<p> and <div> and breaks

[Gus Richter]

Re: &lt;p&gt; and &lt;div&gt; and breaks

Jukka K. Korpela wrote:

Scripsit Gus Richter:
>

>I see the use of typical as "among all available" browsers and not of
>"all" browsers in use out there.

>
I see "typical" as meaning common, usual, or characteristic.

Right, and among all the "different" browsers. Really ridiculous to mean
the total volume of all browsers in use, other than to just be
argumentative.

>>There's no law against a browser having a default style sheet like
>>p { margin: 1.3em 0 0 0; }

>>
>But then it would not follow the "Double Line Leading" convention for
>paragraphs.

>
So what? That convention is bad style anyway ("engineer's paragraphs"),
and, what matters here, just a convention, not part of HTML (or CSS)
specifications.

I happen to not like the Double Line Leading convention for paragraphs,
but it is here and encouraged by W3C. Don't ask for proof for I won't
bother to search for it. You know that this is so.

Your rule for top-margin only as a default for paragraphs would confuse
a document and is truly grasping for argumentative straws.

>The subject in the specification is regarding a DIV being on a "New
>Line" and the method used by the browsers is the placing of
>line breaks before and after DIVs. Example markup is presented and how
>it will be rendered.

>
No, how it is _typically_ rendered.

You are predictable. I left it out on purpose for you.

>Whether it is an error or whatever due to
>carelessness , it is wrong.

>
No, it contains line breaks just where the text says they are. Whether
it also contains vertical spacing is a different issue. And they made a
mistake in confusing the two.

I say that it is an error and that it is wrong. You say No, but hedge
your bet by adding at the end that they made a mistake. It seems to me
that it is you that is confusing the two issues. Vertical spacing is the
issue in this thread as demonstrated in an example where the issue is
about line breaks.

In an example of rendering for line breaks, the example also shows
vertical spacing as "Double Line Leading" in one instance, but not in
the other instance. Perhaps they should have used a different example,
but they didn't and by indicating that it is a typical browser rendering
is wrong, an error and a mistake.

--
Gus

[Lars Eighner]

Re: &lt;p&gt; and &lt;div&gt; and breaks

In our last episode,
<7ooio2tstkcjhh 0daimru6870plkc 2dfcn@4ax.com>,
the lovely and talented Kent Feiler
broadcast on comp.infosystem s.www.authoring.html:

On Tue, 19 Dec 2006 14:09:48 -0600, Kent Feiler <zzzz@zzzz.comw rote:

1. Here's some html from a W3C recommendations page.

<P>aaaaaaaaa<DI V>bbbbbbbbb</DIV><DIV>cccccc cc<P>dddddddd</DIV>
------------------------------------------------------------------------------

I have to say that this newsgroup seems to have a higher percentage of
obnoxious creeps who post to it than most. But maybe I should check
out some of the religious newgroups, it may turn out that everyone
here is really quite polite and courteous!

When you have not made the least effort to learn HTML, what do you expect?

To begin, I was curious as to: (1)why this snippet of html wasn't
producing the results I thought it would, and (2)why IE7 and FF2 were
producing different results. In the end, the answer to (2) is the
usual, "That's the way it is, sukka. Suffer!"

There are plenty of headaches when browsers do not interpret styles the same
way. That they have different defaults for unstyled elements is not really
a problem.

But I'm still confused about (1). Where does the first <pend?

Since P cannot contain a block element and DIV is a block element,
P is closed when the first DIV is opened.

For FF2 it appears to end at the first <div>, but why?

As you would know if you had ever looked at a spec, P cannot contain
a block element. The browser assumes you must have meant to close
P when you opened DIV, because DIV in P is impossible.

Is it because of a syntax error,

It isn't really a syntax error in HTML. In HTML many elements do not
require closing tags. If you are thinking you are putting a DIV in P,
you are just wrong, but <Pblah, blah, blah. <DIVis not incorrect syntax.
It is just syntax that doesn't mean what you think it does.

i.e. <div>s are not allowed within <pso
FF terminates the first <pwith the normal double break and then
continues processing?

