Found Records

[waveyd]

is there a function that allows you to distinguish between found and
omitted records? For example, a certain field = 0 for omitted, 1 for
in the found set.

Thanks.

David Averbach
UC Berkeley

[Matthew Smith]

In article <708065a8.0308211336.6d8a5158@posting.google.com >,
[email]averbach@mac.com[/email] (waveyd) wrote:

> is there a function that allows you to distinguish between found and
> omitted records? For example, a certain field = 0 for omitted, 1 for
> in the found set.

What would you do with such a function? You would only ever have found
records.

--
Matthew Smith
(to reply via email remove xxx)

[Kevin Hayes]

In article <matty_d-823580.22214422082003@duster.adelaide.on.net>,
[email]matty_d@macxxx.com[/email] says...

> In article <708065a8.0308211336.6d8a5158@posting.google.com >,
> [email]averbach@mac.com[/email] (waveyd) wrote:
>

> > is there a function that allows you to distinguish between found and
> > omitted records? For example, a certain field = 0 for omitted, 1 for
> > in the found set.

>
> What would you do with such a function? You would only ever have found
> records.

I would guess so that you can save a "found set" to come back to later.
There isn't but you can write a script to do it:

- Find your record set
- Set the field to 1 using the Replace command.
- Find Omitted
- Set the field to 0 using the Replace command.
- Find Omitted again to get you back to your original found set.

You could get fancy and add a space and a higher number to the found
records to save more than one found set.

Kevin

[Howard Schlossberg]

Actually, such a function would be highly useful to determine of a
record is currently in the found set or not. But it is not possible.

Matthew Smith wrote:

> In article <708065a8.0308211336.6d8a5158@posting.google.com >,
> [email]averbach@mac.com[/email] (waveyd) wrote:
>
>

>>is there a function that allows you to distinguish between found and
>>omitted records? For example, a certain field = 0 for omitted, 1 for
>>in the found set.

>
>
> What would you do with such a function? You would only ever have found
> records.

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Howard Schlossberg (818) 883-2846
FM Pro Solutions Los Angeles, California
Associate Member, FileMaker Solutions Alliance

[Brent Simon]

[posted and mailed]

Unless I don't understand your question correctly, how about a script...


Enter Find Mode
(Insert Query)
Preform Find
Replace Contents ("FoundStatus","1")
Show Omitted
Replace Contents ("FoundStatus","0")


[email]averbach@mac.com[/email] (waveyd) wrote in news:708065a8.0308211336.6d8a5158
@posting.google.com:

> is there a function that allows you to distinguish between found and
> omitted records? For example, a certain field = 0 for omitted, 1 for
> in the found set.
>
> Thanks.
>
> David Averbach
> UC Berkeley

[Lynn allen]

waveyd <averbach@mac.com> wrote:

> is there a function that allows you to distinguish between found and
> omitted records? For example, a certain field = 0 for omitted, 1 for
> in the found set.

Just in case this will do what you want, Status(CurrentFoundCount) gives
you the currently found set.

As to distinguishing whether a specific record is in the found set,
there is no record specific marker or status function that will do that.

You could capture all the currently found record IDs into a global, and
determine by relationship from the global to the Record ID field whether
or not a particular record was part of that relationship or not. This
gives you an unstored calc, but still might do what what you want.

--
Lynn Allen Allen & Allen Semiotics
FSA Associate Filemaker Consulting & Training
[email]lynn@semiotics.com[/email] [url]http://www.semiotics.com[/url]

[Matthew Smith]

In article <vkci9d3b079iab@corp.supernews.com>,
Howard Schlossberg <howard@antispahm.fmprosolutions.com> wrote:

> Actually, such a function would be highly useful to determine of a
> record is currently in the found set or not. But it is not possible.

But how would you use it? You would only ever be working on records in
the found set. Those not in the found set would not be accessible. To
get to those records not found would require you to do a new find, which
would make them found.

I think I would need to see an example of what you want to do to
understand what you want. It just doesn't make sense to me.

