refactoring, added Repository and Factory classes, stream validation support (#6)

refactoring, added Repository and Factory classes, stream validation support

Author: OriHoch

CONTRIBUTING.md

@@ -21,6 +21,8 @@ $ composer test
 
 ## Phpunit - for unit tests
 
+[![Travis](https://travis-ci.org/frictionlessdata/datapackage-php.svg?branch=master)](https://travis-ci.org/frictionlessdata/datapackage-php)
+
 Phpunit is used for unit tests, you can find the tests under tests directory
 
 Running Phpunit directly: `vendor/bin/phpunit --bootstrap tests/autoload.php tests/`
@@ -35,8 +37,22 @@ when running `composer test` phpunit generates coverage report in coverage-clove
 
 ## Scrutinizer-ci - for code analysis
 
-[Scrutinizer-ci](https://scrutinizer-ci.com/) integrates with GitHub and runs on commits.
+[![Scrutinizer-ci](https://scrutinizer-ci.com/g/OriHoch/datapackage-php/badges/quality-score.png?b=master)](https://scrutinizer-ci.com/g/OriHoch/datapackage-php/)
+
+Scrutinizer-ci integrates with GitHub and runs on commits.
 
 It does static code analysis and ensure confirmation to the coding stnadards.
 
 At the moment, the integration with frictionlessdata repo is not working, you can setup a Scrutinizer-ci account for your fork and run against that.
+
+
+## Publishing a release and updating Packagist
+
+[![Packagist](https://img.shields.io/packagist/dm/frictionlessdata/datapackage.svg)](https://packagist.org/packages/frictionlessdata/datapackage)
+[![SemVer](https://img.shields.io/badge/versions-SemVer-brightgreen.svg)](http://semver.org/)
+
+* Publish a release (using SemVer) on GitHub
+* go to https://packagist.org/packages/frictionlessdata/datapackage
+* Login with GitHub which has permissions
+* click "Update"
+* all releases from GitHub appear as releases on Packagist

README.md

@@ -23,8 +23,10 @@ $ composer require frictionlessdata/datapackage
 ```php
 use frictionlessdata\datapackage;
 
+// get a datapackage object
+$datapackage = datapackage\Factory::datapackage("tests/fixtures/multi_data_datapackage.json");
+
 // iterate over the data
-$datapackage = new datapackage\Datapackage("tests/fixtures/multi_data_datapackage.json");
 foreach ($datapackage as $resource) {
     print("-- ".$resource->name()." --");
     $i = 0;
@@ -36,12 +38,12 @@ foreach ($datapackage as $resource) {
     }
 }
 
-// validate the descriptor
-$validationErrors = datapackage\Datapackage::validate("tests/fixtures/simple_invalid_datapackage.json");
+// validate a datapackage descriptor
+$validationErrors = datapackage\Factory::validate("tests/fixtures/simple_invalid_datapackage.json");
 if (count($validationErrors) == 0) {
     print("descriptor is valid");
 } else {
-    print(datapackage\DatapackageValidationError::getErrorMessages($validationErrors));
+    print(datapackage\Validators\DatapackageValidationError::getErrorMessages($validationErrors));
 }
 ```
 

composer.json

@@ -5,7 +5,7 @@
     "require": {
         "php": ">=5.4",
         "justinrainbow/json-schema": "^5.2",
-        "frictionlessdata/tableschema": "^0.1.2"
+        "frictionlessdata/tableschema": "^0.1.3"
     },
     "require-dev": {
         "phpunit/phpunit": "^4.8.35",

src/DataStream.php

@@ -0,0 +1,24 @@
+<?php
+namespace frictionlessdata\datapackage\DataStreams;
+
+/**
+ * Provides a standard interface for streaming
+ *
+ * functionality could mostly be replaced by php generators (http://php.net/manual/en/language.generators.syntax.php)
+ * however, they are only supported on PHP 5.5 and above
+ */
+abstract class BaseDataStream implements \Iterator
+{
+    /**
+     * @param string $dataSource
+     * @param mixed $dataSourceOptions
+     * @throws \frictionlessdata\datapackage\Exceptions\DataStreamOpenException
+     */
+    abstract public function __construct($dataSource, $dataSourceOptions=null);
+
+    /**
+     * @return mixed
+     * @throws \frictionlessdata\datapackage\Exceptions\DataStreamValidationException
+     */
+    abstract public function current();
+}

src/DataStreams/BaseDataStream.php

@@ -0,0 +1,56 @@
+<?php
+namespace frictionlessdata\datapackage\DataStreams;
+
+use frictionlessdata\datapackage\Exceptions\DataStreamOpenException;
+
+/**
+ * streams the raw data without processing - used for default data package resources
+ */
+class DefaultDataStream extends BaseDataStream
+{
+    /**
+     * @param string $dataSource
+     * @throws DataStreamOpenException
+     */
+    public function __construct($dataSource, $dataSourceOptions=null)
+    {
+        try {
+            $this->fopenResource = fopen($dataSource, "r");
+        } catch (\Exception $e) {
+            throw new DataStreamOpenException("Failed to open data source ".json_encode($dataSource).": ".json_encode($e->getMessage()));
+        }
+    }
+
+    public function __destruct()
+    {
+        fclose($this->fopenResource);
+    }
+
+    public function rewind() {
+        if ($this->currentLineNumber == 0) {
+            // starting iterations
+            $this->currentLineNumber = 1;
+        } else {
+            throw new \Exception("DataStream does not support rewinding a stream, sorry");
+        }
+    }
+
+    public function current() {
+        return fgets($this->fopenResource);
+    }
+
+    public function key() {
+        return $this->currentLineNumber;
+    }
+
+    public function next() {
+        $this->currentLineNumber++;
+    }
+
+    public function valid() {
+        return (!feof($this->fopenResource));
+    }
+
+    protected $currentLineNumber = 0;
+    protected $fopenResource;
+}

src/DataStreams/DefaultDataStream.php

@@ -0,0 +1,58 @@
+<?php
+namespace frictionlessdata\datapackage\DataStreams;
+
+use frictionlessdata\datapackage\Exceptions\DataStreamValidationException;
+use frictionlessdata\datapackage\Exceptions\DataStreamOpenException;
+use frictionlessdata\tableschema\DataSources\CsvDataSource;
+use frictionlessdata\tableschema\Schema;
+use frictionlessdata\tableschema\Table;
+use frictionlessdata\tableschema\Exceptions\TableRowValidationException;
+
+
+class TabularDataStream extends BaseDataStream
+{
+    public function __construct($dataSource, $schema=null)
+    {
+        if (empty($schema)) {
+            throw new \Exception("schema is required for tabular data stream");
+        } else {
+            try {
+                $dataSource = new CsvDataSource($dataSource);
+                $schema = new Schema($schema);
+                $this->table = new Table($dataSource, $schema);
+            } catch (\Exception $e) {
+                throw new DataStreamOpenException("Failed to open tabular data source ".json_encode($dataSource).": ".json_encode($e->getMessage()));
+            }
+        }
+    }
+
+    protected $table;
+
+    public function rewind() {
+        $this->table->rewind();
+    }
+
+    /**
+     * @return array
+     * @throws DataStreamValidationException
+     */
+    public function current() {
+        try {
+            return $this->table->current();
+        } catch (TableRowValidationException $e) {
+            throw new DataStreamValidationException($e->getMessage());
+        }
+    }
+
+    public function key() {
+        return $this->table->key();
+    }
+
+    public function next() {
+        $this->table->next();
+    }
+
+    public function valid() {
+        return $this->table->valid();
+    }
+}