bobearl

I found another way of solving my problem, and wanted to share it if anyone is at all interested. It is in no way as flexible or as powerful as Bruce's method, but it solved one problem that I was wrestling with. The problem I couldn't quite fix was the issue of being able to use the list to pass parameters. In this example, I wanted everyone to be able to see who all has tasks to do, filter those tasks if they wish, and then still see who has tasks within the found set. Bruce's solution works beautifully for this. What I struggled with was allowing the user to click on a task-doer, which could then filter the task list to only those tasks in the prior found set by that user. I finally found a solution by adapting Ray Cologon's solution at nightwing.com for "Displaying the Found Set in a Portal - dynamically!" at http://www.nightwing.com.au/FileMaker/demos8/demo804.html. It involves a custom function which captures all the ID's in a found set (in this case, all people in the found set of tasks) and using this set as a global multikey for a portal. My solution has a slight wrinkle, which uses a calc field in the people table to count the number of repeats of a personID in the multikey and show that as the number of tasks for all task-doers in the found set. [i suspect that for really large numbers of records, keeping track of a really large multikey like this will erode performance, and that a different custom function which only adds to the global multikey unique values would be much more elegant. After 3 days stuck on the same goofy problem, though, I can't think of how to do it and still generate my tasks per user field, but I look forward to challenging this - later] At the risk of being sappy, let me reinforce how tremendously valuable it is for novices like me to stand on the shoulders of the profoundly talented and experienced folks through this forum and other resources, in this case Bruce and Ray in particular. To those of you who tirelessly contribute and edify, I suspect you know but I wish to reinforce: people like me are tremendously grateful that you are willing to share your wisdom. Many thanks! SubSumFoundSetPortal.fp7.zip

Very slick, Bruce, thanks - I would definitely not have thought of that. I really appreciate your help!

A sample file might make this clearer. Attached is a simple file of tasks and users. Also is attached a few of the views of the data: I can get the list of people, and the list of tasks, but I'd love to have the list of people above the list of tasks rather than interdigitated as in the 'TaskListByPerson' layout. SubSumPractice.fp7.zip

This seems so simple that I hesitated to post it until I had thrashed around with it for a couple of days unsuccessfully. Any help would be much appreciated. I am making a task list viewer that by default lists all tasks for a group of users. I would like to show all the users and the number of tasks they have in a subsummary report at the top of the page, followed by a list of all of the tasks sorted by user: User 1 - 2 tasks User 2 - 1 task User 3 - 9 tasks --Task A User 1 --Task B User 1 --Task C User 2 --Task D User 3 I am getting stuck, however. If I use the subsummary report and place the task list in the body of the document, I get: User 1 - 2 tasks --Task A User 1 --Task B User 1 User 2 - 1 task --Task C User 2 User 3 - 9 tasks --Task D User 3 This fails to give me the 'heads up' view with all of the tasks at the top (there are 30 users, so scrolling through is a pain). I can get the task view itself with just a subsummary and nothing in the body, but then can't figure out how to get the tasks on the same layout: User 1 - 2 tasks User 2 - 1 task User 3 - 9 tasks I really appreciate any advice anyone might have!

Hello all, Apologies for the basic question, but this is something that I've never quite gotten, but I think Bruce's explanation is as close as I've ever been to understanding how to do this. What I'm not quite certain that I understand is how to use the 'standard FileMaker autocreate relationships'. Bruce, do you happen to have a link or a summary of what you're referring to in general. Your solution seems very elegant, I just am not following exactly when and how the new related record is created. Many thanks!

Very handy, good idea. A very small suggestion: one thing that might save you just a bit of time (and maybe fidelity) with this is instead of recreating all of the relationships, option drag all of the tables that you'd like to create the sub-group from. That will keep all the relationships intact and save you from having to reconnect those TO's. Thanks very much for sharing the tip! Cheers, Bob

It sounds like you have it worked out, so this may not be helpful; however, I had a similar issue where filtered portals were not working in IWP though they worked fine on the local client. It appears that the global fields used for the relationship must be actively set in each IWP session, so one way to do this is to explicitly write your global fields needed for the relationship in a startup script. I found a helpful post about this which I can't find, but I think it may have been from bcooney, in a search for something along the lines of global fields IWP portal. I'm sorry I can't find the link, but

