1
0
mirror of https://github.com/ezyang/htmlpurifier.git synced 2025-07-31 19:30:21 +02:00

Improve auto-paragraph to preserve newlines and handle edge-cases better.

This is a very large commit that includes numerous improvements to the
AutoParagraph injector.  These are:

* Rewritten flow control of the injector to use almost exclusively
  binary conditionals.
* Improved inline documentation with "State" comments, which give concise
  examples of what the token stack looks like at flow points.
* Documentation for all flow branches, even those with no actions.
* Factoring out of common operations to improve readability, especially the
  new iterator private methods.
* Expanded test-suite which covers new flow points, and corrects some errors
  in previous cases.

Signed-off-by: Edward Z. Yang <edwardzyang@thewritingpot.com>
This commit is contained in:
Edward Z. Yang
2008-08-10 00:32:29 -04:00
parent 0423985b45
commit 617f70a8ac
6 changed files with 515 additions and 207 deletions

View File

@@ -72,14 +72,18 @@ class HTMLPurifier_Strategy_MakeWellFormed_InjectorTest extends HTMLPurifier_Str
function testTwoParagraphsContainingOnlyOneLink() {
$this->assertResult(
"http://example.com\n\nhttp://dev.example.com",
'<p><a href="http://example.com">http://example.com</a></p><p><a href="http://dev.example.com">http://dev.example.com</a></p>'
'<p><a href="http://example.com">http://example.com</a></p>
<p><a href="http://dev.example.com">http://dev.example.com</a></p>'
);
}
function testParagraphNextToDivWithLinks() {
$this->assertResult(
'http://example.com <div>http://example.com</div>',
'<p><a href="http://example.com">http://example.com</a> </p><div><a href="http://example.com">http://example.com</a></div>'
'<p><a href="http://example.com">http://example.com</a> </p>
<div><a href="http://example.com">http://example.com</a></div>'
);
}
@@ -92,24 +96,50 @@ class HTMLPurifier_Strategy_MakeWellFormed_InjectorTest extends HTMLPurifier_Str
function testParagraphAfterLinkifiedURL() {
$this->assertResult(
"http://google.com\n\n<b>b</b>",
"<p><a href=\"http://google.com\">http://google.com</a></p><p><b>b</b></p>"
"http://google.com
<b>b</b>",
"<p><a href=\"http://google.com\">http://google.com</a></p>
<p><b>b</b></p>"
);
}
function testEmptyAndParagraph() {
// This is a fairly degenerate case, but it demonstrates that
// the two don't error out together, at least.
// Change this behavior!
$this->assertResult(
"<p>asdf\n\nasdf<b></b></p>\n\n<p></p><i></i>",
"<p>asdf</p><p>asdf</p>"
"<p>asdf
asdf<b></b></p>
<p></p><i></i>",
"<p>asdf</p>
<p>asdf</p>
"
);
}
function testRewindAndParagraph() {
// perhaps change the behavior of this?
$this->assertResult(
"bar\n\n<p><i></i>\n\n</p>\n\nfoo",
"<p>bar</p><p>foo</p>"
"bar
<p><i></i>
</p>
foo",
"<p>bar</p>
<p></p>
<p>foo</p>"
);
}