[Spambayes] SpamBayes Web Interface??

[Michael Kimball]

Please see remarks in-line.  MK

Tony Meyer wrote:
> 
> [...]
> > There are a lot of <br/> tags in the page.  Are these some tag I don't
> > know, or are they mis-generated <br> tags?
> 
> The version of HTML that SpamBayes uses (and has to use, for various
> reasons) is XHTML 1.0 (or above), which means that the HTML has to be valid
> XML.  What this means is that every tag needs a closing tag (so for every
> <p> there is a </p>, for example).  Older HTML versions don't require this
> (and a lot of HTML generators don't do it), and browsers generally do the
> job correctly either way.  Empty tags, like br and hr, can either be written
> <br></br>, or abbreviated <br/>.  It's just the (now) correct way to write a
> <br> tag.

NS 4.79 doesn't seem to act on the <br/>.  I'll have to try <br></br>. 
It should break on the first one and ignore the second.  I can expect
Dreamweaver to flag the </br> though.


> 
> > The style .banner has a padding=5 instead of padding: 5px;
> > When I fixed that the blue background color showed.
> 
> I believe (but haven't checked) that it should default to px if no unit of
> measurement is given, but I don't see any reason why we couldn't generate
> the file with the units explicitly there if it helps.  I'll check this
> change in (in a day or so).  Thanks for finding this.


The point I was trying to make was the use of the equal sign instead of
the colon.  I think you are right that without any unit of measurement
it would default to pixels.


> 
> > When I added width: 100% the background blue stretched across
> > the page.
> 
> I'll check this out, too, thanks.
> 
> > But I couldn't get the 1px solid black borders to show.
> > Nor could I get the borders to show in any of the other tables.


I KNOW I've seen table borders in NS, and I think I've seen them
resulting from styles, rather than from within the tag.  I'll play with
this a bit more.


> 
> I'll look into that.
> 
> > There are also a number of tables with 'id="headedBox"'.  I THINK the
> > ids are supposed to be unique.
> [...]
> 
> I'll look into this - I don't think that the ids are required to be unique,
> but I'm not 100%.  The code that puts the pages together dynamically uses


The more I think about this, the more I think I was recalling someone's
personal preference (i.e an 'id' is an 'identity' and you can't have two
people with the same identity, so don't have two
tables/rows/cells/whatever with identical identities).  I don't recall
that article even mentioning "name".  The same argument would apply to
"name".


> the ids to refer to each piece, so it definitely has to be the id attribute
> and not the name one.
> 
> > From my perspective, none of this is urgent.  I have been able to get
> > SpamBayes to open in Firefox so I don't need to rely on the Netscape
> > formatting.
> 
> Great :)  I still think it's worth fixing, but I'm unlikely to have time to
> look at this until Tuesday.  I'll update then.
> 
> =Tony Meyer
> 
> ---
> Please always include the list (spambayes at python.org) in your replies
> (reply-all), and please don't send me personal mail about SpamBayes. This
> way, you get everyone's help, and avoid a lack of replies when I'm busy.