Skip to content

Navigation Menu

Sign in
Appearance settings

Search code, repositories, users, issues, pull requests...

Provide feedback

We read every piece of feedback, and take your input very seriously.

Saved searches

Use saved searches to filter your results more quickly
Sign up
Appearance settings

Commit 42a59a8

Browse files
Merge pull request #658 from dubinsky/xinclude-demonstration
Demonstrate XInclude work - and fail
2 parents 6486933 + 7b57916 commit 42a59a8

File tree

8 files changed

+88
-3
lines changed

8 files changed

+88
-3
lines changed

‎build.sbt‎

Lines changed: 1 addition & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -113,6 +113,7 @@ lazy val xml = crossProject(JSPlatform, JVMPlatform, NativePlatform)
113113
libraryDependencies += "junit" % "junit" % "4.13.2" % Test,
114114
libraryDependencies += "com.github.sbt" % "junit-interface" % "0.13.3" % Test,
115115
libraryDependencies += "org.apache.commons" % "commons-lang3" % "3.12.0" % Test,
116+
libraryDependencies += "xerces" % "xercesImpl" % "2.12.2" % Test,
116117
libraryDependencies ++= (CrossVersion.partialVersion(scalaVersion.value) match {
117118
case Some((3, _)) =>
118119
Seq()
Lines changed: 3 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,3 @@
1+
<store xmlns:xi="http://www.w3.org/2001/XInclude">
2+
<xi:include href="books/book/author.xml"/>
3+
</store>
Lines changed: 3 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,3 @@
1+
<store xmlns:xi="http://www.w3.org/2001/XInclude">
2+
<xi:include href="author/volume/1.xml"/>
3+
</store>
Lines changed: 1 addition & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1 @@
1+
<collection n="1"/>
Lines changed: 3 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,3 @@
1+
<includee>
2+
<content>Blah!</content>
3+
</includee>
Lines changed: 3 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,3 @@
1+
<includer>
2+
<xi:include xmlns:xi="http://www.w3.org/2001/XInclude" href="includee.xml"/>
3+
</includer>
Lines changed: 3 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,3 @@
1+
<site xmlns:xi="http://www.w3.org/2001/XInclude">
2+
<xi:include href="archive/books.xml"/>
3+
</site>

‎jvm/src/test/scala/scala/xml/XMLTest.scala‎

Lines changed: 71 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -510,8 +510,15 @@ class XMLTestJVM {
510510
}
511511
}
512512

513+
// With both internal and external Xerces now on the classpath, we explicitly disambiguate which one we want:
514+
def xercesInternal: javax.xml.parsers.SAXParserFactory =
515+
javax.xml.parsers.SAXParserFactory.newInstance("com.sun.org.apache.xerces.internal.jaxp.SAXParserFactoryImpl", null)
516+
517+
def xercesExternal: javax.xml.parsers.SAXParserFactory =
518+
javax.xml.parsers.SAXParserFactory.newInstance("org.apache.xerces.jaxp.SAXParserFactoryImpl", null)
519+
513520
/** Default SAXParserFactory */
514-
val defaultParserFactory: javax.xml.parsers.SAXParserFactory = javax.xml.parsers.SAXParserFactory.newInstance
521+
val defaultParserFactory: javax.xml.parsers.SAXParserFactory = xercesInternal
515522

516523
@throws(classOf[org.xml.sax.SAXNotRecognizedException])
517524
def issue17UnrecognizedFeature(): Unit = {
@@ -629,7 +636,7 @@ class XMLTestJVM {
629636
// using namespace-aware parser, this works with FactoryAdapter enhanced to handle startPrefixMapping() events;
630637
// see https://github.com/scala/scala-xml/issues/506
631638
def roundtrip(namespaceAware: Boolean, xml: String): Unit = {
632-
val parserFactory: javax.xml.parsers.SAXParserFactory = javax.xml.parsers.SAXParserFactory.newInstance()
639+
val parserFactory: javax.xml.parsers.SAXParserFactory = xercesInternal
633640
parserFactory.setFeature("http://javax.xml.XMLConstants/feature/secure-processing", true)
634641
parserFactory.setFeature("http://apache.org/xml/features/nonvalidating/load-external-dtd", false)
635642
parserFactory.setFeature("http://apache.org/xml/features/disallow-doctype-decl", true)
@@ -656,7 +663,7 @@ class XMLTestJVM {
656663

