Table Row Height Problems

[stevehayter]

Wonder if anyone can help me solve this problem!

I have a 3x3 table which i'm using to create a table with a rounded
edge using images in the top left, top right, bottom left, and bottom
right cells and lines in the top/left/right and bottom cells (sounds
odd, you'll see what i mean when you see the site). It works fine,
except the top and bottom rows are a lot bigger than i've specified
when they should be flush with the centre cell as the left and right
columns are. I know it's nothing to do with the graphic being too
large because that is the size i have set the height to. Can anyone
shed any light on this!?

The site can be viewed at

http://homepage.ntlworld.com/hayterfamily/borderprob.html

I have left borders on so you can see the cell layout - turning them
off makes no difference.

Many thanks
Steve

[JabbatheHut]

Re: Table Row Height Problems


"stevehayte r" <stevehayter@nt lworld.com> wrote in message
news:370e622b.0 306270658.7ccd0 275@posting.goo gle.com...[color=blue]
> Wonder if anyone can help me solve this problem!
>
> I have a 3x3 table which i'm using to create a table with a rounded
> edge using images in the top left, top right, bottom left, and bottom
> right cells and lines in the top/left/right and bottom cells (sounds
> odd, you'll see what i mean when you see the site). It works fine,
> except the top and bottom rows are a lot bigger than i've specified
> when they should be flush with the centre cell as the left and right
> columns are. I know it's nothing to do with the graphic being too
> large because that is the size i have set the height to. Can anyone
> shed any light on this!?
>
> The site can be viewed at
> http://homepage.ntlworld.com/hayterf...orderprob.html
>
> I have left borders on so you can see the cell layout - turning them
> off makes no difference.
>
> Many thanks
> Steve[/color]

Well, not easy to tell for sure, but here's what I would do:

- remove all height specifications from cells
- replace with pixel specifications for height

That may reveal your problem.

Next:

- rebuild the table, and in any case you have an image sized with percentage
(not always a good idea) replace it with a new table inside the table cell
with the image as a tiled background and the table itself set to 100% to
fill the space.

Failing that, I would get rid of the stripe in the image and use cell bg
colors for all be the corners.

Some things just do not plain work in HTML, you might be trying to do one of
them.

[Sam Hughes]

Re: Table Row Height Problems

stevehayter@ntl world.com (stevehayter) wrote in
news:370e622b.0 306270658.7ccd0 275@posting.goo gle.com:
[color=blue]
> Wonder if anyone can help me solve this problem!
>
> I have a 3x3 table which i'm using to create a table with a rounded
> edge using images in the top left, top right, bottom left, and bottom
> right cells and lines in the top/left/right and bottom cells (sounds
> odd, you'll see what i mean when you see the site). It works fine,
> except the top and bottom rows are a lot bigger than i've specified
> when they should be flush with the centre cell as the left and right
> columns are. I know it's nothing to do with the graphic being too
> large because that is the size i have set the height to. Can anyone
> shed any light on this!?
>
> The site can be viewed at
> http://homepage.ntlworld.com/hayterf...orderprob.html
>
> I have left borders on so you can see the cell layout - turning them
> off makes no difference.[/color]

When trying to perfectly fit images into table cells, you need to leave
no space between the <img> tag and the <td> tags. Browsers think that
you want a space in there.

For example, one of your table cells, in your source code, used this:

<td align=right valign=bottom width=6 height=5>
<img src=./images/corner_red_bott omright.gif>
</td>

Instead, it should use this, with spaces between the tags removed:

<td align=right valign=bottom width=6 height=5><img
src="./images/corner_red_bott omright.gif"></td>

Note that I quoted the path to the file name. This is because the
attribute value included a slash. As the HTML specification says:

"In certain cases, authors may specify the value of an attribute
without any quotation marks. The attribute value may only contain
letters (a-z and A-Z), digits (0-9), hyphens (ASCII decimal 45),
periods (ASCII decimal 46), underscores (ASCII decimal 95), and
colons (ASCII decimal 58). We recommend using quotation marks even
when it is possible to eliminate them."

[from <http://www.w3.org/TR/html4/intro/sgmltut.html#at tributes>,
read more there.]

[Paul Taylor]

Re: Table Row Height Problems

[color=blue]
>
>Instead, it should use this, with spaces between the tags removed:
>
> <td align=right valign=bottom width=6 height=5><img
> src="./images/corner_red_bott omright.gif"></td>
>
>Note that I quoted the path to the file name. This is because the
>attribute value included a slash. As the HTML specification says:
>
> "In certain cases, authors may specify the value of an attribute
> without any quotation marks. The attribute value may only contain
> letters (a-z and A-Z), digits (0-9), hyphens (ASCII decimal 45),
> periods (ASCII decimal 46), underscores (ASCII decimal 95), and
> colons (ASCII decimal 58). We recommend using quotation marks even
> when it is possible to eliminate them."
>
>[from <http://www.w3.org/TR/html4/intro/sgmltut.html#at tributes>,
>read more there.][/color]

If you read more about attributes, you will find that, although height
is a valid attribute for an image, it is NOT a valid attribute for the
td tag, so you shouldn't use it.

You should also use alt="" for these corner images, to make sure that
screenreaders pass over them.

