validate_xml_xsi 0.3.0 → 0.5.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: aeccb816b7e38de2b2cc3b29dd0f3035e56f8bf2190f9d1281ae11ec73573efb
-  data.tar.gz: '0953d2190b837f8b692b468c3e06d77c55cad671a03ed4eaf2e7219e31013cc3'
+  metadata.gz: cda44bf92902eaf418da297019b5390015d244cdb20da04142b01036146a54ff
+  data.tar.gz: c9ac5ba79dd64afc4d47cab9d8de4379e742db5008342f8888dcec63ebac5900
 SHA512:
-  metadata.gz: 323fc0b7cd7b322ff35c72dd77ae1b33b5aee6b5bed953ed7b3e965cc7924bb98c36145a2296e279a6c6e208c9aa056e33809f00f3e9272d15b898425bc3e407
-  data.tar.gz: 8d79a528c2fb9172b2d0d29bcd10f67d241a792fba510b613eed501645d73cb137ed3ba0af0223b5b4cd3510025268184a42e487810428b196d17664778501a1
+  metadata.gz: e5db59f5fbf7c3aaf61c931b0ea5227c9d08aaba1e65de797721f06c73b58939766f4390beac7ef919784d55b51abc4a4b21416f1edafdbe20b9ffec090565ff
+  data.tar.gz: a2217e8efdfb6b6ce53dec4fb7f9ec09147dd95fdc2a241a3ae277b7f3155867c33129780f7b8766b0039422fa76d2440346a72382662cec98cd1b43ae51ce2a

data/CHANGELOG.md

@@ -1,3 +1,8 @@
+## v0.3.0
+
+* Change overall structure to make XML_XSI::Schema a normal class (not just a singleton)
+* Allow the Schema to return the xsd document string (so it can be used externally to build a XSD document tree)
+
 ## v0.2.4
 
 * Fix some silly syntax errors

data/README.md

@@ -1,12 +1,12 @@
 # validate_xml_xsi
 
-This gem validates XML against it's xsi using the xsi:schemaLocation elements to define the location of XSD documents.
+This gem validates XML against it's XSD that is created from xsi:schemaLocation elements.
 
-It does this by first parsing the XML and searching for xsi:schemaLocation elements and then creating a schema document that imports the XSD's identified in the XML.
+It does this by first parsing the XML and searching for elements that include xsi:schemaLocation attributes. It then creates an XSD schema document that includes the root namespace and further imports additional namespaces discovered in the XML.
 
-It then validates the document against that schema and outputs any error messages.
+It then validates the document against that constructed schema document and outputs any error messages.
 
-Note: this gem utilizes the 'nokogiri' gem for XML parsing and schema validation.
+Note: this gem utilizes the 'nokogiri' gem (which in turn relies on Gnome's libxml2) for the XML parsing and schema validation.
 
 ## Installation
 

data/bin/validate_xml_xsi

@@ -3,7 +3,9 @@ require 'validate_xml_xsi'
 
 while ARGV.size > 0
   fname = ARGV.shift
-  errors = XML_XSI::Schema::new(XML_XSI.parse(fname)).validate
+  xml_doc = XML_XSI::parse(fname)
+  xsd = XML_XSI::Schema.new(xml_doc)
+  errors = xsd.validate
   if errors.empty?
     puts "XML Schema check complete - NO ERRORS!"
   else

data/lib/validate_xml_xsi.rb

@@ -1,6 +1,19 @@
 #!/usr/bin/env ruby
 require 'nokogiri'
 
+## Monkey-patch Nokogiri::XML::SyntaxError::aggregate to also keep the ENTIRE LIST of errors - ARGH!
+Nokogiri::XML::SyntaxError.class_eval do
+  class << self
+    alias_method :orig_aggregate, :aggregate
+    def aggregate(errors)
+      agg_err = orig_aggregate(errors)
+      agg_err.instance_variable_set(:@aggregate, errors)
+      agg_err.define_singleton_method(:aggregate) { instance_variable_get(:@aggregate) }
+      agg_err
+    end
+  end
+end
+
 class XML_XSI
   def self.parse(obj)
     filename = nil
@@ -11,8 +24,49 @@ class XML_XSI
     xml_doc
   end
 
