Welcom to the DLDB SDK beta for iOS
- add
pod 'DLDB' , '0.9.8'into your Podfile, - run
Pod install - Add the following import directive
import DLDB_C- download
DLDB.xcframeworkand drag/drop the resulting folder into theFrameworkssection of Project Explorer in Xcode
- add
pod 'DLDB' , '0.9.8'into your Podfile, - run
Pod install - Add the following import directive
#import <DLDB_C/DLDB_C.h>- download
DLDB.xcframeworkand drag the resulting folder into theFrameworkssection of Project Explorer in Xcode
// create a unique instance
let myDLDB: DLDB_C = DLDB_C()
// start DLDB SDK after user consent
myDLDB.start(documentsDirectory.path,
dldbApiKey: "11111111-1111-1111-1111-111111111111",
registerCallback: nil,
dictionary: "{\"button\" : \"t\",\"batteryLevel\" : \"i\"}")
// once per day
myDLDB.heartbeat();
// on a regular basis, when app idle ?
myDLDB.runQueriesIfAny();
// wherever useful
myDLDB.addEvents("{\"button\":\"log in\", \"batteryLevel\" : 55 }");
// location and event
CLLocation loc ...
myDLDB.addEventsWithLocation(
loc
eventsAsJson : '{"batteryLevel" : 5 }');
// location only
CLLocation loc ...
myDLDB.addLocation(
loc,
eventsAsJson: "");Initialize the DLDB SDK with the dldbApiKey and events dictionary.
The dldbApiKey is required and can be obtained from the DLDB dashboard.
| parameter | type | description |
|---|---|---|
| path | string | where to store local db on the device |
| dldbApiKey | string | api key |
| registerCallback | nil | ignored in beta |
| dictionary | string | events dictionary with names and types |
The events dictionary is a json string containing one object per event. Each object defines the event name as a key and the values as the type of events : 't' for text, 'i' for numeric value.
A dictionary { "button" : "t", "batteryLevel": "i" } defines 2 events : button with text value, and batteryLevel with numeric value.
The dictionary is meant to be a constant and can be changed only with new versions of your app.
The event names defined in this dictionary will the only events accepted by the function [addEvents]
Example:
private let myDLDB: DLDB_C = DLDB_C()
...
myDLDB.start(documentsDirectory.path,
dldbApiKey: "1688e135-f5b1-4659-b84a-96ff15c07f57",
registerCallback: nil,
dictionary: "{\"button\" : \"t\",\"batteryLevel\" : \"i\"}")to be called on a low frequency basis compatible with the app usage frequency, in order to provide accurate estimates of query response time to the dashboard.
Example:
myDLDB.heartbeat()to be called on a regular basis compatible with the app usage frequency, in order to provide fast query responses to the dashboard. The more often it is called the faster the queries will be replied on dashboard
Example:
myDLDB.runQueriesIfAny();2 overloads : 2 parameters CLLocation and NSString, or 6 parameters
| parameter | type | description |
|---|---|---|
| location | CLLocation | obtained from CLLocationManager |
| longitudeInDegrees | double | in decimal degrees |
| latitudeInDegrees | double | in decimal degrees |
| horizontalAccuracyInMeters | double | horizontal accuracy in meters aka the radius of the area, defaults to 100m |
| epochUTCInSeconds | Int32_t | UTC epoch in seconds |
| offsetFromUTCInSeconds | Int32_t | offset from UTC in seconds |
| eventsAsJson | NSString | The name and values of the events |
longitudeInDegrees and latitudeInDegrees indicate where the events are happening, and are taken into account if both are present and valid values.
Events eventsAsJson can be any metric or KPI relevant for better understanding of end-user behaviour. All events provided through this call will be attached to the same second. If you do not have access to any location information when this function is called, provide nil as the first parameter.
Event names must be present in the dictionary provided when calling init. Events with unknown names will be discarded.
The events will be attached to exactly the same instant, which will be either the current second provided by epochUTCInSeconds, or the timestamp provided by CLLocation
Example:
// location and event
myDLDB.addEvents(0.0,
latitudeInDegrees: 45.0,
horizontalAccuracyInMeters: 25,
epochUTCInSeconds: 1644229847,
offsetFromUTCInSeconds: 3600,
eventsAsJson: "{\"button\":\"log in\", \"batteryLevel\" : 55 }"
)
// CLLocation and event
CLLocation loc = ...
myDLDB.addEvents(loc,
eventsAsJson: "{\"button\":\"log in\", \"batteryLevel\" : 55 }"
)
// only event
myDLDB.addEvents(nil,
eventsAsJson: "{\"button\":\"log in\", \"batteryLevel\" : 55 }"
)| parameter | type | description |
|---|---|---|
| maxEntries | number | the max number of entries to include in the log |
return the log of maxEntries most recent queries ran on the device.
The log is a json array (most recent last) of json objects, each object representing a query ran on the device
| Property | Description |
|---|---|
| id | uuid of query |
| type | type of query |
| fetched | timestamp in ms when the query was fetched from server |
| answered | timestamp in ms when the query was answered on the device |
| finished | timestamp in ms when the query was finished on the device |
| tries | number of tries to run the query |
| json_in | parameters of the query |
| json_out | results of the query |
Example:
let latestQueriesLog = myDLDB.queriesLog(5)
print(latestQueriesLog ?? "")| parameter | type | description |
|---|---|---|
| durationInSeconds | number | how far back in the past to look for locations |
| maxEntries | number | the max number of entries to include in the log |
| resolution | number | resolution of returned locations |
return the log of at most maxEntries most recent unique locations stored on the device during the last durationInSeconds.
The log is a json array (most recent last) of h3 in hex string format, at resolution resolution.
A call to this function generates an entry into the queries log.
Example:
let latestLocationsLog = myDLDB.locationsLog(3600*24, maxEntries: 100, resolution: 7)
print(latestLocationsLog ?? "")to be called when shutting down the app.
Example:
myDLDB.close() DLDB provides behavioural analytics for mobile applications with privacy by design.
DLDB architecture relies on an SDK to be integrated into your mobile application, and a dashboard https://dashboard.dldb.io/ to build, query, analyze the behaviour of your application users.
For your application, DLDB deploys a distributed database, where each database instance is inside the mobile application scope. All the analytics queries are run by the devices and no raw data ever leaves the devices. Only the statistical KPI-s are sent anonymously to the DLDB dashboard .
From the DLDB dashboard, developers, analysts and app owners can build their own queries and analyze the results. No need to have any additional storage or analytics platform: DLDB provides an end-to-end solution.
DLDB SDK is written in C and has bindings to most common languages - works natively on iOS, Android, React-Native and Flutter. We also have Python binding and C libraries for IoT devices.
- Seamless integration of DLDB SDK into your mobile app source base, in many flavours, on all major platforms
- Define your own schema of collected events and values
- Built-in GDPR compliance on the right to be forgottent: all data belonging to your user stays on the device, so delete the data whenever requested
- Built-in GDPR compliance on the traceability of data usage: all requests processed by the DLDB SDK are traced and can be shown on demand
- No additional online storage
- Rapid scaling
If you face any problems or have any questions, please contact us
support channel in Discord: https://discord.gg/TD4f6p6nUH
email: [email protected]
web: https://dldb.io/
subscribe for product updates and news: https://dldb.io/#Beta