W3C home > Mailing lists > Public > www-dom@w3.org > October to December 2000

Question about Range.surroundContents

From: Lynn Monson <lmonson@flipdog.com>
Date: Tue, 24 Oct 2000 15:17:12 -0600
To: <www-dom@w3.org>
Message-ID: <GAEKJDAJMGADKOMNOAFMGEBHCDAA.lmonson@whizbang.com>
I'm looking for clarification on how DOM 2 Ranges respond to document
mutations triggered by calling Range.surroundContents.  If anyone could
offer some clarification, I would appreciate it.

Suppose I have a text node in an XHTML document:


And I have two ranges, that both lie within this text node:

  start offset=3
  end offset=6
  (r1 selects "Bar")

  start offset=6
  end offset=9
  (r2 selects "Baz")

If I attempt to bold r1 by invoking r1.surroundContents() with a <b/>
element, I get the following result:


It's clear from the DOM 2 spec that r1 then selects <b>Bar</b>, but I'm not
certain what r2 selects.  If I read section 2.10 literally, performing
surroundContents in terms of other operations, and letting r2 respond to
each mutation event, I wind up with r2 as follows:

  start container: Foo
  start offset: 3
  end container: Baz
  end offset: 3
  (r2 selects from the end of Foo, through <b>Bar</b>, to the end of Baz)

It seems counter intuitive that r2 would "jump" to include the (now bolded)
"Bar" text.  In addition it seems to violate the principle stated in section
2.12 of the DOM 2 spec:

"There are two general principles which apply to Ranges under document
mutation: ...
the second is that, as much as possible, all Ranges will select the same
portion of the document after any mutation operation."

That would seem to imply that since my outermost operation was
r1.surroundContents, r2 ought to select "Baz" as it did before.  Have I read
section 2.10 correctly?  If so, this seems in violation of 2.12.

Given that I have multiple Ranges, I'm not certain if a literal reading of
section 2.10 produces the right outcome.  Could anyone clarify what was

Thanks for any feedback
Lynn Monson
Received on Tuesday, 24 October 2000 17:22:15 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 20 October 2015 10:46:07 UTC