Karl Dickhaus

OMG! I'm an idiot. When I copied the database to make the test script revisions I absolutely did forget to name the field in the test layout. Thanks! Although I must now go subject myself to 20 lashes with a wet JCL reference book! Sometimes it is the simplest things. Thanks again. The comment has now been added, too.

I am attempting to de-bug a dynamic sorting routine in FMP 12 Pro that dynamically sorts a list of names when letters are typed into a search field. This is based on the "Spotlight Search" template created by David Graham in 2009. http://bittailor.com. This solution uses a single table ("Contacts") and 2 Table Occurrences - "Contacts" and "Contacts_Filtered" which are related by the Search_Field in "Contacts" equalling the Search_Index in "Contacts_Filtered. Search_Field is a global text field Search_Index is a calculated text field using the custom function "ExplodeToMultiKey" which generates an exploded set of data from the entered text (e.g. Thomas becomes T, Th, Tho, Thom, Thoma, Thomas). The results are shown via a portal on a layout based on the "Contacts" table which displays data from the "Contacts_Filtered" table. In the Spotlight Search Template, there is a script that uses a second layout to simulate the initial layout expanding to show the data as it is found. In my solution that technique is not required. Instead, the display layout shows all of the people in the database and narrows as data is entered into the Search_Field. To test this technique in my solution, I updated the script to set a global text field (called gTest) to the value GetCurrentLayoutObject before completing the rest of the script. The script (as modified) is included here for ease of reference. Set Variable [$currentObject; Value: Get ( ActiveLayoutObjectName ) ] Set Field [Contacts::gTest; $currentObject ] Commit Records / Requests [ ] Go To Object [ Object Name: $currentObject ] Set Selection [ Start Position: Length ( Get ( ActiveFieldContents ) ) +1 ] The purpose of the script is to update the layout contents but keep the user in the field "Search_Field" so that they can type additional letters, delete something already typed etc… Translating this script to my solution, however, this functionality doesn't work. For some reason the script is not setting the variable $currentObject to the value of Get ( ActiveLayoutObjectName ) - and I can't figure out why not. The test modification to the script works flawlessly in the Spotlight Search template file - but not in my solution. To test it, I included the gTest field in the header portion of the layout in both files. In the Spotlight Search template, the field updates to the current object name when data is entered into the field with the script trigger. In my solution it does not. The result is that once you type a single letter in the Search_Field it works, but typing a second letter (or deleting) gives you an error message that you need to select a field. This is proving maddening - so any thoughts are very much appreciated.

Perhaps I am getting confused regarding the naming of the relationships. Is what you are saying that I should create table occurrences for each of the proposed relationships? For example, if there is a relationship Organization to Project, which links the Org_ID field in the Project table to the Org_ID (Key) Field in the Organization Table (using the = relationship) and there is a Contact to Organization relationship, which links the Org_ID field in the Contact to the Org_ID (Key) field in the Organization Table, then wouldn't you need a new table occurrence to create the Project to Project to Organization to Project to Organization to Contact relationship?

I have a three table database in FMP 11 with the following tables: Contacts, Projects and Organizations. I have a layout that shows information on a particular project, based on the Projects table. Each project is associated with a client organization and another unrelated contact from the Contacts table - the salesperson. There is a drop-down menu on the Projects Layout for the contact person from the organization assigned to the project. This drop down list grabs information from the contacts table (a field for the contacts full name) which is sorted by the organization. For example, if Organization A has three project managers Bill Smith, Betty White and George Jones, when the organization is selected, clicking on the drop down field for Project Manager will list only those three choices from the contacts database. This is done using a table occurrence called Contacts_Organizations with a relationship based on the Organization ID field from the Organizations table. All of this works fine. What I want to do is when I click on the project manager (for example George Jones) I want the Project Manager e-mail field to automatically display the project managers e-mail, which is entered on his record in the contacts database. I want that change reflected only on that particular Project Record, not on all the project records linked to that organization. This seems like a simple enough job, but apparently I have been looking at the problem too long, and can't seem to get it to work. The relationships work like this: Contacts is linked to Projects by a join table. Contact ID from the contacts table is linked to Contact ID in the Join Table, Project ID in the Join Table is linked to Project ID in the Projects Table. The Join Table records are created by a script when the salesman is assigned to the project. The Project Table is linked to the Organization table by Project ID. In turn the Contacts_Organization Table Occurrence is linked to the Organization Table by a relationship based on Organization ID. I know its a bit simple, but all help would be greatly appreciated. If you need to see the table occurrences, I am happy to post them. TIA, Karl.