657664
@UnitTest
658665
def useXMLReaderWithXMLFilter(): Unit = {
659-
val parent: org.xml.sax.XMLReader = javax.xml.parsers.SAXParserFactory.newInstance.newSAXParser.getXMLReader
666+
val parent: org.xml.sax.XMLReader = xercesInternal.newSAXParser.getXMLReader
660667
val filter: org.xml.sax.XMLFilter = new org.xml.sax.helpers.XMLFilterImpl(parent) {
661668
override def characters(ch: Array[Char], start: Int, length: Int): Unit = {
662669
for (i <- 0 until length) if (ch(start+i) == 'a') ch(start+i) = 'b'
@@ -682,6 +689,67 @@ class XMLTestJVM {
682689
assertTrue(gotAnError)
683690
}
684691

692+
// Now that we can use XML parser configured to be namespace-aware,
693+
// we can also configure it to be XInclude-aware and process XML Includes:
694+
def check(
695+
parserFactory: javax.xml.parsers.SAXParserFactory,
696+
resourceName: String,
697+
expected: String
698+
): Unit = {
699+
parserFactory.setNamespaceAware(true)
700+
parserFactory.setXIncludeAware(true)
701+
val actual: String = XML
702+
.withSAXParser(parserFactory.newSAXParser)
703+
.load(getClass.getResource(resourceName).toString)
704+
.toString
705+
706+
assertEquals(expected, actual)
707+
}
708+
709+
// Here we demonstrate that XInclude works with both the external and the built-in Xerces:
710+
711+
val includerExpected: String =
712+
s"""<includer>
713+
| <includee xml:base="includee.xml">
714+
| <content>Blah!</content>
715+
|</includee>
716+
|</includer>""".stripMargin
717+
718+
@UnitTest def xIncludeWithExternalXerces(): Unit = check(xercesExternal, "includer.xml", includerExpected)
719+
@UnitTest def xIncludeWithInternalXerces(): Unit = check(xercesInternal, "includer.xml", includerExpected)
720+
721+
// And here we demonstrate that both external and built-in Xerces report incorrect `xml:base`
722+
// when the XML file included contains its own include, and included files are not in the same directory:
723+
// `xml:base` on the `<collection>` element is incorrect
724+
// books/book/author/volume/1.xml instead of the correct
725+
// archive/books/book/author/volume/1.xml!
726+
val siteUnfortunatelyExpected: String =
727+
s"""<site xmlns:xi="http://www.w3.org/2001/XInclude">
728+
| <store xml:base="archive/books.xml" xmlns:xi="http://www.w3.org/2001/XInclude">
729+
| <store xml:base="archive/books/book/author.xml" xmlns:xi="http://www.w3.org/2001/XInclude">
730+
| <collection n="1" xml:base="books/book/author/volume/1.xml"/>
731+
|</store>
732+
|</store>
733+
|</site>""".stripMargin
734+
735+
// Turns out, this is a known Xerces bug https://issues.apache.org/jira/browse/XERCESJ-1102:
736+
// - the bug was reported in October 2005 - more then seventeen years ago;
737+
// - a patch fixing it (that I have not verified personally) was submitted many years ago;
738+
// - the bug is still not fixed in the 2023 release of Xerces;
739+
// - the bug was discussed by the Saxon users in https://saxonica.plan.io/issues/4664,
740+
// and is allegedly fixed in SaxonC 11.1 - although how can this be with Saxon not shipping its own Xerces is not clear.
741+
//
742+
// In my own application, I had to "fix up" incorrect values produced by Xerces, taking into account
743+
// specific directory layout being used. I can only speculate what others do, but none of the alternatives sound great:
744+
// - avoid using nested includes altogether or flatten the directory hierarchy to appease the bug;
745+
// - use privately patched version of Xerces;
746+
// - use Saxon DOM parsing instead of Xerces' SAX.
747+
//
748+
// I find it utterly incomprehensible that foundational library shipped with JDK and used everywhere
749+
// has a bug in its core functionality for years and it never gets fixed, but sadly, it is the state of affairs:
750+
@UnitTest def xIncludeFailWithExternalXerces(): Unit = check(xercesExternal, "site.xml", siteUnfortunatelyExpected)
751+
@UnitTest def xIncludeFailWithInternalXerces(): Unit = check(xercesInternal, "site.xml", siteUnfortunatelyExpected)
752+
685753
@UnitTest
686754
def nodeSeqNs(): Unit = {
687755
val x: NodeBuffer = {

0 commit comments

Comments
(0)

AltStyle によって変換されたページ (->オリジナル) /