[OSM-talk] Python Overpass API wrapper version 0.5.4 is out

[Martijn van Exel]

Hi,

If you code in Python and you like the Overpass API, you may be interested
in the Python Overpass API wrapper. I just released 0.5.4. Find it on PyPi
or github, https://github.com/mvexel/overpass-api-python-wrapper

Don't hesitate to code up / suggest improvements.

Martijn van Exel
http://mvexel.github.io/
___
talk mailing list
talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Data Quality and Mapping for the renderer.

[john whelan]

Whilst the on line tools are useful being able to review the tags in a
spreadsheet I found very useful.  The online tools aren't quite so good at
picking up four different ways that a car rental company's name has been
spelt.  The other part is to do with local knowledge.  Often knowing the
area gives you an edge when looking over things from the data quality side.

Having said that the online tools help enormously to pick out the errors.

Thanks for the input

Cheerio John

On 21 Aug 2016 8:02 pm, "Svavar Kjarrval"  wrote:

> There are also online QA tools which display certain types of errors,
> for example Osmose [1] and Keep right [2]. The users who don't have much
> computer memory installed could use those types of sites instead.
>
> [1] http://osmose.openstreetmap.fr/en/map/
> [2] http://keepright.at/
>
> - Svavar Kjarrval
>
> On sun 21.ágú 2016 17:18, john whelan wrote:
> > Yes I know we shouldn’t but just sometimes it’s nice to think about
> > the people who use the maps.  Can we make life a little easier for them?
> >
> > This post is aimed purely at 64 bit Windows users.  The tools may work
> > on other operating systems but I haven’t tried them.  There are other
> > tools around.
> >
> > These days end users like to type something in on their smartphone or
> > other device and have something display.  It maybe they are looking
> > for a hardware store so what could go wrong?
> >
> > Locally a mall was mapped on the building outline and the stores were
> > added store1=store name, store2 etc.  If you type store1=xyz it will
> > be found.  Problem is most users won’t know this and the renderers
> > will omit non standard tags.  My preference would be nodes with
> > shop=hardware name=xyz and simply adding these to the mall makes it
> > all much more usable.
> >
> > Another example is a double space in a name.  Makes it difficult to
> > find but JOSM will warn about this.  Locally a car rental company had
> > its name spelt in four different ways, one of which was the same as
> > the company’s web site.
> >
> > How do we find them in our local city?
> >
> > Step one is download the relevant part of OSM, download.geofabrik.de
> >  is one place to find it.
> >
> > Then we need to cut out just the bit we are interested in.
> > osmconvert64 can do this but unless you have lots of memory and time
> > I’d first convert the .pbf file to an .osm file.
> >
> > You need the longitude and latitude of the area you’re after.  The
> > easy way is start JOSM and use the slippy map to define the area.  You
> > aren’t going to try to download it, it will be too large.  Click on
> > the bounding box tag and that will give you the minimum and maximum
> > longitude and latitude you need for osmconvert64.
> >
> > http://wiki.openstreetmap.org/wiki/Osmconvert#Clipping_
> based_on_Longitude_and_Latitude
> >
> > If you get a 1k file you’ve probably got the min and max values the
> > wrong way round.
> >
> > This local file can be loaded into JOSM.  It will probably take time
> > to load.  Most cities will load with 64 bit systems these days.  Now
> > run the validator.  If you’re lucky there will be no warnings or errors.
> >
> > The ideal way is to use the JOSM todo plugin and look at each error or
> > warning individually.  Remember the map you’re looking at is probably
> > a day old so for some errors you may wish to download that bit again
> > on a new layer before correcting.  Save the file.
> >
> > Now load the file into Maperitive.  Use the command export-tags to
> > export a list of tags in CSV format.  Load this file into a
> > spreadsheet and look through the sorted tag values.  You may find
> > addr:streetnumber rather than addr:housenumber, government misspelt
> > etc.  Once you know what to look for then you can use JOSM to search
> > for the tag and correct.
> >
> > Have fun
> >
> > Cheerio John
> >
> >
> > ___
> > talk mailing list
> > talk@openstreetmap.org
> > https://lists.openstreetmap.org/listinfo/talk
>
>
>
> ___
> talk mailing list
> talk@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/talk
>
___
talk mailing list
talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Data Quality and Mapping for the renderer.

[Svavar Kjarrval]

There are also online QA tools which display certain types of errors,
for example Osmose [1] and Keep right [2]. The users who don't have much
computer memory installed could use those types of sites instead.

