With regard to validations …
You see a number of cases of <buildingq=*> (& other similar shapes), where somebody (& I’ve done it myself! 
) has drawn a building then incorrectly also immediately hit Q to square it.
Would it be possible to create a validator to check for “q” on the end of a key / tag & pop up a warning?
It’s possible, but I can see that my current approach doesn’t scale well. It’s just that the list of problematic/good keys will quickly grow in the script code.
For individual objects, it’s probably best to query Taginfo/Wiki for each tag and provide some sort of hint. However, for changesets, it’s unclear how many queries will need to be made.
For now, I’m testing the waters for a good solution with these simple checks :)
Plus, I don’t want to flood it with all sorts of checks that mappers will be too lazy to fix.
For “building types that end in q” overpass is pretty capable. This is UK and Ireland (nwr["building"~"q$"]({{bbox}})).
nwr["building"~"q$", i]({{bbox}}) ;-)
Thanks, but does that assume that OT actually works! 
btw, a regular expression to find tags where phone numbers are separated without ;
nwr[phone~"\\+[^;]*\\+"]
Just a heads up.
For phone number validation there is https://github.com/confusedbuffalo/phone-report . He is using libphonenumber-js as well and there seem to be a lot of tweaks necessary to conform to local preferences, that libphonenumber-js is flagging als invalid.
I know about this project, but unfortunately, there are only a few countries involved yet, and it seems to be a bit noisy with formatting alerts.
In my opinion, the only errors that are valuable are those that prevent you from calling a phone number in your navigator, That’s why I only use isPossiblePhoneNumber() in my script.
One of the big advantages to shortbread is that users can easily switch to their own tiles. (yet another switch2osm guide to write). Copy a pmtiles file to a web server or cloud storage and that’s all you need for a hosting setup.
You won’t get the minutely updates of OSMF tiles, but most sites don’t need that.
As a practical matter, the “problematic usage” levels of vector tiles will probably be much higher than raster tiles because of overzoom, 512px tiles, most stuff being rendered in advance, and per-layer rendering.
1.6.4
panoramax|mapillary|... and it will expand to:(
nwr[panoramax];
nwr[mapillary];
nwr[...]
);
or ~panoramax which will expand to nwr[~"panoramax"~".*"].
There are examples in the interface (I hope I will someday remake it from a simple browser alert())
photo previews in search results now support both mapillary and wikimedia_commons
wikimedia_commons=Category:* is now supported:
ref:inaturalist.org tag are also now supported. This is certainly a young tag and it sparked a debate, but the charming sequoias Node: General Sherman Tree (313753662) | OpenStreetMap and yet another way to store photos looked very tempting to meI tested ScriptCat again in Chrome and it seems to be fully compatible with the script now.
It claims to be an open source script manager, so it’s probably a good alternative to Violentmonkey for Chrome.
Try opening the iD and trying again: https://www.openstreetmap.org/edit?editor=id
Now the script is reusing authorization from there and something went wrong.
A small bonus among the latest bug-fix updates (1.6.4.3):
button:
