Skip to main content

Datadots API

Project description

Datadots API for Python

PyPI Build Status

STATUS: What you currently see here is WIP, do not use in production!

This is the Datadots API for Python, allowing you to access and manipulate dots against any Dotmesh cluster. Note that only Python 3 is supported.

Install

You must have Python 3 installed, this is the only version supported. Then do:

$ pip3 install datadots-api==0.1.4

Use

Some usage examples below.

Create a dot

from dotmesh.client import DotmeshClient
d = DotmeshClient(cluster_url=cluster_url, username=user, api_key=api_key)
dot = dmclient.createDot(dotname="test")

Create a dot in a namespace

from dotmesh.client import DotmeshClient
d = DotmeshClient(cluster_url=cluster_url, username=user, api_key=api_key)
dot = dmclient.createDot(dotname="test", ns="mynamespace")

Look up a dot by name

from dotmesh.client import DotmeshClient
d = DotmeshClient(cluster_url=cluster_url, username=user, api_key=api_key)
print(d.getDot(dotname="test"))

Commit to a branch and show the log

from dotmesh.client import DotmeshClient
d = DotmeshClient(cluster_url=cluster_url, username=user, api_key=api_key)
dot = dmclient.createDot(dotname="test")
branch = dot.getBranch("master")
branch.commit("just a test commit")
print(branch.log())

Commit to a branch with some extra metadata

from dotmesh.client import DotmeshClient
d = DotmeshClient(cluster_url=cluster_url, username=user, api_key=api_key)
dot = dmclient.createDot(dotname="test")
branch = dot.getBranch("master")
branch.commit("a commit with some extra metadata", {"fruit": "apples", "color": "red"})
print(branch.log())

Create new branch based on latest master

from dotmesh.client import DotmeshClient
d = DotmeshClient(cluster_url=cluster_url, username=user, api_key=api_key)
dot = dmclient.createDot(dotname="test")
branch = dot.getBranch("master")
branch.commit("just a test commit")
mybranch = branch.createBranch("mybranch", "master")
mybranch.commit("the first commit in my branch")
print(mybranch.log())

End-to-end example

You can see this API in action, for a local setup with Docker running and after you've created a cluster using dm cluster init, by running the following example:

$ DOTMESH_APIKEY=$(cat ~/.dotmesh/config | jq -r .Remotes.local.ApiKey)
$ DOTMESH_APIKEY=$DOTMESH_APIKEY python3 example.py
== Create dot:
--> {"jsonrpc": "2.0", "method": "DotmeshRPC.Create", "params": {"Namespace": "admin", "Name": "test"}, "id": 1}
<-- {"jsonrpc":"2.0","result":true,"id":1} (200 OK)
--> {"jsonrpc": "2.0", "method": "DotmeshRPC.Lookup", "params": {"Namespace": "admin", "Name": "test", "Branch": ""}, "id": 2}
<-- {"jsonrpc":"2.0","result":"32965cef-4310-4291-6f36-c86106655d10","id":2} (200 OK)

ID=32965cef-4310-4291-6f36-c86106655d10, name=test

== Get master branch:
--> {"jsonrpc": "2.0", "method": "DotmeshRPC.Lookup", "params": {"Namespace": "admin", "Name": "test", "Branch": ""}, "id": 3}
<-- {"jsonrpc":"2.0","result":"32965cef-4310-4291-6f36-c86106655d10","id":3} (200 OK)

master

== Do some commit and show log:
--> {"jsonrpc": "2.0", "method": "DotmeshRPC.Commit", "params": {"Namespace": "admin", "Name": "test", "Branch": "", "Message": "just a test commit"}, "id": 4}
<-- {"jsonrpc":"2.0","result":true,"id":4} (200 OK)
--> {"jsonrpc": "2.0", "method": "DotmeshRPC.Commit", "params": {"Namespace": "admin", "Name": "test", "Branch": "", "Message": "and another commit, who would have thought"}, "id": 5}
<-- {"jsonrpc":"2.0","result":true,"id":5} (200 OK)
--> {"jsonrpc": "2.0", "method": "DotmeshRPC.Commits", "params": {"Namespace": "admin", "Name": "test", "Branch": ""}, "id": 6}
<-- {"jsonrpc":"2.0","result":[{"Id":"1007db99-11cb-4bd7-747e-1d121bb9b11a","Metadata":{"author":"admin","message":"just a test commit","timestamp":"1521214312285694689"}},{"Id":"67bb6184-430b-46d7-4337-b858ee85eb1c","Metadata":{"author":"admin","message":"and another commit, who would have thought","timestamp":"1521214312418429042"}}],"id":6} (200 OK)