--
Matthew Smith
(to reply via email remove xxx)

[Howard Schlossberg]

What if you were looking at a record in a portal or in another way
through a relationship? It wouldn't necessarily be in the found set for
this. You might use it to flag a portal row to be highlighted if the
record is currently in the found set.

Matthew Smith wrote:

> In article <vkci9d3b079iab@corp.supernews.com>,
> Howard Schlossberg <howard@antispahm.fmprosolutions.com> wrote:
>
>

>>Actually, such a function would be highly useful to determine of a
>>record is currently in the found set or not. But it is not possible.

>
>
> But how would you use it? You would only ever be working on records in
> the found set. Those not in the found set would not be accessible. To
> get to those records not found would require you to do a new find, which
> would make them found.
>
> I think I would need to see an example of what you want to do to
> understand what you want. It just doesn't make sense to me.

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Howard Schlossberg (818) 883-2846
FM Pro Solutions Los Angeles, California
Associate Member, FileMaker Solutions Alliance

[David Averbach]

To all who are confused as to what the heck would want to do this for,
I wanted to have a portal on the side of my layout that showed the
names of all records that are found. Instead of scrolling through
records, the user would just click on the name to jump to that record.

Many thanks to those who suggested that simple script involving
"replace contents." The functionality is there, but the problem is
cosmetic. It takes about three seconds before the find, show omitted,
and show omitted (again, to return to the original found set) kicks
in. Well, that would be a great addition to the status function
family, in my opinion.

Best,

David Averbach
UC Berkeley

[Foszae]

i've had to do this very thing myself, for a rather complicated
statistical analysis. it's a small cheat but if you add a field for a
flag of some sort, then you can put in a value that demarcates it for a
later, easy to remember find

i) find your original set
ii) replace varFlag = 0 for all records
iii) loop through record next, exit after last
a)if record is to be omitted, set varFlag = 1
b)else just go to next one
iv) perform task with found set
v) find the set where varFlag = 1

waveyd wrote:

> is there a function that allows you to distinguish between found and
> omitted records? For example, a certain field = 0 for omitted, 1 for
> in the found set.
>
> Thanks.
>
> David Averbach
> UC Berkeley

[Mariano Peterson]

Its important to note that none of the suggested scripts for flagging
records in a found set will work in a multi-user environment. As
other users perform their own finds, they'll flag and de-flag the
records the original user had flagged. If the original user were to
do a search for the flagged records, they'd get a different recordset.
Also, if the flag were used to show records in a portal, then the
portal contents would change every time another user on the network
performed a find & flag.

There is a work around for this. First, each user must have a unique
user ID stored in a global. The flag field would have to be a
multi-key: when a user needs to flag records, you first remove their
ID from the flag field in every record where their ID appears in the
flag field (don't clear the flag; you might erase other User's IDs...
just remove the current user's ID). Then append the current user's ID
to the flag field for each record in the found set. Make sure to
seperate user IDs in the flag field with carriage returns.

Alternatively, if you just need to relate to the found records, as
opposed to the omitted records, then you can use a global text field
to store a list of carriage return delimited record IDs. Then you can
create a self-relationship from the global multi-key to the record
IDs. An important caveat the global text multikey is that it can only
hold 64k. If you are using 6 digit padded keys, ie. 000023, then this
key can only store approx 9,100 IDs before you exceed the text block
(memory) limit.

[David Averbach]

What do you suppose a good way to remove the flag field would be? I
know insert text would work, but then you could insert it more than
once and that would be hard to keep track of. Then there's the
problem of removing it.

What I would want would be to check the specific check box
corresponding to the flag field through a button and script, not
actually checking the box. Is there any way to do this?

> There is a work around for this. First, each user must have a unique
> user ID stored in a global. The flag field would have to be a
> multi-key: when a user needs to flag records, you first remove their
> ID from the flag field in every record where their ID appears in the
> flag field (don't clear the flag; you might erase other User's IDs...
> just remove the current user's ID). Then append the current user's ID
> to the flag field for each record in the found set. Make sure to
> seperate user IDs in the flag field with carriage returns.