views:

456

answers:

4

This is really weird... When I open the following simple HTML document in Internet Explorer 7.0.5730.11 (on Windows Server 2003 Web Edition SP2)

<html>
  <body>
    <p>+</p>
  </body>
</html>

it shows me a totally blank page. FWIW, this is just a trivial "repro" sample. In real HTML documents, I observed other, even more bizzarre effects caused by presense of the "plus" character that follows a tag.

NB: The problem appears to be extremely ittermittent. Most of the time it does work properly (i.e. displays the "plus" character), and I still can't find any way to reproduce this problem at will.

Some additional details based on recent comments:

  • There was no server involved. I was opening a file on disk (i.e. used file:// protocol).

  • The file did not contain anything except five lines shown above. No document type declarations, no character encodings, no nothings.

Looks like a bug in IE. Did anybody encounter the same or similar problem?

NB: I appreciate all the responses received so far, but neither of respondednts encountered this problem. Something tells me that 99.(9)% of StackOverflow audience will not be able to reproduce it. :-)

+1  A: 

Does it work if you use the numeric character reference notation?

<html>
  <body>
     <p>&#43;</p>
  </body>
</html>
mhawke
Yes, it does work when encoded this way.
Yarik
A: 

Does it work if you use a Doctype? IE does get a bit picky if you don't use a doctype (insert no-right-to-be-picky pun here).

By intermittent do you mean using the same code it appears and doesn't? That sounds really strange.

CLOSED - NOT REPRO... er I mean I only get the +, no matter how many times I refresh. I suggest using the HTML entity reference - but this might be a problem with your system/browser if others can't reproduce either.

Ross
Yes, that's exactly what I mean by "intermittent"...And today it is totally unreproducable on my system. Yes, it does look like some problem with my system/browser.
Yarik
Sorry, can't try your idea about DOCTYPE right now... Will try as soon as the bug rears its stinky head again.
Yarik
A: 

For whatever it's worth, I just tested this on IE 7 (7.0.5730.13C0) and it consistently displays the "+" even with several refreshes (at least 10 or 12). You didn't mention an OS but in my case it's Windows XP SP2 (Help About displays Version 5.1 (Build 2600.xpsp_sp2_qfe.070227-2300: Service Pack 2). The OS may make a difference in this case.

Onorio Catenacci
True. :-( FWIW, I have Windows Server 2003 Web Edition SP2.
Yarik
A: 

It's possible that this is due to the server, particularly if it's trying to parse the page as a script. To check:

  • What HTTP headers do you see when the effect occurs?
  • When you "View Source" at that point, what do you see?
  • Does the effect ever occur when you load the page directly as a file?
  • Does the effect ever occur in other browsers?
eswald
There was no server involved: I was opening an HTML file on a disk (i.e. using **file://** protocol). "View Source" shows exactly the right source. I did not try any other browsers.
Yarik