What about keeping track of the current visible groups/fields. You could use
a custom property to store names of fields when shown and delete when
didden. This is a horrible workaround but would save a bit of searching time
because before the search you would know what to search in.
> Hi Alan,
>
> Thanks for the suggestion, but both your (adapted) script proposal and a
> similar script I had already tried produce error messages:
>
> Example:
>
> The following script works fine (in the modeless searchstack):
>
> "if word 1 of the owner of fld j of cd i of the topstack is "group" then
> put Return& the owner of fld j of cd i of the topstack after last line
> of SearchResult"
>
> However, if you add "the vis of" or "the ID of" before "owner" as in
>
> "if word 1 of the owner of fld j of cd i of the topstack is "group" then
> put Return& the vis of the owner of fld j of cd i of the topstack after
> last line of SearchResult"
>
> then you get an error message "can't find background" while at the same
> time the name of the not found background is indicated under "bad
> token".
>
> I tried to find out whether the error is maybe produced because of
> exceeding a limit of "of"-nesting.
>
> I went to the card of the stack itself and typed in the message box
>
> "if word 1 of the owner of fld 13 of cd 29 of the topstack is "group"
> then put the vis of the owner of fld 13 of cd 29 of the topstack"
>
> which returned "false" as to be correctly expected. When navigating to
> another card, however, the same script in the message box produces the
> error message (as above) "can't find background" although the name of
> the background is shown in the error message at "bad token".
>
> Running the same script from the modeless searchstack produces error
> messages in both cases, i.e. even if card 29 is the current card.
>
> So I am stuck with my problem.
>
> Is this maybe some kind of a bug?
>
> Regards,
>
> Wilhelm
>
> On 13/2/01 Alan Beattie at <[EMAIL PROTECTED]> wrote:
>
> >How's about:
>
> > if the vis of field x is true then
>
> > if word 1 of the owner of field x is "card" then
>
> > -- add this field to myFieldList
>
> > else
>
> > if the vis of the owner of field x is true then
>
> > -- add this field to myFieldList
>
> > end if
> > end if
>
> > end if
>
> >This script checks the owner of the field is not a card (cards don't
> have a
> >visible property and checking it would cause an error). Then it checks
> the
> >visible of the the owner of the field (which could either be a group or
>
> >stack).
>
> >This assumes all stacks you are searching are visible. Also, if you
> have
> >nested groups (your field inside a group which in turn is inside a
> group),
> >you would need to check further up the hierarchy for visibility.
>
> >Cheers,
>
> >Alan
>
>
> >On 11/2/01 10:04 pm, Wilhelm Sanke at [EMAIL PROTECTED] wrote:
>
> >> All *visible* fields of a stack shall be searched via a modeless
> >> substack to create a list of links for the respective searchstring.
> >>
> >> The problem is that in a stack of a client some visible fields are
> parts
> >> of groups that are only shown under specific conditions and shall
> only
> >> be searched if the group - and so the field - is visible.
> >>
> >> Testing the "vis" of a - not really visible - field of a hidden group
>
> >> yields "true", the same result appears with testing the "effective
> vis"
> >> of the invisible field.
> >>
> >> Is there any way to determine the effective visibility of such a
> field?
>
>
>
>
> Archives: http://www.mail-archive.com/[email protected]/
> Info: http://www.xworlds.com/metacard/mailinglist.htm
> Please send bug reports to <[EMAIL PROTECTED]>, not this list.
>
Archives: http://www.mail-archive.com/[email protected]/
Info: http://www.xworlds.com/metacard/mailinglist.htm
Please send bug reports to <[EMAIL PROTECTED]>, not this list.
