Difference between revisions of "MAPREDUCE basics"

Revision as of 11:16, 27 June 2016

#ENCODING
import io
import sys
sys.stdout = io.TextIOWrapper(sys.stdout.buffer, encoding='utf-16')
#MONGO
from pymongo import MongoClient
client = MongoClient()
client.progzoo.authenticate('scott','tiger')
db = client['progzoo']
#PRETTY
import pprint
pp = pprint.PrettyPrinter(indent=4, width=160)

MapReduce the basics

This tutorial introduces the MapReduce command.

MapReduce examples are available.

Find the total population of the each continent

db.world.mapReduce(
   function(){emit(this.continent,this.population);},
   function(k,v){return Array.sum(v);},
   {out:{inline:1}}
)

db.world.mapReduce(

  function(){emit(this.continent,this.population);},
  function(k,v){return Array.sum(v);},
  {out:{inline:1}}

)

Use the previous answer to find the population of the world to the nearest million

Use the JavaScript round function : Math.round(population/1000000)*1000000

from bson.code import Code temp = db.world.map_reduce(

   map=Code("function(){emit('World Population in Millions', this.population)}"),
   reduce=Code("""function(key, values){
     return Math.round(Array.sum(values)/1000000)*1000000;
   }"""),
   out={"inline":1})

pp.pprint(temp["results"])

Count number of countries by first letter

from bson.code import Code temp = db.world.map_reduce(

   map=Code("""function(){ emit((this.name).substring(0,1), 1)}"""),
   reduce=Code("""function(key, values){
      return Array.sum(values);
   }"""),
   out={"inline":1})

pp.pprint(temp["results"])

Show the number of countries on each continent

from bson.code import Code temp = db.world.map_reduce(

   map=Code("function(){emit(this.continent, 1)}"),
   reduce=Code("""function(key, values){
     return Array.sum(values);
   }"""),
   out={"inline":1})

pp.pprint(temp["results"])

Show the smallest 3 countries name and area (ignore areas of 0 or None)

from bson.code import Code temp = db.world.map_reduce(

       query={"$and":[{"area":{"$ne":None}}, {"area":{"$ne":0}}]},
       sort={"area":1},
       limit=3,
       map=Code("function(){emit(this.name, this.area)}"), 
       reduce=Code("function(key, values){}"),
       out={"inline":1},

)

pp.pprint(

  temp["results"]

)

Return the first and last country based on name order for each continent

from bson.code import Code temp = db.world.map_reduce(

   map=Code("function(){emit(this.continent, {first:this.name,last:this.name})}"),
   reduce=Code("""function(key, values){
     var ret = {first:'ZZZ',last:'AAA'};
     for(var i=0;i<values.length;i++){
       if (ret.first>values[i].first) ret.first=values[i].first;
       if (ret.last<values[i].last) ret.last=values[i].last;
     }
     return ret;
   }"""),
   out={"inline":1})

pp.pprint(temp["results"])

Return country name or capital city that starts with a letter 'M'

from bson.code import Code temp = db.world.map_reduce(

   map=Code("""function(){ 
                if((this.name).startsWith('M')) 
                     emit(this.name,null);
                if((this.capital).startsWith('M')) 
                  emit(this.capital,null);
            }"""),
   reduce=Code("""function(key, values){
      return values;
   }"""),
   
   out={"inline":1})

pp.pprint(temp["results"])

Show the first and last city for each letter and the count of cities

from bson.code import Code temp = db.world.map_reduce(

   map=Code("""function(){ if(this.capital)emit((this.capital).substring(0,1), {first:this.capital, last:this.capital})}"""),
   reduce=Code("""function(key, values){
      var ret = {first:'ZZZ',last:'AAA', count:0};
     for(var i=0;i<values.length;i++){
       if (ret.first>values[i].first) ret.first=values[i].first;
       if (ret.last<values[i].last) ret.last=values[i].last;
       ret.count += 1;
     }

     return ret;
   }"""),
   finalize=Code("""function(key, val){
     if(!val.count){
        val.count = 1;
        return val;
     }else
        return val;
   }"""),
   out={"inline":1})

pp.pprint(temp["results"])

Show country count for countries in the ranges

 0 to 1000000
 1000000 to 2000000
 2000000 to 3000000
 3000000 to 5000000
 5000000 to 10000000
 10000000 to 15000000
 More than 15000000

from bson.code import Code temp = db.world.map_reduce(

   map=Code("""function(){
                 var pop = this.population;
                 switch(true){
                   case pop<1000000:
                      emit("0 TO 1000000", 1);
                      break;
                   case pop<2000000:
                      emit("1000000 TO 2000000", 1);
                      break;
                   case pop<3000000:
                      emit("2000000 TO 3000000", 1);
                      break;
                   case pop<5000000:
                      emit("3000000 TO 5000000", 1);
                      break;
                   case pop<10000000:
                      emit("5000000 TO 10000000", 1);
                      break;
                   case pop<15000000:
                      emit("10000000 TO 15000000", 1);
                      break
                   case pop>15000000:
                      emit("MORE THAN 15000000", 1);
                      break;
                 }
}"""),
   reduce=Code("""function(key, values){ return Array.sum(values); }"""),      
   out={"inline":1})

pp.pprint(temp["results"])