1521214312285694689: just a test commit
1521214312418429042: and another commit, who would have thought

== Create new branch, commit and show log:
--> {"jsonrpc": "2.0", "method": "DotmeshRPC.Commits", "params": {"Namespace": "admin", "Name": "test", "Branch": ""}, "id": 7}
<-- {"jsonrpc":"2.0","result":[{"Id":"1007db99-11cb-4bd7-747e-1d121bb9b11a","Metadata":{"author":"admin","message":"just a test commit","timestamp":"1521214312285694689"}},{"Id":"67bb6184-430b-46d7-4337-b858ee85eb1c","Metadata":{"author":"admin","message":"and another commit, who would have thought","timestamp":"1521214312418429042"}}],"id":7} (200 OK)
--> {"jsonrpc": "2.0", "method": "DotmeshRPC.Branch", "params": {"Namespace": "admin", "Name": "test", "SourceBranch": "master", "NewBranchName": "mybranch", "SourceCommitId": "67bb6184-430b-46d7-4337-b858ee85eb1c"}, "id": 8}
<-- {"jsonrpc":"2.0","result":true,"id":8} (200 OK)
--> {"jsonrpc": "2.0", "method": "DotmeshRPC.Commit", "params": {"Namespace": "admin", "Name": "test", "Branch": "mybranch", "Message": "the first commit in my branch"}, "id": 9}
<-- {"jsonrpc":"2.0","result":true,"id":9} (200 OK)
--> {"jsonrpc": "2.0", "method": "DotmeshRPC.Commits", "params": {"Namespace": "admin", "Name": "test", "Branch": "mybranch"}, "id": 10}
<-- {"jsonrpc":"2.0","result":[{"Id":"565b04ab-abae-476a-75bd-203bce2a56b2","Metadata":{"author":"admin","message":"the first commit in my branch","timestamp":"1521214312967840418"}}],"id":10} (200 OK)


1521214312967840418: the first commit in my branch

Note that the default behavior of the example above is to leave the dot in place, that is, if you do a dm dot show test you should see both the master branch and mybranch with the commits. If you want to automatically clean up, execute it as follows: python3 example.py cleanup:

$ dm list
Current remote: local (use 'dm remote -v' to list and 'dm remote switch' to switch)

  DOT   BRANCH  SERVER            CONTAINERS  SIZE      COMMITS  DIRTY
* test  master  995e5c821c38e7f5              1.00 kiB  1        -

$ dm dot show test
Dot admin/test:
Master branch ID: 7f4a4d67-5bc7-47aa-7829-31d5a02ff70f
Dot is current.
Dot size: 1.00 kiB (all clean)
Branches:
* master
  mybranch

$ dm checkout mybranch && dm log
commit 565b04ab-abae-476a-75bd-203bce2a56b2
Author: admin
Date: 1521214312967840418

    the first commit in my branch

Test

Locally, you can test the Dotmesh JSON-RPC API like shown in the following. It assumes you've got http installed (or alternatively you can use curl) as well as that Docker is running (note: tested on v18.03):

$ dm cluster init
$ DOTMESH_CLUSTERURL=http://localhost:6969/rpc
$ DOTMESH_APIKEY=$(cat ~/.dotmesh/config | jq -r .Remotes.local.ApiKey)
$ http -a admin:$DOTMESH_APIKEY POST $DOTMESH_CLUSTERURL < test/ping.json
HTTP/1.1 200 OK
Content-Length: 57
Content-Type: application/json; charset=utf-8
Date: Fri, 16 Mar 2018 07:06:44 GMT
X-Content-Type-Options: nosniff

{
    "id": 6129484611666146000,
    "jsonrpc": "2.0",
    "result": true
}

Release

To release this project to pypi, check the version (git tag and find the highest one, or check setup.py) then run bump2version --new-version <new-version> <major|minor|patch> && git push --tags

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

datadots-api-0.2.4.tar.gz (9.9 kB view hashes)

Uploaded Source

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page