Jump to content

Dimitrios Fkiaras

  • Content Count

  • Joined

  • Last visited

Community Reputation

0 Neutral

About Dimitrios Fkiaras

  • Rank
  • Birthday 07/30/1972

Profile Information

  • Gender
  • Location

FileMaker Experience

  • Skill Level
  • FM Application

Platform Environment

  • OS Platform
  • OS Version

Recent Profile Visitors

952 profile views
  1. Well, the solution seems to work... What amazes me is how fast FileMaker makes all these calculations. I'd expect that 2000 calc reps x 75,000 records will take a long time. Even in Webdirect there is a delay of just 1 sec (LAN). Maybe my knowledge is a bit limited on this matter also. *New: Even with 20000 reps X 75,000 records there is no delay. Why?? After maybe the 2000th rep all other calc repetitions result to "empty". So we can assume that FileMaker engine can summarize empty repetitions instantly?
  2. Well, if I select a date that lies before or after the 2000 reps I get a "?" in the calculation. I defined 2000 reps for the sake of the test. I am thinking to define maybe 90 reps. If the user selects a date before the initial date or after the final rep date, I am thinking to show a warning like: "You have selected a date range that is not expected. It will take some time to calculate --- OK, Cancel" and then set a different initial date so the 90 reps will cover the desired range As for the $variable... no particular reason... my ignorance... I removed it. I will test the solution on WebDirect and let you know of its performance.
  3. Thanks! I don't know if I am being very dumb, but this version seems to work pretty fast. CrossTabR_Test.fmp12 PS. When I created the test data I didn't take care to avoid overlapping absences, so summarized data could be off sometimes
  4. Thanks a lot for the explanation! As for the second part, you are partly correct. The people responsible for approving leaves etc. want to have the big picture of absences for... lets say -3 / +15 days in order to be able to decide. There's no reason to navigate throughout the past... or the empty future. Just a little back and forth. Unfortunately the Refresh Window script step takes soooo long to complete on Webdirect (FileMaker Pro handles it just fine, even on legacy PC's). All in all we just need to summarize around 150-200 records (100 people with approximately 1-2 leaves in 15 days). That is why I thought of filtering the relationship.
  5. Thank you very much! The calculation is very elegant! The problem is that when I change the gReportDate, the summary field refreshes only for the selected record!! I tried moving the fields gReportDate and cDateR to the Personel table, only to confirm the same behaviour. Then I filtered the relationship by adding gReportDateStart and gReportDateEnd fields, and the summary fields auto refresh. Why don't the repeating calculations evaluate with the change of gReportDate?? CrossTabR_Mod.fmp12
  6. Thanks a lot! Though I feel more comfortable with native FileMaker solutions!
  7. Hello to all, I am building a solution, where the user needs to see a crosstab absence sheet of their personell. The people there can be absent for a variety of reasons, (absence types) which are recorded. The users want to be able to select a date and see the absent personell for a certain date range. I tried very hard to find the a way to do this, so that the app will be responsive. The solution is accessed via WebDirect (so I tried to avoid the Window Refresh script step) So far I have come with two ways to do this. The first way is based on a repeating calc and the second one on a summary field (which summarizes a repeating calc) See the attached files (the records in the real solution, will not be as much. Maybe 1/10th.) How would you approach this problem? In the past I tried to solve this problem with filtered portals in each date, or with virtual list. Both solutions didn't seem to be fast enough (although I admit I haven't spent the time to optimize them, like the current solutions). I also thought about creating a summary table somehow but cannot finish my though on this. I suspect that there may be a better way to do this (maybe without calculations, only relationships). Any help is greatly appreciated. Thank you all! Absence_sheet_demo_1.fmp12 Absence_sheet_demo_2.fmp12
  8. What exactly is a clinic? Can you send a screenshot from you relationship graph?
  9. I want to conditionally format the items (btns, fields) in the active row of the master portal. Thus my formula is : Get (ActivePortalRow) = Get (RecordNumber). This works ok, but when I click outside the portal, the formula no longer validates correctly. (however, in the Data Viewer, the Get(ActivePortalRow)=Get(RecordNumber). I am using FM17 on Windows 7 machine.
  10. Sorry for continuing this thread. I haven't got any success in showing a local .html file thru WebViewer I entered the absolute path, as that was calculated by the example file in post #2 (for a MAC computer) The webviewer seems to find the file but give me a message that a suitable application was not found to open the file What is wrong? ** UPDATE ** Solved : I had a problem with slashes "/" "file://Macintosh HD/Users/JohnDoe/Desktop/Folder/File.html" correct format
  11. Well, I am currently working to accomplish something similar, so although I am quite novice I would like to express my thoughts. I think that its very good you have a clear view of what info you wish to see. Also I think that it is important to imagine how you would like to see all these data and with what interaction by the user. For example : Maybe you see all computers in a portal with 3 columns [Name, Hardware Serial, Building] then you click on a computer and then you sees another portal which has the columns [SoftwareTitle, Lisence, Verion, PO, Date] Its important to have an idea about the user interface and data presentation, because sometimes you have to create new Table Occurences in the Graph just to create the appropriate "paths" between data to present them in the way you imagine (this is vital if you plan to use portals). I don't know if I am correct but I think that every time you wish to display in the same layout lists of data that are related you can't avoid portals. Another trick I learned lately is this : If you want to present a portal with all the computers you can't base the layout on any of the related tables because it will show only related records to (P.O, Locations etc). So you can created a table with just one record (lets call it VIEWER TABLE) and relate that table primary key to a global field in the ASSETS table. Thus all ASSETS will be related to the one and only record in VIEWER TABLE. So when you make a layout based on VIEWER table and a portal based on ASSETS you will see all ASSETS. By filtering the portal by the Asset Type you can get is to show only the Computers Glad if I helped Dimitris
  12. Hi, In an Asset managment solution, do I absolutely need these tables: Asset (e.g. HP Photosmart 102b)------< Individual Assets (e.g. HP Photosmart 102b | s/n:1234567890)---------< Transaction Items >------------Transactions ? Currently I don't have an Individual Assets table and all Individual Assets are created in Transaction Items table and a Serial number is assigned to them there. When I sign out an Asset, I create a new Transaction and when it is returned, I flag the Transaction as VOID. I know I have that Asset at hand if it appears only ONCE in my CURRENT Transactions. (VOID Transactions are excluded via self join) If it appears only once it means that I have only one relative Transaction the one that I created when I first received the item). If I have two, it means that I reveived it and then made another Transaction (signed it out). Is that a correct approach? What are possible pitfalls? I know for sure that this approach is used when handling Inventory, I don't know about assets but if feels a bit odd having this structure of mine... I am a bit confused about it.... Thanks Dimitris
  13. What are the things to have in mind if implementing the supertype/subtype model? I hope this is not a too general question
  • Create New...

Important Information

By using this site, you agree to our Terms of Use.