This library is end-of-life and no longer supported.
This repository will not be updated. The repository will be kept available in read-only mode. If you are interested in continuing development efforts please see the list of forks for alternatives (e.g. #465).
For FAQs and additional information please refer to the Cloudant blog.
Applications use Cloudant Sync to store, index and query local JSON data on a device and to synchronise data between many devices. Synchronisation is under the control of the application, rather than being controlled by the underlying system. Conflicts are also easy to manage and resolve, either on the local device or in the remote database.
Cloudant Sync is an Apache CouchDB™ replication-protocol-compatible datastore for devices that don't want or need to run a full CouchDB instance. It's built by Cloudant, building on the work of many others, and is available under the Apache 2.0 licence.
The API is quite different from CouchDB's; we retain the MVCC data model but not the HTTP-centric API.
This library is for iOS, an Android version is also available.
If you have questions, please join our mailing list and drop us a line.
CDTDatastore is available through CocoaPods, to install it add the following line to your Podfile:
pod "CDTDatastore"
Note: We only support building on the latest stable release of Xcode
CDTDatastore is useable from Swift out of the box with a few small quirks. Install as per the instructions above, and import CloudantSync.h into your bridging header.
The Overview section below has examples in both Objective-C and Swift.
There is an example project in the Project
folder, for iOS 8.0. To get
this up and running independently of the main codebase, a Podfile is
included:
$ cd Project
$ pod install
$ open Project.xcworkspace
In order to run the sample project, edit the URL defined in
CDTViewController.m
: in replicatorURL
. Change this to the
credentials for your own account and database, ensuring you have
_reader
, _writer
, and _replicator
permissions.
Once running you will be able to edit "to-do" items in the app and in your Cloudant database and replicate these changes in both directions.
See CONTRIBUTING.
CDTDatastore gets regularly tested on the following platforms:
Once the libraries are added to a project, the basics of adding and reading a document are:
#import <CloudantSync.h>
// Create a CDTDatastoreManager using application internal storage path
NSError *outError = nil;
NSFileManager *fileManager= [NSFileManager defaultManager];
NSURL *documentsDir = [[fileManager URLsForDirectory:NSDocumentDirectory
inDomains:NSUserDomainMask] lastObject];
NSURL *storeURL = [documentsDir URLByAppendingPathComponent:@"cloudant-sync-datastore"];
NSString *path = [storeURL path];
CDTDatastoreManager *manager =
[[CDTDatastoreManager alloc] initWithDirectory:path
error:&outError];
CDTDatastore *datastore = [manager datastoreNamed:@"my_datastore"
error:&outError];
// Create a document
CDTDocumentRevision *rev = [CDTDocumentRevision revisionWithDocId:@"doc1"];
// Use [CDTDocumentRevision revision] to get an ID generated for you on saving
rev.body = [@{
@"description": @"Buy milk",
@"completed": @NO,
@"type": @"com.cloudant.sync.example.task"
} mutableCopy];
// Add an attachment -- binary data like a JPEG
CDTUnsavedFileAttachment *att1 = [[CDTUnsavedFileAttachment alloc]
initWithPath:@"/path/to/image.jpg"
name:@"cute_cat.jpg"
type:@"image/jpeg"];
rev.attachments[att1.name] = att;
// Save the document to the database
CDTDocumentRevision *revision = [datastore createDocumentFromRevision:rev
error:&error];
// `revision` will be `nil` on failures.
// Read a document
NSString *docId = revision.docId;
CDTDocumentRevision *retrieved = [datastore getDocumentWithId:docId
error:&error];
If you are using Swift, install the library as per the instructions above
and add the use_frameworks!
instruction to your target. A bridging header is
required if the use_frameworks!
instruction is not used.
