teius 0.1.0 → 0.5

data/ext/extconf.rb

@@ -1,5 +1,7 @@
 require 'mkmf'
 
-dir_config('libxml', '/usr/include/libxml2', '/usr/lib');
+dir_config('libxml', '/usr/include/libxml', '/usr/lib');
+dir_config('libxslt', '/usr/include', '/usr/lib');
 have_library('xml2', 'xmlReadFile')
+have_library('xslt', 'xsltParseStylesheetFile')
 create_makefile("teius")

data/ext/teius.c

@@ -1,20 +1,31 @@
+/* NOTES: possibly call xsltCleanupGlobals() and xmlCleanupParser() at some point */
+
 #include <stdio.h>
 #include "ruby.h"
 #include <libxml/parser.h>
 #include <libxml/tree.h>
 #include <libxml/xpath.h>
 
+#include <libxslt/xslt.h>
+#include <libxslt/xsltInternals.h>
+#include <libxslt/transform.h>
+#include <libxslt/xsltutils.h>
+
 static VALUE cParseError;
 static VALUE cXpathError;
 static VALUE cNodeNotFound;
 static VALUE cNodeNotUnique;
 static VALUE cNode;
 static VALUE cDocument;
+static VALUE cStylesheet;
 static ID    sRequired;
 static ID    sFirst;
 static ID    sAll;
 static ID    sUnique;
 
+/* Suppress error messages to stderr */
+static void null_handler(void *ctx, const char *msg, ...) {}
+
 static void doc_free(void *p) {
   xmlFreeDoc(p);
 }
@@ -151,7 +162,31 @@ static VALUE node_pointer(VALUE self) {
   return INT2NUM((int)node);
 }
 
-static VALUE parse_file(VALUE self, VALUE rFilename) {
+/* Returns the line number the node starts on */
+static VALUE node_line(VALUE self) {
+  xmlNodePtr node = NULL;
+  Data_Get_Struct(self, xmlNode, node);
+  return INT2NUM(node->line);
+}
+
+/* Returns the recursive content of the node in string format */
+static VALUE node_content(VALUE self) {
+  xmlNodePtr node = NULL;
+  const xmlChar *mem;
+  xmlBufferPtr buf;
+  int size;
+  
+  Data_Get_Struct(self, xmlNode, node);
+  buf = xmlBufferCreate();
+  size = xmlNodeDump(buf, node->doc, node, 0, 1);
+  mem  = xmlBufferContent(buf);
+  VALUE rStr = rb_str_new2(mem);
+  xmlBufferFree(buf);
+  
+  return rStr;
+}
+
+static VALUE document_parse_file(VALUE self, VALUE rFilename) {
   xmlDocPtr doc = xmlReadFile(StringValuePtr(rFilename), NULL, 0);
   if (doc == NULL) {
     xmlErrorPtr  err = xmlGetLastError();
@@ -165,7 +200,7 @@ static VALUE parse_file(VALUE self, VALUE rFilename) {
   return rDoc;
 }
 
-static VALUE parse_string(VALUE self, VALUE rDocString) {
+static VALUE document_parse_string(VALUE self, VALUE rDocString) {
   xmlDocPtr doc = xmlReadDoc(StringValuePtr(rDocString), NULL, NULL, 0);
   if (doc == NULL) {
     xmlErrorPtr  err = xmlGetLastError();
@@ -178,8 +213,8 @@ static VALUE parse_string(VALUE self, VALUE rDocString) {
   return rDoc;
 }
 
-static VALUE document_to_s(VALUE self, VALUE rDocString) {
-  xmlDocPtr  doc  = NULL;
+static VALUE document_to_s(VALUE self) {
+  xmlDocPtr  doc = NULL;
   xmlChar *mem;
   int size;
   
@@ -191,9 +226,96 @@ static VALUE document_to_s(VALUE self, VALUE rDocString) {
   return rStr;
 }
 