[1] http://osmose.openstreetmap.fr/en/map/
[2] http://keepright.at/

- Svavar Kjarrval

On sun 21.ágú 2016 17:18, john whelan wrote:
> Yes I know we shouldn’t but just sometimes it’s nice to think about
> the people who use the maps.  Can we make life a little easier for them?
>
> This post is aimed purely at 64 bit Windows users.  The tools may work
> on other operating systems but I haven’t tried them.  There are other
> tools around.
>
> These days end users like to type something in on their smartphone or
> other device and have something display.  It maybe they are looking
> for a hardware store so what could go wrong?
>
> Locally a mall was mapped on the building outline and the stores were
> added store1=store name, store2 etc.  If you type store1=xyz it will
> be found.  Problem is most users won’t know this and the renderers
> will omit non standard tags.  My preference would be nodes with
> shop=hardware name=xyz and simply adding these to the mall makes it
> all much more usable.
>
> Another example is a double space in a name.  Makes it difficult to
> find but JOSM will warn about this.  Locally a car rental company had
> its name spelt in four different ways, one of which was the same as
> the company’s web site.
>
> How do we find them in our local city?
>
> Step one is download the relevant part of OSM, download.geofabrik.de
>  is one place to find it.
>
> Then we need to cut out just the bit we are interested in. 
> osmconvert64 can do this but unless you have lots of memory and time
> I’d first convert the .pbf file to an .osm file.
>
> You need the longitude and latitude of the area you’re after.  The
> easy way is start JOSM and use the slippy map to define the area.  You
> aren’t going to try to download it, it will be too large.  Click on
> the bounding box tag and that will give you the minimum and maximum
> longitude and latitude you need for osmconvert64.
>
> http://wiki.openstreetmap.org/wiki/Osmconvert#Clipping_based_on_Longitude_and_Latitude
>
> If you get a 1k file you’ve probably got the min and max values the
> wrong way round.
>
> This local file can be loaded into JOSM.  It will probably take time
> to load.  Most cities will load with 64 bit systems these days.  Now
> run the validator.  If you’re lucky there will be no warnings or errors.
>
> The ideal way is to use the JOSM todo plugin and look at each error or
> warning individually.  Remember the map you’re looking at is probably
> a day old so for some errors you may wish to download that bit again
> on a new layer before correcting.  Save the file.
>
> Now load the file into Maperitive.  Use the command export-tags to
> export a list of tags in CSV format.  Load this file into a
> spreadsheet and look through the sorted tag values.  You may find
> addr:streetnumber rather than addr:housenumber, government misspelt
> etc.  Once you know what to look for then you can use JOSM to search
> for the tag and correct.
>
> Have fun
>
> Cheerio John
>
>
> ___
> talk mailing list
> talk@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/talk



___
talk mailing list
talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] MAPS.ME edits - partly sub-standard

[Michał Brzozowski]

Recently, multilingual name support was rolled out. Still, seems that
the new name: tags get abused, or rather filled with
useless data.
Eg. http://www.openstreetmap.org/changeset/41589376

Michał

___
talk mailing list
talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] [OSM-dev] http://overpass-api.de/ seems to be down on Saturday 20.08.2016 at 19:09

[Michał Brzozowski]

The areas feature of overpass-api.de doesn't work and Achavi returns 404.

___
talk mailing list
talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk

[OSM-talk] battlegrid fixed

[Martijn van Exel]

Hi all,

I fixed the TIGER battlegrid, it now shows nice Stamen tiles in the
background. Thanks Stamen for providing a great service. (You can donate to
help them provide this service: https://stamen.com/opensource/)

For those who don't know what Battlegrid is, it is a tool that shows
differences between TIGER (US Census) data and OSM in a grid form, so you
can focus on areas that probably still need work from the original TIGER
import.

There is an old blog post for if you want to know more:
https://openstreetmap.us/2013/09/battlegrid/

Or just get to work :)
http://184.73.220.107/battlegrid/#10/35.1603/-106.5324

Right now the backend service does not update the squares, so they may not
represent current reality. I may or may not fix that, it was a bit of a
hack.. It should still give you a good hint of where fixing is still needed.

Thanks & happy mapping

Martijn van Exel
http://mvexel.github.io/
___
talk mailing list
talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk

[OSM-talk] Data Quality and Mapping for the renderer.

[john whelan]

Yes I know we shouldn’t but just sometimes it’s nice to think about the
people who use the maps.  Can we make life a little easier for them?

This post is aimed purely at 64 bit Windows users.  The tools may work on
other operating systems but I haven’t tried them.  There are other tools
around.

These days end users like to type something in on their smartphone or other
device and have something display.  It maybe they are looking for a
hardware store so what could go wrong?

Locally a mall was mapped on the building outline and the stores were added
store1=store name, store2 etc.  If you type store1=xyz it will be found.
Problem is most users won’t know this and the renderers will omit non
standard tags.  My preference would be nodes with shop=hardware name=xyz
and simply adding these to the mall makes it all much more usable.

Another example is a double space in a name.  Makes it difficult to find
but JOSM will warn about this.  Locally a car rental company had its name
spelt in four different ways, one of which was the same as the company’s
web site.

How do we find them in our local city?

Step one is download the relevant part of OSM, download.geofabrik.de is one
place to find it.

Then we need to cut out just the bit we are interested in.  osmconvert64
can do this but unless you have lots of memory and time I’d first convert
the .pbf file to an .osm file.

You need the longitude and latitude of the area you’re after.  The easy way
is start JOSM and use the slippy map to define the area.  You aren’t going
to try to download it, it will be too large.  Click on the bounding box tag
and that will give you the minimum and maximum longitude and latitude you
need for osmconvert64.

http://wiki.openstreetmap.org/wiki/Osmconvert#Clipping_based_on_Longitude_and_Latitude

If you get a 1k file you’ve probably got the min and max values the wrong
way round.

This local file can be loaded into JOSM.  It will probably take time to
load.  Most cities will load with 64 bit systems these days.  Now run the
validator.  If you’re lucky there will be no warnings or errors.

The ideal way is to use the JOSM todo plugin and look at each error or
warning individually.  Remember the map you’re looking at is probably a day
old so for some errors you may wish to download that bit again on a new
layer before correcting.  Save the file.

Now load the file into Maperitive.  Use the command export-tags to export a
list of tags in CSV format.  Load this file into a spreadsheet and look
through the sorted tag values.  You may find addr:streetnumber rather than
addr:housenumber, government misspelt etc.  Once you know what to look for
then you can use JOSM to search for the tag and correct.

Have fun

Cheerio John
___
talk mailing list
talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk