v0.38.3
⚠️ gcloud
has been deprecated
$ npm uninstall --save gcloud
- require('gcloud')
🔀 It has been renamed to google-cloud
$ npm install --save google-cloud
+ require('google-cloud')
🎉 Now you can install only the APIs you need
$ npm install --save @google-cloud/storage
var gcs = require('@google-cloud/storage')();
No more wasting time installing dependencies from the other modules that your project doesn't require.
How will this all work?
All of the code for the modules will live under this single repository. Issues and PRs for any of the APIs can be filed here.
google-cloud
will work exactly as gcloud
did. It wraps all of the other modules and can share your config data between them. Your code should continue to work exactly the same. You will continue to receive new features and bug fixes under the normal semver release pattern. The first version of google-cloud
picks up where gcloud
left off, at 0.38
.
The service modules, e.g. @google-cloud/bigquery
, have all been published at 0.1
, and will likely reach 1.0
status shortly.
How about the docs?
Our documentation website has expanded to support showing only the docs for the module you are using. If you are just using the bundled package, google-cloud
, the docs should look and behave exactly as you're used to.
If you are using a service module, you can access its docs from the title bar on the top of the page:
0.38.3
Release Notes
Hello, Natural Language API!
Using the all-in-one module
$ npm install --save google-cloud
var gcloud = require('google-cloud');
var language = gcloud.language;
Using the Natural Language API module
$ npm install --save @google-cloud/language
var language = require('@google-cloud/language');
Preview
var languageClient = language({
projectId: 'grape-spaceship-123',
keyFilename: '/path/to/keyfile.json'
});
// Get the entities from a sentence.
languageClient.detectEntities('Stephen of Michigan!', function(err, entities) {
// entities = {
// people: ['Stephen'],
// places: ['Michigan']
// }
});
// Create a document if you plan to run multiple detections.
var document = languageClient.document('Contributions welcome!');
// Analyze the sentiment of the document.
document.detectSentiment(function(err, sentiment) {
// sentiment = 100 // Large numbers represent more positive sentiments.
});
// Parse the syntax of the document.
document.annotate(function(err, annotations) {
// annotations = {
// language: 'en',
// sentiment: 100,
// entities: {},
// sentences: ['Contributions welcome!'],
// tokens: [
// {
// text: 'Contributions',
// partOfSpeech: 'Noun (common and proper)',
// partOfSpeechTag: 'NOUN'
// },
// {
// text: 'welcome',
// partOfSpeech: 'Verb (all tenses and modes)',
// partOfSpeechTag: 'VERB'
// },
// {
// text: '!',
// partOfSpeech: 'Punctuation',
// partOfSpeechTag: 'PUNCT'
// }
// ]
// }
});
Features
- Bigtable (#1451, #1456): Specify column families when creating a table.
- Compute (#1442): Allow using a disk from your own project when creating a VM.
- Compute (#1073, #1435): Subnetwork resources can now be listed and created. (Thanks, @ashleyschuett!)
- Compute (#1073, #1441): List and interact with MachineType resources. (Thanks again, @ashleyschuett!)
- Core (#1443): Allow streaming large responses from gRPC APIs (Bigtable, Datastore, Language, Logging, and Pub/Sub).
Fixes
- Bigtable (#1458, #1459): Condition filters don't require a fail condition.
- Datastore (#1453, #1454): Allow saving empty objects.
Thank you!
There are bound to be some kinks for us to work out from all of these changes. If you catch anything, please open an issue and we'll be right on it.