Re: [HOT] building=yes both on the way and on the nodes

2018-07-08 Thread Jean-Marc Liotier
On 2018-07-08 21:22, john whelan wrote: > I suspect its more general than just Chad since its more than one mapper. At 2,657 in Chad I'd say that's liveable with unless some kind soul could zap them. As I said, https://www.openstreetmap.org/changeset/60516490 took care of that. > I very much

Re: [HOT] building=yes both on the way and on the nodes

2018-07-08 Thread Jean-Marc Liotier
On 2018-07-08 22:22, Pierre Béland wrote: > A more generic solution where it shows child nodes of a building [..] > way({{bbox}})["building"]->.b; > node(w.b)[!"source"](if:count_tags()>0); Nice, I was indeed wondering if Overpass had a concept of child nodes like JOSM. Very useful, thanks !

Re: [HOT] building=yes both on the way and on the nodes

2018-07-08 Thread john whelan
Thanks your emails came in fine. Sounds like yahoo.fr and Google spam filters are now aligned. Cheerio John On Sun, 8 Jul 2018, 4:22 pm Pierre Béland, wrote: > It seems that you have not received my previous answer to John on this > list where I provided a solution to query a bbox from

Re: [HOT] building=yes both on the way and on the nodes

2018-07-08 Thread Pierre Béland
It seems that you have not received my previous answer to John on this list where I provided a solution to query a bbox from Overpass.See http://overpass-turbo.eu/s/A9c A more generic solution where it shows child nodes of a building with any key except source.In this example around Freetown,

Re: [HOT] building=yes both on the way and on the nodes

2018-07-08 Thread john whelan
Thank you. Cheerio John On 8 July 2018 at 15:36, Jean-Marc Liotier wrote: > On 2018-07-08 21:22, john whelan wrote: > > I suspect its more general than just Chad since its more than one mapper. > At 2,657 in Chad I'd say that's liveable with unless some kind soul could > zap them. > > > As I

Re: [HOT] building=yes both on the way and on the nodes

2018-07-08 Thread john whelan
I suspect its more general than just Chad since its more than one mapper. At 2,657 in Chad I'd say that's liveable with unless some kind soul could zap them. I very much doubt there is a valid reason for their existence. Cheerio John On 8 July 2018 at 15:12, Jean-Marc Liotier wrote: > On Sun,

Re: [HOT] building=yes both on the way and on the nodes

2018-07-08 Thread Jean-Marc Liotier
On Sun, July 8, 2018 4:09 pm, john whelan wrote: > I'm seeing a fair number of these especially in CHAD. Could someone do an > overpass or whatever to see how big a problem it is? I loaded the following query into JOSM: [out:xml]; {{geocodeArea:chad}}->.searchArea; (

Re: [HOT] building=yes both on the way and on the nodes

2018-07-08 Thread john whelan
​There is a batch here in Chad: "name"="Mandi" "place"="village" Coordinates: 13.4288186, 14.7635571​ users starting BUR with a number appear to have some training issues and not just on buildings. Cheerio John On 8 July 2018 at 10:46, Jorieke Vyncke wrote: > Where in Chad is this

Re: [HOT] building=yes both on the way and on the nodes

2018-07-08 Thread john whelan
I hate to say it but I'm not sure. I just grab chunks these days and scan them for missing tags etc. I came across at least three instances by different mappers using JOSM so not the building_tool, and I have cleaned up some 500 manually but I think an automated edit might save my fingers. It

Re: [HOT] building=yes both on the way and on the nodes

2018-07-08 Thread Jorieke Vyncke
Where in Chad is this John? Jorieke 2018-07-08 15:09 GMT+01:00 john whelan : > I'm seeing a fair number of these especially in CHAD. Could someone do an > overpass or whatever to see how big a problem it is? > > and they were done months ago by mappers who haven't mapped recently. > > Thanks

Re: [HOT] building=yes both on the way and on the nodes

2018-07-08 Thread Pierre Béland
Hi John, In Overpass-turbo, with the following Query it is possible to select any bbox and extract.  The query will retain only buildings where duplicated tag on the chlid nodes. For this example in N'Djamena, other then the duplicate, we see that the OSM contributor added various unnecessary