Un Data API

This is where support questions get answered.

Hello, and thanks for this genuinely handy API!

I spotted a couple of small problems

1) There's a small mistake in the documentation here:
http://www.undata-api.org/wiki/filteredquery

the query only accepts
http://undata-api.appspot.com/data/query/{dataset name1}/{country3}/{year2}
and not
http://undata-api.appspot.com/data/query/{dataset name1}/{year2}/{country3}

2) Also, for some reason every record comes back twice on queries like these:

http://undata-api.appspot.com/data/query/Population%20annual%20growth%20rate%20%28percent%29/any/2000?user_key=xxx

or

http://undata-api.appspot.com/data/query/Population%20annual%20growth%20rate%20%28percent%29/France?user_key=xxx

The responses look like this — each year shows up twice. Could the data be half-yearly?

<record>
<field name="Country or Area">France</field>
<field name="Year(s)">1990</field>
<field name="Value">0.5</field>
</record>
−
<record>
<field name="Country or Area">France</field>
<field name="Year(s)">1990</field>
<field name="Value">0.5</field>
</record>
−
<record>
<field name="Country or Area">France</field>
<field name="Year(s)">1991</field>
<field name="Value">0.5</field>
</record>
−
<record>
<field name="Country or Area">France</field>
<field name="Year(s)">1991</field>
<field name="Value">0.5</field>
</record>

Thanks

Oops, the forum formatting engine mangled my text a little :)
Hope it still makes sense…

Issue: the country name "Côte d’Ivoire" isn't UTF8-encoded in the "Population annual growth rate (percent)/" query (and possibly in other queries too, if they all share the same country table).

Issue: broken XML on this query:
http://undata-api.appspot.com/data/query/Population%20living%20below%20the%20poverty%20line%20%28percent%20living%20on%20less%20than%20US$1%20per%20day%29/any?user_key=xxx

the text reads "<2.0" where it should read "2.0"

this shows up for the first country, "Albania", in the poverty-line query

Oops — actually it affects every country whose value is "2.0", since that means "less than 2%"

which raises a question: what should be used in place of this special character?

The <2.0 problem webving raised is a serious one. It breaks XML parsers because a < is being emitted unescaped. Malformed XML can't be parsed by the endpoints unless it's cleaned up first.