Skip to content

3 API reference

Jump to bottom Edit New page
Remy Lalanne edited this page Jun 10, 2020 · 46 revisions

The damas-core API is implemented as modules for Python and Javascript programming languages. You could read 2 Getting Started to setup a scripting environment, or try the demo site https://demo.damas.io

Notes on types

The client modules use the built-in types: Python expose elements as dictionaries and JavaScript expose elements as Objects. Python None, True, False, are equivalent to JavaScript null, true, false, and are translated to/from JSON to communicate with the server. Python lists, tuples or sets can be used as arrays.

Sync / Async

The JavaScript API supports both synchronous and asynchronous requests. If the optional callback argument is provided, the request will run asynchronously and the response will be given as an argument to the specified callback. If the callback is not provided, the request is made synchronously and the return value holds the response. The Python API uses synchronous requests only (a bit of work is required to make them async ready).

Graphs

Since version 2.3, and in the NodeJS server, the edges (the directed links between nodes) are objects as nodes, wearing key/values, with the reserved keys src_id and tgt_id referring to the _id of the nodes to link.

Specifications

The API implementations follow the specifications described in the Specifications page to communicate with the server, that you could read if you would like have more details about the underlying architecture.

Table of contents

Generic elements

The elements are identified using unique identifiers stored in the reserved _id key. If the _id key is not provided at creation, a default unique value for _id is assigned. The other properties are stored next to it as key/value pairs using the JSON types.

{
    "_id": "your_custom_id",
    "number": 1234,
    "string":"hello world",
    "boolean": true
}

A set of generic CRUD functions is provided to create, read, modify and delete elements:

  • create - insert new element(s)
  • read - retrieve element(s) using identifiers
  • update - modify existing element(s)
  • upsert - create or modify element(s)
  • delete - delete element(s)

User Management

The users are described as elements wearing some reserved keys: username, password, class plus optional keys.

{
    "class": "user",
    "email": "usermail@address.com",
    "fullname": "Firstname Lastname String",
    "password": "13d3a2a16c0cd2f7bf115d471999377e",
    "username": "userlogin"
}

The Authentication page gives more details about the authentication mechanism.

  • signIn- request an authentication token from the server
  • signOut - revoke a token
  • verify - ask the server for the authentication status and user

Search queries

  • graph - retrieve connected edges and nodes (recursive)
  • search - find elements matching a query string
  • search_one - find first element matching a query string
  • search_mongo - find elements using a MongoDB query object (if MongoDB is the back-end)

Functions list

create

Create element(s) in the database. Elements have an _id key being their unique identifier in the database. This key can be specified during creation, but can't be updated afterwards without first deleting the element. The server may add some other arbitrary keys (like author, time at creation) depending on its configuration.

create ( elements [, callback] )

parameters

  • elements {object | array} : object(s) to insert in the database
  • callback {function} (js only, optional) : if specified, the request is asynchronous

return values

  • returns an object or an array of objects (depending on the input) on success
  • returns null (Javascript) or None (Python) on failure

examples

Python

# create a new element wearing an auto-generated unique identifier
>>> damas.create({"key1":"value1"})
{u'_id': u'55ae0b1ed81e88357d77d0e9', u'key1': u'value1', u'time': 1437469470133, u'author': u'demo'}

# create a new element with a specified identifier
>>> damas.create({"_id":"/project/folder/file", "additional_key":"value"})
{u'_id': u'/project/folder/file', u'additional_key': u'value', u'time': 1480586620449, u'author': u'demo'}

# create two elements
>>> damas.create([{"label":"element1"}, {"label":"element2"}])
[{u'_id': u'583ff5a747e759beb73bde32', u'time': 1480586663024, u'label': u'element1', u'author': u'demo'}, {u'_id': u'583ff5a747e759beb73bde33', u'time': 1480586663024, u'label': u'element2', u'author': u'demo'}]

# create two elements using an array as _id
>>> damas.create({"_id":["identifier1","identifier2"],"key":"keyvalue"}])

# create a new graph edge element
>>> damas.create({"src_id":"/project/folder/file1","tgt_id":"/project/folder/file2"})
{u'tgt_id': u'/project/folder/to/file2', u'_id': u'583ff67647e759beb73bde34', u'time': 1480586870826, u'src_id': u'/project/folder/to/file1', u'author': u'demo'}

Javacsript

// create a new element
damas.create({key1: "value1"});
▸ Object { author: "demo", time: 1437469470133, key1: "value1", _id: "55ae0b1ed81e88357d77d0e9" }

// create a new element wearing an auto-generated unique identifier (asynchronously)
damas.create({key1: "value2"}, function (element) {
    console.log(element.time);
});

