MongoDB - Sorting in Local Language

MongoDB uses unicode, instead of the coding for a certain local language, to sort data in this language (i.e. Chinese). Together with esProc, MongoDB can realize sorting in local language conveniently (i.e. sort Chinese according to Chinese phonetic alphabet). The following will teach you the method in detail by taking Chinese as an example.

person - a collection in MongoDB - stores names and genders as follows:

> db.person.find()

{ "_id" : ObjectId("544e4e070f03ad39eb2bf498"), "name" : "å®æ±", "gender" : "ç·"}

{ "_id" : ObjectId("544e4e070f03ad39eb2bf499"), "name" : "æéµ", "gender" : "ç·"}

{ "_id" : ObjectId("544e4e070f03ad39eb2bf49a"), "name" : "å´ç¨", "gender" : "ç·"}

{ "_id" : ObjectId("544e4e070f03ad39eb2bf49b"), "name" : "æç", "gender" : "ç·"}

{ "_id" : ObjectId("544e4e070f03ad39eb2bf49c"), "name" : "å¬å­è", "gender" : "ç·" }

{ "_id" : ObjectId("544e4e070f03ad39eb2bf49d"), "name" : "é²æºæ·±", "gender" : "ç·" }

{ "_id" : ObjectId("544e4e070f03ad39eb2bf49e"), "name" : "æ­¦æ¾", "gender" : "ç·"}

{ "_id" : ObjectId("544e4e070f03ad39eb2bf49f"), "name" : "é®å°äº", "gender" : "ç·" }

{ "_id" : ObjectId("544e4e070f03ad39eb2bf4a0"), "name" : "æ¨å¿", "gender" : "ç·"}

{ "_id" : ObjectId("544e4e070f03ad39eb2bf4a1"), "name" : "å­äºå¨", "gender" : "女" }

{ "_id" : ObjectId("544e4e070f03ad39eb2bf4a2"), "name" : "æä¸å¨", "gender" : "女" }

{ "_id" : ObjectId("544e4e080f03ad39eb2bf4a3"), "name" : "çé", "gender" : "ç·"}

â¦

Sort the data using MongoDBâs sort function rather than the Chinese phonetic alphabet:

> db.person.find({},{"name":1,"gender":1,"_id":0}).sort({"name":1})

{ "name" : "å¬å­è", "gender" : "ç·" }

{ "name" : "å´ç¨", "gender" : "ç·" }

{ "name" : "å­äºå¨", "gender" : "女" }

{ "name" : "å®æ±", "gender" : "ç·" }

{ "name" : "æä¸å¨", "gender" : "女" }

{ "name" : "æç", "gender" : "ç·" }

{ "name" : "æéµ", "gender" : "ç·" }

{ "name" : "æ¨å¿", "gender" : "ç·" }

{ "name" : "æ­¦æ¾", "gender" : "ç·" }

{ "name" : "çé", "gender" : "ç·" }

{ "name" : "é®å°äº", "gender" : "ç·" }

{ "name" : "é²æºæ·±", "gender" : "ç·" }

â¦

A1ï¼Connect to the MongoDB database. The IP and port number is localhost:27017, the database name is test and both the user name and the password are test. If any other parameters are needed, write them in line with the format mongo://ip:port/db?arg=value&â¦

A2ï¼Fetch data from the MongoDB database using find function to create a cursor. The collection is person. The filtering criterion is null and the specified keys are name and gender. It can be seen that this find function is similar to the find function of MongoDB. By fetching and processing data in batches, the esProc cursor can avoid the memory overflow caused by big data importing.

A3ï¼Since the data here are small, fetch function will fetch them all at once.

A4ï¼Sort the data by name in ascending order, using sort function. Chinese is used in the data sorting. For the other localized languages esProc supports, please see below.

One thing to note is that esProc doesnât provide the java driver of MongoDB. To access MongoDB with esProc, the latter (a driver of 2.12.2 version or above is required, i.e. mongo-java-driver-2.12.2.jar) should be put into the [esProc installation directory]\common\jdbc beforehand.

The script for computation in MongoDB with the assistance of esProc is easy to integrate with Java program. By adding another line of code - A5, which is result A4, the result in the form of resultset can be output to Java program. For detailed code, please refer to esProc Tutorial. In the same way, to access MongoDB by calling esProc code with Java program also requires putting the java driver of MongoDB into the classpath of Java program.