source: other-projects/maori-lang-detection/mongodb-data/4counts_tentativeNonProductSites.json@ 33823

/*

The websites that have some MRI detected AND which are either in NZ or with NZ TLD
or (so if they're from overseas) don't contain /mi or mi.* in URL path.
We'll include Australia, to get the valid "kiwiproperty.com" website,
otherwise the sole exception, included in the result list.


db.getCollection('Websites').find({$and: [
                {numPagesContainingMRI: {$gt: 0}},
                {$or: [{geoLocationCountryCode: /(NZ|AU)/}, {domain: /\.nz$/}, {urlContainsLangCodeInPath: false}]}
            ]}).count()

397

Aggregate results by a count of country codes:

db.Websites.aggregate([
    {
        $match: {
            $and: [
                {numPagesContainingMRI: {$gt: 0}},
                {$or: [{geoLocationCountryCode: /(NZ|AU)/}, {domain: /\.nz$/}, {urlContainsLangCodeInPath: false}]}
            ]
        }
    },
    { $unwind: "$geoLocationCountryCode" },
    {
        $group: {
            _id: {$toLower: '$geoLocationCountryCode'},
            count: { $sum: 1 }
        }
    },
    { $sort : { count : -1} }
]);

*/

/* 1 */
{
    "_id" : "us",
    "count" : 181.0
}

/* 2 */
{
    "_id" : "nz",
    "count" : 89.0
}

/* 3 */
{
    "_id" : "au",
    "count" : 21.0
}

/* 4 */
{
    "_id" : "de",
    "count" : 19.0
}

/* 5 */
{
    "_id" : "fr",
    "count" : 17.0
}

/* 6 */
{
    "_id" : "nl",
    "count" : 16.0
}

/* 7 */
{
    "_id" : "dk",
    "count" : 8.0
}

/* 8 */
{
    "_id" : "ca",
    "count" : 7.0
}

/* 9 */
{
    "_id" : "es",
    "count" : 6.0
}

/* 10 */
{
    "_id" : "gb",
    "count" : 5.0
}

/* 11 */
{
    "_id" : "cz",
    "count" : 4.0
}

/* 12 */
{
    "_id" : "at",
    "count" : 3.0
}

/* 13 */
{
    "_id" : "it",
    "count" : 3.0
}

/* 14 */
{
    "_id" : "ro",
    "count" : 3.0
}

/* 15 */
{
    "_id" : "il",
    "count" : 2.0
}

/* 16 */
{
    "_id" : "ch",
    "count" : 2.0
}

/* 17 */
{
    "_id" : "bg",
    "count" : 1.0
}

/* 18 */
{
    "_id" : "sg",
    "count" : 1.0
}

/* 19 */
{
    "_id" : "mx",
    "count" : 1.0
}

/* 20 */
{
    "_id" : "ir",
    "count" : 1.0
}

/* 21 */
{
    "_id" : "cn",
    "count" : 1.0
}

/* 22 */
{
    "_id" : "ie",
    "count" : 1.0
}

/* 23 */
{
    "_id" : "jp",
    "count" : 1.0
}

/* 24 */
{
    "_id" : "fi",
    "count" : 1.0
}

/* 25 */
{
    "_id" : "gr",
    "count" : 1.0
}

/* 26 */
{
    "_id" : "ru",
    "count" : 1.0
}

/* 27 */
{
    "_id" : "unknown",
    "count" : 1.0
}