doltharz

Is there a way to set the default value for a "global number[35]" field? I want to set [1]=1, [2]=2, [3]=3 I'm using FM7 I want something else than Set Field(f, 1) Set Field(f[2], 2) Set Field(f[3], 3) ... Set Field(f[35], 35) Set Field(f=Split("1;2;3;...;35")) would be marvelous.

If I'm right, the magic stuff with $ and $$ to specify scope weren't yet "invented" in the mystic world of FM7 ?

Spreading parameters fields all around the database makes it messy so moving all "parameters" to a "session" table would be the best solution but the way FM works, it's a bit complicated then to access those parameters for calculation in all tables. To do so, I have to set an extra "constant" field in each table with the only goal to link it to another "constant" field in the "session" table with a [=] relationship. Each table must then have it's own instance of the "session" table to avoid having multiple relationship path between two table objects. Every day, i realyse how SQL is, compared to FileMaker. Something that requires 2 minutes of coding in SQL, takes 2 days in FileMaker.

The concat approach seems interesting, but I'm not sure the big merged report will correctly span over pages with page header and footer. BTW, I'm not sure at all it will need to. What i mean by "unique id" is simply a way to know exactly for whom the temporary data are stored there. Kinda "Session ID" on an ASP application. The "Username" is a setting in the client app and can be the same for multiple clients, "Login name" can also be the same for multiple connections till we have role based accounts. Both are useless. I can generate 5 random numbers and expect the sequence to be unique but it's not as safe as the identifier used internally by FileMaker.

Using a temporary (volatile) table should also be a solution but it becomes a bit tricky when dealing with concurent users. Is there a way to create a "UNIQUE CONNECTION ID"? I've tried "MacAddress" but some machines doesn't reveal this information, "Username" isn't safe either.

Can you elaborate a bit more on how it can be achieved? I think of a N:N join with calculation and filters but there may be an easier way. Without using additional tables and fields. I don't expect UNION like query to be indexed and/or fast and it's mainly for specific reports where "Other specify: ____________" are dumped and maybe counted. We've been asked to include more than one of these fields in the same table. Good design would have made this a separate table but it's not the case here. Thanks

Is there a way to reproduce such technique in FileMaker. SELECT t.Field1 AS Field FROM Table t WHERE t.Field1 IS NOT NULL UNION ALL SELECT t.Field2 FROM Table t WHERE t.Field2 IS NOT NULL /*[...]*/ UNION ALL SELECT t.FieldN FROM Table t WHERE t.FieldN IS NOT NULL ORDER BY Field /* If I'm correct, aliases are taken from first subquery and "ORDER BY" clause from the last one.*/ I want to be able to switch from : ID, "BMW", "Audi", "GM", NULL, NULL ID, "Ford", "Audi", NULL, NULL, NULL ID, "Toyota", "Pontiac", NULL, NULL, NULL To: "Audi" "Audi" "BMW" "Ford" "GMC" "Pontiac" "Toyota" Merci ;)

Versions: FileMaker Pro 7 7.0.3.320 FileMaker Dev 7 7.0.3.320

I do have a relation between 2 TO (Tables) that way: STUDENT (ID, Name, GroupID_fk, GroupName (Calculation: Unstored: GROUP::Name), ...) GROUP (ID, Name, ...) STUDENT::GroupID_fk ----[=]---- GROUP::ID When I sort using the "Sort Records..." menu command using "STUDENT::GroupName" it doesn't sort and produce the same output as "Unsort" but when I do use "GROUP::Name" it does sort correctly. Till it's a calculation field using a field from another table, I can't force it to be stored rather than "unstored and calculated when needed" Is there something I missed?

Your "philosophy" makes me a bit confused but I'm not working with templates, samples or tutorials from FM but with our "legacy" database. Even though building it again from scratch is an alternative, I doubt i'll achieve more. My concern is not the data nor the database but the engine itself. Maybe I'm wrong the way I do it and it's the kind of tips I am looking for. Should I try to achieve the save using only "find..." instead of "Constrain" ? Would it makes any difference ?

Our database design has a major flaw till it's a big "spreadsheet" rather than a true relation database with multiple tables. I doubt the structure of the database has anything aware with what I am looking for. I'm prolly not the first one to use multiple "Constrain Found Set" and "Sort Records" commands to get desired result. I "MUST" use "Show All Records" cause I can't expect data to be already "unfiltered". If it's a FM "Feature" (I wont say bug)... right, I'll deal with it but it makes layout loading time way way longer.

I want to do something like that Go To Layout [Any Layout Name Here] Show All Records Constrain Found Set Constrain Found Set Again Check for additionnal condition then Constrain Found Set one more time Sort etc. My main concern is that engine seems to start processing summaries and other stuffes behind the hood before I actually filter records out. Is there a way to prevent any processing before I actualy ask it to do so? Kinda Set Engine [Off] Go To Layout [...] Show All Records [...] Filter Out Sort Change my mind and Filter Out Again Sort Again Change my mind one more time and Filter Out [...] Set Engine [On] Switch to Preview Mode Thanks

Till my version in in french, I have to translate function names but it seems it wasn't the right one... French = Recapitulatif() English = GetSummary() Sorry...

Reading a bit more in help files i got my answer. It appears it's not possible. I'll have to find something else.

Can I have records sorted by an ALIAS calculated field but summaries and recap still refereing the "original" field? e.g. [color:red]ALIAS (Calculated) = [color:green]SALESMAN SomeField total (Calculated) = Recap(somefield, [color:green]SALESMAN) In the layout i do have Subsummary by [color:red]ALIAS You'll agree that order is the same...