Jump to content

balooka

Members
  • Content count

    128
  • Joined

  • Last visited

Community Reputation

0 Neutral

About balooka

  • Rank
    member
  • Birthday 10/04/1972

Profile Information

  • Gender
    Not Telling
  1. text field vs numeric field in sorting issue

    Ouch! I was looking at the wrong place for that setting then. I assumed that field formatting overruled any and all preference. Just as a side question... If you make a new calc field where the formula is 'A_NUM_T' and the output is text, it shows the original value again. Does that mean that the value as entered remains the same, even if the numeric field is not displaying it?
  2. text field vs numeric field in sorting issue

    I'm not using the systems settings as far as I know - but the separator is a comma. Is this the cause of all this?
  3. text field vs numeric field in sorting issue

    OK, I have made a file with two tables, A and B. In table A I want to insert an Edition (I only made the edition 'A' so anything else will not match). In table A I also add the number I need and what this is all about. There are two fields in table A where you should put the same value (like 1.1 or 11 or whatever). The A_NUM field is numeric, the A_NUM_T is a textfield. I have made two relations, one for each. The result on the Table_A layout should be the name of the related file. You can view both results in the Table_A layout. The values in Table_B are unique, no duplicates. match.fmp12
  4. text field vs numeric field in sorting issue

    Thanks Comment, But I have no clue left where to find it to be honest. Somehow if it's a textfield it does recognize '1.1' correctly. if I use it in number fields it does not. If I calculate the contents of a textfield into a number field the result is 1.1 again. I'm confused about why it recognizes the 1.1 to be a valid outcome in a calc field and not when I use a number field.
  5. text field vs numeric field in sorting issue

    That could very wel be it... but I have found another way to tackle this. I added a calc field with numeric result and as a calc I used the textfield. The value of that calc is 1.1 where is was 1.1 in the textfield... odd. Anyway, I can use that to sort my records again.
  6. Hi I have a table with a 'lesson-ID' that looks like 1.1, 1.2, 3.2 etc, these are subcategories of field 'type'. Each ID also jas a name to it. For instance 1.1 has is named ONE and 1.2 is named TWO. I have an exams table that looks up values from the ID table using a global (type A, B or C) plus the subcat (1.1, 1.2 etc). I need to look up values from the ID table using the type and ID -> A 1.1 / A 1.2 The fields are of the same kind in both tables, type is test and ID is number. My issue is that when I do a match it sees the '1.1' as '11'. It will match up to both 1.1 AND 11 it looks like because it refuses to show me the name (there is more than one match!). If the ID is 3.13 I have no errors since there is no ID 313 and therefore there is no double record. I can fix this by change the fields to text fields, then it all works just fine, however, my sorting goes bananas because it now sorts 3.10 right after 3.1 Hope that's clear? How can I get it to match 1.1 as 1.1 and not 11? I'm not sure where to look (took me an hour to find the source of the problem!). Thanks!!
  7. Filemaker Server and Apple Script

    Thanks Wim, I assumed that because I was able to select the script in the server (using my admin username/password) I would be able to run it. I didn't think about the user rights on the machine itself as script maker ran the script on the FMS just fine. I'll look into the this now. Thanks for helping!
  8. Hello I need to run a daily script to unmount and mount a volume using applescript. The script runs fine in AppleScript but does nothing (as far is I can tell) in a scheduled script in FM server. Is this a problem of location of the files (on the server) that causes the AppleScript to not function? Testing the same on a local copy of the file works just fine. The script I'm trying to run inside FM: tell application "Finder" try eject disk "directie" end try end tell set my_volumes to "directie" set the_volumes to list disks if my_volumes is not in the_volumes then try mount volume "smb://192.168.0.169/directie" as user name "directie" with password "Vekabestadm2016!" end try end if Thanks!
  9. Hello Wim, I think I might have misread your solution. I already have a web viewer on my layouts and somehow I only half-read your solution. I thought you meant to view the whole thing inside a webviewer.... my mistake! I'm going to try that right now!
  10. Hmmm I think I have it working now... looking for the flaws that I missed. I have an on commit trigger to set 2 $$vars with record ID and timestamp if there was a modification in the record (empty otherwise), the account name is the current user so no need for a $$var there, So every time it is triggered by a modification and commit it will set these two $$vars to the exact same values. The second trigger is the Onload script trigger that will check if both $$vars are empty. If both are in use, meaning there has been a modification in the previous record, it will set a new line in the log using the $$vars and the account name. Clearing both $$vars after that will reset the script trigger. One flaw I already see is that a user could close the database after the last modification. ETA The two Vars somehow did not transfer as I would like them to. Solved that by using two globals to hold the content of the recordID and the timestamp. The vars are only used when adding the line because I needed to clear the globals before the end of the script (started looping).
  11. Thanks Wim I seem to always assume that some things are pretty basic and would not involve that much hassle. I'm not in the position (time wise) to redo my designs, unfortunately. I tried to work around this by having am 'on commit' script trigger to set a field to 'yes'. So even if there are 10 modifications in a single record, the value would still be just one 'yes'. I then made a second script that was triggered by the 'OnLoadRecord' script trigger (not sure what the English translation is) and using $$vars to see if the previous record was modified and if so, write a line to the log. But there are many flaws in this (if a user is not going to a next record or when the next action is a search etc). Why isn't there a script trigger for leaving a record... Thanks for the help!
  12. Hi Comment, You are correct when a user has modified a record ten different times it would need to show up ten times in the list (with a different timestamp). I have that now, but I saw a massive record increase in the new table and found that when a user opens a record and modifies a field and clicks somewhere on the layout in the same record, it commits. If a user modifies 10 fields and clicks on the layout between modifying the fields, it would commit 10 times. The result I now get is the same user modified the record 10 times when all that the user did was modify the whole record just once. In other words, this user only should have been mentioned once because he/she modified the record once, multiple fields, but just one record. Problem I have seems to be that the record commits when you leave a field in a record, rather than when you leave a record. A list of ten would indeed be much more efficient, I have to figure that out then (unless you have a working snippet for me?)
  13. Hi I would like to have a log or list of users that modified a record, but I do not really need to see what they have modified. Anyway, I have tried using the on commit script trigger to fill a table with record ID, timestamp and account name, which in a way works but the problem with this is that every time something is modified in a record, like a radiobutton the script will trigger if the user clicks somewhere else on the same layout and same record. So it commits a lot of times, filling the table with countless and useless entries. I would like to see only the ten (or so) last users to modify the record (not those that only viewed it!). I can use a timer to do a nightly run and delete all entries that are over 10, but I have no idea how to stop it from filling up the table. My script is now: if Get(ModifiedFields) = "", do nothing else set current record and user values in a var and create a new record in the acces table. It's triggered by the on commit script trigger. (My FM is Dutch so the translation might be a bit off). I am aware that there a solution that track everything, but all I need is so see who actually accessed and modified a record. Thanks!
  14. But to make use of the sub-summary I must sort on the A-field values, correct? I need to sort on another value and not disturb the current layout. ETA: Solved it in another manner using pattern count. This way I can sort on whatever field I need and see the individual summary values. FIELD PATTERN ONE would be: PatternCount ( Field_A ; "ONE" ) This will look for the pattern in field A and a resume total of that field FIELD SUMMERY ONE would be: Resume total of FIELD PATTERN ONE
  15. Hahah that would indeed get very boring, but this solution might just work fine for 5 options. I'll give this a shot - thank you very much!
×

Important Information

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