Paul Taylor

[brucie]

Re: Table Row Height Problems

In post <vi7tfvg118785o n7cus69quq6rfe8 0feiv@4ax.com>
Paul Taylor said...
[color=blue]
> If you read more about attributes, you will find that, although height
> is a valid attribute for an image, it is NOT a valid attribute for the
> td tag,[/color]

yes it is although it depends on the DTD

http://www.w3.org/TR/html401/struct/tables.html#edef-TD

[color=blue]
> You should also use alt="" for these corner images, to make sure that
> screenreaders pass over them.[/color]

that doesn't ensure anything. a screen reader may say "image without
alt text image name dot PNG". the same with alt=" "

if you want an image ignored don't put it on the page, stick it in as
a background on something.

--
brucie a. blackford. 29/June/2003 06:52:22 pm kilo.

http://loser.brucies.com/

[Stephen Poley]

Re: Table Row Height Problems

On Mon, 30 Jun 2003 20:05:47 +1000, brucie <brucie@loser.b rucies.com>
wrote:
[color=blue]
>In post <h9ltfvcima6k0v edmvuji34qkbolu ndmhb@4ax.com>
>Stephen Poley said...
>[color=green][color=darkred]
>>>> You should also use alt="" for these corner images, to make sure that
>>>> screenreaders pass over them.[/color][/color]
>[color=green][color=darkred]
>>>that doesn't ensure anything. a screen reader may say "image without
>>>alt text image name dot PNG". the same with alt=" "[/color][/color]
>[color=green]
>> Well it *might*,[/color]
>
>there is no might. ibm home page reader is one audio browser that
>*does*
>[color=green]
>> but web-authors surely need not be worried about browsers that are so broken
>> as that -[/color]
>
>why is a browser broken if it has the option to identify images
>without alt text and notify the user?[/color]

The HTML spec says: "User agents must render alternate text when they
cannot support images, they cannot support a certain image type or when
they are configured not to display images." So if alt=" " is specified,
then the correct behaviour is for the browser to render a space. (If alt
is erroneously omitted, then the behaviour you mention is fairly common
and quite reasonable.)
[color=blue]
>if the user wants that why cant
>they have it?[/color]

If the users really want that, of course they can have it. My statement
was that web-authors should not have to worry about that situation, and
that statement still stands. (On reflection, perhaps I should withdraw
the word "broken" and substitute "non-standard").
[color=blue]
>this is just another example of the author thinking they know better
>than the visitor what they want or should have.[/color]

What point do you actually want to make? You originally said "if you
want an image ignored don't put it on the page, stick it in as a
background on something" - but now you seem to be saying that an author
shouldn't want to do anything of the sort.

Normally speaking, if an author considers an image decorative, he
specifies alt="" or alt=" ", and expects text browsers / screen readers
to render an empty string or space respectively. If however the reader's
preferred browser behaviour is for it to start playing Chopin's
Fantaisie-Impromptu every time it encounters an image with alt="", that
doesn't come under the heading of things an author should worry about.

--
Stephen Poley

http://www.xs4all.nl/~sbpoley/webmatters/

[Alan J. Flavell]

WAI guidelines, was Re: Table Row Height Problems

On Mon, Jun 30, Stephen Poley inscribed on the eternal scroll:
[color=blue][color=green]
> >why is a browser broken if it has the option to identify images
> >without alt text and notify the user?[/color]
>
> The HTML spec says: "User agents must render alternate text when they
> cannot support images, they cannot support a certain image type or when
> they are configured not to display images."[/color]

Yes, they must at least do that.
[color=blue]
> So if alt=" " is specified, then the correct behaviour is for the
> browser to render a space.[/color]

It's not in dispute that this is a good default choice for normal
circumstances. Jim will be along shortly to present some special
requirements...
[color=blue][color=green]
> >if the user wants that why cant they have it?[/color][/color]

I'd say that the WAI guidelines for client agents indeed encourage
client agents to support a rich repertoire of configuration options.
Start at http://www.w3.org/TR/UAAG10/guidelin...content-access
and read on.
[color=blue]
> If the users really want that, of course they can have it.[/color]

I must say I found it rather annoying that IBM HPR (at least, the
version I tried) didn't seem to have the option: if it encountered
alt="", then it insisted on announcing the relative URL
(foobar-dot-PNG etc.) of the image.
[color=blue]
> My statement was that web-authors should not have to worry about
> that situation, and that statement still stands.[/color]

Agreed. Web authors ought to be able to follow _their_ part of the
guidelines, and rely on users selecting user agents which sufficiently
follow _their_ part of the guidelines to meet with their own
satisfaction.

Unfortuantely, reality is messier than that...
[color=blue][color=green]
> >this is just another example of the author thinking they know better
> >than the visitor what they want or should have.[/color][/color]

I couldn't exactly agree with that. The author knows whether they
intend an image to be a substantive part of the content, or a mere
decoration. True, they can't know what their users want, but they
nevertheless need a markup convention which describes their _content_
in order to give the client agent a chance of doing something
sensible with it.

And up till now, alt="" has been accepted as a convention for marking
up their content (i.e as being insignificant to text mode or
non-visual browsing). If you're now saying that images that are only
for visual presentational purposes ought to be supplied via CSS since
they aren't part of the substantive content, then you'd have a fair
debating point.

best