Thinking about "double breaks" is misleading. A browser could reasonably
have a default that puts a single break when P closes and and a break with
an indent when P opens. It is not about how browsers choose to render
paragraphs. It is about the structure of the document.

IE7 looks like it doesn't end the first <puntil it hits the second
one and processes the two internal <div>s with a single break. I would
have thought it would discard the syntactically invalid <divtags and
produce something like:

The <DIV>s are not syntactically incorrect (in HTML). They just do
not mean what you think they do.

aaaaaaaaaabbbbb bbbbbcccccccccc

dddddddddd

...but instead it seems to ignore the idea the <div>s aren't allowed
within <p>s. Maybe it didn't attend the meeting. Or maybe it doesn't
think it should be in the business of enforcing syntax errors.

If IE doesn't assume P is closed when DIV opens, it is wrong.

Does that sound like what's happening?

--
Lars Eighner <http://larseighner.com/ <http://myspace.com/larseighner>
Consider what might be fertilizing the greener grass across the fence.

[Kent Feiler]

Re: &lt;p&gt; and &lt;div&gt; and breaks

On 20 Dec 2006 22:42:25 GMT, Lars Eighner <usenet@larseig hner.com>
wrote:

In our last episode,
<7ooio2tstkcjhh 0daimru6870plkc 2dfcn@4ax.com>,
the lovely and talented Kent Feiler
broadcast on comp.infosystem s.www.authoring.html:

On Tue, 19 Dec 2006 14:09:48 -0600, Kent Feiler <zzzz@zzzz.co m>

wrote:

1. Here's some html from a W3C recommendations page.

<P>aaaaaaaaa<DI V>bbbbbbbbb</DIV><DIV>cccccc cc<P>dddddddd</DIV>
>

------------------------------------------------------------------------------

I have to say that this newsgroup seems to have a higher percentage
of obnoxious creeps who post to it than most. But maybe I should
check out some of the religious newgroups, it may turn out that
everyone here is really quite polite and courteous!

----------------------------------------------------------------------
When you have not made the least effort to learn HTML, what do you
expect?


I rest my case.



Regards,


Kent Feiler

https://www.KentFeiler.com

[Toby Inkster]

Re: &lt;p&gt; and &lt;div&gt; and breaks

Kent Feiler wrote:

It looks like in IE7 a <pCAN contain a <divwhich accounts for the
difference between #1 and #2 when we add the </p>s. The <pin #1
probably doesn't end until it hits the second <p>.

I'm never surprised by IE getting things wrong.

Firefox is right, and is consistant with the W3C Recommendation, as is
Opera, Safari and iCab. IE 5 for Mac gets things right too.

--
Toby A Inkster BSc (Hons) ARCS
Contact Me ~ http://tobyinkster.co.uk/contact

[maxonline.sabeen@gmail.com]

Free Website Designing solution.

Welcome to MAX web Solution's Website, a full service affordable
offshore web site design, Search Engine marketing and web development
company based in Kathmandu. We focus on e business solutions at low
prices serving the needs of small to medium businesses all over the
world. We offer a wide range of custom web site design development and
seo services at affordable prices starting from small presentation
sites to complex multifunctional web portals and advanced custom
e-commerce business solutions.
Service is everything - your satisfaction is guaranteed
That is why we would gladly accept the responsibility of taking care of
your site. We want to save your time, so that you could devote yourself
to the most important thing - your business. Allow us to maintain and
support your web site.
Money back guarantee
A 30 Day money back guarantee for the first monthly fee.
You can also response to this message.

Fillup the below information for further inquarry.
Contact Name :
Organization Name:
E-mail Address :
Phone no:
Call us for Demostration
Call : 2390177 / 9841206820
Check it out for more information

http://www.max-online.biz/idevaffiliate/idevaffiliate.php?id=749

[Andy Mabbett]

Re: &lt;p&gt; and &lt;div&gt; and breaks

In message <1NmdnQ-lYY9TGhTYnZ2dnU VZ_rGinZ2d@gold en.net>, Gus Richter
<gusrichter@net scape.netwrites

>Don't ask for proof for I won't bother to search for it. You know that
>this is so.