I said it was something basic I was missing. I had made it too difficult, interposing an unnecessary TO, I can't figure out from my notes why. Thank you for helping me past stuck.

I am trying to create a filtered value list based on a property from a related table. The first table is Products. The second table is Customers. I want a value list on the customers layout that shows only the products associated with that customer's SIC Code. There is a field in the customers table that lists their SIC code. For purposes of this solution, there is only one SIC code per customer. I want a value list that will show only the products from the products table that have the same SIC code as the one shown on the customer's layout. I won't bore you with the 200 approaches I have taken - I'm sure I am missing something basic, I have just been looking at the problem too long and am stuck.

I have an export script that exports a "Merge" file of data for use in a Word merge document. The script exports using Windows (ASCI) method, and outputs a comma delimited text file with the field values in quotations. On the Mac clients, Word will use this file to create the merge document with no issues. On the Windows machines, however, the file is not useable for merges because it contains two record delimiters. I suspect that FMP is using a LFCR delimiter to end the record, instead of just a CR. However, I cannot figure out how to solve the problem on either end. I cannot how to get FMP 11 to export with just a CR instead of the LFCR it uses - or how to tell Word for Windows (2010) that the end of record delimiter is LFCR or 2 CR's or something similar. There is no problem on the Mac side - everything works as intended.

I am having the same difficulty - but only on Windows platforms. I traced the issue back to the merge file created in FMP. When opened in Word, with the "Show Paragraph Formatting" turned on, it shows an extra return after the last record. Remove this extra return, and it will merge no problem. This result did not change when "Use Smart Quotes" was checked or unchecked (I suppose the previous poster was suggesting you un-check it, on the basis that the "smart quotes" would goof-up Word). I have not found a work-around, but need one. My solution runs on both platforms - but this merge functionality is critical.

I have a solution that is shared between two users - both of whom need to be able to add, delete, change and manage all the data in the solution. User 2 does not have access to the scripting functions - but can add, delete or modify any data. The solution is stored in Dropbox - in a shared folder. Both users have a copy of FMP 10 installed on their machines. They open the file and log-in as unique users (different access). The problem is that if both change records at the same time, two copies of the solution are created. One is database and the other is database (User 2's conflicted copy dated today). I want to be able to warn the second user that the first user is currently logged in - so this won't happen. I tried using a Get ( UserCount ) function - but this only seems to work across an ethernet network. Anyone know if what I want to do is possible - and how to do it? I would prefer to do it without using custom functions, but any solution is fine. TIA, Karl.

I am trying to run a single script that prints first to a laser printer over the network and then to a local Dynmo 450 Label Writer. There are separate scripts for Print to Dynmo and Print to Laser Jet - these are called from the primary script. FMP 9 ignores the Print Settings, though, and just prints to the last used printer. I have searched for non-AppleScript solutions, but haven't found any that worked. This client will not upgrade to FMP 11, so I am stuck on FMP 9. Anyone have a good work-around for this issue? It doesn't occur on the PC side - same client, same solution - only on the Mac side using 10.6.4.

I am trying to print a report in Preview Mode, OSX Leopard, that has 5 pages. When I select print, printing to an HP 4000 TN, it will only print the first page. It does not matter which page is displayed on the screen. The "print records being browsed" button is selected. This behavior repeats across all reports in all of my FMP databases. I am at a loss as to how to get it to print all of the pages.