All Activity

"The idea is to lock the parent record only for the duration of issuing a number." ja cool thats what were doing. "In both scenarios, a number once generated should never be reused. " Not quite I think; in my scenario if a user does not finish his editing of the record (ie he abandons it by reverting the record so the record disappears); the number of the reverted record is stored for later re-use. That actually works well; and it is a further advantage of the method with the auto-entered serial number. ja? I just noted I didnt actually say thank you to you Comment. Whoops. Thank you for the help, it was much appreciated. best

How do you interact with your clients in such a way that you can discover their actual needs? Richard discusses his personal tips, as well as those from several senior engineers at RCC! Future live streams: https://fmtraining.tv/#LIVE support@rcconsulting.com www.fmtraining.tv www.fmstartingpoint.com www.rcconsulting.com

We're having another discussion on licensing in FileMaker, specifically, the cost of bundling FileMaker with a plug-in or add-on product! Future live streams: https://fmtraining.tv/#LIVE support@rcconsulting.com www.fmtraining.tv www.fmstartingpoint.com www.rcconsulting.com

I have just designed a simple Webviwer calendar (using FM19 add-on). I want to use it on my iPad, desktop and iPhone and making sure that the information keyed in are synced. Can anybody advise what is the best and most economical way to do this? Someone suggested AWS but its only 1 app and for my own use.Using AWS seems to be like an overkill. Hope to hear from someone soon. Thank you all in advance.

Thanks comment and bcooney, you right, its just the matter of right clicking the script and choosing "Grant Full Access Privileges". For security measures, I just assign no access to other users to this script except the one who's gonna use it. Thanks again.

Non-admin accounts can be given the ability to manage all user accounts, except full access. see: https://help.claris.com/en/pro-help/content/other-privileges.html

True. Once you have defined the relationship, you can even delete the match fields and it will keep working. P.S. Kudos for using the term "Cartesian Product Relationship" instead of the so often misused "Cartesian Relationship".

Sorry about that--I should've written "Cartesian Product Relationship"; I do have the "X" selected between tables but used the ID as the connecting fields...which is moot because if I recall correctly, one can use any two (connecting) fields in such a relationship; it doesn't have to be the ID field. "I am afraid I did not understand the "default" filter thing. If you have a predicate that you want to apply always, regardless of other conditions, why don't you implement it as part of the relationship?" Simple answer: because I didn't think of it. : | Thanks for the suggested fix (it works well), and I'll post separately about the script triggers...and sorry about posting picture instead of code--I won't make that mistake again.

Hi, We have used FileMaker Server 16 for hosting our databases. Is there any way to get the information of the logged-in users in our DB in the server? We have planned to add log details for the logged-in users. However, we need to know who has connected this DB for the last 10 days. Is there any log that will show us the list of users connected to the FileMaker DB hosted in FileMaker Server 16? Please advice. Regards, Selva

The idea is to lock the parent record only for the duration of issuing a number. This would happen when user attempts to commit the record, not before. Otherwise a user taking their time over creating a record will prevent other users from doing their work on other records belonging to the same parent. This is similar to issuing an auto-entered serial number, with the option to generate the number 'On commit' selected. This is what one would use when generating invoice numbers and other series that aren't allowed to have gaps. In both scenarios, a number once generated should NEVER be reused.

SOLVED! Once again you pointed me in the right direction. The key was this: "...it would be much simpler to accomplish it from the Ads table, after performing a find for the issue number..." It took a little head-scratching and a few trial and error runs, but I finally got it all working exactly the way I want it to run. Thank you again Mr. Comment!

