cmcfarling

I have a shared webhosting account through a hosting provider. PHP is installed on the shared web server so I have no problem creating PHP pages. However since this is a shared server (running Linux) I can't install any 3rd party software. In addition I have an in-house server running FileMaker Server. With this setup will I be able to utilize PHP custom web publishing? What's unclear to me is that the manual states that the "Filemaker Web Server Module" must be installed on the web server. What exactly is the Web Server Module. Is that nothing more than a set of PHP files, in which case I assume this would work. Or am I missing something?

What's wrong with creating records via a portal? I've used this method many times with no unexpected side effects.

I did in fact test this by sharing the database over the network and trying to edit the same portal row from 2 machines at once. FMP won't allow it. I took this a step further and discovered something more. I created two separate portals that both access the same group of related records. One related record, and therefore one portal row, already existed. That was Number 1. On ComputerA I edited a new portal row in Portal A but did not commit it. On ComputerB I edited a new portal row in PortalB. At this point in time both portals had uncommited changes in them. The expected behavior, once both were commited, would be that I would end up with duplicate Number fields. Two Number 2's in this case. That was not the case in my testing however. It appears that FMP did not assign the Number field until the commit event. So the premise that duplicate Number fields are a potential issue appears to be false, at least in this situation. Note that the relationship has "Allow creation of records in this table via this relationship" enabled.

I understand LaRetta's point. My point is that Filemaker won't let that happen though, so it's not an issue. In this case the user creates all records through single portal. The only way to create a record is by typing into the portal's empty row. Filemaker won't let 2 users create a new portal row at the same time. I misspoke when I said I minded numbering gaps, wasn't think apparently. Also,I have considered the scenario where record 6 gets deleted, record 5 becomes the MAX and record 6 gets issued again. I don't think that's going to be a problem, I'll ponder it some more though.

In fact I do mind numbering gaps and ongoing serial numbers are not an option. Each group of related records has to begin with Number 1 and count up from there. I actually solved the issue anyway. Deselecting "Do not replace existing value for field (if any)" and "Do not evaluate if all referenced fields are empty" allows the following formula to populate the Number field Max ( TableSELFJOIN::Number ) + 1 The big hangup was not having "Do not evaluate if all referenced fields are empty" unchecked initially. After unchecking that, I thought all would be well. Oddly, even though the Number field is emtpy upon creation, "Do not replace existing value for field (if any)" has to be unckecked as well in order for this to work. As for multi-user concerns, the records will be created through a portal. There is no chance of User-A creating a record and not committing it before User-B creates a record. FMP will display an message box if User-B tries to alter the same record that User-A is actively in.

Regarding the multi-user mode issue, the question is when does FMP evaluate an auto-enter calculation, on record creation or record commit. If it's on commit then I dont see that as a problem. Does anyone know the answer to that? Here is why I need to do this. When a record is created the Number field is significant because that will be used as an identifier for that record. The first record created is Number 1, and always needs to be Number 1. Say 3 records exist with Number field set to 1, 2 & 3 repectively. Now the second record is deleted leaving record 1 - Number = 1 record 2 - Number = 3 I need to be able to reference the second record by it's Number field, 3 in this case. Using the @ method does not allow that.

I'm not clear on what you mean. Why can it break in multi-user mode? Please clearify "simply display the related numbers you need"

Evaluates to an empty field as well

I need to number records in a self join relationship. A simplified table might look like this: JobNumber Department Number 12345 Sales 1 12345 Sales 2 12345 Shipping 1 12345 Sales 3 A self join is created on the JobNumber AND Department. The Number field is an auto-enter calculation. The first occurance of a particular relationship has to be Number 1, the second Number 2 and so on. The problem I'm having is calculating Number 1 for the first occurance in a relationship. Obviously no records match the relationship prior to the first record being created. I need to test for that condition where no records exist in the relationship and then set the Number field accordingly, as an auto-enter calculation mind you. If I use If ( Count ( TableSELFJOIN::JobNumber ) = 0 ; 1 ; Max ( TableSELFJOIN::Number ) + 1 ) Filemaker appears to not evaluate the formula because the Number field ends up empty. To simplify and test this, I shortened the formula to simply Count ( TableSELFJOIN::JobNumber ) however that returns an empty field as well. So I tried IsEmpty(Count ( TableSELFJOIN::JobNumber )) to see what happens and that return zero. Likewise IsValid(Count ( TableSELFJOIN::JobNumber )) returns 1. So, something is being returned from Count ( TableSELFJOIN::JobNumber ) in this case but it's not a value that I can test against it appears. I've also tried Count ( TableSELFJOIN::JobNumber ) = "" and Count ( TableSELFJOIN::JobNumber ) < 1 to no avail.

