this is not a GIS issue, or rather it is, but one that cannot be fixed
Well, you're underestimating us and Map capabilities, so I give up right now, sir.
Alfy, he's all yours...
That is precisely what I wanted to do and it seems to work, at least for some bits.
I'm having problems that some of my features have intersects that the topology doesn't seem to like, but I've tried on simpler versions and it seems quite happy. I may have to do some more cleaning for those others, but I seem to be on the right track and actually creating polygons. As long as I can get it to work on the larger files and map sets I think this could be the solution. Thank you.
>> I'm having problems that some of my features have intersects that the topology doesn't seem to like
Upload a DWG with one or some of the polylines that make troubles so I can go "into them" and look what to do.
- alfred -
Try this. It's rather different to most of what you are used to dealing with I would think as it is all about archaeolgy. It was created some time ago without real thought for incorporation in GIS, hence the combination of open and closed lines etc. The fit and smoothing don't help either. What I'm after is a drawing with preferably nothing but closed polylines in it, but ones that represent the line widths etc currently expressed by the EWIDTH attribute. I hope that makes sense and I'm most grateful for your help.
am I right that - while creating the buffer - you get the message that a value=0 is not ok for that operation?
If so it's the question what should happen with these polylines. And whatever should be done with them ==> select them with command _QSELECT and make the changes you need (delete them, give them a minimal global-width, move them to a separate layer for a different handling).
For selecting out the closed polylines you also can use _QSELECT to get them selected, push them to another layer, freeze this layer and start then your polyline-width buffering. When this is done you can thaw the previous layer and you have only closed polylines.
If you get another message then please let me know the error you get and upload a drawing "with one or some" entities (few entities, not hundreds, so it's easier to identify the source of the problem).
- alfred -
Thanks, that is just what I'm after.
However, there appear to be some anomalies in that some features have been turned only into single polylines e.g. the blue sub-rectangular feature at c232300,61745 and some have not worked at all e.g. the green "no 7" just to the north at c232292,61827. Also some have been moved e.g. the set of lines at c232657,61949 where the created polygons have "bent" so that they no longer follow the original alignment. The difference is only 1m maximum so is within the tolerances of a +2.5m accuracy, but it is still slightly worrying.
The other thing is that I've just tried again with another file and even though I checked that all it had in in were polylines and 2Dpolylines and using QSelect I checked they had a Globalwidth <>0, when I ran the buffer I got an error saying "Zero offset. Can't build buffer". When I reduce the file by cutting it into chunks it works, but that defeats the object as I have 200+ files I need to process and so want to be able to use MapSets to minimise effort.
Anyway I hope that all made sense and thanks again for all your help.
Like I said that seems to answer my query, but I still don't understand why I get the "zero offset" error, when I am quite certain everything has a width. As I say, it is particularly puzzling as when I cut the file down it works, but as I said that somewhat defeats the object as I have 200+ files I'd like to deal with simultaneously using MapSets.
So if anyone has any ideas I'd be most grateful.
I attach an example file with only a small number of features for anyone to look at and see if they can get it to work and explain what I'm doing wrong. As you can see it has only polylines and 2D polylines, all of which have a width of 0.5 or 1, yet I still get an error telling me I have zero offset.
for a short test: yes I could reproduce it with 2012, but I can't in 2013. So 2013 seems to better work with your data.
Why 2012 does not do what it should? :... I have no idea (currently).
- alfred -