Refactor Resource class according to v1 specs (#44)

lib/datapackage/helpers.rb

@@ -70,9 +70,9 @@ def join_paths(base_path, resource)
       if base_path.nil? || base_path.empty?
         resource
       elsif base_path =~ /\A#{URI::regexp}\z/
-        URI.join(base_path, resource)
+        URI.join(base_path, resource).to_s
       elsif File.directory?(base_path)
-        File.join(base_path, resource)
+        File.join(base_path, resource).to_s
       elsif File.file?(base_path)
         base_path
       else
@@ -90,7 +90,7 @@ def is_fully_qualified_url?(string)
     def is_safe_path?(string)
       path = Pathname.new(string)
       return false if path.absolute?
-      return false unless /^\.+$/.match(path.split[0].to_s).nil?
+      return false unless /^\.+$/.match(path.to_s.split('/').first).nil?
       true
     end
 

lib/datapackage/resource.rb

@@ -2,23 +2,55 @@ module DataPackage
   class Resource < Hash
     include DataPackage::Helpers
 
-    attr_reader :data
+    attr_reader :name, :profile, :source, :source_type, :valid, :errors
 
     def initialize(resource, base_path = '')
       resource = dereference_descriptor(resource, base_path: base_path,
         reference_fields: ['schema', 'dialect'])
-      if resource.fetch('data', nil)
-        @data = resource['data']
-      elsif resource.fetch('path', nil)
-        @data = open(join_paths(base_path, resource['path'])).read
-      else
-        raise ResourceException.new 'A resource descriptor must have a `path` or `data` property.'
-      end
       self.merge! resource
+      @profile = DataPackage::Profile.new(self.fetch('profile', 'data-resource'))
+      @name = self['name']
+      get_source!(base_path)
     end
 
     def table
-      @table ||= TableSchema::Table.new(CSV.parse(data), self['schema']) if self['schema']
+      @table ||= TableSchema::Table.new(self.source, self['schema']) if tabular?
+    end
+
+    def tabular?
+      tabular_profile = 'tabular-data-resource'
+      return true if @profile.name == tabular_profile
+      return true if DataPackage::Profile.new(tabular_profile).valid?(self)
+      false
+    end
+
+    alias :tabular :tabular?
+
+    def valid?
+      validate
+      @valid
+    end
+
+    def validate
+      @errors = @profile.validate(self)
+      @valid = @profile.valid?(self)
+    end
+
+    private
+
+    def get_source!(base_path)
+      if self.fetch('data', nil)
+        @source = self['data']
+        @source_type = 'inline'
+      elsif self.fetch('path', nil)
+        unless is_safe_path?(self['path'])
+          raise ResourceException.new "Path `#{self['path']}` is not safe"
+        end
+        @source = join_paths(base_path, self['path'])
+        @source_type = is_fully_qualified_url?(@source) ? 'remote' : 'local'
+      else
+        raise ResourceException.new 'A resource descriptor must have a `path` or `data` property.'
+      end
     end
 
