validate_xml_xsi 0.3.0 → 0.4.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: aeccb816b7e38de2b2cc3b29dd0f3035e56f8bf2190f9d1281ae11ec73573efb
-  data.tar.gz: '0953d2190b837f8b692b468c3e06d77c55cad671a03ed4eaf2e7219e31013cc3'
+  metadata.gz: dbe79de11a54ebfff237869fbef6f4ccb9ce8b063dc3305299cb1fcb7856298b
+  data.tar.gz: 549031f8aa72adca7434f9294bfef40e2b2ce70b8980b04960cd20ab320dace4
 SHA512:
-  metadata.gz: 323fc0b7cd7b322ff35c72dd77ae1b33b5aee6b5bed953ed7b3e965cc7924bb98c36145a2296e279a6c6e208c9aa056e33809f00f3e9272d15b898425bc3e407
-  data.tar.gz: 8d79a528c2fb9172b2d0d29bcd10f67d241a792fba510b613eed501645d73cb137ed3ba0af0223b5b4cd3510025268184a42e487810428b196d17664778501a1
+  metadata.gz: 1336127c5245a94b7a8915f5d87372e17ab52bad43c32d1c07ef165d26db0ca1182baefcc826bd58cd401e213a2f7c04129dfea28d7c950d9fe1e06e2bb6780a
+  data.tar.gz: b60f8dcc6904c9b706f338a7afa7438fd98b7029a2a06b9f07010a63e3cddd1f0913bad0bc5ec2d608c3446c137339bb2ae8640251e7b4775ac006b9799e0c75

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/lib/validate_xml_xsi.rb

@@ -35,41 +35,53 @@ class XML_XSI
     end
 
     attr_reader :xsd
-    def initialize(xml_doc)
+    def initialize(xml_doc, parent_xml_doc = nil)
       unless xml_doc.is_a?(Nokogiri::XML::Document)
         raise DocumentError.new("invalid Nokogiri::XML::Document - #{xml_doc.class.name}")
       end
+      unless parent_xml_doc.nil? || parent_xml_doc.is_a?(Nokogiri::XML::Document)
+        raise DocumentError.new("invalid parent Nokogiri::XML::Document - #{parent_xml_doc.class.name}")
+      end
       @document = xml_doc
       ## Determine default/top/root namespace
       target_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')
       end
-      raise DocumentError.new("Unable to determine default (xmlns) namespace!") if target_ns_href.nil? || target_ns_href.empty?
+      if target_ns_href.nil? || target_ns_href.empty?
+        raise DocumentError.new("Unable to determine a default (xmlns) namespace!")
+      end
+
+      ## Determine schema locations, optionally inheriting their location declarations from a parent document
+      schema_locations = parent_xml_doc.nil? ? {} : self.class.find_schema_locations(parent_xml_doc)
+      schema_locations.merge!(self.class.find_schema_locations(@document))
+
+      ## If we still don't have a file location for the target namespace, attempt to look for
+      ## one based on the name of the root node (assuming that where the namespace was declared).
+      if !schema_locations.include?(target_ns_href) &&
+          @document.root.namespace.href.eql?(target_ns_href)
+        root_file_xsd = "#{@document.root.name}.xsd"
+        schema_locations[target_ns_href] = root_file_xsd if File.exist?(root_file_xsd)
+      end
+
+      unless schema_locations.include?(target_ns_href)
+        ## XXX - Another possibility would be to default to a file named after the node name declaring the xmlns
+        raise DocumentError.new("Unable to locate a source/file for the default (xmlns) namespace schema!")
+      end
+
       ## 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")
+
+      ## Minimally we need the target namespace location or we have nothing to include
+      target_ns_file = schema_locations.delete(target_ns_href)
+      @xsd << "  <xsd:include schemaLocation=\"#{target_ns_file}\"/>\n" unless target_ns_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"
 
@@ -89,5 +101,32 @@ class XML_XSI
       end
       errors
     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
   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       = "0.4.0"
   spec.authors       = ["David Hansen"]
   spec.email         = ["david@hansen4.net"]
 

metadata

@@ -1,14 +1,13 @@
 --- !ruby/object:Gem::Specification
 name: validate_xml_xsi
 version: !ruby/object:Gem::Version
-  version: 0.3.0
+  version: 0.4.0
 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 +51,6 @@ dependencies:
     - - ">="
       - !ruby/object:Gem::Version
         version: 1.13.2
-description: 
 email:
 - david@hansen4.net
 executables:
@@ -76,7 +74,6 @@ 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 +88,7 @@ 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: []