If you are using a bridging header, include the CloudantSync.h
and you
should be good to go:
#import <CDTDatastore/CloudantSync.h>
To add, and read documents in Swift, the basics are:
import CDTDatastore
do {
let fileManager = FileManager.default
let documentsDir = fileManager.urls(for: .documentDirectory, in: .userDomainMask).last!
let storeURL = documentsDir.appendingPathComponent("cloudant-sync-datastore")
let path = storeURL.path
let manager = try CDTDatastoreManager(directory: path)
let datastore = try manager.datastoreNamed("my_datastore")
// Create a document
let rev = CDTDocumentRevision(docId: "doc1")
rev.body = ["description":"Buy Milk",
"completed": false,
"type":"com.cloudant.sync.example.task"]
// Add an attachment - binary data like a JPEG
let att1 = CDTUnsavedFileAttachment(path: "/path/to/image/jpg",
name: "cute_cat.jpg",
type: "image/jpeg")!
rev.attachments[att1.name] = att1
let revision = try datastore.createDocument(from: rev)
// Read a document
let docId = revision.docId
let retrieved = try datastore.getDocumentWithId(docId!)
} catch {
print("Encountered an error: \(error)")
}
Read more in the CRUD document.
You can subscribe for notifications of changes in the database, which is described in the events documentation. It's still a bit raw right now:
Replication is used to synchronise data between the local datastore and a remote database, either a CouchDB instance or a Cloudant database. Many datastores can replicate with the same remote database, meaning that cross-device synchronisation is achieved by setting up replications from each device to the remote database.
Replication is simple to get started in the common cases:
#import <CloudantSync.h>
// Create and start the replicator -- -start is essential!
CDTReplicatorFactory *replicatorFactory =
[[CDTReplicatorFactory alloc] initWithDatastoreManager:manager];
NSString *s = @"https://apikey:apipassword@username.cloudant.com/my_database";
NSURL *remoteDatabaseURL = [NSURL URLWithString:s];
CDTDatastore *datastore = [manager datastoreNamed:@"my_datastore"];
// Replicate from the local to remote database
CDTPushReplication *pushReplication = [CDTPushReplication replicationWithSource:datastore
target:remoteDatabaseURL];
NSError *error;
CDTReplicator *replicator = [replicatorFactory oneWay:pushReplication error:&error];
//check error
// Start the replicator
[replicator start];
import CDTDatastore
do {
let datastore = try manager.datastoreNamed("my_datastore");
// Replicate from the local to remote database
let remote = URL(string: "https://apikey:apipassword@username.cloudant.com/my_database")!
datastore.push(to: remote) { error in
if let error = error {
print("Encountered an error: \(error)")
} else {
print("Replication complete")
}
}
} catch {
print("Encountered an error: \(error)")
}
Read more in the replication docs including how to use IAM authentication instead of username/password.
Once you have thousands of documents in a database, it's important to have efficient ways of finding them. We've added an easy-to-use querying API. Once the appropriate indexes are set up, querying is as follows:
NSDictionary *query = @{
@"name": @"John", // name equals John
@"age": @{ @"$gt" : @25} // age greater than 25
};
CDTQResultSet *result = [datastore find:query];
[result enumerateObjectsUsingBlock:^(CDTDocumentRevision *rev, NSUInteger idx, BOOL *stop) {
// do something
}];
let query = [
"name" : "John", // name equals John
"age" : ["$gt" : 25] // age greater than 25
]
let result = datastore.find(query)
result?.enumerateObjects { rev, idx, stop in
// do something
}
As of version 0.16.0 the indexing and querying code has been re-written and has more features than the previous implementation. For details about migrating to a 0.16.0+ indexing and query version from a previous version see Index and Querying Migration.
An obvious repercussion of being able to replicate documents about the place is that sometimes you might edit them in more than one place at the same time. When the databases containing these concurrent edits replicate, there needs to be some way to bring these divergent documents back together. Cloudant's MVCC data-model is used to do this.
A document is really a tree of the document and its history. This is neat because it allows us to store multiple versions of a document. In the main, there's a single, linear tree -- just a single branch -- running from the creation of the document to the current revision. It's possible, however, to create further branches in the tree. At this point your document is conflicted and needs some surgery to resolve the conflicts and bring it back to full health.
All requirements are included in the source code or pulled in as dependencies
via pod install
.
See CONTRIBUTORS.
See CONTRIBUTING.
See LICENSE
Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.