17

How do I find the document with the maximum uid field with map-reduce in pymongo?

I have tried the following but it prints out blanks:

from pymongo import Connection
from bson.code import Code

db = Connection().map_reduce_example
db.things.insert({  
    "_id" : "50f5fe8916174763f6217994", 
    "deu" : "Wie Sie sicher aus der Presse und dem Fernsehen wissen, gab es in Sri  Lanka mehrere Bombenexplosionen mit zahlreichen Toten.\n", 
    "uid" : 13, 
    "eng" : "You will be aware from the press and television that there have been a  number of bomb explosions and killings in Sri Lanka." 
})

db.things.insert({  
    "_id" : "50f5fe8916234y0w3fvhv234", 
    "deu" : "Ich bin schwanger foo bar.\n", 
    "uid" : 14, 
    "eng" : "I am pregnant foobar." 
})

db.things.insert({  
    "_id" : "50f5fe8916234y0w3fvhv234", 
    "deu" : "barbar schwarz sheep, haben sie any foo\n", 
    "uid" : 14, 
    "eng" : "barbar black sheep, have you any foo" 
})

m = Code("function () {emit(this.uid,{'uid':this.uid,'eng':this.eng})}")

r = Code("function (key, values) {var total = 0;for (var i = 0; i < values.length; i++) {total += values[i];}return total;}")


result = db.things.inline_map_reduce(m, r)

for r in result:
    print

An example document that look like these:

{ 
    "_id" : ObjectId("50f5fe8916174763f6217994"), 
    "deu" : "Wie Sie sicher aus der Presse und dem Fernsehen wissen, gab es mehrere Bombenexplosionen mit zahlreichen Toten.\n", 
    "uid" : 13, 
    "eng" : "You will be aware from the press and television that there have been a 
             number of bomb explosions and killings." 
}
alvas
  • 115,346
  • 109
  • 446
  • 738

1 Answers1

48

You can use find_one to find the doc with the maximum uid by sorting on that field descending:

db.things.find_one(sort=[("uid", -1)])

or using the defined constant:

db.things.find_one(sort=[("uid", pymongo.DESCENDING)])
JohnnyHK
  • 305,182
  • 66
  • 621
  • 471
  • isn't it more expensive to do `find()` or `find_one()` without the reduced map, especially when i have close to 2,000,000 documents in the database. – alvas Jan 22 '13 at 05:39
  • 1
    No, but you'd still want to make sure you had an index on `uid` if you need this to be performant. – JohnnyHK Jan 22 '13 at 06:02
  • 3
    Yo may use more descriptive predifined constants to point direction argument: `pymongo.DESCENDING` instead of `-1` and `pymongo.ASCENDING` instead of `1` – userlond Jun 08 '16 at 01:49