On 2011-11-04, Friedrich Romstedt wrote:
> 2011/11/4 Guenter Milde <mi...@users.berlios.de>:
>>> It appears by inspection of the html source code, that indeed the
>>> sublist is not contained in a <p> tag. This appears like a mistake to
>>> me. Because it should be a paragraph, since it is an entity on its
>>> own to be rendered such. It also generates a rather wrong-looking
>>> visual impression of the rendered html output.
>> I don't think that a sub-list of a list should be nested in a paragraph
>> element.
> OK, I don't think I'm knowledgable enough to decide that. But to a
> LaTeX user, a list makes a paragraph, although it might contain
> paragraphs in its items.
In Docutils, lists are "paragraph like block level objects" and the
content of a list item is always put in a paragraph (as the pseudoxml
output of the example shows):
<paragraph>
This not:
<bullet_list bullet="*">
<list_item>
<paragraph>
item 1
<bullet_list bullet="-">
<list_item>
<paragraph>
subitem 1.1
<list_item>
<paragraph>
subitem 1.2
<list_item>
<paragraph>
item 2
<paragraph>
second paragraph
The relationship between the objects is defined in
http://docutils.sourceforge.net/docs/ref/doctree.html#element-hierarchy
The standard HTML writer omits the paragraph tags around list item content.
(The html-strict writer is more consistent here and adjusts the CSS style
sheet to get the compact/non-compact distinction.)
In LaTeX, text is not structured by putting it in a "paragraph object"
<p> first paragraph </p>
<p> second paragraph </p>
(the \paragraph macro is actually a section heading).
Rather a "paragraph separator" is used:
first paragraph
<new-paragraph/>
second paragraph
Hence, the question whether a list is an "instance" of a paragraph or should
be embedded into one does not apply.
BTW, in LaTeX, you can write both, lists that are separate from the
context and lists that are "embedded" - depending on whether the \begin
\end pair is offset from the pre/post text with blank lines.
In rst, embedded lists are not possible, the syntax requires blank lines
around a list. (To embed a list in a paragraph, you need a "compound"
directive.)
...
>>> I would appreciate a solution, since I'm
>>> observing this mistake since years now; although I seem to be a rare
>>> case in constructing such lists, it apprears regularly to me.
>> I suppose that the visual impression of the rendered html output is due
>> the used CSS style sheet and can be fixed in a style sheet (either as a fix
>> in the default Sphinx style or as a workaround in a custom style sheet).
> I also agree that it seems to be style sheet related. I think it can
> be solved fully via this. I see that the ``simple`` as well as the
> ``first`` classes are missing at least in the default + nature style
> sheets; it is the ``ol.simple, ul.simple`` that introduces
> ``margin-bottom:1em``.
> Now that I know what I have to work on I can try a fix.
Attention: AFAIK the "! important" feature is not supported by ePub. It
might be better to list the relevant objects in the specifier (divided by
``,``), e.g. ::
.first {
margin-top: 0 }
.last, ol.simple.last, ul.simple.last, .with-subtitle {
margin-bottom: 0 }
> P.S.: I was apparently too quick in believing that the <p> tag is
> necessarily the only solution. Although I still believe it is a more
> elegant solution. The browser can decide how much space to place, and
> the stylesheet does not need to say ``1em`` explicitly. I have put my
> thoughts in a comprehensive form on
> http://friedrichromstedt.github.com/lists1.html, with an example how
> it could work out in terms of resulting HTML.
IMV, not a good idea: what happens, if you decide to separate
paragraphs by indentation like
p {
margin-bottom: 0
}
p + p {
text-indent:1.5em;
margin-top:0
}
?
Besides this, the change to add the <p> tag would be required in the
Docutils html writer that does not exhibit the problem.
Günter
--
You received this message because you are subscribed to the Google Groups
"sphinx-dev" group.
To post to this group, send email to sphinx-dev@googlegroups.com.
To unsubscribe from this group, send email to
sphinx-dev+unsubscr...@googlegroups.com.
For more options, visit this group at
http://groups.google.com/group/sphinx-dev?hl=en.
