela, Check http://www.vectordepot.com/PlugIns1.shtml and also the marketplace of the site for useful add-ons. I think the one that might help you is called Broken Line; I've been thinking of publishing my 'Marker Poly' which does pretty much the same, but with polygons. Stay tuned to VectorDepot - the premium 'shame list' of NNA's shortcomings...

John, Let's say it is doable, but quite complex. If I had hundreds of walls to create, I might write a script, but (after thinking about it a bit more), only might.

Too many uncertainties to my humble taste. If the polygons were rectangular & have exactly 4 (or 5) vertices, it might be feasible, if not, then not. In either case, in the time you join the walls, you draw them. Someone of course may have resolved the somewhat complex trigonometry and logic already, but starting from scratch is not really appealing.

Do the ungrouping in a 3D view.

quote: Originally posted by propstuff: Exactly the same thing that happens now if you use the Undo command 1,2, 3, or whatever times. Yes, I know Adobe's approach, but can't really see the benefits. The implementation of the undo interface won't make the operations themselves any faster - if anything, it could be slower. After a series of operations, you don't really know what you did at a particular stage. Even in PhotoShop, I know I make dozens of changes in a minute and have to use simple undo steps. Not that I would oppose a history palette, I just don't see it as a priority. Also, I did make a false assumption of your particular request: the issue has been raised earlier several times, often with 'non-linear' undos in mind, so my tone was unduly blunt.

quote: Originally posted by propstuff:Being able to jump back to the desired instance directly would be a joy.While this is a nice idea, more often than not, undoing an undesired action in the past would also undo more recent actions, with unknown consequences. Consider the matter from time travel point of view: the last action is a solid subtraction. If you undo the creation of one of the objects used, what would happen to the subtraction?