Google maps : Is it possible to draw a polygon and retrieve all street names within it?

Question

Using Google Maps' API v3, is it possible to draw a polygon on a Google maps and get a list of all the street names that fall within that polygon?

Answer 1

So I used the solution (OpenStreetMap) from @Spacedman:

1. Download the .pbf for your country https://download.geofabrik.de

2. Download the polygon file for the desired municipality from https://github.com/JamesChevalier/cities or create your polygon file (.poly) here : http://share.mapbbcode.org/

3. Download lasted version from Osmosis https://wiki.openstreetmap.org/wiki/Osmosis#Latest_stable_version

4. Do this:

   osmosis --read-pbf-fast you_file_country-latest.osm.pbf file="your_file_country-latest.osm.pbf" --bounding-polygon file="your_municipality_polygon.poly" --write-xml file="your_file_out.osm"

5. In the .osm file, you obtain all the street names, to extract the names use the option (regex) of notepad++

It's complex but it works, but I still have to find a way to do it directly from Google API

Answer 2

Sandra's solution put me into the right direction and helped me a lot. However, I am convinced the way how she "parses" the output file is not correct and the regular expression doesn't cover all streets - at least in Spain as a street can start with Calle and this kind of streets don't have the addr:street key.

This is a Linux solution which bases on Sandra's answer:

bin/osmosis -q ... --bp file="My.poly" \
            --tf accept-nodes way=* --wk keyList=highway --wx file=- \
    | sed -e 's/[ ]*<nd [^>]*>//g' \
    | sed -r '/^\s*$/d' \
    | perl -lne 'print $1 if /k="name" v="(.+)"/' \
    | sort \
    | uniq

What happens:

1. Get only the way nodes which contain the highway key.
2. Remove the <nd..> elements. Here I refer to this question regarding removing it.
3. Remove empty lines.
4. Run perl in order to get the street name from the element.
5. Sort the result.
6. Remove duplicates which can be, when for example the street is a one-way and has multiple entries.