Ok, got it. I have to add the releated fields from the Customes table (Name, Address) to the layout for the related table that I will export from. Easy enough just not obvious at first I guess. Thnx

As an example I have a Customes table. The main fields in that table are Name & Address. In a related table are individual contacts. The main fileds there are FirstName,LastName,Email. Each customer has multiple contacts. I want to export the data from the Customers table as a tab delimted file to be opened in Excel. Each line (record) in the .tab file should include all of the fields noted above. So, the .tab file should look like this: Name,Address,FirstName,LastName,Email Name,Address,FirstName,LastName,Email Name,Address,FirstName,LastName,Email Name,Address,FirstName,LastName,Email However, when I do the export, FMP exports a record from the customers table and separate records for each of the customer contacts. So the .tab actually looks like this: Name,Address , ,FirstName,LastName,Email , ,FirstName,LastName,Email , ,FirstName,LastName,Email , ,FirstName,LastName,Email Name,Address , ,FirstName,LastName,Email , ,FirstName,LastName,Email , ,FirstName,LastName,Email , ,FirstName,LastName,Email Name,Address , ,FirstName,LastName,Email , ,FirstName,LastName,Email , ,FirstName,LastName,Email , ,FirstName,LastName,Email Is there a way to make FMP join all of that data on export so it is formatted like the fist example?

I have a number field formatted as a decimal with 2 fixed decimal digits. So if I type 100 into the field, it displays as 100.00 Is there a way to get the displayed value out of that field (100.00 as opposed to the actual field value of 100) to display in a text field? Or, do I have to monkey with the number by adding decimal digits (via a calc) if they weren't manually typed?

Looks like this will do it. (Note that I'm only accounting for numeric characters in this example) Let ([ list= RepeatingField[1] & "¶" & RepeatingField[2] & "¶" & RepeatingField[3] & "¶" & RepeatingField[4] & "¶" & RepeatingField[5] & "¶" & RepeatingField[6] & "¶" & RepeatingField[7] & "¶" & RepeatingField[8] & "¶" & RepeatingField[9] & "¶" & RepeatingField[10] & "¶" & RepeatingField[11] & "¶" & RepeatingField[12] & "¶" & RepeatingField[13] & "¶" & RepeatingField[14] & "¶" & RepeatingField[15] & "¶" & RepeatingField[16] & "¶" & RepeatingField[17] & "¶" & RepeatingField[18] & "¶" & RepeatingField[19] & "¶" & RepeatingField[20]; string= Substitute( RepeatingField[1] & "x" & RepeatingField[2] & "x" & RepeatingField[3] & "x" & RepeatingField[4] & "x" & RepeatingField[5] & "x" & RepeatingField[6] & "x" & RepeatingField[7] & "x" & RepeatingField[8] & "x" & RepeatingField[9] & "x" & RepeatingField[10] & "x" & RepeatingField[11] & "x" & RepeatingField[12] & "x" & RepeatingField[13] & "x" & RepeatingField[14] & "x" & RepeatingField[15] & "x" & RepeatingField[16] & "x" & RepeatingField[17] & "x" & RepeatingField[18] & "x" & RepeatingField[19] & "x" & RepeatingField[20] & "x" ;["0";"z"] ;["1";"z"] ;["2";"z"] ;["3";"z"] ;["4";"z"] ;["5";"z"] ;["6";"z"] ;["7";"z"] ;["8";"z"] ;["9";"z"] )]; MiddleValues ( list; 1; 20 - ( Length(string) - (Position(string;"z"; Length(string);-1) +1) ) ) )

Those "........" are just part of the string, they are period characters. Any one field would contain just one string. I was just giving 3 separate examples. The field would be text. Ultimately I need to do this... I have a repeating filed with 20 reps. Some reps may be blank and some may not. For example: rep1: blank rep2: blank rep3: data rep4: blank rep5: blank rep6: data rep7: blank rep8: blank rep9: blank rep10: blank rep11: blank rep12: blank rep13: blank rep14: blank rep15: blank rep16: blank rep17: blank rep18: blank rep19: blank rep20: blank I need to format the data from this repeating field in a way that allows it to shrink down for printing. However, I also need to maintain any leading blank spaces. Repeating fields won't shrink so I need to create a calc field that contains the info in the repeating field up to the last data value. So my calc field would be a text field that looks like this: blank blank data blank blank data My thinking was to concatenate all of the repetions together into one big string while substituting blank reps with periods (or some other character). Then I would somehow find the first non-period character in the string and go from there. However, I think that that won't work now. What will work I think though, is to still concatenate all reps while substituting blank reps with "x" for example. I'll also substitute all alphanumeric characters with "z" for example. I'll then find the 1st occurance of "z" from the right side of the string. That will tell me which rep is the last one to contain any data. Finally I can use MiddleValues to gather all the values of the repeating field up to the last data value. I think I'm on the right track, but if someone has a better way I'd be happy to hear it.

Any ideas how I would find the position of the first alphanumeric character in a string? For example, say I have the following 3 strings: .......abc..... ..123.......... 0.............. I need to find the position of a, 1, and 0 in each of the strings with a single calculation. Thanks