Mailing List Archive

Init value ignored in SELECT
Hi,

SELECT fields don't get initialized in the same way as INPUTs:

[- $fdat{test1} = 100 -]
<input name=test1 value=3>

would show 3 (which is correct and intuitive I believe) while

[- $fdat{test2} = 100 -]
<select name=tst2>
<option value=222>222
<option value=333>222
<option selected value=444>444
</select>

would show 100 instead 444 which is a bit confusing.

Even more confusing is that $optDisableTableScan disables SELECT
handling too: sometime I need to turn off table scan (often because
table head and foot are in different files), but I still want Embperl to
take care of the form handling. Maybe $optDisableTableScan could disable
OL and UL together with tables stuff, but disabling SELECTs that way is
a bit counter-intuitive.

- Robert

P.S. I'm running with EMBPERL_OPTIONS = 16
RE: Init value ignored in SELECT [ In reply to ]
>
> SELECT fields don't get initialized in the same way as INPUTs:
>
> [- $fdat{test1} = 100 -]
> <input name=test1 value=3>
>
> would show 3 (which is correct and intuitive I believe) while
>
> [- $fdat{test2} = 100 -]
<select name=test2>
<option value=100>100
> <option value=222>222
> <option value=333>222
> <option selected value=444>444
> </select>
>
> would show 100 instead 444 which is a bit confusing.
>

Yes, that's because Embperl can't know that 444 is selected, when it
processes option value=100. That's due to the current architecture of
Embperl and can't be easily changed.

> Even more confusing is that $optDisableTableScan disables SELECT
> handling too: sometime I need to turn off table scan (often because
> table head and foot are in different files), but I still want Embperl to
> take care of the form handling. Maybe $optDisableTableScan could disable
> OL and UL together with tables stuff, but disabling SELECTs that way is
> a bit counter-intuitive.
>

That's because Embperl handles SELECT internaly like a table, but you are
right two options may make sense. I add this to the TODO list, but Embperl 2
will change a lot in this area, so it shouldn't be neccessary at all to use
the optDisableTable scan, only because start end end of the table is in two
different files.

Gerald

-------------------------------------------------------------
Gerald Richter ecos electronic communication services gmbh
Internetconnect * Webserver/-design/-datenbanken * Consulting

Post: Tulpenstrasse 5 D-55276 Dienheim b. Mainz
E-Mail: richter@ecos.de Voice: +49 6133 925151
WWW: http://www.ecos.de Fax: +49 6133 925152
-------------------------------------------------------------