Table of contents
  1. Transfer database from one server to another server
  2. Install mongodb in Centos7
  3. Aggregation pipeline
  4. update array using filter
  5. restart MongoDB
  6. projection in pyMongo
  7. Copy one collection from one database to another
  8. Remove one filed in collection
  9. Rename the filed in collection
  10. append an item to an array in mongoDB

Transfer database from one server to another server

  1. Copy the database: mongodump : mongodump --db db_name --archive=./mongodbBackup/db_name.dump --gzip
  2. Scp transfer the dump file: scp db_name.yml server_name@ip_address:path_to_save
  3. login another server
  4. Restore the database: mongorestore --gzip --archive=db_name.dump

It will restore the database to the default mongodb save path. You can find the save path at /etc/mongod.conf

Install mongodb in Centos7

  1. Create a file : sudo vi /etc/yum.repos.d/mongodb-org.repo
  2. Enter the content at /etc/yum.repos.d/mongodb-org.repo: 
     [mongodb-org-6.0]
 name=MongoDB Repository
 baseurl=https://repo.mongodb.org/yum/redhat/7/mongodb-org/6.0/x86_64/
 gpgcheck=1
 enabled=1
 gpgkey=https://www.mongodb.org/static/pgp/server-6.0.asc
  3. Install the package: sudo yum install mongodb-org
  4. Start : sudo systemctl start mongod
  5. Check Status: sudo systemctl status mongod
  6. If you’d like to change the /etc/mongo.conf, such as change the dbPath, then you should use the command to restart the Mongodb service: sudo service mongod restart
  7. If you change the dbPath to a location that the mongodb have no permission to modify, then the status of mongodb is failed. you need to change it back or change the folder permission.

Uninstall the Mongodb :

  1. Stop : sudo service mongod stop
  2. Remove the installed package: sudo yum erase $(rpm -qa | grep mongodb-org)
  3. Remove data files: 
    sudo rm -r /var/log/mongodb
sudo rm -r /var/lib/mongo

Aggregation pipeline

Check the official document for pipeline stages

You can use aggregation pipeline to get the documents based on complex conditions.

def searchFusions():
    tumorType = request.form.get('tumorType',None)
    geneName1 = request.form.get('geneName1',None)
    geneName2 = request.form.get('geneName2',None)
    tier = request.form.get('tier',None)
    
    
    listAnd = []
    if geneName1:
        listAnd.append({"$regexMatch":{ "input": "$$fusion.gene" , "regex": f"{geneName1}", "options": "i" }})
    # because gene is the second level of dict, so use double $: $$
    # Do Not use f"/{geneName1}/" as the CLI format
    if geneName2:
        listAnd.append({"$regexMatch":{ "input": "$$fusion.gene" , "regex": f"{geneName2}", "options": "i" }})
    if tier:
        listAnd.append({"$eq":["$$fusion.tier",tier]})

    stageMatchTumorType = {}
    if tumorType:
        stageMatchTumorType = {
            '$match':{'dictCaseInfo.tumorType':{'$regex':tumorType,'$options':'i'}}
        }

    stageProjectFilter = {
        "$project":{
            "listFusion":{
                "$filter":{
                    "input":"$listFusion",
                    "as":"fusion",
                    "cond": {"$and":listAnd},
                },
            },
            'dictCaseInfo':True
            
        
        }
    }

    stageProjectFileds = {
        "$project":{
            '_id':True,
            'listFusion':True,
            'dictCaseInfo':True,
            "numberOfFusion":{"$size":"$listFusion"},

        }
    }

    stageMatchRemoveEmptyResult = {
        "$match":{
        "numberOfFusion":{"$gt":0}
        }
    }

    # narrow down the documents and arrnge the results
    pipeline = [stageMatchTumorType,stageProjectFilter,stageProjectFileds,stageMatchRemoveEmptyResult]
    pipeline = [element for element in pipeline if len(element) > 0]

    listRecord = collectionReport.aggregate(pipeline)
    
    listRecord = list(listRecord)

    return jsonify(listFormatRecord)

update array using filter

if you’d like to update an element in an array

query = {'_id':ObjectId(ID)}
update = {"$set":{f'listArray.$[element]':updateStuff}}
arrayFilter = [{f"element.ID2":ID2}]
result = collectionImportant.update_one(filter=query,update=update,array_filters=arrayFilter)

restart MongoDB

When the server crashed down, and you failed to connect to the mongoDB server after typing mongosh. Then you need to restart the mongoDB server as shown below.

sudo service mongod restart

projection in pyMongo 

query = {'status':'active', 'testOrdered':testOrdered,'panel':panel,'specimen':specimen}
list_tag = list(collection_rule_extraction.find(query, projection={'_id':False,'tag':True}))
list_tag = [rule['tag'] for rule in list_tag]

You need to use True or False to specify which filed needed to be returned.

Copy one collection from one database to another

  1. Install mongosh
  2. In the shell using the command line to backup: mongodump -d some_database -c some_collection (e.g. mongodump -d comment -c comment_ST)
  3. restore the collection to another database: mongorestore -d some_other_db -c some_or_other_collection dump/some_collection.bson (e.g. mongorestore -d report -c comment_ST dump/comment/comment_ST.bson)
  4. Done!

Remove one filed in collection 

db.example.update({}, {$unset: {words:1}} , {multi: true});

Add one filed in collection:

db.example.update({}, {$set: {'status':'active'}} , {multi: true});

Rename the filed in collection

go into the mongosh command line interface

db.collectionName.update({}, {
    $rename: {
        "old": "new"
    }
}, false, true);


//e.g.
db.foo.update({}, {
    $rename: {
        "name.additional": "name.last"
    }
}, false, true);

The false, true in the method above are: { upsert:false, multi:true }. You need the multi:true to update all your records.

append an item to an array in mongoDB 

coll.update({'ref': ref}, {'$push': {'tags': new_tag}})