
Davesmind
Members-
Posts
28 -
Joined
-
Last visited
Profile Information
-
Gender
Not Telling
FileMaker Experience
-
Skill Level
Intermediate
-
Application
18
Platform Environment
-
OS Platform
Windows
-
OS Version
10
Recent Profile Visitors
The recent visitors block is disabled and is not being shown to other users.
Davesmind's Achievements
-
FMWebFrame Container Issues/Comments
Davesmind replied to Davesmind's topic in FM Easy Suite Framework
I agree. I really wish Filemaker put more effort into the documentation and abilities regarding the PHP API's. I know their focus is now on webdirect, but licensing for that aspect just does not work when you need to build intranet services that do not need the features of Webdirect. I built a scanning system for Filemaker (using NAPS) and now i need a way for the end user to perform a lookup for a file that was scanned. Giving them Filemaker Pro or even Webdirect is a waste of a license. Giving them a simple find page is perfect and where PHP fills the gap. I still retain the speed of development on my Filemaker applications, but gain the robust abilities of PHP. On a side note I copied over all of the files and restructured my demo application. I made only 2 changes to the entire demo application ( single quotes around the $fm_databases section and defines the HTTP_ROOT_URL ( define ( 'HTTP_ROOT_URL', 'http://' . $_SERVER['HTTP_HOST'] . '/test' );). I did not change any other settings, but the end result is the same. I think the issue is either a setting on the database or a setting regarding the container. I do not get any errors when the pages load so i believe the database is set correctly. I am under the impression that when a page loads and the system cannot pull the file from the FM db that it will end the code and result in a 404 error. Instead the page loads without issue, but i am unable to open the file itself and the link returns the 404 page. I did see all of the temp files (.info ) and they do reference the files within the FM DB. I think a cool feature to add in a status page that a lot of PHP drop ins provide (OSTicket is a good example). This page could check the functionality and accessibility of all of the defined parts of the settings.php and determine if they work or not. I know Tim is doing this all on his own and I commend him on his progress and abilities. This is not to complain about anything, but more to help the programmer see what the end user deals with and possible fixes to make the next user experience all the better. I did use Tim's debug.php file, but that seems to check more for server settings. The server works and most of the application features work. This appears to be more of a settings issue in my case. I am continuing to debug. -
To start off this is by far an awesome addition to the PHP side of Filemaker. I have been so frustrated with Filemaker's own PHP Api and how little information there is available. This has given me almost everything I need to build a larger system with Filemaker as the main db access and the PHP to support the users who really don't need anything but some data stored on the system. I am using version 13e of FMWebFrame I have installed the demo application and after fighting with the settings here and there I have a working demo system. I think an easier approach is to build set framework locations (similar to CodeIgniter/Cake PHP and just have the end user put all of the required files into a specific location (IE Filemaker API goes here). I know you want to allow flexibility, but If you keep a standard then the end user will have to follow your direction and issues that come up will be easier to identify. In my case i ran into issues such as the nickname without the quotes, various setting issues where i thought something was supposed to be part of the file location was actually something you defaulted for your own setup. In my case once removed everything started to work. Unfortunately i am continuing to run into issues and it is with the container section of the system. If I upload a document via Container Uploading. Everything works fine until the system generates the link to the new file. That link leads to a Not Found Page. Above is the link created. The file is uploaded and found in the FMWebFrame demo database, but the link itself does not exist. I did notice that it generates a similar cache file in the temp folder, but there is less than 1kb of data. I also noticed that it adds an html extension to all file types as part of the link. The second issue is the Container Publishing. The entire page just does not function. This may be linked to the Container uploading, but it is starting to become needle in a haystack as i don't want to break what does work. As you can see below the default image does not get loaded. The two demo links also go to folders that do not exist, but even if they are directed to the correct temp folder there is no file found in that folder. The links shown on the screen are also different then the actual links they direct you to. In both cases the link on the page shows the eagle/test/../temp/item.png. The actual link is eagle/temp/item.png. Any help would be appreciated. I've also included my settings.php as an attachment below. settings_copy.php
-
UPDATE AND POSSIBLE SOLUTION/WORKAROUND It is obvious that there is something wrong with the 64 Bit version of Filemaker and pulling in data via a 64 Bit ODBC connector from a Informix database system. I've attempted different ways to pull in the data and concluded that a Integer field appears to be the issue. Again i have had no issue with any version of Filemaker that is 32 bit. I also have no issue using other applications that pull data through the same 64bit Odbc connector. I also have not noticed the same issue with other field types such as SMALLINT. SOLUTION/WORKAROUND Since i cannot wait for Filemaker to fix the issue or doubt they will fix an issue for one particular database there is a workaround. SELECT CAST ( fieldname as CHAR(10)) as fieldname FROM mydatabase If you run into this issue try using the CAST statement. I'm sure you can use other data types. In my case i had no issue importing the Integer cast as a char into a Filemaker number field. I'm sure that the main interest is to maintain the datatype as much as possible. At least now I have the data in the Filemaker side and now can work with it.