   end

lib/profiles/data-resource.json

@@ -0,0 +1,278 @@
+{
+  "$schema": "http://json-schema.org/draft-04/schema#",
+  "title": "Data Resource",
+  "description": "Data Resource.",
+  "type": "object",
+  "oneOf": [
+    {
+      "required": [
+        "name",
+        "data"
+      ]
+    },
+    {
+      "required": [
+        "name",
+        "path"
+      ]
+    }
+  ],
+  "properties": {
+    "profile": {
+      "propertyOrder": 10,
+      "title": "Profile",
+      "description": "The profile of this descriptor.",
+      "context": "Every Package and Resource descriptor has a profile. The default profile, if none is declared, is `default`. The namespace for the profile is the type of descriptor, so, `default` for a Package descriptor is not the same as `default` for a Resource descriptor.",
+      "type": "string",
+      "default": "default",
+      "examples": [
+        "{\n  \"profile\": \"tabular-data-package\"\n}\n",
+        "{\n  \"profile\": \"http://example.com/my-profiles-json-schema.json\"\n}\n"
+      ]
+    },
+    "name": {
+      "propertyOrder": 20,
+      "title": "Name",
+      "description": "An identifier string. Lower case characters with `.`, `_`, `-` and `/` are allowed.",
+      "type": "string",
+      "pattern": "^([-a-z0-9._/])+$",
+      "context": "This is ideally a url-usable and human-readable name. Name `SHOULD` be invariant, meaning it `SHOULD NOT` change when its parent descriptor is updated.",
+      "examples": [
+        "{\n  \"name\": \"my-nice-name\"\n}\n"
+      ]
+    },
+    "path": {
+      "propertyOrder": 30,
+      "title": "Path",
+      "description": "A reference to the data for this resource, as either a path as a string, or an array of paths as strings. of valid URIs.",
+      "oneOf": [
+        {
+          "title": "Path",
+          "description": "A fully qualified URL, or a POSIX file path..",
+          "type": "string",
+          "examples": [
+            "{\n  \"path\": \"file.csv\"\n}\n",
+            "{\n  \"path\": \"http://example.com/file.csv\"\n}\n"
+          ],
+          "context": "Implementations need to negotiate the type of path provided, and dereference the data accordingly."
+        },
+        {
+          "type": "array",
+          "minItems": 1,
+          "items": {
+            "title": "Path",
+            "description": "A fully qualified URL, or a POSIX file path..",
+            "type": "string",
+            "examples": [
+              "{\n  \"path\": \"file.csv\"\n}\n",
+              "{\n  \"path\": \"http://example.com/file.csv\"\n}\n"
+            ],
+            "context": "Implementations need to negotiate the type of path provided, and dereference the data accordingly."
+          },
+          "examples": [
+            "[ \"file.csv\" ]\n",
+            "[ \"http://example.com/file.csv\" ]\n"
+          ]
+        }
+      ],
+      "context": "The dereferenced value of each referenced data source in `path` `MUST` be commensurate with a native, dereferenced representation of the data the resource describes. For example, in a *Tabular* Data Resource, this means that the dereferenced value of `path` `MUST` be an array.",
+      "examples": [
+        "{\n  \"path\": [\n    \"file.csv\",\n    \"file2.csv\"\n  ]\n}\n",
+        "{\n  \"path\": [\n    \"http://example.com/file.csv\",\n    \"http://example.com/file2.csv\"\n  ]\n}\n",
+        "{\n  \"path\": \"http://example.com/file.csv\"\n}\n"
+      ]
+    },
+    "data": {
+      "propertyOrder": 230,
+      "title": "Data",
+      "description": "Inline data for this resource."
+    },
+    "schema": {
+      "propertyOrder": 40,
+      "title": "Schema",
+      "description": "A schema for this resource.",
+      "type": "object"
+    },
+    "title": {
+      "propertyOrder": 50,
+      "title": "Title",
+      "description": "A human-readable title.",
+      "type": "string",
+      "examples": [
+        "{\n  \"title\": \"My Package Title\"\n}\n"
+      ]
+    },
+    "description": {
+      "propertyOrder": 60,
+      "format": "textarea",
+      "title": "Description",
+      "description": "A text description. Markdown is encouraged.",
+      "type": "string",
+      "examples": [
+        "{\n  \"description\": \"# My Package description\\nAll about my package.\"\n}\n"
+      ]
+    },
+    "homepage": {
+      "propertyOrder": 70,
+      "title": "Home Page",
+      "description": "The home on the web that is related to this data package.",
+      "type": "string",
+      "format": "uri",
+      "examples": [
+        "{\n  \"homepage\": \"http://example.com/\"\n}\n"
+      ]
+    },
+    "sources": {
+      "propertyOrder": 140,
+      "options": {
+        "hidden": true
+      },
+      "title": "Sources",
+      "description": "The raw sources for this resource.",
+      "type": "array",
+      "minItems": 1,
+      "items": {
+        "title": "Source",
+        "description": "A source file.",
+        "type": "object",
+        "required": [
+          "title"
+        ],
+        "properties": {
+          "title": {
+            "title": "Title",
+            "description": "A human-readable title.",
+            "type": "string",
+            "examples": [
+              "{\n  \"title\": \"My Package Title\"\n}\n"
+            ]
+          },
+          "path": {
+            "title": "Path",
+            "description": "A fully qualified URL, or a POSIX file path..",
