An Erlang/OTP library for interfacing with Google Cloud Storage. Named after the smoke monster of Japanese folklore, for no particular reason. This library provides basic CRUD operations on buckets and objects, including uploading and downloading of object media.

• Erlang/OTP R18 or R19
  • R20 has backward incompatible API changes
• rebar3 3.0.0 or higher

$ rebar3 compile
$ export GOOGLE_APPLICATION_CREDENTIALS=~/.your_gcp_credentials.json
$ rebar3 ct

Note that the test suite expects to find an environment variable named GOOGLE_APPLICATION_CREDENTIALS which specifies the path to your Google Cloud Platform service credentials. This JSON formatted file is created via the Cloud Console, as described in the setup section below. This file contains your private key, so be sure to store this file with permissions that prevent exposure to third parties.

To include enenra as a dependency in your release, add it to the list of dependencies in your rebar.config file, like so:

{deps, [
    {enenra, {git, "https://github.com/nlfiedler/enenra", {tag, "0.3.0"}}}
]}.

To have the enenra application started automatically, be sure to include enenra in the applications list of your application configuration before building a release. You may also want to add jiffy to the list of included_applications, and hackney to the list of applications.

Below is a simple example in which a bucket is created and a file is uploaded to that bucket. This also demonstrates loading the credentials from a file and computing an MD5 checksum of a file to be uploaded to a bucket.

1> rr("include/enenra.hrl").
[access_token,bucket,credentials,object]
2> application:ensure_all_started(enenra).
{ok,[idna,mimerl,certifi,ssl_verify_fun,metrics,hackney,enenra]}
3> Credentials = os:getenv("GOOGLE_APPLICATION_CREDENTIALS").
"/Users/nfiedler/.gcloud/testing.json"
4> {ok, Creds} = enenra:load_credentials(Credentials).
{ok,#credentials{type = <<"service_account">>,
                 project_id = <<"a-project">>,
                 private_key_id = <<"a-private-key-id">>,
                 private_key = <<"a-private-key">>,
                 client_email = <<"an-email-address">>,
                 client_id = <<"a-client-id">>}}
5> enenra:insert_bucket(#bucket{
    name = <<"0136d00f-a942-11e6-8f9a-3c07547e18a6-enenra-1234">>,
    location = <<"US-WEST1">>,
    storageClass = <<"STANDARD">>}, Creds).
{ok,#bucket{id = <<"a-bucket-id">>,
            projectNumber = <<"a-project-number">>,
            name = <<"0136d00f-a942-11e6-8f9a-3c07547e18a6-enenra-1234">>,
            timeCreated = <<"2016-11-18T22:25:54.239Z">>,
            updated = <<"2016-11-18T22:25:54.239Z">>,
            location = <<"US-WEST1">>,
            storageClass = <<"STANDARD">>}}
6> {ok, Md5} = enenra:compute_md5("test/enenra_SUITE_data/IMG_5745.JPG").
{ok,<<"kq56YDAH2p4mzAqrQw84kQ==">>}
7> enenra:upload_file("test/enenra_SUITE_data/IMG_5745.JPG", #object{
    name = <<"my_image">>,
    bucket = <<"0136d00f-a942-11e6-8f9a-3c07547e18a6-enenra-1234">>,
    contentType = <<"image/jpeg">>,
    md5Hash = Md5,
    size = 107302}, Creds).
{ok,#object{id = <<"a-bucket-id/name/an-object-id">>,
            name = <<"my_image">>,
            bucket = <<"0136d00f-a942-11e6-8f9a-3c07547e18a6-enenra-1234">>,
            contentType = <<"image/jpeg">>,
            timeCreated = <<"2016-11-18T22:28:01.232Z">>,
            updated = <<"2016-11-18T22:28:01.232Z">>,
            storageClass = <<"STANDARD">>,
            size = <<"107302">>,
            md5Hash = <<"kq56YDAH2p4mzAqrQw84kQ==">>}}

1. Visit https://console.cloud.google.com/ and log in with your account.
2. Select an existing project or create a new one.
3. Note the Project ID as that will be necessary when connecting via enenra.
4. Using the menu, select API Manager, then Credentials.
5. On the Credentials page, select the Create credentials drop-down, then select Service account key.
6. From the Service account drop-down, select an existing service account or create a new one.
7. For Key type, select the JSON key option, then select Create. The file automatically downloads to your computer.
8. Put the *.json file you just downloaded in a directory of your choosing. This directory must be private, but accessible to your application.

The credentials file must be readable by the application, but should not be readable by casual users. Likewise, it is possible, if the server process crashes, that the private keys will end up in the log file (they are an argument to the API, after all). As such, the log files should be protected from unintended exposure to third parties.

Docker can be used to build and test the code without affecting your development environment, which may have a different version of Erlang/OTP installed. The use of docker-compose, as shown in the example below, is optional, but it makes the process very simple. Note that GOOGLE_APPLICATION_CREDENTIALS should have a path reachable within the container, such as /src/creds.json

$ cd docker
$ docker-compose build
$ docker-compose run enenra
$ rebar3 clean
$ rebar3 compile
$ rebar3 ct

BSD 3-Clause, see the LICENSE file.