Jim Lah

I find it difficult to follow a script that bounces around up and down records in a loop. Why not try a more mathematical approach whereby you count (in a calc field) the quantity of items for each shot and simply subtract from that the quantity of those items from the previous shot. Jim

Many thanks - that does help my understanding a lot. Obviously I would like to find a way of doing this that minimises the amount of effort. If I do as you say and export to separate flat files for each table - then I end up with a series of files that the user has to ensure he keeps together. Unless there's a way of combining them into another file or package. The purpose is to be able to archive and share data between users, eg by sending files via email for import into someone elses client. There's no server available. My challenge is that the real application has about 10 tables, some related data, some not. Each table may have 20 or so fields. So writing XML and XSL this way is likely to be error prone. What would be the recommended approach to this?

No, I wish to create a single export XML file that contains the data of multiple tables, that I can then re-import as needed to the relevant tables one at a time. Jim

I am new to the world of XML and XSL and am struggling with how to transform a Filemaker output which is a series of table outputs all related back to a single record of the form: <?xml version="1.0" encoding="UTF-8"?> 0 249 250 251 252 253 254 255 256 257 258 259 260 261 249 250 251 252 253 254 255 256 257 258 259 260 261 MLP John Bill James Andy Jamie Danny Chris Tom Arianna George Beth Charlie 249 249 249 249 250 251 252 252 252 253 253 253 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 M1 M2 M3 JJ1 JJ2 JJ3 B1 B2 B3 J1 J2 J3 A1 A2 A3 JJJ1 JJJ2 JJJ3 D1 D2 D3 C1 C2 C3 249 249 249 250 250 250 251 251 251 252 252 252 253 253 253 254 254 254 255 255 255 256 256 256 Into a form that will simply import one of the sets of Data into a table of the form: <?xml version="1.0" encoding="UTF-8"?> 0 249 MLP 250 John 249 251 Bill 249 252 James 249 253 Andy 249 254 Jamie 250 255 Danny 251 256 Chris 252 257 Tom 252 258 Arianna 252 259 George 253 260 Beth 253 261 Charlie 253 Ie data from a single table. Can anyone point me to an easy reference that will get me there. Thanks, Jim

I would like to resume a paused script. My challenge is that I wish to use a script trigger (rather than a button script) to accomplish this. This is because I am using a web-viewer to create an HTML system button with hover functions etc on a pop-up window. Is it possible to restart a paused script from within another script - or is there another work-around? Thanks in anticipation. Jim

Thanks both. You are right, it is documented now I read it (twice ) again ! Jim

I have just tracked down spurious behaviour in my database to the MiddleValues function. Using it to extract a single value out of a list results in the single value plus a trailing return character. Whilst this can be removed with substitute function I was wondering if this is deliberate behaviour of MiddleValues as it does not seem consistent with the function description. Can anyone advise me? Regards Jim

OK - file is attached. The drop down field will select different tabs. You'll notice that the webviewers appear in the wrong tabs using the drop down, but in the correct tabs when selecting the tabs directly. Also note that the incorrect webviewers have different movements when resizing the window. Tab_Problem.fp7.zip

I have a tabbed layout with Webviewers on some of the tabs. The tabs are named objects and I have a script which uses Go To Object to open the desired tab. If I click on the tab controls, everything displays correctly, but if I use the script move between the tabs, all of the webviewers display on all the tabs. I have checked the obvious, ensuring tab controls are inside the body, webviewers are inside the tabs etc but cannot find the problem. Does anyone have any ideas I should try? Thanks. Jim

I am trying to use the webviewer to view stored documents (pdfs and jpgs) by using the filepath and filename stored in a field ( appropriately modified to give me a "//drive/folder/filename" format. The pdfs and jpgs are stored in a folder in the same folder that the FM7 files is stored and I use Get (FilePath) and Get (Filename) to get me the path to the documents. It works fine for documents held on my hard drive but when I try to refer to the path for documents held on my iDisk (OSX 10.5) the webviewer reports that the document is not found. Has anyone experienced the same or found a solution? Regards Jim

Ah, yes found it on Technet, thanks. Jim

I have noticed different behaviours between using PopUp field types and other field types (Drop Down, Calendar etc), especially as to the values which can be grabbed by the Get (ActiveFieldContents) script step. In all other field types - the script step grabs the previous field value, whereas with a popup style, the new entered value is grabbed. Also noted is that PopUp requires an explicit commit (either scripted or by mouse clicking) and therefore has a different result when using Tab Orders on a layout. The attached file may be of interest to those trying to use script triggers as it demonstrates the impact of different field styles and different script triggers. If anyone can explain why PopUp is different to the rest, and whether it's a feature or a bug it would be interesting to know. Currently this issue prevents me from using the more desirable PopUp in some of my programmes. Jim POC_ScriptTriggers___Popups.fp7.zip

