Curves and Lines tool creates topology erros?

[Zeke]

I've been running some practice cases migrating our data to the parcel fabric, and ran into a question about topology.

Our data is in two separate polygon layers - parcels and subdivisions. Parcels in subdivisions should be within the subdivision, and share a border when appropriate. I can run and validate a topology on the existing data to make sure this is correct. So far, so good.

My question is how to maintain this after creating lines from the polygons and then running the Curves and Lines Tool. I've been using a suggested practice of recreating the polygon layer from the lines after the Curves and Line tool, which ensures that the polygons and lines match topologically. This works fine on individual feature types, but wouldn't that workflow result in errors for parcels correctly fitting into subdivisions when the Curves and Lines tool makes different adjustments for parcels than subdivisions? Then I'd have to fix errors for the polygon and line parcel class.

Am I thinking about this incorrectly, or is there a way to deal with this issue that I'm not seeing? Thanks.

[TiffanyPuett]

I'm not sure why you'd want to create polys from the lines other than to weight the lines topologically. I may be missing something...I would suggest keeping your polys--along with their attributes--but continuing to run the curves and lines tool. Then run your topology on the subs & the parcels at the same time: boundary must be covered by...parcel line.

FYI, the curves & lines tool does not like bezier curves. It will turn them into straight lines. These are generated when you run a smooth on your data, typically on r/w's.

[Zeke]

Thanks Tiffany. Creating the poly layer from the lines was a method shown in an ESRI Meetup seesion; I'd link to it, but it's a login site. The reason for doing it this way, if I understand correctly, is so that after running the C&L tool on the lines, generating the poly layer from them means they will automatically match up correctly. This is different than the method in the data migration white paper.

I'm not sure I understand what you mean by running the topology on both subs & parcels at the same time. Do you mean the 6 rule topology? My understanding is that it's run on the parcel line and poly layer, then on the sub line & poly layer. Also, if I run the rule (sub) 'boundary must be covered by...parcel line', I'm going to get a lot of exceptions to mark where parcel lines don't cover the sub boundary (streets, alleys, etc.), although maybe there's no way around that. Guess I'm confused.

[ChristineLeslie]

You are correct, if you have modified your lines using the C&L tool, you may want to regenereate the polygons from the lines so that they match correctly, with the correct curves. You can use the Integrate geoprocessing tool to enforce coincidence between boundaries - this should help you line up your overlapping boundaries better and minimize topology errors.

Christine

[TiffanyPuett]

...I'm not sure I understand what you mean by running the topology on both subs & parcels at the same time. Do you mean the 6 rule topology? My understanding is that it's run on the parcel line and poly layer, then on the sub line & poly layer. Also, if I run the rule (sub) 'boundary must be covered by...parcel line', I'm going to get a lot of exceptions to mark where parcel lines don't cover the sub boundary (streets, alleys, etc.)...

What Christine said about Intergrate is a good tip!


About the topo errors:  If you're not changing your existing cluster tolerance or you don't have bezier curves, you shouldn't have many errors on parcels and parcel lines no matter which way you go.  The main point I was trying to make about topology and migrating to the fabric is that the parcels and subs should be coincident when appropriate. If they are already coincident as you mentioned, then I would create my lines from parcels as I mentioned before and as suggested in the white paper. If the object here is to keep the good lines, then do that. Again the errors shouldn't be that bad if you're already using a topology with the same cluster tolerance. With respect to the parcels being split by subs, I'm pretty sure those lines get created during the migration process anyway, so I wouldn't mark them as exceptions.  Remember all polys migrated to the parcel fabric is a parcel feature and is only distinguished by type, which means it must have lines, has the same attributes, and has to adhere to the same topology rules.

It just seems like it would save time to run the boundary topology (boundary must be covered by & must be covered by boundary of) on subs & polys at the same time.

Hope that helps...