gears / nosqlite
A NoSQL Database, with the portability of SQLite
Requires
- php: >=5.3
Suggests
- gears/arrays: You will be dealing with arrays lots so why not make your job a little easier.
README
I recently found MongoDB and fell in love with it, but I wanted something like SqLite, something where I didn't have to worry about database servers. This is what this project is all about.
The NoSQL database engine is built with PHP code. Thus the server and client are one just like SqLite. You simply give it a path to your database folder and away you go, no server setup, no user accounts, no passwords.
Originally the project just used plain old JSON to store the data. However it now contains many diffrent backends, so your database can be XML based or it might use the built in PHP serialize functionality for example.
Perhaps one day someone much clever than I may build a PHP extension. Also in the documentation there is (will be) a specification document outlining every detail so that other programmers can build their own drivers to read the database format much the same way MongoDB works.
How to Install
Installation via composer is easy:
composer require gears/nosqlite:*
How do I use it?
Once you have the src included in your project you can use it like this:
<?php //////////////////////////////////////////////////////////////////////////////// // __________ __ ________ __________ // \______ \ |__ ______ / _____/ ____ _____ ______\______ \ _______ ___ // | ___/ | \\____ \/ \ ____/ __ \\__ \\_ __ \ | _// _ \ \/ / // | | | Y \ |_> > \_\ \ ___/ / __ \| | \/ | ( <_> > < // |____| |___| / __/ \______ /\___ >____ /__| |______ /\____/__/\_ \ // \/|__| \/ \/ \/ \/ \/ // ----------------------------------------------------------------------------- // Designed and Developed by Brad Jones <brad @="bjc.id.au" /> // ----------------------------------------------------------------------------- //////////////////////////////////////////////////////////////////////////////// /* * This script is used as an example of how to use the NoSqLite component. * Where appropriate the following examples have the SQL that you may have * written to do the same thing. */ // Load our classes - note we not actually using composer // for this example, hence these manual require statements. require('../Db.php'); require('../Collection.php'); require('../Result.php'); require('../Backends/Driver.php'); require('../Backends/Json.php'); // To create or connect to an existing database. $db = new Gears\NoSqLite\Db('/tmp/gears-nosqlite-test'); /** * Section: CREATE * ============================================================================= * The first method is "create", this is how you add stuff to your database. * Note that the creation of collections or tables in sql speak is implicit. * Collections are really just folders on the filesystem in our case. * * Example of SQL: * ----------------------------------------------------------------------------- * INSERT INTO staff (name, email, office) * VALUES (Brad Jones, bj@gravit.com.au, Highton) * * Returns: * ----------------------------------------------------------------------------- * The unique id of the new document you just created, this is in fact the * file name of the JSON, XML, BSON, etc document. If it could not create the * file for whatever reason it will either throw an exception or return false. */ echo "Create some documents:\n"; echo "================================================================================\n"; echo $db->staff->create(['name' => 'Brad Jones','email' => 'bj@gravit.com.au','office' => 'Highton'])."\n"; echo $db->staff->create(['name' => 'Daniel Strumpel','email' => 'ds@gravit.com.au','office' => 'Highton'])."\n"; echo $db->staff->create(['name' => 'Kath Strumpel','email' => 'ks@gravit.com.au','office' => 'Highton'])."\n"; echo $db->staff->create(['name' => 'Benjamin Beshara','email' => 'bb@gravit.com.au','office' => 'Highton'])."\n"; echo "\n"; /** * Section: READ * ============================================================================= * Once we have created some documents or rows in sql speak. We usually want to * search for and read that data at some point in the future. This next function * is the eqilivant of the sql SELECT keyword. * * Note anywhere you see an empty array you can omit it but it is shown here * for completeness sake, just so you have a better idea on what is happening. * * The first array provided to the read function is your search query. * The second array is the values you want out of the matched documents. * * Example of SQL: * ----------------------------------------------------------------------------- * SELECT * FROM staff * SELECT name FROM staff * SELECT * FROM staff WHERE office = Highton * SELECT email FROM staff WHERE office = Highton * * Returns: * ----------------------------------------------------------------------------- * A multi-dimensional array - ie: exactly what you created above. * Well not exactly the same, we also return the id of the document in case * you want that as well. */ echo "Read some documents:\n"; echo "================================================================================\n"; print_r($db->staff->read([])); print_r($db->staff->read([], ['name'])); print_r($db->staff->read(['office' => 'Highton'])); print_r($db->staff->read(['office' => 'Highton'], ['email'])); echo "\n"; /** * Section: UPDATE * ============================================================================= * Next up is the "update" function. Basically this just does a read and then a * create automatically for you. The first array provided to the update function * is your search query, this is passed directly to the read function. * * The second array is the values you want to update or add to the documents * that were matched by the read function. The data is modified and then written * back to their original files. * * Example of SQL: * ----------------------------------------------------------------------------- * UPDATE staff SET office = Melbourne WHERE name = Brad Jones * * Returns: * ----------------------------------------------------------------------------- * This will return the updated data that was successfully * written back to the filesystem. */ echo "Update some documents:\n"; echo "================================================================================\n"; print_r($db->staff->update(['name' => 'Brad Jones'], ['office' => 'Melbourne'])); echo "\n"; /** * Section: DELETE * ============================================================================= * Finally this is how we delete something in our database. * * Example of SQL: * ----------------------------------------------------------------------------- * DELETE FROM staff WHERE name = Brad Jones * DROP TABLE staff * DROP DATABASE dbname * * Returns: * ----------------------------------------------------------------------------- * This will return the updated data that was successfully * written back to the filesystem. */ echo "Delete some documents:\n"; echo "================================================================================\n"; var_dump($db->staff->delete(['name' => 'Brad Jones'])); var_dump($db->staff->delete([])); var_dump($db->delete()); echo "\n"; /** * Section: Advanced Examples * ============================================================================= * The above is easy stuff, below are some more examples that you might not use * as often as the above but still just as valid. * * Example of SQL: * ----------------------------------------------------------------------------- * This time some of the examples don't really have any sort of equivalent in * SQL thus I won't be providing any. At this point I am assuming your smart * enough to work it out for your self :) */ echo "Advanced Examples:\n"; echo "================================================================================\n"; // First up you will note we need to create a new database // because we just deleted the one from above. $db = new Gears\NoSqLite\Db('/tmp/gears-nosqlite-test'); // First lets create some data to work with - yep you have seen this before... easy right? $db->staff->create(['name' => 'Brad Jones','email' => 'bj@gravit.com.au','office' => 'Highton','postcode' => '3220']); $db->staff->create(['name' => 'Daniel Strumpel','email' => 'ds@gravit.com.au','office' => 'Highton','postcode' => '3220']); $db->staff->create(['name' => 'Kath Strumpel','email' => 'ks@gravit.com.au','office' => 'Highton','postcode' => '3221']); $db->staff->create(['name' => 'Benjamin Beshara','email' => 'bb@gravit.com.au','office' => 'Highton','postcode' => '3221']); // We also use these down the road too. $db->articles->create(['name' => 'NoSQL Release','author' => 'Brad Jones','tags' => ['something','cool','nosql']]); $db->articles->create(['name' => 'Server 2012','author' => 'Daniel Strumpel','tags' => ['something','cool','server-2012']]); $db->articles->create(['name' => 'My Apple Mac','author' => 'Benjamin Beshara','tags' => ['something','cool','apple-mac']]); // We need something with numbers $db->counter->create(['x' => 1]); $db->counter->create(['x' => 2]); $db->counter->create(['x' => 3]); $db->counter->create(['x' => 4]); $db->counter->create(['x' => 5]); $db->counter->create(['x' => 6]); $db->counter->create(['x' => 7]); $db->counter->create(['x' => 8]); $db->counter->create(['x' => 9]); $db->counter->create(['x' => 10]); /* * Now just to make it very clear this is not a relational database. * There are no primary keys, there is no "JOIN" keyword or anything of * the sort. But we can still achieve the same sort of functionality. */ $db->staff->managers->create(['name' => 'Daniel Strumpel']); /* * All I have done is create a new collection with one document. * This new collection is in no way linked to the parent staff collection. * Now watch and learn... */ echo "JOIN Example\n"; echo "================================================================================\n"; $managers = $db->staff->managers->read([], ['name']); $staff_that_are_managers = $db->staff->read(['in' => ['name' => $managers]]); print_r($staff_that_are_managers); /* * And for those that like one liners. Also note that I don't have to actually * specify the exact value in the first query as the "in" query and most other * querys are smart enough to read a returned array from another read call. */ print_r($db->staff->read(['in' => ['name' => $db->staff->managers->read()]])); echo "\n"; /* * Next up are the diffrent operators, you have already seen "in". * I know the syntax of these aren't the same as how you would read it, * eg: normally you would say "is x in y" but in our case we say "is in y x". * One day it may change but for now this is how it is. */ echo "IN Example\n"; echo "================================================================================\n"; print_r($db->staff->read(['in' => ['name' => ['Daniel Strumpel', 'Brad Jones']]])); echo "\n"; echo "NOT IN Example\n"; echo "================================================================================\n"; print_r($db->staff->read(['nin' => ['name' => ['Daniel Strumpel', 'Brad Jones']]])); echo "\n"; echo "OR Example\n"; echo "================================================================================\n"; print_r($db->staff->read(['or' => [['name' => 'Daniel Strumpel'],['name' => 'Brad Jones']]])); echo "\n"; echo "NOR Example\n"; echo "================================================================================\n"; print_r($db->staff->read(['nor' => [['name' => 'Daniel Strumpel'],['name' => 'Brad Jones']]])); echo "\n"; echo "All Example\n"; echo "================================================================================\n"; print_r($db->articles->read(['all' => ['tags' => ['something','cool']]])); print_r($db->articles->read(['all' => ['tags' => ['something','cool','nosql']]])); print_r($db->articles->read(['all' => ['tags' => ['something','cool','somethingelse']]])); echo "\n"; echo "AND Example\n"; echo "================================================================================\n"; print_r($db->articles->read(['and' => [['tags' => 'something'],['tags' => 'cool']]])); echo "\n"; echo "> Example\n"; echo "================================================================================\n"; print_r($db->counter->read(['gt' => ['x' => 5]])); echo "\n"; echo "< Example\n"; echo "================================================================================\n"; print_r($db->counter->read(['lt' => ['x' => 5]])); echo "\n"; echo ">= Example\n"; echo "================================================================================\n"; print_r($db->counter->read(['gte' => ['x' => 5]])); echo "\n"; echo "<= Example\n"; echo "================================================================================\n"; print_r($db->counter->read(['lte' => ['x' => 5]])); echo "\n"; echo "!= Example\n"; echo "================================================================================\n"; print_r($db->counter->read(['neq' => ['x' => 5]])); echo "\n"; /* * Now lets say you know exactly what document you want. * Eg: you already know in advance what the generated id of the document is. */ echo "_id Example\n"; echo "================================================================================\n"; $id = $db->staff->create(['name' => 'Bill Brown']); print_r($db->staff->read(['_id' => $id])); // The same works if you want to update it print_r($db->staff->update(['_id' => $id], ['name' => 'Fred Green'])); // And we can delete him to print_r($db->staff->delete(['_id' => $id])); echo "\n"; /* * Okay now we are going to show you how dot notation works. * Unlike an SQL database we can have a document with many levels. * This is how we access those levels. */ echo "dot Notation Example\n"; echo "================================================================================\n"; $db->staff->create ([ 'name' => 'Jim Brown', 'email' => 'jb@gravit.com.au', 'office' => 'Melbourne', 'address' => [ 'street' => [ 'no' => '567', 'name' => 'Bourke', 'type' => 'ST' ] ] ]); // Now you can't do that with SQL, pretty neat huh! // Now how do we say get all staff that work on Bourke street. // Easy right something like this... print_r($db->staff->read(['address' => ['street' => ['name' => 'Bourke']]])); // Unfortantly the above doesn't work, what the above is // asking for is a document that looks like this exactly: [ 'address' => [ 'street' => [ 'name' => 'Bourke' ] ] ]; // What we really need to do is this. print_r($db->staff->read(['address.street.name' => 'Bourke'])); echo "\n"; // Regular expressions echo "Regular Expression Example\n"; echo "================================================================================\n"; print_r($db->staff->read(['email' => '/^[\w-\.]+@([\w-]+\.)+[\w-]{2,4}$/'])); echo "\n"; /* * Alternative Collection Syntax * Maybe sometimes you may like to provide the path to the collection directly * as a string or maybe an array. This is how you can do it. */ echo "Alternative Collection Syntax\n"; echo "================================================================================\n"; print_r($db->c('staff/managers')->read()); print_r($db->c(['staff','managers'])->read()); echo "\n"; /* * Skip, Limit, Count and Size */ echo "Limit and Skip Examples\n"; echo "================================================================================\n"; print_r($db->staff->read()->limit(4)); print_r($db->staff->read()->skip(2)->limit(4)); echo $db->staff->read()->skip(2)->limit(4)->size()."\n"; echo $db->staff->read()->skip(2)->limit(4)->count()."\n"; echo "\n"; /* * Distinct */ echo "Distinct Example\n"; echo "================================================================================\n"; print_r($db->staff->read()->distinct('postcode')); echo "\n"; /* * Sorting examples. */ echo "Sorting Example\n"; echo "================================================================================\n"; print_r($db->staff->read()->sort(['name' => 1])); echo "\n"; // Delete everything again $db->delete();
The above example can be found as a real working example in /Example
Developed by Brad Jones - brad@bjc.id.au