Yep, but now we can see where streets are missing.
Also, I’m thinking about deleting old addr info withing tel-aviv borders (with source other then muni data), otherwise we’ll have many duplicates.
Please don’t delete automatically… It took me soo much time to collect them
(I know its not that much, but still)
Let us consider deleting them, and let me check if every manual insert from me is the same in the Tel Aviv Dataset.
Probably there is no change but still. Give me some time.
If you see only some of the numbers, refresh the image, because I uploaded in two batches and the image shows only the first batch.
Now we should finish adding buildings to the city where possible, and align the existing buildings and addresses where necessary so that they match each other.
Also, in Kfar Shalem (southeast corner of the city) there are entire streets missing, the addresses (including street name as well as number!) should allow us to reconstruct the streets there. Possibly also in Yafo too.
Lets export and delete all old address data in TelAviv. I don’t see a reason to keep it anymore. The new data seems very accurate and should be used in the future.
####################
Before we start adding buildings, we should look if there are any issues with the new data:
There are for example a lot of double nodes on the same place now with housenumbers:
If I understand correctly there are house numbers only available for Tel Aviv municipality.
Meaning that this is the most accurate information so far what are the administrative borders for the Tel Aviv municipality.
Can we adjust the relation of TelAviv by this information?
This is after exporting the XLS file to CSV, reordering the columns, doing some find/replace in the CSV to account for quotes in the street name, and deleting extra spaces/tabs in the CSV…
When exporting the CSV, I had to make sure the Hebrew encoding was correct (Unicode…)
#!/usr/bin/perl
sub trim($)
{
my $string = shift;
$string =~ s/^\s+//;
$string =~ s/\s+$//;
return $string;
}
open(SHAPES, "<c3.csv") or die ("Could not load file'\n");
open(OUT, ">out.osm") or die ("Could not open output file\n");
$count=0;
# Opening XML
print OUT "<?xml version='1.0' encoding='UTF-8'?>\n";
print OUT "<osm version='0.6' generator='addshapes.pl'>\n";
# Read/process lines
while(my $line = <SHAPES>) {
$count++;
$line =~ s/\r//g;
chomp $line;
unless($line) { next; }
my @linearray = split(/,/,$line);
@linearray[0]=~ s/\'/'/g; # Escape single quote in street name
print OUT "<node id=\"-".$count."\" lat=\"".@linearray[4]."\" lon=\"".@linearray[3]."\"><tag k='addr:street' v='".@linearray[0]."' /><tag k='addr:housenumber' v='".@linearray[1].trim(@linearray[2])."' /><tag k='source' v='Tel Aviv municipality' /><tag k='contact:website' v='http://www.tel-aviv.gov.il/OnlineServices/DataTLV/Pages/Data%20Tel-Aviv.aspx' /></node>\n";
}
# Closing XML
print OUT "</osm>\n";
close(SHAPES);
close(OUT);
Then I checked and uploaded the .osm file with JOSM.
Haven’t even tried.
A list of stations, their addresses and coordinations is published under the Tel-Aviv muni data site, under the same license so I see no problen using it.
There is a relation for the city of Tel-Aviv Yafo that marks the border of the city.
Applications and search engines based on OSM are checking for this relations and if exist, they know the city name and country.
Adding the addr:city and addr:country tag is only mandatory if there is no city relation- what is not the case in Israel.
We have very well defined city definitions all over Israel. No issue with that
I’d like to better understand how this can be done.
For example, is there a way to query the OSM DB for all the streets in Tel-Aviv based on this relation?
Similarly, is there a way to query the DB for items in Israel only based on its borders relations?