+static VALUE document_root(VALUE self) {
+  xmlDocPtr  doc = NULL;
+  xmlNodePtr node = NULL;
+  VALUE rResult;
+  
+  Data_Get_Struct(self, xmlDoc, doc);
+  node = xmlDocGetRootElement(doc);
+  rResult = Data_Wrap_Struct(cNode, 0, 0, node);
+  
+  return rResult;
+}
+
+/* LibXslt Stuff */
+static void stylesheet_free(void *p) {
+  xsltFreeStylesheet(p);
+}
+
+static VALUE stylesheet_parse_file(VALUE self, VALUE rFilename) {
+  xsltStylesheetPtr xsl = xsltParseStylesheetFile(StringValuePtr(rFilename));
+  if (xsl == NULL) {
+    xmlErrorPtr  err = xmlGetLastError();
+    rb_raise(cParseError, "could not parse file %s: %s", 
+      StringValuePtr(rFilename), err->message);
+  }
+
+  /* Load new document */
+  VALUE rStylesheet = Data_Wrap_Struct(cStylesheet, 0, stylesheet_free, xsl);
+
+  return rStylesheet;
+}
+
+static VALUE stylesheet_apply(VALUE self, VALUE rDocument) {
+  xsltStylesheetPtr stylesheet = NULL;
+  xmlDocPtr  doc = NULL, res = NULL;
+  
+  Data_Get_Struct(self, xsltStylesheet, stylesheet);
+  Data_Get_Struct(rDocument, xmlDoc, doc);
+  res = xsltApplyStylesheet(stylesheet, doc, NULL); 
+
+  VALUE rDoc = Data_Wrap_Struct(cDocument, 0, doc_free, res);
+  
+  return rDoc;
+}
+
+static VALUE stylesheet_save_to_file(VALUE self, VALUE rResult, VALUE rFilename) {
+  xsltStylesheetPtr stylesheet = NULL;
+  xmlDocPtr result = NULL;
+  int st;
+  
+  Data_Get_Struct(self, xsltStylesheet, stylesheet);
+  Data_Get_Struct(rResult, xmlDoc, result);
+
+  st = xsltSaveResultToFilename(StringValuePtr(rFilename), result, stylesheet, 0);
+  if (st == -1) {
+    xmlErrorPtr  err = xmlGetLastError();
+    rb_raise(rb_eIOError, "could not save stylesheet result file %s: %s", 
+      StringValuePtr(rFilename), err->message);
+  }
+           
+  return Qnil;
+}
+
+static VALUE stylesheet_save_to_string(VALUE self, VALUE rResult) {
+  xsltStylesheetPtr stylesheet = NULL;
+  xmlDocPtr result = NULL;
+  int st;
+  xmlChar *mem;
+  int size;
+  
+  Data_Get_Struct(self, xsltStylesheet, stylesheet);
+  Data_Get_Struct(rResult, xmlDoc, result);
+
+  st = xsltSaveResultToString(&mem, &size, result, stylesheet);
+  if (st == -1) {
+    xmlErrorPtr  err = xmlGetLastError();
+    rb_raise(rb_eStandardError, "could not save stylesheet result to string: %s", 
+      err->message);
+  }
+  VALUE rStr = rb_str_new2(mem);
+  xmlFree(mem);
+  
+  return rStr;
+}
+
 void Init_teius() {
 /*  LIBXML_TEST_VERSION   */
 
+  xmlSetGenericErrorFunc(NULL, null_handler);
+  xsltSetGenericErrorFunc(NULL, null_handler);
+
   /* Symbols */
   sRequired = rb_intern("required");
   sFirst    = rb_intern("first");
@@ -220,10 +342,20 @@ void Init_teius() {
   rb_define_method(cNode, "name",  node_name, 0);  
   rb_define_method(cNode, "xpath",  node_xpath, 0);  
   rb_define_method(cNode, "pointer",  node_pointer, 0);  
+  rb_define_method(cNode, "line",  node_line, 0);  
+  rb_define_method(cNode, "content",  node_content, 0);  
 
   /* Document */
   cDocument = rb_define_class_under(mTeius, "Document", cNode);
-  rb_define_singleton_method(cDocument, "parse_file",  parse_file, 1);  
-  rb_define_singleton_method(cDocument, "parse_string",  parse_string, 1);  
-  rb_define_method(cNode, "to_s",  document_to_s, 0);  
+  rb_define_singleton_method(cDocument, "parse_file",  document_parse_file, 1);  
+  rb_define_singleton_method(cDocument, "parse_string",  document_parse_string, 1);  
+  rb_define_method(cDocument, "to_s", document_to_s, 0);  
+  rb_define_method(cDocument, "root", document_root, 0);  
+  
+  /* XSL Stylesheet */
+  cStylesheet = rb_define_class_under(mTeius, "Stylesheet", cNode);
+  rb_define_singleton_method(cStylesheet, "parse_file",  stylesheet_parse_file, 1);  
+  rb_define_method(cStylesheet, "apply", stylesheet_apply, 1);  
+  rb_define_method(cStylesheet, "save_to_string", stylesheet_save_to_string, 1);  
+  rb_define_method(cStylesheet, "save_to_file", stylesheet_save_to_file, 2);  
 }