Yup, that's convinced me.

--
Andy Mabbett
* Say "NO!" to compulsory ID Cards: <http://www.no2id.net/>
* Free Our Data: <http://www.freeourdata .org.uk>
* Are you using Microformats, yet: <http://microformats.or g/?

[Andy Mabbett]

Re: &lt;p&gt; and &lt;div&gt; and breaks

In message <v30ko2t2dl4chr jo9rktfg8fnflg3 847og@4ax.com>, Kent Feiler
<zzzz@zzzz.comw rites

>I have to say that this newsgroup seems to have a higher percentage
>of obnoxious creeps who post to it than most. But maybe I should
>check out some of the religious newgroups, it may turn out that
>everyone here is really quite polite and courteous!

>----------------------------------------------------------------------
>When you have not made the least effort to learn HTML, what do you
>expect?
>
>
>I rest my case.

It clearly hasn't occurred to you that many people might regard someone
who comes in, demanding answers to questions which they can easily find
on-line, and then complaining when they're given helpful answers anyway,
as impolite and discourteous, or even as an obnoxious creep.

--
Andy Mabbett
* Say "NO!" to compulsory ID Cards: <http://www.no2id.net/>
* Free Our Data: <http://www.freeourdata .org.uk>
* Are you using Microformats, yet: <http://microformats.or g/?

[Kent Feiler]

Re: &lt;p&gt; and &lt;div&gt; and breaks

On Wed, 20 Dec 2006 18:50:27 +0000 (UTC), Darin McGrew
<mcgrew@stanfor dalumni.orgwrot e:

Kent Feiler <zzzz@zzzz.comw rote:

<P>aaaaaaaaa<DI V>bbbbbbbbb</DIV><DIV>cccccc cc<P>dddddddd</DIV>

In http://www.w3.org/TR/REC-html40/stru...l.html#h-7.5.4 the
example
is actually

<P>aaaaaaaaa<DI V>bbbbbbbbb</DIV><DIV>ccccc< P>ccccc</DIV>

but I like your version better.

There is no syntax error in the HTML. Since DIV elements are not
allowed inside P elements, and since the closing </Ptag is optional,
the closing of the P element is implied by the opening of the DIV
element.
---------------------------------------------------------------------

I apoligize for dragging this out. I understand that DIV elements are
not allowed inside P elements and I understand that the closing </p>
tag is optional, but unless it's somewhere stated what a browser
should DO when it finds a DIV that appears to be intended to be inside
a P, it seems like a browser-philosopy issue. Should the browser
assume that the html writer just forgot to insert the </pbefore the
<divand put one in for him, or should it assume that the html writer
didn't know that <div>s weren't allowed within <p>s and forget about
the W3C recomendation? On the third hand, if a browser really wants to
enforce the W3C recomendation maybe it should discard the two <div>s.

In the end, the browser is just trying to make a sensible screen out
of whatever piece of html crap is presented to it. I'd vote for the FF
method, but I have some sympathy for the IE idea, that may be what the
html writer really intended.


Regards,


Kent Feiler

https://www.KentFeiler.com

[Lars Eighner]

Re: &lt;p&gt; and &lt;div&gt; and breaks

<lidjp2lsk8ja3r 7us283eopf5l0ih 3ipur@4ax.com>,
the lovely and talented Kent Feiler
broadcast on comp.infosystem s.www.authoring.html:

I apoligize for dragging this out. I understand that DIV elements are
not allowed inside P elements and I understand that the closing </p>
tag is optional, but unless it's somewhere stated what a browser
should DO when it finds a DIV that appears to be intended to be inside
a P, it seems like a browser-philosopy issue.

How could the browser possibly judge that that a DIV "appears to be intended
to be inside a P"?

