224988 - microsoft.com - JS throwing an error on unescaped "/" character in RegExp literals

Status: RESOLVED DUPLICATE of bug 309840

(Tech Evangelism Graveyard :: English US, defect)

Description

[Olivier Cahagne]

Build ID: 20031105

Not sure if this is a regression, I'll test an earlier build, pre-patch for bug
223273.

On URL:
http://msdn.microsoft.com/workshop/author/dhtml/overview/ccomment_ovw.asp

JavaScript console shows:
Error: unterminated character class 
Source File: http://msdn.microsoft.com/workshop/code/browdata.js
Line: 26, Column: 31
Source Code:
	else if
(this.userAgent.match(/Mozilla[/].*(95[/]NT|95|NT|98|3.1).*Opera.*(\d+)\.(\d+)/))

As this is new to me (I hadn't noticed this before on msdn) and look related to
bug 223273, this could be a regression.

I reduced it to:

js> /foo[/].(\d+)/
Error: unterminated character class 

Not being a JS expert, I'm not sure if it's mozilla's fault, I'll check with an
earlier build asap.

Comment 1

[Phil Schwartau]

JS is doing the right thing in throwing this error. See bug 173067:
'Using "/" character unescaped in RegExp literals causes an error'

Also see bug 98306 comment 31, and following comments.

Bug 173067 is ready to be marked Invalid, but has been held open
due to discrepancies we noticed in our error text.

We filed an Evangelism bug on Microsoft for this issue: bug 101070,
'microsoft.com - Using "/" character unescaped in RegExp literals is non-ECMA'

That bug was marked fixed, because it seemed like MS had fixed things.
But it looks like they haven't gotten all instances of this. 

Olivier: thank you for this report. I'm going to reassign this to 
Tech Evangelism for disposition. It is either a dupe of bug 101070,
which would then have to be reopened, or it could stand alone -

Comment 2

[Felix Miata]

Conforming summary to TFM item 10 at 
http://www.mozilla.org/projects/tech-evangelism/site/procedures.html#file-new

Comment 3

[Jesse Ruderman]

Fixed in Firefox, see bug 309840.

Comment 4

[Olivier Cahagne]

*** This bug has been marked as a duplicate of 309840 ***