Re: [docbook-apps] <itemizedlist><variablelist><funcsynopsis>

[Bob Stayton <bobs@sco.com>]

On Tue, Nov 25, 2003 at 06:13:45PM -0600, John Himpel wrote:
> On Tue, 2003-11-25 at 11:51, Bob Stayton wrote:
> > On Tue, Nov 25, 2003 at 06:22:11AM -0600, John Himpel wrote:
> > > On Mon, 2003-11-24 at 21:00, Bob Stayton wrote:
> > > > On Mon, Nov 24, 2003 at 09:10:47PM -0600, John Himpel wrote:
> > > > > When I use the following markup:
> > > > > 
> > > > > <itemizedlist>
> > > > >   <listitem>
> > > > >     <variablelist>
> > > > >       <varlistentry>
> > > > >         <term>
> > > > >           <funcsynopsis>
> > > > >             <funcprototype>
> > > > >               <funcdef>
> > > > >                 int
> > > > >                 <function>xf86GetVerbosity</function>
> > > > >               </funcdef>
> > > > >               <void />
> > > > >             </funcprototype>
> > > > >           </funcsynopsis>
> > > > >         </term>
> > > > >         <listitem>
> > > > >           <simpara>which returns the value of
> > > > >             <varname>xf86Verbose</varname>
> > > > >           </simpara>
> > > > >         </listitem>
> > > > >       </varlistentry>
> > > > >     </variablelist>
> > > > >   </listitem>
> > > > >   <listitem>
> > > > >     <variablelist>
> > > > >       <varlistentry>
> > > > >         <term>
> > > > >           <funcsynopsis>
> > > > >             <funcprototype>
> > > > >               <funcdef>
> > > > >                 int
> > > > >                 <function>xf86GetDepth</function>
> > > > >               </funcdef>
> > > > >               <void />
> > > > >             </funcprototype>
> > > > >           </funcsynopsis>
> > > > >         </term>
> > > > >          <listitem>
> > > > >            <simpara>which returns the               
> > > > > <parameter>-depth</parameter> command line 
> > > > >              If not set on the command line, <constant>-1</constant>
> > > > >              is returned.
> > > > >            </simpara>
> > > > >         </listitem>
> > > > >       </varlistentry>
> > > > >     </variablelist>
> > > > >   </listitem>
> > > > > </itemizedlist>
> > > > > 
> > > > > This yields something similar to this:
> > > > > 
> > > > > int xf86GetVerbosity():
> > > > > (circle)
> > > > >      which returns the value of xf86Verbose
> > > > > 
> > > > > I think the stylesheets cannot cope with the level of nesting of the
> > > > > funcsynopsis.
> > > > > 
> > > > > Can anyone else replicate the problem.
> > > > > 
> > > > > I am using the latest xsltproc/4.2 dtd/1.62.4 xsl-stylesheets
> > > > 
> > > > You didn't mention if this problem is in the
> > > > HTML or print output.
> > > 
> > > Bob,
> > > 
> > > Sorry about that!
> > > 
> > > I have only tried HTML.  I haven't tried PRINT yet.
> > > 
> > > John
> > 
> > Hmm, I get the following with your example when I view it
> > with IE5 or Netscape 4.7:
> > 
> > o
> >     int xf86GetVerbosity():
> >            which returns the value of xf86Verbose
> > 
> > where "o" is the bullet.  That is, the bullet is not out
> > of order as in your example.  What browser are you using?
> > Is CSS involved?
> > 
> > The bullet is above the term, but that's expected because
> > the variablelist in the itemizedlist/listitem starts a
> > <div>, which will start on a new line.
> > 
> > I don't think you can get the term to line up with
> > the bullet without using CSS to adjust the
> > div margins.
> > 
> > I do see one oddity: the right paren in the
> > expression is slightly lower than the left
> > paren.
> 
> I have three browsers available to me:
> 1) Ephiphany
> 2) Mozilla
> 3) IE - Latest with all patches
> 
> The first two render as I describe (makes since because they use the
> same rendering engine) and IE renders as you describe.
> 
> I have stored the results of the transformation at:
> http://fun.bestdsl.net/~jlhimpel/jwh.html
> 
> You can view the resulting source, but it looks just like you describe.
> 
> I fed that URI to the HTTP validator at w3.org and it fails to validate
> at HTML 4.1 Transitional.  I recall, but can't seem to locate, a
> previous discussion on this list something to the effect that DocBook
> will generate working HTML (what a browser will display correctly)
> rather than correct HTML (validates properly).  Is this one of those
> instances?

Probably.
I wish HTML were a science instead of an art.  I think
we should concentrate on getting DocBook XHTML output
to conform to the standards. The stylesheets can produce
valid XHTML, but they can also produce invalid XHTML under
certain circumstances.


-- 

Bob Stayton                                 400 Encinal Street
Publications Architect                      Santa Cruz, CA  95060
Technical Publications                      voice: (831) 427-7796
The SCO Group                               fax:   (831) 429-1887
                                            email: bobs@sco.com