Merge Points: Inconsistent Behavior?

[jcarlson]

Has anyone encountered this? I use Merge Points in an area where I know the topology is good. No gaps, no overlaps, everything is lined up nicely. The preview shows exactly what I want.

But when I complete the merge, one of the parcels behaves as though it was the other point, while the rest of the features align properly.

With more points and parcels involved, the results get even wackier:

These examples are easy enough to correct with topological editing tools, but I shouldn't have to do that every time I merge points.

It gets worse. Sometimes the parcels behave as though none of the other vertices are "pinned" by fabric points, even if they are!

This leads to "alignments" in which the entire parcel is transformed in the same was as the target segment, leaving me with literally every other vertex now being misaligned.

Is anyone else experiencing this?

Merge Points used to be my favorite tool, now I avoid it like the plague.

[TimHodson]

Hi @NickN - This one's a bit tricky to work out without the data. What is the scale of the screen-shots above? Are they zoomed in to 1:1? Trying to understand the size of the gaps shown.

Yes, that preview is definitely odd, can you please confirm that those lines are actually two-point lines? I'd expect to see something like that if the green and purple lines were polylines, and had not been split at the bend. For example, in the second screen-shot, are there 2 purple line features, or 1 purple line feature in the data extent shown?

Does the problem only happen when merging points that involve these specific lines, or does it happen in all cases for all of your data?

Thanks.

-Tim

[NickN]

Hi Tim,

Yep, you were right. I just happened to be working in an area where stuff wasn't 2-point lines. Now, at least, I know what to look for when I have this issue. As for what scale I was at, I can't say for sure but when I looked at it again today it looked like I was around 1:3 or 1:5.

Thanks!

[TimHodson]

Hi @NickN - thanks for the update. Glad to hear the issue is resolved by splitting the polylines into 2-point lines.