class Nokogiri::XML::Schema

Parent:: Object

Nokogiri::XML::Schema is used for validating XML against a schema (usually from an xsd file).

Synopsis

Validate an XML document against a Schema. Loop over the errors that are returned and print them out:

xsd = Nokogiri::XML::Schema(File.read(PO_SCHEMA_FILE))
doc = Nokogiri::XML(File.read(PO_XML_FILE))

xsd.validate(doc).each do |error|
  puts error.message
end

The list of errors are Nokogiri::XML::SyntaxError objects.

NOTE: As of v1.11.0, Schema treats inputs as UNTRUSTED by default, and so external entities are not resolved from the network (‘http://` or `ftp://`). Previously, parsing treated documents as “trusted” by default which was counter to Nokogiri’s “untrusted by default” security policy. If a document is trusted, then the caller may turn off the NONET option via the ParseOptions to re-enable external entity resolution over a network connection.

Attributes

errors[RW]

Errors while parsing the schema file

parse_options[RW]

The Nokogiri::XML::ParseOptions used to parse the schema

Public Class Methods

from_document(doc) Show source

static VALUE
from_document(int argc, VALUE *argv, VALUE klass)
{
  VALUE document;
  VALUE parse_options;
  int parse_options_int;
  xmlDocPtr doc;
  xmlSchemaParserCtxtPtr ctx;
  xmlSchemaPtr schema;
  VALUE errors;
  VALUE rb_schema;
  int scanned_args = 0;
  xmlExternalEntityLoader old_loader = 0;

  scanned_args = rb_scan_args(argc, argv, "11", &document, &parse_options);

  Noko_Node_Get_Struct(document, xmlDoc, doc);
  doc = doc->doc; /* In case someone passes us a node. ugh. */

  if (scanned_args == 1) {
    parse_options = rb_const_get_at(rb_const_get_at(mNokogiriXml, rb_intern("ParseOptions")), rb_intern("DEFAULT_SCHEMA"));
  }
  parse_options_int = (int)NUM2INT(rb_funcall(parse_options, rb_intern("to_i"), 0));

  if (has_blank_nodes_p(DOC_NODE_CACHE(doc))) {
    rb_raise(rb_eArgError, "Creating a schema from a document that has blank nodes exposed to Ruby is dangerous");
  }

  ctx = xmlSchemaNewDocParserCtxt(doc);

  errors = rb_ary_new();
  xmlSetStructuredErrorFunc((void *)errors, Nokogiri_error_array_pusher);

#ifdef HAVE_XMLSCHEMASETPARSERSTRUCTUREDERRORS
  xmlSchemaSetParserStructuredErrors(
    ctx,
    Nokogiri_error_array_pusher,
    (void *)errors
  );
#endif

  if (parse_options_int & XML_PARSE_NONET) {
    old_loader = xmlGetExternalEntityLoader();
    xmlSetExternalEntityLoader(xmlNoNetExternalEntityLoader);
  }

  schema = xmlSchemaParse(ctx);

  if (old_loader) {
    xmlSetExternalEntityLoader(old_loader);
  }

  xmlSetStructuredErrorFunc(NULL, NULL);
  xmlSchemaFreeParserCtxt(ctx);

  if (NULL == schema) {
    xmlErrorPtr error = xmlGetLastError();
    if (error) {
      Nokogiri_error_raise(NULL, error);
    } else {
      rb_raise(rb_eRuntimeError, "Could not parse document");
    }

    return Qnil;
  }

  rb_schema = Data_Wrap_Struct(klass, 0, dealloc, schema);
  rb_iv_set(rb_schema, "@errors", errors);
  rb_iv_set(rb_schema, "@parse_options", parse_options);

  return rb_schema;

  return Qnil;
}

Create a new Schema from the Nokogiri::XML::Document doc

new(string_or_io, options = ParseOptions::DEFAULT_SCHEMA) Show source

# File lib/nokogiri/xml/schema.rb, line 46
def self.new(string_or_io, options = ParseOptions::DEFAULT_SCHEMA)
  from_document(Nokogiri::XML(string_or_io), options)
end