// create a new element with a specified identifier
damas.create({"_id":"/project/folder/file", "additional_key":"value"});
▸ Object {_id: "/project/folder/file", additional_key: "value", time: 1480586620449, author: "demo"}

// create two elements
damas.create([{"label":"element1"}, {"label":"element2"}]);
▸ Array [{_id: "583ff5a747e759beb73bde32", time: 1480586663024, label: "element1", author: "demo"}, {_id: "583ff5a747e759beb73bde33", time: 1480586663024, label: "element2", author: "demo"}]


// create two elements with an array as _id
damas.create({"_id":["identifier1","identifier2"],"key":"keyvalue"}]);

// create a new graph edge element
damas.create({"src_id":"/project/folder/file1","tgt_id":"/project/folder/file2"})
▸ Object {tgt_id: "/project/folder/to/file2", _id: "583ff67647e759beb73bde34", time: 1480586870826, src_id: "/project/folder/to/file1", author: "demo"}

read

Retrieve one or more elements given their identifiers

read ( identifiers [, callback] )

parameters

  • identifiers {string | array} : identifier(s) string(s) to read
  • callback {function} (js only, optional) : if specified, the request is asynchronous

return values

  • returns an object or an array of objects (depending on the input) on success
  • returns null or None on failure

For multiple mode, The resulting array is sorted in the same order as the input array of identifiers. If some identifiers are not found, the result array is filled with None / null values for that position.

examples

Python

# read a file
>>> damas.read("/project/folder/file")
{u'additional_key': u'value', u'_id': u'/project/folder/file', u'time': 1480586620449, u'author': u'demo'}

# read 2 elements
>>> damas.read(["583ff5a747e759beb73bde32","583ff5a747e759beb73bde33"])
[{u'_id': u'583ff5a747e759beb73bde32', u'time': 1480586663024, u'label': u'element1', u'author': u'demo'}, {u'_id': u'583ff5a747e759beb73bde33', u'time': 1480586663024, u'label': u'element2', u'author': u'demo'}]

Javascript

// read a file
damas.read("/project/folder/file");
▸ Object {additional_key: "value", _id: "/project/folder/file", time: 1480586620449, author: "demo"}

// read 2 elements
damas.read(["583ff5a747e759beb73bde32", "583ff5a747e759beb73bde33"]);
▸ Array [_id: "583ff5a747e759beb73bde32", time: 1480586663024, label: "element1", author: "demo"}, {_id: "583ff5a747e759beb73bde33", time: 1480586663024, label: "element2", author: "demo"}]

update

Modify and remove keys of the specified element(s)

update ( elements [, callback] )

parameters

  • elements {object | array} : object(s) to update
  • callback {function} (js only, optional) : if specified, the request is asynchronous

return values

  • returns an object or an array of objects (depending on the input) on success
  • returns null (None in Python) on failure
  • returns an array of objects or null on mixed success/failures

The specified keys overwrite the existing keys on the server. Unspecified keys are left untouched on the server. A null or None value removes the key.

If multiple nodes are specified, the resulting array is sorted in the same order as the input array of identifiers.

The _id key can accept an array of identifiers, which means that the update will be performed on each specified element

examples

Javascript

// update an element
damas.update({_id: "55ae0b1ed81e88357d77d0e9", key1: null, newkey: "value2"});
▸ Object { author: "demo", time: 1437469470133, newkey: "value2", _id: "55ae0b1ed81e88357d77d0e9" }

// update multiple elements
damas.update([{_id: "583ff5a747e759beb73bde32", "label":"elementA"}, {_id: "583ff5a747e759beb73bde33", "label":"elementB"}]);
▸ Array [{_id: "583ff5a747e759beb73bde32", time: 1480586663024, label: "elementA", author: "demo"}, {_id: "583ff5a747e759beb73bde33", time: 1480586663024, label: "elementB", author: "demo"}]

// update multiple elements, condensed
damas.update({_id: ["583ff5a747e759beb73bde32","583ff5a747e759beb73bde33"], "newkey":"value"});
▸ Array [{_id: "583ff5a747e759beb73bde32", time: 1480586663024, label: "elementA", author: "demo", newkey:"value"}, {_id: "583ff5a747e759beb73bde33", time: 1480586663024, label: "elementB", author: "demo", newkey:"value"}]

In Python, the None value is used to remove a key. Python

# update an element
>>> damas.update({'_id': '55ae0b1ed81e88357d77d0e9', 'key1': None, 'newkey': 'value2'})
{u'newkey': u'value2', u'time': 1437469470133, u'_id': u'55ae0b1ed81e88357d77d0e9', u'author': u'demo'}

# update multiple elements
damas.update([{'_id': '583ff5a747e759beb73bde32', 'label':'elementA'}, {'_id': '583ff5a747e759beb73bde33', 'label':'elementB'}])
[{u'_id': u'583ff5a747e759beb73bde32', u'time': 1480586663024, u'label': u'elementA', u'author': u'demo'}, {u'_id': u'583ff5a747e759beb73bde33', u'time': 1480586663024, u'label': u'elementB', u'author': u'demo'}]

# update multiple elements, condensed
damas.update({'_id': ['583ff5a747e759beb73bde32','583ff5a747e759beb73bde33'], 'newkey':'value'})
[{u'newkey': u'value', u'_id': u'583ff5a747e759beb73bde32', u'time': 1480586663024, u'label': u'elementA', u'author': u'demo'}, {u'newkey': u'value', u'_id': u'583ff5a747e759beb73bde33', u'time': 1480586663024, u'label': u'elementB', u'author': u'demo'}]

upsert

Create or update element(s) when identifier(s) are specified and found

upsert ( elements [, callback] )

parameters

  • elements {object | array} : object(s) to insert and/or update in the database
  • callback {function} (js only, optional) : if specified, the request is asynchronous

return values

  • returns an object or an array of objects (depending on the input) on success
  • returns null (Javascript) or None (Python) on failure

examples

# Python
# create a new node with a specified id
>>> project.upsert({"_id":"55ae0b1ed81e88357d77d0e9"})
{u'time': 1437469470133, u'_id': u'55ae0b1ed81e88357d77d0e9', u'author': u'demo'}

# update an existing node
>>> project.upsert([{"_id":"55ae0b1ed81e88357d77d0e9", "additional_key":"value"}])
{u'additional_key': u'value', u'time': 1437469470133, u'_id': u'55ae0b1ed81e88357d77d0e9', u'author': u'demo'}

# create a new node without specifying id
>>> project.upsert({"_id":"null"})
{u'_id': u'57ae0b1ed81e88357d77d0b4', u'time': 1480586620449, u'author': u'demo'}

# create and update
>>> project.upsert([{"_id":["55ae0b1ed81e88357d77d0e9", "null"], "additional_key":"hello"}])
[{u'additional_key': u'hello', u'time': 1437469470133, u'_id': u'55ae0b1ed81e88357d77d0e9', u'author': u'demo'}, {u'additional_key': u'hello', u'_id': u'56ae0b1ed81e88357d77d0f9', u'time': 1480586620449, u'author': u'demo'}]

Javascript

// create a new node
damas.upsert({_id: "null"});
>> Object { author: "damas", time: 1480588505449, _id: "583ffcd947e759beb73bde39" }

// update a node
damas.upsert({_id: "583ffcd947e759beb73bde39", a : "test"})
>> Object { a : "test", author: "damas", time: 1480588505449, _id: "583ffcd947e759beb73bde39" }

//create and update
damas.upsert({_id: ["583ffcd947e759beb73bde39", "null"], a : "damas"})
>> Object { a : "damas", author: "damas", time: 1480588505449, _id: "583ffcd947e759beb73bde39" }
>> Object { a : "damas", author: "damas", time: 1480588505449, _id: "583ffcd947e759beb73ple85" }

// create a new node using an asynchronous call
damas.upsert({key1: "value2"}, function (node) {
    // asynchronous mode
    console.log(node.time);
});

delete

Permanently remove element(s) from the database

delete ( identifiers [, callback] )

parameters

  • identifiers {string | array} : identifier(s) to delete from the database
  • callback {function} (js only, optional) : if specified, the request is asynchronous

return values

  • returns true on success, false otherwise
  • returns a boolean or an array of booleans (depending on the input)

the returned array is sorted as the input array

examples

Javascript

damas.delete("55ae0b1ed81e88357d77d0e9");
▸ true

signin

signIn ( username, password [, expiresIn, callback] )

parameters

  • username {string} : the username or email string
  • password {string} : the user secret password string
  • expiresIn {string} (optional) : time before a new connection is required (learn more)
  • callback {function} (js only, optional) : if specified, the request is asynchronous

return values

  • returns an object containing an authentication token on success, false otherwise

Sign in using the server embedded authentication system

Javascript

// classic authentication
damas.signIn("axel", "password");
>> Object { _id: "55ae0b1ed81e88357d77d0e9", address: "::ffff:127.0.0.1", class: "user", lastActivity: 1561237341643, lastlogin: 1561237340643, username: "axel" }

// signIn using an asynchronous call
// res is the user node (the result of authentication)
let callback = function (res) {
    console.log(res);
}
damas.signIn("axel", "password", callback());
>> Object { _id: "55ae0b1ed81e88357d77d0e9", address: "::ffff:127.0.0.1", class: "user", lastActivity: 1561237341643, lastlogin: 1561237340643, username: "axel" }

// you can set token lifespan
damas.signIn("axel", "password", "1d" ,callback());
>> Object { _id: "55ae0b1ed81e88357d77d0e9", address: "::ffff:127.0.0.1", class: "user", lastActivity: 1561237341643, lastlogin: 1561237340643, username: "axel" }

signout

signOut ( [callback] )

parameters

  • callback {function} (js only, optional) : if specified, the request is asynchronous

return values

  • returns true on success, false otherwise

verify

Ask the server for the authentication status and user

verify ( [callback] )

parameters

  • callback {function} (js only, optional) : if specified, the request is asynchronous

return values

  • returns the authenticated user element on success, false otherwise

search

Find elements wearing the specified key(s) using a query string.

search ( query [, callback] )

parameters

  • query {string} : search query string
  • callback {function} (js only, optional) : if specified, the request is asynchronous

return values

  • returns an array of matching identifiers or null if no match was found

  • format: "keyname1:value keyname2:value"

  • operators list: :, <, <=, >, >=

In case of : operator, you can use a regular expression as value.

examples

Python

# list every png file containing "floor" in the file name, case insensitive
>>> damas.search("file:/floor.*png$/i")

# List every file containing the word rabbit and wearing the `type` key = `char`
>>> damas.search('file:/rabbit/ type:char')

Javascript

// list every png file containing "floor" in the file name, case insensitive
damas.search("file:/floor.*png$/i");

// List every file containing the word rabbit and wearing the `type` key = `char`
damas.search('file:/rabbit/ type:char');

search_one

Search nodes, returning the first matching occurrence as a node object (not as index as in search). The search string format is the same as for the search method.

search_one ( query [, callback] )

parameters

  • query {string}
  • callback (js only, optional) if specified, the request is asynchronous

return values

  • returns the first matching identifier or null if no match was found

search_mongo

We expose the MongoDB find and cursor methods here in order to provide a powerful search with many options. It is only available when the server runs a MongoDB database to store the data.

search_mongo ( query [, sort, limit, skip, callback] )

parameters

return values

  • returns arrays of matching indexes

In order to use regular expressions, and because the JSON format only accept strings and has no type for regular expressions, we use strings with the prefix REGEX_ to indicate to the server that it must convert it to a RegExp object before executing the Mongo query. To add options to regular expressions, prefer the syntax RX_expression_RXoptions.

For example: the /.*/ regular expression.

is written "REGEX_.*" as string format in the JSON messages

Example with case-insensitive option: /.*/i

is written "RX_.*_RXi" as string format in the JSON messages

Javascript

// Search every file paths starting with /DATABASE/02_ASSETS using a regular expression   
damas.search_mongo({_id:"REGEX_/DATABASE/02_ASSETS.*"},{}, 0,0)

Python

# get the 10 most recent files (having the higher `time` key on nodes)
>> damas.search_mongo({"file":{"$exists": True}}, {"time":-1}, 10, 0)
[u'56701f266899505c6d82ffc4', u'56701e2583cfa5c16c1a2f78', u'56701b791da266d26bc11126', u'56701cf1570a32f16be5bb60', u'56701923a26510e96969e277', u'5670305b40c1a51070f3356f', u'567026b69b2d56016f663242', u'56702671a86238e76e08f600', u'56701fb92fdef89f6dcb8c6c', u'55b36829cc6742a30da59b98']

Javascript

// get the 200 last indexed files, sorted by descending time key, and display a table in html format as output
damas.search_mongo({'time': {$exists:true}}, {"time":-1},200,0, function(res){
    damas.read(res, function(assets){
        var out = document.querySelector('#contents');
        var str = '<table><tr><th>author</th><th>file</th><th>time &xutri;</th><th>comment</th></tr>';
        for(var i=0; i<assets.length; i++)
        {
            str +=  '<tr>';
            str +=  '<td>'+assets[i].author+'</td>';
            str +=  '<td>'+assets[i].file+'</td>';
            str +=  '<td>'+new Date(parseInt(assets[i].time))+'</td>';
            str +=  '<td style="white-space:normal">'+assets[i].comment+'</td>';
            str +=  '</tr>';
        }
        str += '</table>';
        out.innerHTML = str;
    });
});

graph

Recursively get all source nodes and edges connected to the specified node

graph ( identifiers, [, callback] )

parameters

  • identifiers {string | array} : identifier(s) string(s) to read
  • callback {function} (js only, optional) : if specified, the request is asynchronous

return values

  • @returns {Array} array of element indexes

examples

Javascript

// retrieve graph as an array containing the elements (nodes and edges)
damas.graph("55687e68e040af7047ee1a53");
Add a custom footer
Add a custom sidebar

Clone this wiki locally