Fix duplicate responses in XPath following, following-sibling, preceding, preceding-sibling

[naitoh]

Why?

See: #251 (comment)

Expected values

• XPath : a/d/preceding::* => ["d", "c", "b"]

<a>
  <b/> <!-- a/d/preceding::b -->
  <c/> <!-- a/d/preceding::c -->
  <d/> <!-- a/d/preceding::d -->
  <d/> <!-- self -->
  <e/>
  <f/>
</a>

• XPath : a/d/following::* => ["d", "e", "f"]

<a>
  <b/>
  <c/>
  <d/> <!-- self -->
  <d/> <!-- a/d/following::d -->
  <e/> <!-- a/d/following::e -->
  <f/> <!-- a/d/following::f -->
</a>

• XPath : a/b/x/following-sibling:* => ["c", "d", "e"]

<a>
  <b>
    <x/> <!-- self -->
    <c/> <!-- a/b/x/following-sibling::c -->
    <d/> <!-- a/b/x/following-sibling::d -->
  </b>
  <b>
    <x/> <!-- self -->
    <e/> <!-- a/b/x/following-sibling::e -->
  </b>
</a>

• XPath : a/b/x/following-sibling:* => ["c", "d", "x", "e"]

<a>
  <b>
    <x/> <!-- self -->
    <c/> <!-- a/b/x/following-sibling::c -->
    <d/> <!-- a/b/x/following-sibling::d -->
    <x/> <!-- a/b/x/following-sibling::x -->
    <e/> <!-- a/b/x/following-sibling::e -->
  </b>
</a>

• XPath : a/b/x/preceding-sibling::* => ["e", "d", "c"]

<a>
  <b>
    <c/>  <!-- a/b/x/preceding-sibling::c -->
    <d/>  <!-- a/b/x/preceding-sibling::d -->
    <x/>  <!-- self -->
  </b>
  <b>
    <e/>  <!-- a/b/x/preceding-sibling::e -->
    <x/>  <!-- self -->
  </b>
</a>

• XPath : a/b/x/preceding-sibling::* => ["e", "x", "d", "c"]

<a>
  <b>
    <c/>  <!-- a/b/x/preceding-sibling::c -->
    <d/>  <!-- a/b/x/preceding-sibling::d -->
    <x/>  <!-- a/b/x/preceding-sibling::x -->
    <e/>  <!-- a/b/x/preceding-sibling::e -->
    <x/>  <!-- self -->
  </b>
</a>

• XPath : //a/following-sibling:*[1] => ["w", "x", "y", "z"]

<div>
  <div>
    <a/> <-- self -->
    <w/> <-- //a/following-sibling:*[1] -->
  </div>
  <a/> <-- self -->
  <x/> <-- //a/following-sibling:*[1] -->
  <a/> <-- self -->
  <y/> <-- //a/following-sibling:*[1] -->
  <a/> <-- self -->
  <z/> <-- //a/following-sibling:*[1] -->
</div>

[tompng]

It has a breaking change.

d = REXML::Document.new('<div>
  <div>
    <a/><w/>
  </div>
  <a/><x/>
  <a/><y/>
  <a/><z/>
</div>')
# Finds a node flowing <a/>
REXML::XPath.match(d, '//a/following-sibling::*[1]')
# Before: [<w/>, <x/>, <y/>, <z/>]
# After: [<w/>, <x/>]

For reference, the behavior in master branch is same as JavaScript xpath.

document.body.innerHTML = `
<div>
  <div><a></a><w></w></div>
  <a></a><x></x>
  <a></a><y></y>
  <a></a><z></z>
</div>`
result = document.evaluate('//a/following-sibling::*[1]', document, null, XPathResult.ORDERED_NODE_SNAPSHOT_TYPE, null)
for(i=0;i<result.snapshotLength;i++) console.log(i, result.snapshotItem(i))
// 0 <w></w>
// 1 <x></x>
// 2 <y></y>
// 3 <z></z>

[naitoh]

It has a breaking change.

Thanks!

I have corrected the issue you mentioned by removing the duplicate.

[naitoh]

I have corrected the points you pointed out.
What do you think?

[naitoh]

I have corrected the points you pointed out.
What do you think?

[kou]

• Let's use XXX(YYY) style not XXX( YYY ) style for newly added codes
• Let's use doc or document instead of d because we have <d/> in this context

Suggested change

	d = REXML::Document.new( source )
	doc = REXML::Document.new(source)

[naitoh]

OK.
I see.

[kou]

Suggested change

	matches = REXML::XPath.match(d, "a/d/preceding::*")
	matches = REXML::XPath.match(doc, "a/d/preceding::*")

[naitoh]

OK.
I see.

[naitoh]

I have corrected the points you pointed out.
What do you think?

[kou]

Thanks.