I was stumped by your very first sentence: If the self-join is by recordID, then surely there will ever be only one related record, which is the same record? Perhaps your relationship uses the x relational operator (in which case it's irrelevant which fields are the match fields)? Here is a simplified example using 2 global fields: ( IsEmpty ( Parent::gColor ) or Parent::gColor = Child::Color ) and ( IsEmpty ( Parent::gSize ) or Parent::gSize = Child::Size ) This filtering expression will apply filtering by color if - and only if - the global field gColor has a value. Otherwise it will pass all colors. As you can see, it takes exactly one line per filter to cover all possible combinations. I am afraid I did not understand the "default" filter thing. If you have a predicate that you want to apply always, regardless of other conditions, why don't you implement it as part of the relationship? I suggest you post a separate question regarding the script triggers. -- P.S. Next time please post code, not pictures of code.

Howdy, howdy: I have a self-join portal (by recordID) with five global fields above it that I want to use as filters, with the "default" filter being EVENT_PARENT_X_ID::EventDate__lxd ≥ Get ( CurrentDate ) so that if no values are chosen as filters, the portal will display the records with the current date, forward. The problem is how to set up the filter calculation so that if one or more filters have values chosen, the portal will filter records based on those values. The problem is--at least how I'm looking at it currently--is that I'd need to create many permutations for the filter calc to cover all conditions. Surely, there has to be a better way of coding this! Also, here's an odd thing: I attached a script trigger to each of the five filter fields so that when the value within is changed/modified, it fires the script to flush cached join and external data and refresh the window. The weird thing is, that none of the script trigger options will fire the script, but when I manually fire the script, it works as it should.

Absolutely FileMaker is faster in terms of idea to app. Many refer to FMP as a RAD (rapid application development ) And on top of it the foundation of building an FMP app the developer doesn't necessarily have to be fully disciplined every aspect such as data modeling. (it does help) We get a lot of freebies that other platforms can't take advantages of. FMP developers tend to be generalists where they can build the logic, the structure and the interface where as other platforms I imagine there is more specialization one guy that does the database and then another team that does the interface. Just look at the Claris YouTube videos that spotlight this, I don't think a solution like this could have been developed and deployed in the time frame needed on another conventional platforms.

Hi Forum! Im unsure if this belongs here please feel free to move this entry to the right place if need be. In the last 35 years, I have programmed in SQL, MySQL, Oracle and Filemaker. Native as well as Web Interfaces. I consistenty find that for writing a native interface for a local inhouse database, the development time required is *way* faster with filemaker compared to the other comparable technologies. (Yes we all agree if you want a zillion clicks a second we need Oracle, but for the rest of the 99% of us who dont need that...) I estimate the development time for a local inhouse database is about three times faster. I dont need to create multiple Screens/Masks, I dont need routines to deal with Umlauts and accents and so on, the list just goes on and on. I have explained this to a client but alas, he does not really believe me. I have seen other Filemaker developers make similar claims (rightly so in my opinion). So my question is to all of you who have programmed in SQL, MySQL or Oracle as well as Filemaker is: Do you agree that the same solution to build a local in house database, takes a lot less time to develop in Filemaker? My clients always get estimates for the same work in SQL or MySQL that are 2-3 times higher than mine. This always seems to verify the axiom. Unfortunately there dont seem to be any studies related to this. Can you tell us your experience and comment on the statement above? We would be most grateful. If someone knows a study...thanks! Spongebob

hey dudes, just for info I tried Comments approach to try to lock the record and if that succeeds proceed, if not wait a random time. Sofar this works like a charm. I did several Multi-User tests with multiple machines clicking on "create record" "at the same time" and it works really well. In 200 Tests not a single failure. I extended on this idea that I use a specially created parent table to the table I wanto change (the one with the group entries and the running number for each group). By trying to lock the parent record and the subrecord, we get the situation that all remains locked and I can make multiple changes to fields in the subrecord, before committing and freeing the whole thing up again. This way the record remains locked through all changes until I commit. (It makes the changes undivisible; the only way Filemaker supports that concept as far as I know.) This method also allows me to track when a number should be reused. If a user REVERTS the record during entry, I have a revert script. That script checks if the number of records before the revert is the same as after the revert. If not, it stores the number for reuse in the group record that also keeps the latest running number. From there it can be reused. Sofar this seems to work well. Thank you Ocean West also for your suggestion. Have a good time everyone! If something goes wrong I will keep you posted! lol ;) Luv Spongebob

