A tiny implementation of NoSQL database with RESTful API support.

from nosql import Database

Database

• Create a database: db = Database('./YOUR_DATABASE_DIRECTORY/')
• Show collections in database: db.show_collections()
• Create a new collection: db.create_collection(NEW_NAME)
• Get the collection object from a database: col = db.get_collection(NAME)
  • Alternatively, you can also call a collection as an attribute. For example, db.main will return the collection with name main.
• Delete a collection from a database: db.drop_collection(NAME)
• Close and save database: db.close()

Collection

• Return all documents from collection: col.all()
• CRUD Operations: Please see CRUD section

Start the nosql.py with corresponding argument. By default the database will run on localhost:9020, and the examples below are based on this.

Suppose there exists a collection named fruits.

curl -X "POST" "http://localhost:9020/fruits/_insert" \
     -H 'Content-Type: application/json' \
     -d $'{
  "type": "banana",
  "price": 10
}'

Supported document types: a JSON object or Python dictionary is acceptable. The valid types include float, integer, string and nested array or object.

To retrieve all documents in a collection.

A UUID-format document ID will be generated every time a document is inserted. The document ID will be returned after the insertion.

curl "http://localhost:9020/fruits/_find" \
     -H 'Content-Type: application/json' \
     -d $'{
  "_id": "DOC_ID"
}'

curl "http://localhost:9020/fruits/_find" 
     -H 'Content-Type: application/json' 
     -d $'{
  "type": "banana"
}'

The query criteria must be a query string, JSON object, or Python dictionary. Below are examples of JSON object criteria.

• _id: UUID string. Specify document id to find the exact document.
• Equal: Use field/value pair to find all documents with field matching the value.
• Contain: Use field/value pair to find all documents with lists containing the value.
  • e.g. {"type": "orange", "taste": ["sweet", "sour"]} will be matched with criterion {"taste": "sweet"}
• Advanced search
  • _eq: equal. E.g. {"_eq": {"taste": "sour"}} will only match documents with taste equal to sour.
  • _lt, _le: less than, less or equal to.
  • _gt, _ge: greater than, greater or equal to.
  • _ne: not equal to.
  • _in: is in.
  • _nin: is not in.

For example, if you want to add 1 dollars to all products with price lower than 10 dollars, this contains a query (find all documents with price <10), and an update operation (increase price by 1 ). The JSON needs to contain to object: "criteria" and "operation".

curl -X "PUT" "http://localhost:9020/fruits/_update" \
     -H 'Content-Type: application/json' \
     -d $'{
  "operation": {
    "_increment": [
      "price"
    ]
  },
  "criteria": {
    "price": {
      "_lt": 10
    }
  }
}'

• _set: key/value paris of fields to update and new values
• _unset: list of fields to be deleted
• _increment: list of fields to increase by one. Need to be numeric values.
• _append: key/value paris of fields and values to be appended

curl -X "DELETE" "http://localhost:9020/fruits/_remove" \
     -H 'Content-Type: application/json' \
     -d $'{
  "operation": {
    "_increment": [
      "price"
    ]
  },
  "criteria": {
    "price": {
      "_lt": 100
    }
  }
}'