If I understand your question correctly, I believe the dotted lines you are referring to are designed to show which fields are enterable. If you change this in 'Field Behavior' for any given field, the dotted lines disappear and you can no longer enter the field in browse and/or Find mode. Apologies and please let me know if I've misunderstood the question. Screen_shot_2010-02-13_at_1.17.43_PM.pdf

Hi ignotum, I was wrestling with a similar problem, but I'm one step behind you. What is your script step that allows the Enter to resume the script? The only way I've figured out to resume a script is with the pulldown menu in the button setup dialog (this may be different if you are resuming the script itself rather than a parent script). In any case, thanks for your thoughts. Cheers, Bob

That's correct, a button.

I'm wrestling with resuming a paused script. Background: Script 1 checks user-input fields for blank entries; if found, calls script 2 Script 2 opens a window with an explanation of the problem, a field in which to enter text for input into the blank field, and a "Cancel" button [script 3: Close window, halt script] and a "Correct" button [script 3: Close window, return text to appropriate field]. Both buttons resume Script 2. Problem: I would like to have the ability to resume script 2 not only from the Cancel or Correct buttons, but also by a script trigger (on object Save) from the field; this would allow the user to avoid reaching for the mouse after having keyed in the correction. I have looked over the forums, and it seems that others have run up against this and get the impression that there is no resume step that can be invoked from a script trigger in the same way it is from the button. http://fmforums.com/forum/showtopic.php?tid/202899/post/323522/#323522 My question: I could fix this if there is a scripted way to invoke a button click, by having a script trigger which invokes a script to <>. Is there a way to do this, perhaps for example by using the button's "Object Name"?

I'm a novice, so you should take this with a grain of salt, and my apologies if I'm either completely off base or stating the obvious. Seems like one way to come at this is to use a stored calc on the child side of the relationship. I think that you may have to refresh the values (I just posted a couple of days ago requesting help on the most efficient way to do this) since they are stored, but then you are able to make the relationship. Another option would be to do a find by script. If you have an example of your setup, it may be helpful to post for the folks here to troubleshoot.

I think that yours may be the best strategy; I'm not using Server yet, but hope to be soon. Do you have advice for the most efficient way to refresh? I wasn't aware of a built-in way to run a script at a given time each day, which is why I was thinking of a script on opening the file, but a daily script would probably have less adverse effect on performance. Should I just make the script go to the layout with the calc on it and view it in List View, and do I need to do it with the screen unfrozen? Pardon the novice questions, it's a new area for me. Many thanks for your help.

I'm having difficulty with creating a filtered portal on the criterion of age. I am trying to make a portal provide choices filtered by input from the user in fields of LastName, FirstName, Gender, and (DOB or estimated_age). It is actually working to filter by user input, but the age calculation is incorrect because it currently is stored. This is not a big deal for a 50 year old (the estimated age looks up +/- 20%), but is a substantial issue for a 14d old if the calculation is not refreshed for a few weeks. The example file is attached. The user can put in the LastName, FirstName, Gender, and estimated age. The user input is placed into a global field and a calculation creates the multikey for the relationship (for text, "text" & ⁋ & "textzzz"; for estimated_age, estimated_age * 0.8 & ⁋ & estimated_age*1.2). However, the multikey looks at the child table for the exact age in days, a stored calculation using Get (CurrentDate) - DOB. As a stored calculation, the date gets stale unless the record is refreshed; as an unstored calculation, the relationship breaks. Looking on other posts, I see that the child side of this relationship cannot be unstored, and since the parent side of the relationship uses a calculation from a global, I don't think I can reverse the relationship. Can anyone suggest a workaround? I have considered refreshing all the records on opening the file, but it may be many thousands of records, and I worry that this will be a significant performance hit. Many thanks in advance for your wisdom. UnstoredFilterPortal.fp7.zip

Brilliant! Thanks for the suggestion! I don't think I would have ever thought of that.