-
So, I tracked down the issue to the versions of Filemaker that are 64Bit Applications. I am now using Filemaker 14 Pro Advanced 64Bit and am getting the same issue as the FM 13 Server. It appears that Filemaker is not able to pull an integer from an Informix database. This is not an issue with the ODBC driver itself as I have a 64 Bit SQL query tool using the same ODBC driver and I can run exactly the same query. I am also able to pull any other data from Informix via the ODBC connector into Filemaker without issue. This issue also does not affect Filemaker applications running as 32 Bit as all 32 Bit ODBC connectors return the correct information. This is becoming a real drag on my development. If I can't get this to work I may have to abandon Filemaker altogether or find an alternative method to pull data from our Informix system. I am very frustrated at the moment as this prevents me from running anything on the server via FM 13 or 14. I cannot continue developing everything on the client side as this hinders me in many other ways. Has anyone ever come across an issue like this? Here is my table data on Informix: Here is the FM data after import: Here is the same data via an SQL Query tool:
-
Peter, I would take a look at the Employee_ID field's type. Are both Employee_ID fields Numbers or Text. If they are different this can result in weird behavior or no data at all.
-
There are no validations within this table. This is a simple table created just to pull this data. I matched the sql database (Informix) exactly. I even checked the fields on the Informix side just to make sure that the doc_no field type matches filemaker. It is an integer field so their should be no issues. No Errors, that is the crazy part of this issue. If i run the import on the client, there are no issues and the doc_no is populated. If I run the import on the server only that field is not populated. I even changed doc_no to a text field and again was able to import via the client and not via the server. The only direction I can think to look is that the FM Server is 64bt (FMS13) while the client is 32bit. I am using two different ODBC drivers and getting two different results. It just makes no sense that both drivers pull the same data except for that field.
-
Can you show a screen capture of your database relationships for those tables as well as the layout that is supposed to show the signatures. It is hard to come up with a solution to your problem without seeing it. The container field should work regardless of the store settings. I have multiple databases running external and secure and no issues displaying them within the tables. How are Employee to Sales connected? On the Sales layout did you drop a container field that shows that it is from the Employee table? Make sure the layout is using the correct "View Records From:" that you are trying to relate the data from. I have had instances when I used the wrong connection to for the layout basis and fields would show up blank because the data was not related the same way the correct table instance would even though both table instances were from the same table.
-
I am building a simple ODBC Import to report mechanism. In building the connection and pulling the data everything works. I ran the scheduled script on the server and for some unknown reason a specific field is not being pulled. I even ran the script via perform on server and I get the same missing field. I can't imaging how the server and the client can both pull data, but the result is different. Any one have an idea as to what is happening? Below is an example of both the client and server side scripts. The import script is exactly the same script run under both instances except, for the server side, I added a "Perform on Server" to trigger the script. As you can see the doc_no does not populate when importing on the server. Yet it still pulls data without issue as no errors are reported on the server. Client Side Server Side
-
Cannot see my DSN source in Filemaker 13 pro Advance
Davesmind replied to zindahayat's topic in FileMaker Server 13
Filemaker is a 32bit Application and thus needs a 32 bit ODBC connector. You are probably using Windows 7 64 bit. Windows 7 by default only shows the 64bit ODBC connector. In order to create a 32 Bit connector you will need to run c:WindowsSysWOW64Odbcad32.exe - Create your connector in this window and you should see the connector. -
Trouble adding ORDER BY to SQL statement
Davesmind replied to DeathRobot's topic in FileMaker Query Language or FQL
I'm not the goto SQL guru, but your Select statement pulls one specific field (People.id and id_people), Yet your Order By statement is referring to People.name. I don't know what your data set looks like when you run it without the Order By. Are you actually getting that field (people.name) to refer to? -
I know it is confusing. Our company has a culture of Excel spreadsheets. Many of my reports need to be from left to right and include data from a past year for comparison. I have managed to create a work-around for many issues. In most cases, because the data is usually static, I can even use just SQL to pull the data into the fields needed. I was really looking for a solid FM relationship, but because I am working backwards with the data and they want to see multiple criteria on the same screen, it seems to get overly complex. The sub-summary will not work since I need the data on the same line for the year to year. The second issue is that while it is easy to just pull the top 25 customers and list them, I then need to find out what the yearly total was, remove the Top 25 from that total and then provide a that number so that a plan and target can be created on the catchall alone. In my schema I am using the Customer_ID table as my viewing table. The sales_plan -> super_stats is where I pull the customer and the yearly totals. The Customer_ID -> Sales_team -> Fiscal_Year is where I pull the sales person. Normally this would be simple, but each sales person can have multiple ID's (poor DB construction of the past). I resolved this by pulling with a field that has each ID followed by a carriage return. In the end, in order to view the data on the Customer_ID layout i created a Global field to trigger the year, so I can switch between years, or even show past years. I then need to pull all of this into a single report this is where the biggest issues will arise. The report will consist of a line below: Salesperson (Sub Summary) Body (List layout) Customer: Alpha | Forecast for Current Year [2015] (pulled from sales_plan | Sales to current plan current sales - plan | current sales [2015] | Last Year sales [2014] That is a simplified version. I will need to perform that line for each customer and then add in the same for the catchall. I will also have to add a breakdown of booking, billings, GP%, GP$ and so forth comparing to plan, current and last years totals. I will probably run a bunch of SQL queries to pull the data. I was looking for a more solid FM approach, but in the end SQL may be the only way to approach this without generating a huge headache for myself. Thanks for your help. I love these kinds of problems as it forces me to research and develop my FM skills beyond the normal setups.
