W3C home > Mailing lists > Public > www-style@w3.org > March 2013

Re: [css3-page] Add way for webpage to signal UA to disable default footer/header

From: Julian Viereck <julian.viereck@googlemail.com>
Date: Sun, 17 Mar 2013 22:03:23 +0100
Message-ID: <51462F9B.3040001@googlemail.com>
To: Simon Sapin <simon.sapin@exyr.org>, spec:w3c:style <www-style@w3.org>, Daniel Glazman <daniel.glazman@disruptive-innovations.com>, Robert O'Callahan <robert@ocallahan.org>
On 3/4/13 5:01 PM, Simon Sapin wrote:
> Le 03/03/2013 00:54, Julian Viereck a écrit :
>> - Does adding a flag like "nomarginboxes" makes sense?
>
> What’s a flag in this context? Is it something that can be switched 
> from JavaScript? From CSS? 
In the current implementation, it's something on the HTML structure and 
therefore can be toggled by JS. However, I think this flags makes more 
sense to be set from within CSS. E.g. how about:

``` CSS
html {
   default-margin-boxes: no;
}
```

... yes, I know this is not the way it should end up but I cannot think 
of a better naming ATM. Better ideas welcome :)


On 3/5/13 10:38 PM, Simon Sapin wrote:
> Le 05/03/2013 21:44, Daniel Glazman a écrit :
>> On 04/03/13 17:01, Simon Sapin wrote:
>>
>>> So another option could be, as you say, to completely disable the
>>> default headers and footers in some situation:
>>>
>>> a. If at least one page-margin box is generated [1] on the current 
>>> page.
>>> (I don’t like the inconsistency between pages of the document)
>>>
>>> b. … on any page of the same document (costly to implement, you need to
>>> look for "propagated" values of the 'page' property in the box tree.)
>>>
>>> c. If any margin at-rule is used in author stylesheets (simpler)
>>>
>>> d. With an explicit switch, maybe a new property in the @page context.
>>> (Not very elegant.)
>>>
>>> e. Some other condition?
>>>
>>> [1] http://dev.w3.org/csswg/css3-page/#generated
>>>
>>>
>>> What do you think?
>>
>> I think we're on a dangerous path here, where author's settings
>> could override user defined settings... In our case, have a document
>> tell the UA to override the header/footer settings specified in the
>> print dialog. Careful, careful.
>>
>> In fact, the best option would be to
>>
>> 0. implement @page and the margin boxes rules
>> 1. have browsers implement the print dialog's current 6 margin boxes
>>      through a UA @page and page margin boxes
>> 2. a radio in the print dialog "let the document specify headers
>>      and footers" and another radio "override document headers/footers"
>>
>> That way all is based on the cascade and the user can always have
>> all the various settings applied.
>
> How about this? The user still ultimately has control on what headers 
> and footers are printed, but <some condition, see above> only changes 
> the default.
How about this (only slightly different to previous raised approach):

   Assume there is at least one margin-box defined in the document. The 
margin-box might have `content: none`. Then:

   - the UA should disable it's default header/footer but let the user 
enable them if wanted
   - the UA should provide a way to disable all the margin-boxes defined 
on the page. This might be done by using a checkbox "Ignore document 
specify headers and footers".

With this, there is no need for an extra "disable the margin-boxes by 
default" flag like 'nomarginboxes'.

Problem: What is a "document". E.g. the user might select only a range 
of pages to be printed - is a "document" then only the selected pages OR 
all the possible pages that can be printed?
> In both cases, I don’t think the spec should have normative 
> requirements for UAs, only suggestions.
Having only suggestions in the spec sounds good to me.

- Julian
Received on Sunday, 17 March 2013 21:03:53 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 17:21:07 GMT