+  def self.find_schema_locations(xml_doc)
+    ## Include the default xml namespace
+    schema_locations = {}
+
+    ## Determine if the document has reference to the namespace "http://www.w3.org/2001/XMLSchema-instance"
+    ## which is used for defining schemaLocations
+    xsi_prefix = xml_doc.namespaces.invert['http://www.w3.org/2001/XMLSchema-instance']&.delete_prefix('xmlns:')
+
+    ## Iterate over all the elements and find any xsi:schemaLocation attributes
+    ## and build a hash of all of the namespaces and locations
+    unless xsi_prefix.nil?
+      xsi_loc = "#{xsi_prefix}:schemaLocation"
+      xml_doc.search("//*[@#{xsi_loc}]").each do |elem|
+        elem[xsi_loc].scan(/(\S+)\s+(\S+)/).each do |ns_set|
+          if ns_loc = schema_locations[ns_set.first]
+            unless ns_loc.eql?(ns_set.last)
+              raise DocumentError.new("MISMATCHING namespace: #{ns_set.first} -> #{ns_loc} VS #{ns_set.last}")
+            end
+          else
+            schema_locations[ns_set.first] = ns_set.last
+          end
+        end
+      end
+    end
+    schema_locations
+  end
+
   class Schema
     class DocumentError < StandardError; end
+    class NamespaceError < StandardError
+      def initialize
+        super("Unable to determine default/target (xmlns) namespace!")
+      end
+    end
+    class LocationError < StandardError
+      def initialize(ns_href = nil)
+        msg = ns_href.nil? ?
+          "No schema locations defined or provided" :
+          "No location for the default (xmlns) namespace schema: #{ns_href}"
+        super(msg)
+      end
+    end
+    ## Unified Error reportign class for both XSD and XML errors
     class ValidationError < StandardError
       attr_reader :type, :file, :line, :column, :level, :message, :description, :error
       def initialize(type, err, filename = nil)
@@ -34,60 +88,68 @@ class XML_XSI
       end
     end
 
+    class Erroneous
+      attr_reader :errors
+      def initialize(ex)
+        errs = ex.respond_to?(:aggregate) ? ex.aggregate : [ex]
+        @errors = errs.map { |err| ValidationError.new(:XSD, err) }
+      end
+    end
+
     attr_reader :xsd
-    def initialize(xml_doc)
+    def initialize(xml_doc, schema_locations = {})
+      raise ArgumentError.new("Provided schema locations must be a Hash") unless schema_locations.is_a?(Hash)
       unless xml_doc.is_a?(Nokogiri::XML::Document)
         raise DocumentError.new("invalid Nokogiri::XML::Document - #{xml_doc.class.name}")
       end
       @document = xml_doc
       ## Determine default/top/root namespace
-      target_ns_href = nil
+      @ns_href = nil
       @document.namespaces.each do |ns_prefix, ns_href|
-        target_ns_href = ns_href if ns_prefix.nil? || ns_prefix.empty? || ns_prefix.eql?('xmlns')
+        @ns_href = ns_href if ns_prefix.nil? || ns_prefix.empty? || ns_prefix.eql?('xmlns')
       end
-      raise DocumentError.new("Unable to determine default (xmlns) namespace!") if target_ns_href.nil? || target_ns_href.empty?
+      raise NamespaceError.new if @ns_href.nil? || @ns_href.empty?
+
+      ## Determine schema locations found in the source document
+      schema_locations.merge!(XML_XSI::find_schema_locations(@document))
+
+      raise LocationError.new if schema_locations.empty?
+      raise LocationError.new(@ns_href) unless schema_locations.include?(@ns_href)
+
       ## Build an all-in-one XSD document that imports all of the separate schema locations
       @xsd = "<?xml version=\"1.0\"?>\n"
       @xsd << "<xsd:schema xmlns:xsd=\"http://www.w3.org/2001/XMLSchema\"\n" \
-                 "            targetNamespace=\"#{target_ns_href}\"\n" \
-                 "            version=\"1.0\">\n"
-      ## Include the default xml namespace
-      schemata_by_ns = {}
-      ## Iterate over all the elements and find any xsi:schemaLocation attributes
-      ## and build a hash of all of the namespaces and locations
-      @document.search('//*[@xsi:schemaLocation]').each do |elem|
-        elem['xsi:schemaLocation'].scan(/(\S+)\s+(\S+)/).each do |ns_set|
-          if ns_loc = schemata_by_ns[ns_set.first]
-            unless ns_loc.eql?(ns_set.last)
-              raise DocumentError.new("MISMATCHING namespace: #{ns_set.first} -> #{ns_loc} VS #{ns_set.last}")
-            end
-          else
-            schemata_by_ns[ns_set.first] = ns_set.last
-          end
-        end
-      end
-      schemata_by_ns.each do |ns_href, ns_file|
-        @xsd << (ns_href.eql?(target_ns_href) ?
-                    "  <xsd:include schemaLocation=\"#{ns_file}\"/>\n" :
-                    "  <xsd:import namespace=\"#{ns_href}\" schemaLocation=\"#{ns_file}\"/>\n")
+              "            targetNamespace=\"#{@ns_href}\"\n" \
+              "            version=\"1.0\">\n"
+
+      ## Minimally we need the target namespace location or we have nothing to include
+      @xsi_file = schema_locations.delete(@ns_href)
+      @xsd << "  <xsd:include schemaLocation=\"#{@xsi_file}\"/>\n" unless @xsi_file.nil?
+
+      ## Now add imports for the other defined schemaLocations
+      schema_locations.each do |ns_href, ns_file|
+        @xsd << "  <xsd:import namespace=\"#{ns_href}\" schemaLocation=\"#{ns_file}\"/>\n"
       end
       @xsd << "</xsd:schema>\n"
 
