105640 - Expose the mode [strict/quirks] to web content

Status: RESOLVED FIXED

(Core :: DOM: Core & HTML, defect, P3)

Description

[Fabian Guisset]

From Bugzilla Helper:
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:0.9.4+)
Gecko/20010919
BuildID:    20011019

I filed this bug after Stephan Tolksdorf's newsgroup post "Strict and standards
compliant mode" of oct 17th.
It links to an article (see the URL field) that claims that the best way to know
whether we're in strict or quirks mode is this:

var d=document.doctype
strict=(d&&d.systemId)?(d.systemId.indexOf("strict")>-1)?true:(d.publicId.indexOf("transitional")>-1)?true:false:(d.publicId.indexOf("transitional")==-1)?true:false

If this is true (and I believe it is), it is my opinion that we should expose
the mode as a simple property of the document.
IE has a property called document.compatMode. It takes two values:
CSS1Compat and BackCompat (strings). Maybe we could implement that.

I'm just filing this bug to make sure we take a decision instead of ignoring it.
Feel free of course to mark wontfix.

Reproducible: Always
Steps to Reproduce:
1. Load above
2.
3.

Actual Results:  Blah

Expected Results:  Bleh

Comment 1

[Tim Powell]

It'd be nice to provide this information to JavaScript/DOM. I'm not sure that
Microsoft's property and string results make sense for Mozilla, but it may be
helpful to web developers if Mozilla provides the same property. Using the same
property and BackCompat result would probably be sufficient. We could return
something different when in strict mode.

Microsoft's documentation about compatMode is here:
http://msdn.microsoft.com/workshop/author/dhtml/reference/properties/compatmode.asp

Comment 2

[Fabian Guisset]

Attachment: Proposed fix. Let me know if we want this for XML docs as well.

Comment 3

[Jonas Sicking (:sicking) No longer reading bugmail consistently]

Comment on attachment 55377 [details] [diff] [review]
Proposed fix. Let me know if we want this for XML docs as well.

IMHO we don't want this on XML documents. They should always be in strict mode AFAIK

r=sicking

Comment 4

[Fabian Guisset]

Attachment: Better fix.

Comment 5

[Johnny Stenback (:jst)]

Comment on attachment 56787 [details] [diff] [review]
Better fix.

sr=jst

Comment 6

[Jonas Sicking (:sicking) No longer reading bugmail consistently]

Comment on attachment 56787 [details] [diff] [review]
Better fix.

r=sicking

Comment 7

[Fabian Guisset]

-> Mozilla 0.9.7, not a feature critical for 0.9.6.
Reassigning to myself, though I hope jst will check it in when the tree opens :-)

Comment 8

[Hixie (not reading bugmail)]

I don't think we want to return CSS1Compat. That would leave no clear way for
authors to determine if the layout was IE's CSS1Compat or Gecko's CSS1Compat. I
think it would be better to return "GeckoStrict". or some such.

Comment 9

[Fabian Guisset]

I think it's a good idea.

Comment 10

[Johnny Stenback (:jst)]

Determining if a script runs in IE or Gecko is something script writers do any
way, I think we should be compatible with IE wrt this string so that people
don't need to check for two strings if they don't need to distinguish between IE
and Gecko, which they can do already anyway. The biggest reason for adding
support for this property was to be compatible with IE, and if we start using
different strings we're not exactly compatible wrt this property.

I vote for using the patch we already have.

Comment 11

[Fabian Guisset]

Marking, like, FIXED!
thanks for the reviews and the checkin :-)

Comment 12

[Henrik Gemal]

FYI: http://gemal.dk/browserspy/document.html now also shows the "compatMode"
mode...