Should the browser assume that the html writer just forgot to insert the
</p>

The closing tag is *optional*. Many authors know the rules and simple don't
waste the keystrokes to put in the optional tags. They did not forget. The
knew the rules and expected a conforming browser to abide by them. You seem
to be saying that authors who study the rules and know how to use them
should have there documents second-guessed because some people just put any
old slop on the web and have never looked at a DTD in their lives.

before the <divand put one in for him, or should it assume that the html
writer didn't know that <div>s weren't allowed within <p>s and forget
about the W3C recomendation? On the third hand, if a browser really wants
to enforce the W3C recomendation maybe it should discard the two <div>s.

Nonsense. This is from the W3C 4.01 spec, a document you still show no sign
of having ever read:

+ A user agent must ensure that rendering is unchanged by the
presence or absence of start tags and end tags when the HTML
DTD indicates that these are optional. See the section on
element definitions for introductory information on SGML
elements.

See that word "must"? A conforming browser *must* treat:

<p>blah blah blah<div>yadda yadda</div>

exactly like:

<p>blah blah blah</p><div>yadda yadda</div>

In the end, the browser is just trying to make a sensible screen out
of whatever piece of html crap is presented to it. I'd vote for the FF
method, but I have some sympathy for the IE idea, that may be what the
html writer really intended.

--
Lars Eighner <http://larseighner.com/ <http://myspace.com/larseighner>
An effective way to deal with predators is to taste terrible.

[VK]

Re: &lt;p&gt; and &lt;div&gt; and breaks

Kent Feiler wrote:

I apoligize for dragging this out. I understand that DIV elements are
not allowed inside P elements and I understand that the closing </p>
tag is optional, but unless it's somewhere stated what a browser
should DO when it finds a DIV that appears to be intended to be inside
a P, it seems like a browser-philosopy issue.

There is no philosophy here: purely technical aspect plus a leftover of
the "HTML childhood". That was explained N times already but let's make
N+1th time:

>From the very beginning and until relatively recently P element was not

a container: it was an element w./o content marking _paragraph break_

This way originally in HTML there were paired elements "small" line
break BR and "big" paragraph break P. Respectively BR marked the
beginning of the next line and P marked the beginning of the next
paragraph.

When later it was decided to promote P into container the task was do
not break millions of already existing pages. The compromise rule was
made as simple as moo-cow:

New paragraph starts with the opening <Ptag and ends by closing </P>
tag or by opening tag of a block element - whatever comes first.

I fail to find any "browser philosophy" in here.

[Darin McGrew]

Re: &lt;p&gt; and &lt;div&gt; and breaks

VK <schools_ring@y ahoo.comwrote:

From the very beginning and until relatively recently P element was not
a container: it was an element w./o content marking _paragraph break_

Do you really consider HTML 2.0 "relatively recently" in the history of
HTML? The P element has been a container (with an optional closing tag)
since then.
--
Darin McGrew, mcgrew@stanford alumni.org, http://www.rahul.net/mcgrew/
Web Design Group, darin@htmlhelp. com, http://www.HTMLHelp.com/

"I'd love to make time, if only I could find the recipe."

[VK]

Re: &lt;p&gt; and &lt;div&gt; and breaks

Darin McGrew wrote:

Do you really consider HTML 2.0 "relatively recently" in the history of
HTML? The P element has been a container (with an optional closing tag)
since then.

This comment is correct: but only from the modern point of view.

The time we are talking about RFC HTML specs were of the same public
interest as the average density of Jupiter is right now. Besides of a
very few extremely curious people around the world - everyone else was
reading Netscape specs and testing on the current Netscape release. To
the honor of Netscape they stated from the beginning that "despite
closing </Ptag has no sense, we encourage you to use it in case of
future changes" (by memory quote). But an average developer even at
that far time was pretty much the same as it is right now: skip on any
keystroke one can skip and to use only absolutely necessary things :-)

The first more or less wide interest to HTML standards appeared only
closer to the end of Browser Wars when standard/non-standard topic was
used as a propaganda tool to fight the rival. By that time it was
already absolutely irrelevant what <Pis or should be - the only
really important matter was to never break millions of legacy pages.