When you do Go to Related Record [Show only related records ], you will create a found set of all child records related to the parent record. If you want to see only some of these records, you will need to either constrain the found set or use a relationship that makes only some records related - e.g. Advertisers::AdvertiserID = Ads 2 ::AdvertiserID AND Advertisers::gIssueNumber = Ads 2::IssueNumber I am afraid I could not fully understand what are you trying to accomplish here - but I still suspect it would be much simpler to accomplish it from the Ads table, after performing a find for the issue number. There is a reason for this: you are trying to isolate a specific record among the "many"; I say that when you are on the "many" side, there is only one "one" and you already have all you need to get data from it.

The way I built this database, an advertiser has only a single record in the AD HISTORY table for any specific Run Number of the magazine. The Run Number field contains that information. If an advertiser has more than one ad in the issue, the Ad History record for that Run Number will have multiple line items contained within the record. The "report" is actually a page of small ads that will appear in the back of a magazine. You helped me with the FINDS related to this problem back in June in THIS THREAD. That part of the script now works perfectly, thank you very much! The "ad" (report) contains basic contact information for each advertiser in the Run Number selected (Name, Address, Phone, and Web URL) along with a calculation field (MAGADPAGE) in the Advertiser Table that says, "See Our Ad on Page nn." It is that "page nn" that is giving me the problem. As the script is set now, the page of ads will properly display a page number in the MAGADPAGE field, however the page number displayed is from the FIRST (oldest) related record in the AD HISTORY table. I want it to display the data from the ONE record in the AD HISTORY table that matches the $$RunNumber variable. That is what I cannot figure out how to do. Here is the script: It is line 4, right after the first "Loop" command, where the problem is. Everything else in this script works OK.

I am a little confused about what is the actual result you are looking for. You mention a report; I suspect the easiest way to produce the report you want is to go to a layout of the AD HISTORY table, find all the ads in the issue of interest, sort them by their parent advertiser and show them in layout with a sub-summary part by advertiser showing data from the ADVERTISERS table, and the body part showing the ad details. Note that you speak of "the ONE record in the AD HISTORY table that matches the $$RunNumber variable" - but at least in theory an advertiser could have more than one ad in the same issue.

You can set the script to run with full access privileges - see: https://help.claris.com/en/pro-help/content/creating-editing-scripts.html

Is there any way I can assign a user (non-administrator) to create other users? I want to keep the admin account, but I want to allow a certain user his role to create other user accounts and not giving him the full admin right? I have a working script to create users but it seems to be ONLY working when logged in as an Administrator. I was just wondering if it's possible to just allow this privilege to a non-admin. Thanks.

I have a one-to-many relationship in play here. The ADVERTISERS table contains the ad clients (the "one"). The AD HISTORY table contains all the ads a client has run over the years (the "many"). The tables are related via an account number field. At the beginning of the script, I generate a variable containing the issue for which I want to see a report. This variable is named $$RunNumber. I can successfully find all the ADVERTISERS that have placed an ad in the chosen Run Number. I want to display one field of data from the AD HISTORY table in the report. The use the "Go to Related Record" function to get to the AD HISTORY table. Here's where the problem starts. I want to find the ONE record in the AD HISTORY table that matches the $$RunNumber variable, and display a field from that record on another layout. There is a field in the AD HISTORY table that contains the Run Number. The script finds the FIRST record for each client in the AD HISTORY table. The report layout properly displays the AD HISTORY data, but it's the wrong data. How do I find the specific record in the AD HISTORY table that matches the $$RunNumber variable?

Hey, no need to trouble yourself further with this. The solution you offered works perfect. Thank you once again!

Thanx Matt for the steps provided. If it works, then that's great!! I will definitely try it. In the meantime Im still looking for alternatives. I was hoping for a faster/easier step that doesn't rely on other external utilities/software or timings... maybe a filemaker plugin?