-
Thank you for responding. I have been working on the issue. Please refer to my attachment for the structure of what I have done so far. We have a set list of customers that each sales person is assigned. Because of this I setup relationship to filter out by the customer ID and the specific fiscal year. The fiscal year also contains a target forecast for each customer. On that end it goes data->salesplan->customer. Outside of this I also need to know the totals for the entire year. We then take that total and subtract the totals for the customer list. This will give us the customers that they do not track. The below table is exactly what they are looking for.  Salesperson Name  Customer                Last Years Sales         Current Year Sales       Forecast Target          To Target CustNo.1                    $500,000                  $100,000            $550,000              $450,000 CustNo.2                    $350,000                  $50,000             $375,000              $325,000  Then a separate line on the same screen:  Other Customers             $100,000                  $10,000             $110,000             $100,000  In this area, in order to get the total, we need to now create a relationship that tracks the 2 id's most salespersons have. I use a single field and just added each unique identifier as a line (salesperson_codes -> fk_sales_codes). It isn't optimal, but it works. The fiscal year table is where we chart the remaining customers Forecast Target.  This report has been a little frustrating as each time I think I have the solution, I find another variable that creates even more complexity. Either way when this is done it should save our sales staff hours upon hours of time. Win/Win for Filemaker! Thanks for any help!
-
I need a little help getting my relationship filtering to work. I've searched through many of the examples and issues others have had, but for some reason i did not find an answer to my problem. I've tried join tables and multipe one to many tables and while I can get some of the results i'm looking for, i know my end result is incorrect and there is a better approach to the data. To start with; I am given a table of transactions that contact a salesperson and customer (around 200,000 records). My data table will have a record looking like this: salesperson_id ; customer_id ; sales amount ; date Bob1 ; Widget100 ; 100.00 ; 1/2/2015 Bob2 ; Widget100 ; 100.00 ; 1/2/2015 Carl1; Widget200; 100.00 ; 1/2/2015 I need to break this table down according to the salesperson and their assigned customers. The issue is each salesperson can have multiple salesID's. I do not believe that the different salespersons can have the same customer, but due to the multiple salesperson id's i guess it is almost the same. Unfortunately since I am inheriting bad data, an older system generates these records, I am finding not only an issue with tying the multiple salesIDs to a single person, but the sales id can be similar as in BobE and BobEl. I did find that BobE can see BobEl data but not vice versa. I am looking to build a sales tracking viewer. In this i want to create a layout with the salesperson displayed. I want to be able to add all of the salesID's this person can have and all of the assigned customers. When I add a customer I want to then total all of the sales for that customer. SalesPerson (pk_id, salesperson) ----> salesperson_codes (fk_id, sales_id) ----> customer_codes (fk_id; customer_id) data_table ( salesperson_id; customer_id; amount) I need to be able to create a sales person, assign multiple salesperson ID's to their account, then add in the customer's assigned to them. I figure I can add both the salesperson_codes and customer_codes as portals on the salesperson layout. I want to assign the salesperson_id's in the one portal. The second portal would have the customer_codes and then summary fields that will total the sales for that customer. So far I've been able to get one side working, but not both. Any help would be appreciated.
-
Getting percentages in a sub-summary field
Davesmind replied to dgroyal's topic in Calculation Engine (Define Fields)
Thank you for updating your post as this was the same problem I was having. -
I was able to obtain a favorable result using sub-summary and ExecuteSQL. In this instance I have the data layout and then a viewing layout. The viewing layout is just a duplicate of my data layout. I then remove the body of the viewing layout and create 2 sub summary parts. I sort by the first and then the second sub summary. I then use ExecuteSQL with Calculation fields to query the data based on the criteria necessary for each collection of data. This allows me to see exactly the data I wanted and is pretty simple. I would still like to know if there is a much better alternative to this approach. I did notice that I had to refresh the ExecuteSQL fields if data change on the layout. I changed each field calculation storage to calculate as needed. If I switched between layouts the data updated. If data changed on the layout, without switching, then the sql fields did not update. A quick refresh of the layout forced them to update. Just an FYI.