+            "type": "string",
+            "examples": [
+              "{\n  \"path\": \"file.csv\"\n}\n",
+              "{\n  \"path\": \"http://example.com/file.csv\"\n}\n"
+            ],
+            "context": "Implementations need to negotiate the type of path provided, and dereference the data accordingly."
+          },
+          "email": {
+            "title": "Email",
+            "description": "An email address.",
+            "type": "string",
+            "format": "email",
+            "examples": [
+              "{\n  \"email\": \"example@example.com\"\n}\n"
+            ]
+          }
+        }
+      },
+      "examples": [
+        "{\n  \"sources\": [\n    {\n      \"name\": \"World Bank and OECD\",\n      \"uri\": \"http://data.worldbank.org/indicator/NY.GDP.MKTP.CD\"\n    }\n  ]\n}\n"
+      ]
+    },
+    "licenses": {
+      "description": "The license(s) under which the resource is published.",
+      "propertyOrder": 150,
+      "options": {
+        "hidden": true
+      },
+      "title": "Licenses",
+      "type": "array",
+      "minItems": 1,
+      "items": {
+        "title": "License",
+        "description": "A license for this descriptor.",
+        "type": "object",
+        "properties": {
+          "name": {
+            "title": "Open Definition license identifier",
+            "description": "MUST be an Open Definition license identifier, see http://licenses.opendefinition.org/",
+            "type": "string",
+            "pattern": "^([-a-zA-Z0-9._])+$"
+          },
+          "path": {
+            "title": "Path",
+            "description": "A fully qualified URL, or a POSIX file path..",
+            "type": "string",
+            "examples": [
+              "{\n  \"path\": \"file.csv\"\n}\n",
+              "{\n  \"path\": \"http://example.com/file.csv\"\n}\n"
+            ],
+            "context": "Implementations need to negotiate the type of path provided, and dereference the data accordingly."
+          },
+          "title": {
+            "title": "Title",
+            "description": "A human-readable title.",
+            "type": "string",
+            "examples": [
+              "{\n  \"title\": \"My Package Title\"\n}\n"
+            ]
+          }
+        },
+        "context": "Use of this property does not imply that the person was the original creator of, or a contributor to, the data in the descriptor, but refers to the composition of the descriptor itself."
+      },
+      "context": "This property is not legally binding and does not guarantee that the package is licensed under the terms defined herein.",
+      "examples": [
+        "{\n  \"licenses\": [\n    {\n      \"name\": \"odc-pddl-1.0\",\n      \"uri\": \"http://opendatacommons.org/licenses/pddl/\"\n    }\n  ]\n}\n"
+      ]
+    },
+    "format": {
+      "propertyOrder": 80,
+      "title": "Format",
+      "description": "The file format of this resource.",
+      "context": "`csv`, `xls`, `json` are examples of common formats.",
+      "type": "string",
+      "examples": [
+        "{\n  \"format\": \"xls\"\n}\n"
+      ]
+    },
+    "mediatype": {
+      "propertyOrder": 90,
+      "title": "Media Type",
+      "description": "The media type of this resource. Can be any valid media type listed with [IANA](https://www.iana.org/assignments/media-types/media-types.xhtml).",
+      "type": "string",
+      "pattern": "^(.+)/(.+)$",
+      "examples": [
+        "{\n  \"mediatype\": \"text/csv\"\n}\n"
+      ]
+    },
+    "encoding": {
+      "propertyOrder": 100,
+      "title": "Encoding",
+      "description": "The file encoding of this resource.",
+      "type": "string",
+      "default": "utf-8",
+      "examples": [
+        "{\n  \"encoding\": \"utf-8\"\n}\n"
+      ]
+    },
+    "bytes": {
+      "propertyOrder": 110,
+      "options": {
+        "hidden": true
+      },
+      "title": "Bytes",
+      "description": "The size of this resource in bytes.",
+      "type": "integer",
+      "examples": [
+        "{\n  \"bytes\": 2082\n}\n"
+      ]
+    },
+    "hash": {
+      "propertyOrder": 120,
+      "options": {
+        "hidden": true
+      },
+      "title": "Hash",
+      "type": "string",
+      "description": "The MD5 hash of this resource. Indicate other hashing algorithms with the {algorithm}:{hash} format.",
+      "pattern": "^([^:]+:[a-fA-F0-9]+|[a-fA-F0-9]{32}|)$",
+      "examples": [
+        "{\n  \"hash\": \"d25c9c77f588f5dc32059d2da1136c02\"\n}\n",
+        "{\n  \"hash\": \"SHA256:5262f12512590031bbcc9a430452bfd75c2791ad6771320bb4b5728bfb78c4d0\"\n}\n"
+      ]
+    }
+  }
+}

lib/profiles/registry.json

@@ -26,5 +26,19 @@
     "schema": "https://specs.frictionlessdata.io/schemas/table-schema.json",
     "schema_path": "table-schema.json",
     "specification": "https://specs.frictionlessdata.io/table-schema/"
+  },
+  {
+    "id": "data-resource",
+    "title": "Data Resource",
+    "schema": "https://specs.frictionlessdata.io/schemas/data-resource.json",
+    "schema_path": "data-resource.json",
+    "specification": "https://specs.frictionlessdata.io/data-resource"
+  },
+  {
+    "id": "tabular-data-resource",
+    "title": "Tabular Data Resource",
+    "schema": "https://specs.frictionlessdata.io/schemas/tabular-data-resource.json",
+    "schema_path": "tabular-data-resource.json",
+    "specification": "https://specs.frictionlessdata.io/tabular-data-resource/"
   }
-]
+]