data/lib/teius.rb

@@ -1,17 +1,34 @@
-require 'teius.so'
+require File.join(File.dirname(__FILE__), '..', 'ext', 'teius.so')
 
 module Teius
+  class Document
+  
+    def self.content(xml)
+      parse_string(xml).content
+    end
+    
+    def content
+      root.content
+    end
+  end
+  
   class Node
     def fetch(xpath)
-      node = find(:unique, xpath, :required => true)
+      node = find(:unique, xpath, :required => false)
       return node ? node.value : nil
     end
     
+    def [](key)
+      attributes[key]
+    end
+    
     def ==(other)
+      return false unless other.kind_of? Node
       self.pointer == other.pointer
     end
 
     def eql?(other)
+      return false unless other.kind_of? Node
       self.pointer == other.pointer
     end
     
@@ -31,5 +48,25 @@ module Teius
       self.find :all, 'child::*'
     end
     
+    def to_s
+      "#{name}:#{line}"
+    end
+    
+  end
+  
+  class Stylesheet
+    
+    # If no filename is provided, will return the result as a String
+    def transform(doc, filename = nil)
+      result = apply doc
+      if filename
+        out = save_to_file result, filename   # returns nil
+      else
+        out = save_to_string result
+      end      
+      out
+    end
+    
   end
+  
 end

data/test/input.xml

@@ -0,0 +1,11 @@
+<?xml version="1.0"?>
+<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.1//EN" "http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd">
+<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en">
+     <head>
+          <title>Test document</title>
+     </head>
+     <body>
+          <h1>Test document</h1>
+          <p>This paragraph indicates the rest of the document.</p>
+   </body>
+</html>

data/test/teius_test.rb

@@ -1,6 +1,7 @@
 $:.unshift File.join(File.dirname(__FILE__), '..', 'lib')
 require 'test/unit'
 require 'teius'
+require 'tempfile'
 
 class TeiusTest < Test::Unit::TestCase
   include Teius
@@ -135,4 +136,61 @@ class TeiusTest < Test::Unit::TestCase
     assert_equal 9, sibs.size
   end
   
+  def test_xslt
+    current_dir = File.dirname(__FILE__)
+    test_doc_path = File.join(current_dir, 'input.xml')
+    xsl_path = File.join(current_dir, 'test.xsl')
+    doc = Document.parse_file test_doc_path
+    xsl = Stylesheet.parse_file xsl_path
+    assert_no_match %r{menu}, doc.to_s
+    assert_match %r{menu}, xsl.transform(doc)
+  end
+  
+  def test_xslt_write_to_file
+    current_dir = File.dirname(__FILE__)
+    test_doc_path = File.join(current_dir, 'input.xml')
+    xsl_path = File.join(current_dir, 'test.xsl')
+    doc = Document.parse_file test_doc_path
+    xsl = Stylesheet.parse_file xsl_path
+    Tempfile.open('xslt_test_file') do |tf|
+      outpath = tf.path
+      xsl.transform doc, outpath
+      str = tf.read
+      assert_match %r{menu}, str
+    end
+  end
+    
+  def test_xml_bad_file
+    assert_raise(Teius::ParseError) {
+      Document.parse_file File.join(File.dirname(__FILE__), 'bogus.xsl')
+    }
+  end
+
+  def test_xslt_bad_file
+    assert_raise(Teius::ParseError) {
+      Stylesheet.parse_file File.join(File.dirname(__FILE__), 'bogus.xsl')
+    }
+  end
+  
+  def test_root
+    root = @doc.root
+    assert_equal 'sports-content:2', root.to_s
+  end
+  
+  def test_node_content
+    node = @doc.find :first, '//team-metadata'
+    expected = [ '<team-metadata team-key="l.nfl.com-t.24" alignment="away">',
+      '  <name first="Green Bay" last="Packers"/>', '</team-metadata>' ].join "\n"
+    assert_equal expected, node.content
+  end
+  
+  def test_doc_content
+    assert_match /^<sports-content/, @doc.content
+  end
+
+  def test_doc_class_content
+    content = Document.content '<library><book/></library>'
+    assert_equal "<library>\n  <book/>\n</library>", content
+  end
+  
 end

