Individual Boundaries from Intersecting Offset Curves

Replies to This Discussion

Permalink Reply by peter fotiadis on April 24, 2015 at 4:25am

I wonder: is this (another) case for the black pill and the DarkSide? (or it's doable with components?)

let's wait and see

he he

Permalink Reply by Andrew M on April 24, 2015 at 4:34am

Peter! I mean Lord... your image is tantalizing! I envision having a growing number of street widths for this, in addition to these two... I wish I could feasibly say "anything I can do to help?"... haha

Permalink Reply by Pieter Segeren on April 24, 2015 at 4:58am

Hi Andrew, here is a way using Clipper components:

Attachments:

• Boundaries_PSG.gh, 29 KB

Permalink Reply by Andrew M on April 24, 2015 at 8:01am

Hi Pieter, thanks for the file as well. It looks like this also does the trick. I have a question though: I've added some more road types (I connected them in the same way as you did), and this is a situation where the roads may change widths occasionally (e.g. a 21m road may become an 12m road, etc.). Though when I try this, the index numbers that are associated with each plot changes. This is a problem because various characteristics may be associated with each individual plot (e.g. land use) and I would like the plot index to remain stable even if the road widths change. Illustrated in the images below. Is there any way to maintain this index when making changes to the roads?

Initial plot index:

Plot index changes after road addition/change:

Attachments:

• Boundaries_PSG-andrew.gh, 50 KB
• Variable Road Network-aa2.3dm, 26 KB

Permalink Reply by Pieter Segeren on April 25, 2015 at 10:29am

Hi Andrew, I'm afraid I cannot help you to find a way to sort these 'plots' correctly, but I do know a way to create them. See attachment.

I hope someone can come up with a way to sort them!

Attachments:

• Boundaries_PSG2.gh, 30 KB

Permalink Reply by peter fotiadis on April 25, 2015 at 11:33am

A very rough indication (I repeat: indication) to approach that issue: see attached.

Attachments:

• Order_breps_V2.gh, 29 KB

Permalink Reply by Andrew M on April 29, 2015 at 8:49am

Hi Peter! Thanks for providing this method. I think it clears a lot of things up. I've been experimenting with both your script and the multiple polyoffset components representing different road types. They work well together. 

I am however trying to link your work with some help that was provided earlier, as far as keeping this inded intact if a new road was added or if an existing line was changed. Basically this set of components would identify the new plot and set its index id as a new number, not interfering with the old plot numbers. E.g. there are 11 plots and I add a new curve after the second plot, the new plot would be numbered 12 and the original plots would continue with their old index numbers. That way no land use changes would be made if a new road was added. I'm trying to fit that into this new system. Obviously, I'm having some issues. If you're interested in taking a look, I've uploaded it here. 

Hope all is well Peter!

Thanks to Pieter as well for providing some polyoffset insight!

Attachments:

• Boundaries and Index.gh, 55 KB

Permalink Reply by peter fotiadis on April 29, 2015 at 9:08am

My dear Watson

This IS NOT elementary (he he) > that's why I've used the 3 magic words: rough + indication + approach.

There's several other constrains/"factors" that cross my mind for plot urban management (anyway for that type of management).

Things like: blah, blah, blah ... and blah. And after that (quite) explicit enumeration of ...er ...."factors" ... you can post the case.

best, Lord of SardineLand

Permalink Reply by Andrew M on April 29, 2015 at 9:44pm

Hi Peter, I hope I can earn the title Watson! I agree, this is complicated stuff.

The case is attached, I have a road network and the plots are numbered based on your initial index. If I add a road or several roads, since it creates a new plot and thus a new index corresponding to its location within the site, all plots numbered after the new plot jump up a number and thus the whole land use assignment is thrown off. That is the crux of the issue since every other task I perform is related to this initial land use assignment. 

Example below: I add a road at plot 45. The new plot gets assigned '52', and everything after '52' has a number added to it. Ideally I'd like that '52' plot to get a completely new number, i.e. 135 in the case of this site. 

I'm not sure how to connect the dots on this one. Also, adding roads may be an iterative process so for each new set of roads (assuming other intervening changes were made so I can't group the first and second set of new roads together), I'll have to create a new curve component for them. 

Plots before road addition:

Plots after road addition:

Thanks again for your insight. 

Andrew

Attachments:

• Boundaries and Index-2.gh, 47 KB
• Road Network and Index-2.3dm, 69 KB

Permalink Reply by peter fotiadis on April 30, 2015 at 12:45am

Hmm (+ hmm) ... I say

1. Since N45 is affected wouldn't be more "logical" to create child indices for the resulting mini-plots? kinda 45.1 + 45.2 (or 45.a or 45.whatever). It's all about propagation of things, you know. After all if this happens in real-life ... well ... people living (for instance) in N52 they'll find living in N 53 (is this a good thing?). Revolutions started due to less important reasons you know (and we want Order at any cost don't you agree?).

2. In the broad aspect of things: Instead of meshing with the whole combo > examine what plot is affected (or plots) and .... blah blah.

3. A topology assistant is required in order to prevent stupid (or illogical) plot sub-divisions. For instance: min area allowed for a building permission per property and/or topology constrains per property and/or cost VS benefits VS a given x subdivision ... etc etc.  

4 ....

I'll try to figure out a pseudo code (some kind of "algorithm outline" in plain English) that could "handle" that mess. Because if a x code is generated for addressing this (easy) ... and minds change (Oops this is stupid > can I have the other way please?) ... we could continue at infinitum writing stuff the likes: if(this) doThat(a,b,c);else goVacations(now); 

I'll be back, Terminator 2, Arnold S

Permalink Reply by Andrew M on April 30, 2015 at 9:09am

Hi Peter,

Thank you for the reply! I'll explain a little more about the basis for these changes. They would take place during the design process, so the entire road network would be 'under consideration' and nothing concrete. So if plot 45 was subdivided into two, for all intents and purposes it would be as if the original plot 45 never existed, save that whatever remained would still be that specific land use. But yes if this were an existing city with a developed plot, your assumption would be correct.
As for your #2, any plot divisions that are made would be desired for reasons of the design outside of building permission, etc. (again since in reality nothing yet exists).

What I'm going for is basically to allow for multiple iterations on a design without compromising the basic land use structure of plots which aren't affected by road additions or changes.

Haha, I hope that makes things easier. I think you have a really intricate understanding of how data structures in Grasshopper work and I'm envious about that! Thanks again Peter.

Permalink Reply by peter fotiadis on April 30, 2015 at 9:43am

OK got the gist of the thing.

Whilst I'm thinking some strategy for that mess + testing some weird things (under the guidance of 4 friendly horsemen) ... the 1M Q: are you doing business in Planet Flat? (like Earth for instance - as everyone intelligent enough knows: Earth is actually flat [kinda a peperoni pizza]).

• ‹ Previous
• 1
• 2
• 3
• Next ›
• Page