[VK]

Re: &lt;p&gt; and &lt;div&gt; and breaks


VK wrote:

From the very beginning and until relatively recently P element was not

a container: it was an element w./o content marking _paragraph break_
>
This way originally in HTML there were paired elements "small" line
break BR and "big" paragraph break P. Respectively BR marked the
beginning of the next line and P marked the beginning of the next
paragraph.
>
When later it was decided to promote P into container the task was do
not break millions of already existing pages. The compromise rule was
made as simple as moo-cow:
>
New paragraph starts with the opening <Ptag and ends by closing </P>
tag or by opening tag of a block element - whatever comes first.

I guess the question "why P is not like everyone else?" and similar
will stay forever, so I will bookmark this segment of the thread for
future quick references. For the same purpose I'll include in here the
answer on your original question:

Both
<P>aaa<DIV>bb b</DIV><DIV>ccc<P> ddd</DIV>
and
<P>aaa</P><DIV>bbb</DIV><DIV>ccc<P> ddd</P></DIV>
produce the same tree fragment:

#parent_node
|_
#P
|_#text aaa
|
#DIV
|_#text bbb
|
#DIV
|_#text ccc
|_#P
|_#text ddd

because P element will be auto-closed before any block-level element
even if closing tag is ommited. It is important that block-level vs
inline detection goes by default display and not by the current display
style.
This way
<P>aaa<DIV style="display: inline">bbb</DIV><DIV
style="display: inline">ccc<P>d dd</DIV>
doesn't change anything in the parsing process. If you are really
targeted to be surprised by something in HTML :-) then the latter is
much more appropriate subject :-)
At the same time say this source:
<P>aaa<DIV>bb b</DIV></P>
will produce tree fragment:

#parent_node
|_
#P
|_#text aaa
|
#DIV
|_#text bbb

because P will be automatically closed before opening <DIV>, so besides
other "surprises" you are getting unmatched closing </Ptag after
</DIV>. While HTML doesn't care of such details, XHTML parser will
complain.

[Andy Dingley]

Re: &lt;p&gt; and &lt;div&gt; and breaks


Kent Feiler wrote:

I understand that DIV elements are
not allowed inside P elements and I understand that the closing </p>
tag is optional, but unless it's somewhere stated what a browser
should DO when it finds a DIV that appears to be intended to be inside
a P, it seems like a browser-philosopy issue.

It is stated, although it's not directly stated and it's not clearly
stated.

"Browser philosophy" issues would be things like the rendering of
invalid HTML (and whether to even try), default stylesheets and error
recoveries. The parsing of tags into the elements that structure the
document model though is simple SGML behaviour pre-dating HTML, isn't
an error and is clearly defined with a single clear interpretation (for
most cases). Admittedly finding this definition requires knowing that
HTML is an SGML application and knowing where to find the SGML
definitions themselves.

Omitted closing tags aren't in any way "wrong" in the SGML context,
because SGML has always defined its parsing behaviour from a partial
set of tags plus a DTD into a complete document structure as a tree of
elements. Obviously the DOM elements are always closed and well nested
(the DOM can't hold a structure that isn't) and so every parsing
operation _must_ transform one to the other (or raise an error).
According to SGML's rules and for HTML that's still valid (even if
stripped of many closing tags) this mapping is a defined and
non-ambiguous process.

If you read the release notes for Tidy then there's a little
explanation of how far this strategy can go and what the awkward cases
are that can break it.

SGML set out with a design principle of allowing tags to be omitted
where practical, as a way of reducing verbosity. XML takes the opposite
approach and requires that documents (tag to element parsing anyway)
are parseable in isolation from their DTDs. This also requires them to
not use the tag omission rules, as in the abbsence of a DTD there's no
way to tell what the implied closures would be. XML has the
side-effect of making thhe document more easily human-readable too, or
at least more accurately human readable. Humans have historically been
bad at correctly interpreting terse SGML documents with much
implication in them.