data/test/test.xsl

@@ -0,0 +1,27 @@
+<?xml version="1.0"?>
+<xsl:stylesheet version="1.0"  xmlns:xhtml="http://www.w3.org/1999/xhtml"  xmlns="http://www.w3.org/1999/xhtml"  xmlns:xsl="http://www.w3.org/1999/XSL/Transform" xmlns:xs="http://www.w3.org/2001/XMLSchema" xmlns:fn="http://www.w3.org/2005/02/xpath-functions" xmlns:xdt="http://www.w3.org/2005/02/xpath-datatypes" exclude-result-prefixes="xhtml xsl fn xs xdt"> 
+<xsl:output method="xml"  version="1.0" encoding="UTF-8" doctype-public="-//W3C//DTD XHTML 1.1//EN" doctype-system="http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd" indent="yes"/>
+
+<!-- the identity template -->
+<xsl:template match="@*|node()">
+     <xsl:copy>
+          <xsl:apply-templates select="@*|node()"/>
+     </xsl:copy>
+</xsl:template>
+
+<!-- template for the head section. Only needed if we want to change, delete or add nodes. In our case we need it to add a link element pointing to an external CSS stylesheet. -->
+<xsl:template match="xhtml:head">
+     <xsl:copy>
+          <link rel="StyleSheet" href="xhtml_test.css" type="text/css"/>
+          <xsl:apply-templates select="@*|node()"/>
+     </xsl:copy>
+</xsl:template>
+
+<!-- template for the body section. Only needed if we want to change, delete or add nodes. In our case we need it to add a div element containing a menu of navigation. -->
+<xsl:template match="xhtml:body">
+     <xsl:copy>
+          <div class="menu"><p><a href="home">Homepage</a> &gt; <strong>Test document</strong></p></div>
+          <xsl:apply-templates select="@*|node()"/>
+     </xsl:copy>
+</xsl:template>
+</xsl:stylesheet>

metadata

@@ -3,11 +3,12 @@ rubygems_version: 0.8.11
 specification_version: 1
 name: teius
 version: !ruby/object:Gem::Version 
-  version: 0.1.0
-date: 2006-05-05 00:00:00 +02:00
+  version: "0.5"
+date: 2006-06-28 00:00:00 +03:00
 summary: Light-weight Ruby API to LibXML.
 require_paths: 
 - lib
+- dll
 email: joshmh@gmail.com
 homepage: http://teius.rubyforge.org
 rubyforge_project: teius
@@ -28,17 +29,15 @@ cert_chain:
 authors: 
 - Joshua Harvey
 files: 
-- ext/extconf.rb
-- ext/Makefile
-- ext/teius.c
-- ext/teius.o
-- ext/teius.so
-- lib/teius.rb
-- lib/teius.so
+- test/input.xml
 - test/teius_test.rb
+- test/test.xsl
 - xml/xt.3608774-update-mid-event.xml
 - xml/xt.3608787-update-post-event.xml
 - doc/LICENCE.txt
+- ext/extconf.rb
+- ext/teius.c
+- lib/teius.rb
 test_files: 
 - test/teius_test.rb
 rdoc_options: []