Create a new Nokogiri::XML::Schema object using a string_or_io object.

read_memory(string) Show source

static VALUE
read_memory(int argc, VALUE *argv, VALUE klass)
{
  VALUE content;
  VALUE parse_options;
  int parse_options_int;
  xmlSchemaParserCtxtPtr ctx;
  xmlSchemaPtr schema;
  VALUE errors;
  VALUE rb_schema;
  int scanned_args = 0;
  xmlExternalEntityLoader old_loader = 0;

  scanned_args = rb_scan_args(argc, argv, "11", &content, &parse_options);
  if (scanned_args == 1) {
    parse_options = rb_const_get_at(rb_const_get_at(mNokogiriXml, rb_intern("ParseOptions")), rb_intern("DEFAULT_SCHEMA"));
  }
  parse_options_int = (int)NUM2INT(rb_funcall(parse_options, rb_intern("to_i"), 0));

  ctx = xmlSchemaNewMemParserCtxt((const char *)StringValuePtr(content), (int)RSTRING_LEN(content));

  errors = rb_ary_new();
  xmlSetStructuredErrorFunc((void *)errors, Nokogiri_error_array_pusher);

#ifdef HAVE_XMLSCHEMASETPARSERSTRUCTUREDERRORS
  xmlSchemaSetParserStructuredErrors(
    ctx,
    Nokogiri_error_array_pusher,
    (void *)errors
  );
#endif

  if (parse_options_int & XML_PARSE_NONET) {
    old_loader = xmlGetExternalEntityLoader();
    xmlSetExternalEntityLoader(xmlNoNetExternalEntityLoader);
  }

  schema = xmlSchemaParse(ctx);

  if (old_loader) {
    xmlSetExternalEntityLoader(old_loader);
  }

  xmlSetStructuredErrorFunc(NULL, NULL);
  xmlSchemaFreeParserCtxt(ctx);

  if (NULL == schema) {
    xmlErrorPtr error = xmlGetLastError();
    if (error) {
      Nokogiri_error_raise(NULL, error);
    } else {
      rb_raise(rb_eRuntimeError, "Could not parse document");
    }

    return Qnil;
  }

  rb_schema = Data_Wrap_Struct(klass, 0, dealloc, schema);
  rb_iv_set(rb_schema, "@errors", errors);
  rb_iv_set(rb_schema, "@parse_options", parse_options);

  return rb_schema;
}

Create a new Schema from the contents of string

Public Instance Methods

valid?(thing) Show source

# File lib/nokogiri/xml/schema.rb, line 68
def valid?(thing)
  validate(thing).empty?
end

Returns true if thing is a valid Nokogiri::XML::Document or file.

validate(thing) Show source

# File lib/nokogiri/xml/schema.rb, line 55
def validate(thing)
  if thing.is_a?(Nokogiri::XML::Document)
    validate_document(thing)
  elsif File.file?(thing)
    validate_file(thing)
  else
    raise ArgumentError, "Must provide Nokogiri::Xml::Document or the name of an existing file"
  end
end

Validate thing against this schema. thing can be a Nokogiri::XML::Document object, or a filename. An Array of Nokogiri::XML::SyntaxError objects found while validating the thing is returned.

© 2008–2023 by Mike Dalessio, Aaron Patterson, Yoko Harada, Akinori MUSHA, John Shahid,
Karol Bucek, Sam Ruby, Craig Barnes, Stephen Checkoway, Lars Kanis, Sergio Arbeo,
Timothy Elliott, Nobuyoshi Nakada, Charles Nutter, Patrick MahoneyLicensed under the MIT License.
https://nokogiri.org/rdoc/Nokogiri/XML/Schema.html