I'm hoping to improve an interface issue I'm running up against. I've created a portal which shows a checkbox list with a global filter. What I can't seem to do is to make the checkboxes go away completely. Pardon the dopey sample file, but what it is designed to do is: allow the user to select which colors of fruits are to be described in a given bowl from a checkbox list (global value) -> choose what size of fruit we want to describe first -> choose the color of that fruit -> choose one or more fruits that fit the description. If there are no fruits that fit the description, I get the response. With conditional formatting, I am able to make a condition that if the value list is not valid, set the text color to the background color. However, unless the color is black, I can't hide the first checkbox because it seems it is always black. Pardon the poor description; the attached image file makes it more clear. I've seen other people avoiding checkboxes altogether using repetitions, scripts, and buttons. I'm a novice and this seems quite involved, but I can go that route if there's no simpler way, but as this seems like a fairly small item, I wanted to ask for other suggestions. Thanks in advance. Regards, Bob Screen_shot_2009-10-26_at_2.05.08_PM.pdf BowlFruit8.fp7.zip

Incredibly useful series of posts on conditional value lists; many thanks - I believe I'm starting to get an inkling. I do have one question, which I'm asking so that I might better understand the various alternatives. Instead of having one TO of table and two of values as comment did, I got similar results from creating one table each of the values themselves and of a value catalog. Is there a way to assess which approach will be most efficient/clear? VeggieTales.fp7.zip

Any help at all with the above would be incredibly appreciated!

I am still looking for any advice on portal sorting in a value list. I have made slight progress since my prior posts, but am stumped and would welcome advice. As it stands, my example file does allow me to select a size from a correctly ordered value list, and then displays correct options from the conditional value list in the name field within the portal. What I cannot figure out is how to get the portal to sort in the same order as the value list, and to aggregate multiple name values for each size. This model, which is quirky, is to allow users to describe something with checkboxes; this model would be useful if the user were asked to catalog types of fruit in a fruit bowl by size. I've attached the fp7 file as well as a mockup the desired output. Any assistance would be greatly appreciated. BowlFruit5.fp7.zip desired.pdf

Apologies, first attempt to upload a file didn't go well. Par for the course for me. BowlFruit2.fp7.zip

I hit one additional snag - the SORTtable that Bruce and Michael made with the calculation works for the dropdown selection lists. However, when I try to use the same Value List to sort a portal, it doesn't seem to work. I'm attaching a file, in this case trying to sort fruit in a bowl by size. When I use the calculation for the value list for size when choosing a size in the Fruit layout, the dropdown works flawlessly. When I try to sort the portal, however, it only sees the size value from the fruit table and reverts to alphabetical order. Is there some clever way to make the fruit in the portal appear sorted by size. Pardon my being so thick!

This is an incredibly powerful and useful tool that Michael and Bruce came up with. It is a solution to a similar problem I had. I have one additional question as a neophyte. How do you all recommend annotating this fix in a solution? I can fix my problem here, but what is the standard way to mark it up so that the next person who takes a look at this (even me!) doesn't wonder what the heck the calculation field is for?

Hi there, I also have an open case with Filemaker, and also have been unable to open ports 1608-1620 for FMS 9. Has anyone figured this puzzle out?

This is probably not the way to go, but I'm using a complex find I modeled after Matt Petrowsky's tutorial which seems to work well to easily create a query. Specifically, I take the 750K records, create value lists which capture all of the different types of values (in this case, they are documents which have an identified Author, Document Type, and location entered). After I do the find, I take what is not a much more manageable 10K or so records and use the portal to generate lists of documents for a given day (or week or month) and perform a count. I can then export the count of documents by month to Excel to graph how the trends are going for a given Author, Document Type, etc. As it is, the complex find takes about 2 minutes to run, but because I haven't figured out how to work on just the found files without deleting the 740K others (importing of which takes about an hour), the fastest solution I have found is to Export the found set, and then reimport it into a new database. Bizarre, clunky, unwieldy for sure - I have just cobbled this together, but I definitely think there must be more straightforward ways of doing it. The portal may be dumb or unnecessary - I just picked it because it seems handy to be able to coast over to Aug 12 and look at the complete list of documents from the found set that were done that day. If anyone has some advice on how to simplify this mess, please let me know. Thanks!

Thanks very much for the helpful thoughts. I agree, the forums are just tremendous - I have to say, though, I frequently feel like a chimp trying to fix a VCR when I look at the example files, but I probably just need a little more patience. I think I just need to do exercises to get sharper at this, because just about everything seems conceptually simple, but getting things to do what I want them to do . . . another matter.