data/ext/Makefile

@@ -1,138 +0,0 @@
-
-SHELL = /bin/sh
-
-#### Start of system configuration section. ####
-
-srcdir = .
-topdir = d:/msys/local/lib/ruby/1.8/i386-mingw32
-hdrdir = $(topdir)
-VPATH = $(srcdir);$(topdir);$(hdrdir)
-
-DESTDIR = d:
-prefix = $(DESTDIR)/msys/local
-exec_prefix = $(DESTDIR)/msys/local
-sitedir = $(prefix)/lib/ruby/site_ruby
-rubylibdir = $(libdir)/ruby/$(ruby_version)
-archdir = $(rubylibdir)/$(arch)
-sbindir = $(exec_prefix)/sbin
-datadir = $(prefix)/share
-includedir = $(prefix)/include
-infodir = $(prefix)/info
-sysconfdir = $(prefix)/etc
-mandir = $(prefix)/man
-libdir = $(DESTDIR)/msys/local/lib
-sharedstatedir = $(prefix)/com
-oldincludedir = $(DESTDIR)/usr/include
-sitearchdir = $(sitelibdir)/$(sitearch)
-bindir = $(exec_prefix)/bin
-localstatedir = $(prefix)/var
-sitelibdir = $(sitedir)/$(ruby_version)
-libexecdir = $(exec_prefix)/libexec
-
-CC = gcc
-LIBRUBY = lib$(LIBRUBY_SO).a
-LIBRUBY_A = lib$(RUBY_SO_NAME)-static.a
-LIBRUBYARG_SHARED = -l$(RUBY_SO_NAME)
-LIBRUBYARG_STATIC = -l$(RUBY_SO_NAME)-static
-
-CFLAGS   =  -g -O2  
-CPPFLAGS = -I. -I$(topdir) -I$(hdrdir) -I$(srcdir)  -Id:/msys/local/include/libxml2 
-CXXFLAGS = $(CFLAGS) 
-DLDFLAGS =   -Wl,--enable-auto-import,--export-all 
-LDSHARED = gcc -shared -s
-AR = ar
-EXEEXT = .exe
-
-RUBY_INSTALL_NAME = ruby
-RUBY_SO_NAME = msvcrt-ruby18
-arch = i386-mingw32
-sitearch = i386-msvcrt
-ruby_version = 1.8
-ruby = d:/msys/local/bin/ruby
-RUBY = $(ruby)
-RM = rm -f
-MAKEDIRS = mkdir -p
-INSTALL = /bin/install -c
-INSTALL_PROG = $(INSTALL) -m 0755
-INSTALL_DATA = $(INSTALL) -m 644
-COPY = cp
-
-#### End of system configuration section. ####
-
-preload = 
-
-libpath = d:/msys/local/lib $(libdir)
-LIBPATH =  -L"d:/msys/local/lib" -L"$(libdir)"
-DEFFILE = 
-
-CLEANFILES = 
-DISTCLEANFILES = 
-
-extout = 
-extout_prefix = 
-target_prefix = 
-LOCAL_LIBS = 
-LIBS = $(LIBRUBYARG_SHARED) -lxml2  -lwsock32  
-SRCS = teius.c
-OBJS = teius.o
-TARGET = teius
-DLLIB = $(TARGET).so
-STATIC_LIB = 
-
-RUBYCOMMONDIR = $(sitedir)$(target_prefix)
-RUBYLIBDIR    = $(sitelibdir)$(target_prefix)
-RUBYARCHDIR   = $(sitearchdir)$(target_prefix)
-
-TARGET_SO     = $(DLLIB)
-CLEANLIBS     = $(TARGET).so $(TARGET).il? $(TARGET).tds $(TARGET).map
-CLEANOBJS     = *.o *.a *.s[ol] *.pdb *.exp *.bak
-
-all:		$(DLLIB)
-static:		$(STATIC_LIB)
-
-clean:
-		@-$(RM) $(CLEANLIBS:/=\) $(CLEANOBJS:/=\) $(CLEANFILES:/=\)
-
-distclean:	clean
-		@-$(RM) Makefile extconf.h conftest.* mkmf.log
-		@-$(RM) core ruby$(EXEEXT) *~ $(DISTCLEANFILES:/=\)
-
-realclean:	distclean
-install: install-so install-rb
-
-install-so: $(RUBYARCHDIR)
-install-so: $(RUBYARCHDIR)/$(DLLIB)
-$(RUBYARCHDIR)/$(DLLIB): $(DLLIB)
-	$(INSTALL_PROG) $(DLLIB) $(RUBYARCHDIR)
-install-rb: pre-install-rb install-rb-default
-install-rb-default: pre-install-rb-default
-pre-install-rb: Makefile
-pre-install-rb-default: Makefile
-$(RUBYARCHDIR):
-	$(MAKEDIRS) $@
-
-site-install: site-install-so site-install-rb
-site-install-so: install-so
-site-install-rb: install-rb
-
-.SUFFIXES: .c .m .cc .cxx .cpp .o
-
-.cc.o:
-	$(CXX) $(CXXFLAGS) $(CPPFLAGS) -c $<
-
-.cxx.o:
-	$(CXX) $(CXXFLAGS) $(CPPFLAGS) -c $<
-
-.cpp.o:
-	$(CXX) $(CXXFLAGS) $(CPPFLAGS) -c $<
-
-.c.o:
-	$(CC) $(CFLAGS) $(CPPFLAGS) -c $<
-
-$(DLLIB): $(OBJS)
-	@-$(RM) $@
-	$(LDSHARED) $(DLDFLAGS) $(LIBPATH) -o $@ $(OBJS) $(LOCAL_LIBS) $(LIBS)
-
-
-
-$(OBJS): ruby.h defines.h