Jean-Christophe Boggio wrote:
Hello,
I have really strange behaviours in my generated pages and I suspect
embperl to be "modifying" my html. I'd like to prevent this.
The case is this (I use Embperl Object) :
base.epl:
<div id="upperband">
<a href="/" class="noborder"> [- Execute('upperband.epl'); -]</a>
</div>
upperband.epl:
<img src="/img/upperband_home.png" class="noborder" />
subdir/upperband.epl:
<img src="/img/upperband_subdir.png" class="noborder" />
As you see, I want a different image in the upper band depending on
where the
user is.
The problem lies in base.epl : immediately after the <a href="/"
class="noborder">
and before the [- Execute... -]
* If I leave the space, I get... a (vertical) space after the block.
* If I remove the space, the generated code is this :
<a href="/" class="noborder"<img src="/img/upperband_home.png"
class="noborder" />
></a>
So my "<a ...>" tag is broken and I see the ">" character.
Also, I take the care of having only one space between 'href="/"' and
'class=...'
and I'd like embperl not to add useless spaces. On a big page with lots
of tags
it becomes significant.
Is there a way to disable embperl's reinterpretation of the HTML code ?
I never
use the auto-table-like features.
Thanks,
This does look like an Embperl bug, based on your description of what
changes in the output when you change your code. However, there is a way
to modify what Embperl does in terms of processing HTML tags - assuming
you are compiling Embperl from source. I wanted to keep the forms
processing (i.e. have Embperl still process input tags) but didn't want
it messing with TABLE and TR etc, which it was barfing on because of the
nesting being messed up by conditional blocks. So I just went and edited
the EmbperlHTML syntax source file and removed all the bits that looked
like being related to tables. Under the Embperl source tree, this file
is in Embperl/Syntax/EmbperlHTML.pm. You need to recompile an reinstall
Embperl, obviously. This is (to me) a hack, but at least it might let
you experiment with a workaround...
Neil
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
