Difference between revisions of "Round"
From NoSQLZoo
m |
|||
(9 intermediate revisions by the same user not shown) | |||
Line 1: | Line 1: | ||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
{{TopTenTips}} | {{TopTenTips}} | ||
− | Rounding is easy inside a <code>MapReduce</code> as it is possible to use the <code>Math</code> object provided by JavaScript.<br/> | + | <div style="min-height:25em"> |
+ | There are various ways of rounding in MongoDB.<br/> | ||
+ | Rounding is easy inside of a <code>MapReduce</code> as it is possible to use the <code>Math</code> object provided by JavaScript.<br/> | ||
More information on <code>Math</code> can be found [https://developer.mozilla.org/en/docs/Web/JavaScript/Reference/Global_Objects/Math here].<br/><br/> | More information on <code>Math</code> can be found [https://developer.mozilla.org/en/docs/Web/JavaScript/Reference/Global_Objects/Math here].<br/><br/> | ||
− | + | In MongoDB versions prior to 3.2 the <code>aggregation()</code> method has no rounding functions, though it is still doable with <code>$mod</code> and <code>$multiply</code>, as shown [http://www.kamsky.org/stupid-tricks-with-mongodb/rounding-numbers-in-aggregation-framework here] by the lead project manager at MongoDB.<br/> | |
− | + | As of MongoDB 3.2, it is possible to round up, down, and truncate via <code>$ceil</code>, <code>$floor,</code> and <code>$trunc</code> respectively. | |
− | < | + | </div> |
− | <div class=q data-lang=" | + | ==MapReduce== |
− | <pre class=def> | + | <div class="q nonum" data-lang="mongo"> |
− | + | <pre class="def"><nowiki> | |
− | + | db.world.mapReduce( | |
− | + | function () { | |
− | + | emit(this.name, this.population); | |
− | + | }, | |
− | + | function (key, values) { | |
− | + | return values; | |
− | + | }, | |
− | + | { | |
− | ) | + | finalize: function (key,values) { |
− | </pre> | + | return { |
+ | "population": values, | ||
+ | "population in millions": Math.round(values / 1000000) | ||
+ | }; | ||
+ | }, | ||
+ | out: {inline: 1}, | ||
+ | query: {"name": "United Kingdom"} | ||
+ | } | ||
+ | );</nowiki></pre> | ||
</div> | </div> | ||
− | <div class=q data-lang=" | + | ==Aggregate== |
− | <pre class=def> | + | <div class="q nonum" data-lang="mongo"> |
− | + | <pre class="def"><nowiki> | |
− | + | db.world.aggregate([ | |
− | + | {"$match": {"name": "United Kingdom"}}, | |
− | + | {"$project": { | |
− | + | "_id": 1, | |
− | + | "name": 1, | |
− | + | "population": {"$divide": ["$population", 1000000]} | |
− | + | }}, | |
− | + | {"$project": { | |
− | + | "_id": 0, | |
− | + | "name": 1, | |
− | + | "population in millions": '$population', | |
− | + | "population in millions (floored)": {"$floor": "$population"}, | |
− | + | "population in millions (ceiling)": {"$ceil": "$population"}, | |
− | + | "population in millions (truncated)": {"$trunc": "$population"}, | |
− | ) | + | "population in millions (1dp with multiply / mod)":{ |
− | </pre> | + | "$divide":[ |
+ | {"$subtract":[ | ||
+ | {"$multiply": ['$population', 100]}, | ||
+ | {"$mod": [{"$multiply": ['$population', 100]}, 1]} | ||
+ | ]}, | ||
+ | 100 | ||
+ | ] | ||
+ | } | ||
+ | }} | ||
+ | ]).pretty();</nowiki></pre> | ||
</div> | </div> |
Latest revision as of 16:45, 18 July 2018
There are various ways of rounding in MongoDB.
Rounding is easy inside of a MapReduce
as it is possible to use the Math
object provided by JavaScript.
More information on Math
can be found here.
In MongoDB versions prior to 3.2 the aggregation()
method has no rounding functions, though it is still doable with $mod
and $multiply
, as shown here by the lead project manager at MongoDB.
As of MongoDB 3.2, it is possible to round up, down, and truncate via $ceil
, $floor,
and $trunc
respectively.
MapReduce
db.world.mapReduce( function () { emit(this.name, this.population); }, function (key, values) { return values; }, { finalize: function (key,values) { return { "population": values, "population in millions": Math.round(values / 1000000) }; }, out: {inline: 1}, query: {"name": "United Kingdom"} } );
Aggregate
db.world.aggregate([ {"$match": {"name": "United Kingdom"}}, {"$project": { "_id": 1, "name": 1, "population": {"$divide": ["$population", 1000000]} }}, {"$project": { "_id": 0, "name": 1, "population in millions": '$population', "population in millions (floored)": {"$floor": "$population"}, "population in millions (ceiling)": {"$ceil": "$population"}, "population in millions (truncated)": {"$trunc": "$population"}, "population in millions (1dp with multiply / mod)":{ "$divide":[ {"$subtract":[ {"$multiply": ['$population', 100]}, {"$mod": [{"$multiply": ['$population', 100]}, 1]} ]}, 100 ] } }} ]).pretty();