I am having difficulty reconciling the results of the attached file. My ultimate purpose is to simply keep two tables in synch for a data separation model. The attached file is an extract to show the problem I am having. I am using a calculation c_current ID's to generate a list of the current ID's in the target table. I've added a "null" value in the first record so that it still works when no data is present. That works fine. I use that calculation with a not equal to relationship to show those records in the source table which are missing. But I get very variable results. Adding a new record to the target table and filling in an ID results in the c_current ID's being updated correctly, but the portals showing the missing records don't update. They do update when a new record is added to the target. Scrolling back through the records also shows a strange effect. Even though the c_current ID's calculation always shows the same result, the resulting records in the portals are different. Opening and closing the edit layout resets everything to be as expected, but adding a few more records to the Target causes the same problem to occur. What am I missing? Is it a refresh or commit problem somehow? POC_Copy_Records.fp7.zip

It's certainly confused me as I have been using IsEmpty () thinking it was a reliable way of testing for a record's presence as well as for content of the field. In the attached I've taken it a step further by comparing results with using If ( x = "" ; "Empty" ; "Not Empty ). The results are even more confusing - none of them simulate the (correct) result as shown in the Portals. If I check the box to evaluate when records are not present, then it simulates the IsEmpty function - giving me Virtual records of some kind. If I check the box not to evaluate if records are not present, then is does what you would expect. Attachment shows the results. POC_Table_Error.fp7_2.zip

Yes, Correct. That's all the calc is doing in this example. If there's nothing there is should say "Empty" - which it does in the Primary field My problem is that where there is no valid relationship it also shows "Empty" in the secondary field - even though there should be no record there.

Please can someone help my understanding of a simple relationships issue. The attached illustrates the problem. I have a single table, with an ID and Parent ID to relate the tables occurrences together - 4 tocs are included each related in the same way. I have a simple calculation which looks at the ID of the child toc. If the field is empty, it assigns "empty", if not it assigns "not empty" - ie simply detecting the presence or otherwise of a related record. When I show the Portals for the 4 tocs in the layout, they display "Empty", "Not Empty" or nothing (if no record there) as expected. However, if I show the associated field values, they always show "Empty" or "Not Empty" - even when there is no associated record in that relationship. How can this be if there is no record present to actually display the calculated result? I'm sure someone out there will be able to explain this "feature". Jim POC_Table_Error.fp7.zip

Søren You are of course quite right - this is a much faster way to create a list of events or dates. My challenge is that I need records so that I can fill those records with other information and relate those records to other tables. Unless I am missing something I don't think there's a way of doing that with repetitions or lists of data unless new records are created. Yet creating new records seems to take an age (or two). Unless of course you know otherwise... Jim

I have the same issue with Filemaker speed - particularly with creating new records. Generally I'm lucky if it will add 50-100 records per second (and that's if I keep the number of auto enter calculations down to a minimum). On a processor that is running at 5,000,000,000 operations per second that seems a tad slow. I know its doing other things but is there no other way to speed life up. Yes, freezing windows and having no fields on a layout speeds up slightly - but I was expecting "click of a finger" types of speed rather than "have another cup of tea" speeds. Am I missing something? Jim

Thanks - Verdana is certainly much more usable.

It does appear to be a font rendering problem - it looks like a different system is used for pop-ups than drop downs. It looks like a bit mapped font of old. The font used in both fields is Helvetica - both of the same style. I have tried differing fonts and run the same example on different Macs - same issue. I have tried FM10 and FM9 - same issue. I don't seem to be able to find a font that is anything but scrawny on a pop-up at around 12pt. Interesting that the problem appears not to occur on XP. The file has not been converted through Win at any stage. I would really like to use the pop-up feature but this font rendering make the whole application look ugly.

Example is attached - very simple - it shows that Drop Down's include nicely formatted text, whilst the PopUp generates ugly text. PNG's are included to show what it looks like on my Mac. DropDownEx.fp7.zip

I'm not aware of any system utilities installed - but thanks for the reply. Does anyone else have any ideas? Jim

Can anyone explain to me why the fonts being displayed in my pop-up and drop-down value lists are different? The drop-down list provides a nice clean font, whilst using a pop-up for the same list provides a scrunched up font - barely readable. The attached screen shots show the problem.

Ah, got it now. Couldn't see the spaces in your sort field. But now it works. Thanks.

I have been struggling with this too - want to use the value list in a drop down but sorted - not alphabetically. Closest I have found is to reverse the order of the fields in a value list so that the sort field is the second field displayed. By making the sort field a long series of blanks with the sort number appended at the end then the sort number doesn't appear within the field on the layout - though it does appear within the drop down which still looks ugly. I feel there must be an obvious answer out there somewhere ...