       ## Create the Schema objects
-      @schema = Nokogiri::XML::Schema.new(@xsd)
+      begin
+        @schema = Nokogiri::XML::Schema.new(@xsd)
+      rescue Nokogiri::XML::SyntaxError => ex
+        ## Trap the errors so we can finish initialization and then
+        ## report the errors in a sane manner
+        @schema = Erroneous.new(ex)
+      end
     end
 
+    def errors; @schema.errors; end
+
     def validate
-      errors = []
-      @schema.errors.each do |err|
-        errors << ValidationError.new(:XSD, err)
-      end
       errs = @schema.validate(@document)
-      errs.each do |err|
-        fname = (err.file.nil?) ? @document.filename : err.file
-        errors << ValidationError.new(:XML, err, fname)
+      errs.map do |err|
+        fname = err.file.nil? ? @document.filename : err.file
+        ValidationError.new(:XML, err, fname)
       end
-      errors
     end
   end
 end

data/validate_xml_xsi.gemspec

@@ -1,6 +1,6 @@
 Gem::Specification.new do |spec|
   spec.name          = "validate_xml_xsi"
-  spec.version       = "0.3.0"
+  spec.version       = "0.5.0"
   spec.authors       = ["David Hansen"]
   spec.email         = ["david@hansen4.net"]
 
@@ -26,4 +26,5 @@ Gem::Specification.new do |spec|
   spec.add_development_dependency "rake", ">= 13.0"
 
   spec.add_dependency "nokogiri", ">= 1.13.2"
+  spec.add_dependency "libxml-ruby", ">= 5.0.0"
 end

metadata

@@ -1,14 +1,13 @@
 --- !ruby/object:Gem::Specification
 name: validate_xml_xsi
 version: !ruby/object:Gem::Version
-  version: 0.3.0
+  version: 0.5.0
 platform: ruby
 authors:
 - David Hansen
-autorequire: 
 bindir: bin
 cert_chain: []
-date: 2022-03-25 00:00:00.000000000 Z
+date: 1980-01-02 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: bundler
@@ -52,7 +51,20 @@ dependencies:
     - - ">="
       - !ruby/object:Gem::Version
         version: 1.13.2
-description: 
+- !ruby/object:Gem::Dependency
+  name: libxml-ruby
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: 5.0.0
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: 5.0.0
 email:
 - david@hansen4.net
 executables:
@@ -76,7 +88,6 @@ metadata:
   homepage_uri: https://github.com/d-hansen/validate_xml_xsi
   source_code_uri: https://github.com/d-hansen/validate_xml_xsi
   changelog_uri: https://github.com/d-hansen/validate_xml_xsi/blob/master/CHANGELOG.md
-post_install_message: 
 rdoc_options: []
 require_paths:
 - lib
@@ -91,8 +102,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
-rubygems_version: 3.3.8
-signing_key: 
+rubygems_version: 4.0.2
 specification_version: 4
 summary: Validate XML against it's embedded XSI elements that define the XSD's.
 test_files: []