몇 가지 유용한 Pymongo 조각

#display all databases available
db_list = list(client.list_databases())
print(db_list)

#or

for db in client.list_databases():
    print(db)

#display all collections available
for db in client.list_databases():
    name = db['name']
    for col in client[name].list_collections():
        print(col)

#display one document from "Papers" collection
db = client.arxiv
papers_col = db.papers
doc = papers_col.find_one()
print(doc)

#display number of documents in the collection
number_of_doc = papers_col.count_documents({})
print(number_of_doc)

#display 10 articles titles
articles = list(papers_col.find({}, {'title': 1}).limit(10))

from pymongo import ASCENDING
#"Submitter" attribute is author's name
#Display 10 first authors name in ascending alphabetical order

# sort, get 10
articles = list(
    papers_col.find({}, {'submitter': 1})
    .limit(10)
    .sort([('submitter', ASCENDING)])
)
print(articles)

#Display the quantity of articles that has not published by "Damien Chablat"
pattern = re.compile(r'Damien Chablat')
articles = papers_col.count_documents({ 'submitter': { '$not': pattern } } )
print(articles)

#"update_date" attibute contain documents upload date informations (yyyy-mm-dd format)
# Display number of article upload on 2014

from datetime import date

first_date = date.isoformat(date(2014,1,1))
last_date = date.isoformat(date(2015,1,1))

articles_count = papers_col.count_documents({'update_date':{'$gte':first_date,'$lt':last_date}})
print(articles_count)

# Display an article title where "Machine Learning" is metionned in the abstract

papers_col.create_index([("abstract", TEXT)])
articles = papers_col.find({"$text":{"$search": "Machine Learning"}},{'abstract':1})

# Display an article title where "Machine Learning" is metionned in the abstract

pattern = re.compile(r'Machine Learning')
articles = papers_col.find({ 'abstract': { '$regex': pattern } } )

print(list(articles))

#Display the amount of publications/articles for the 10 best submitters

pipeline = [
    { "$group": {"_id": "$submitter", "count": {"$sum": 1}} },
    { "$sort": { "count": -1 } },
    { '$limit': 10 }
]

articles = list(papers_col.aggregate(pipeline))
print(articles)

def  update_or_create_paper(paper_data):
    # update 'data' if custom 'id' exists otherwise insert new document
    return collection.find_one_and_update({"id": paper_data['id']},
                               {"$set": {"data": {**paper_data}}},
                               upsert=True)