Cassandra walker

Rust version in here

Unix philosophy

Rule of Generation: Developers should avoid writing code by hand and instead write abstract high-level programs that generate code. This rule aims to reduce human errors and save time.

Cassandra walker is and ORM and query builder, for Cassandra,

cassandra-walker is command-line tool to generate idiomatic Golang code for Cassandra keyspaces (databases).

with pointing cassandra-walker to Cassandra cluster, cassandra-walker find tables in each keyspaces and for each tables create golang types, and idiomatic go codes.

This tools makes working with Cassandra very easy, it works great with IDEs, catch most of bugs at compile time, makes code base much more scalable and maintainable.

Demo of how you will code with the result of this tool

Create, Update, Delete for a Row:

Building query against Cassandra:

Quickstart

Install cassandra-walker with:

go get -u github.com/jozn/cassandra-walker

Then point to cassandra node to genrate code for a keyspace (ex twitter):

cassandra-walker twitter

This will produce xc folder in current directory, and puts generated golang codes in this folder.

Command Line Parameters

Use cassandra-walker -h to see parameter options.

Usage: cassandra-walker [--host HOST] [--port PORT] [--verbose] [--dir DIR] [--package PACKAGE] [--minimize] [KEYSPACES [KEYSPACES ...]]

Positional arguments:
  KEYSPACES              cassandra keyspaces to build

Options:
  --host HOST, -c HOST   cassandra cluster address (default 127.0.0.1)
  --port PORT, -p PORT   cassandra port (default 9042)
  --verbose, -v          verbosity Log
  --dir DIR, -d DIR      output of generated codes (default './')
  --package PACKAGE      package of go
  --minimize, -m         minimize docs
  --help, -h             display this help and exit

Guides

Lets see how to use this tool. We will follow twitter sample in sample directory

Assume you have this cassandra keyspace:

CREATE KEYSPACE twitter
  WITH REPLICATION = {
   'class' : 'SimpleStrategy',
   'replication_factor' : 1
 };

CREATE TABLE twitter.tweet (
	user_id bigint,
	tweet_id varchar,
	body varchar,
	create_time int,
	PRIMARY KEY (user_id,tweet_id)
);

CREATE TABLE twitter.user (
	user_id int,
	user_name varchar,
	full_name varchar,
	created_time bigint,
	PRIMARY KEY (user_id)
);

Run the following command:

cassandra-walker twitter --host 127.0.01 --port 9042

This will generates codes in xc directory.

See the result in godoc or in go files.

package xc

type Tweet struct {
	Body       string // body  regular
	CreateTime int    // create_time  regular
	TweetId    string // tweet_id  clustering
	UserId     int    // user_id  partition_key

	_exists, _deleted bool
}

/*
:= &xc.Tweet {
	Body: "",
	CreateTime: 0,
	TweetId: "",
	UserId: 0,
*/

type User struct {
	CreatedTime int    // created_time  regular
	FullName    string // full_name  regular
	UserId      int    // user_id  partition_key
	UserName    string // user_name  regular

	_exists, _deleted bool
}

/*
:= &xc.User {
	CreatedTime: 0,
	FullName: "",
	UserId: 0,
	UserName: "",
*/

Now we have in type-safe can build queries, look at this simple script:

package main

import (
	"github.com/gocql/gocql"
	"github.com/jozn/cassandra-walker/samples/twitter/xc"
)

func main() {
	// create cassandra session
	cluster := gocql.NewCluster("127.0.0.1")
	cluster.Keyspace = "twitter"
	cluster.Consistency = gocql.One
	session, _ := cluster.CreateSession()
	defer session.Close()

	// Create
	tweet1 := xc.Tweet{
		Body:       "Hello World",
		CreateTime: 1566000000,
		TweetId:    "1",
		UserId:     1,
	}

	err := tweet1.Save(session)

	// Delete one object
	tweet1.Delete(session)

	//////////////// For Selector
	tweets, err := xc.NewTweet_Selector().UserId_Eq(1).Limit(5).GetRows(session) // returns and array of tweets ( []*tweet ,err )

	tweet, err := xc.NewTweet_Selector().UserId_Eq(1).Limit(5).GetRows(session) // returns a single tweet ( *tweet ,err )

	//can use clustering columns too
	tweets, err = xc.NewTweet_Selector().UserId_Eq(1).And_TweetId_In("1", "25", "68").GetRows(session)

	//can select just some columns, it will returns *[]Tweet, with just selected columns sets
	tweets, err = xc.NewTweet_Selector().Select_UserId().Select_Body().UserId_Eq(1).And_TweetId_In("1", "25", "68").Limit(12).GetRows(session)

	//for when need to use filtering
	tweets, err = xc.NewTweet_Selector().UserId_LT_Filtering(100).Limit(10).AllowFiltering().GetRows(session)

	//////////////// For Updater
	err = xc.NewTweet_Updater().
		Body("new tweet text").UserId_Eq(1).And_TweetId_In("1", "2", "3").Update(session)

	//////////////// For Deleter
	err = xc.NewTweet_Deleter().UserId_Eq(1).And_TweetId_In("1", "2", "3").Delete(session)
	err = xc.NewTweet_Deleter().UserId_Eq(1).Delete(session)

	_ = err
	_ = tweets
	_ = tweet
}

/* log output - this is produced CQL queries to cassandra:

2018/09/18 22:35:54 CQL:  [insert into tweeter.tweet (body,create_time,tweet_id,user_id) values (?,?,?,?)  [Hello World 1566000000 1 1]]
2018/09/18 22:35:54 CQL:  [DELETE FROM tweeter.tweet WHERE  user_id = ? And tweet_id = ?  [1 1]]
2018/09/18 22:35:54 CQL:  [SELECT * FROM tweeter.tweet WHERE  user_id = ?  LIMIT 5 [1]]
2018/09/18 22:35:54 CQL:  [SELECT * FROM tweeter.tweet WHERE  user_id = ?  LIMIT 5 [1]]
2018/09/18 22:35:54 CQL:  [SELECT * FROM tweeter.tweet WHERE  user_id = ? And tweet_id IN (?,?,?)  [1 1 25 68]]
2018/09/18 22:35:54 CQL:  [SELECT user_id, body FROM tweeter.tweet WHERE  user_id = ? And tweet_id IN (?,?,?)  LIMIT 12 [1 1 25 68]]
2018/09/18 22:35:54 CQL:  [SELECT * FROM tweeter.tweet WHERE  user_id < ?  LIMIT 10  ALLOW FILTERING [100]]
2018/09/18 22:35:54 CQL:  [UPDATE tweeter.tweet SET body = ?  WHERE  user_id = ? And tweet_id IN (?,?,?)  [new tweet text 1 1 2 3]]
2018/09/18 22:35:54 CQL:  [DELETE FROM tweeter.tweet WHERE  user_id = ? And tweet_id IN (?,?,?)  [1 1 2 3]]
2018/09/18 22:35:54 CQL:  [DELETE FROM tweeter.tweet WHERE  user_id = ?  [1]]

*/

Todos

• Add twitter sample play code.
• Add docs for logging , and better docs for *_Selector() ,*_Updaterr() and *_Deleter().
• Add for Batching.
• Add AllowFiltering to Deleter
• Modify .Save(...) and add .SaveCompact(...)
• Do final cleanups ( remove double cql whitespaces, some unused codes, ... )

Related Projects

Similar projects for SQL databases.

• db_walker
• xo
• Android Orma