When leaving a field that a user validation function_name

I am getting errors saying:

The validation requirement enforced by the User Conversion Subroutine "Function_name"

The only option is OK.

Seems to happen first time the control is used.

This is new to OI 8.03 as best as I can tell.

Any workarounds?

The dialog btw is titled "Data Validation Error".

and the data is should be OK.

It's caused by having a validation of fn_name but a conversion of ].

This is new to OI 8.03.

Jim,

What kind of control is this?

dbakke@srpcs.com

SRP Computer Solutions, Inc.

Editline

That btw is not the solution, I see it happening the first time a validation of fn_name is called, makes no difference if a conversion is set or not.

If I create test form with two editlines and add TEST as the validation for the first editline (see code below).

Then run the form and enter data in the first editline and tab to the second editline I get a "Data validation error" dialog.

Why?

If I repeat the steps, go back to the first editline and tab away the error does not occur.

If I close and re-run the form I do not get the error.

Can anyone reproduce this? If I don't get a quick solution I need to move back to 8.01.

The following is a simplification of the code in my actual error checking.

Subroutine [/color]TEST(Type, Indata, Branch, Ret_Val) 
[/color]If [/color]Len(Indata) ] [/color]45 [/color]Then  
  [/color]Status()=[/color]3 
[/color]End   
[/color]Ret_Val=Indata 
[/color]Return 
[/color][/color][/size]         
               

Jim,

I tried your test and it works fine. If I enter 46 characters the cursor will not leave the prompt and no message is displayed. If under 46 then the cursor can jump to the next prompt with not message.

HTH,

John

Are you running 8.03?

I see the dialog the first time I run form.

I enter 2 char hit tab and bang the dialog pops up.

Here is a screen shot of the dialog:

http://www.waterloo-software.com/sshot.jpg

I think the following code fixes the problem. I think it's something to do with the value of Status() being 2 the first time the form is called.

What is going on here? Any ideas anyone?

Subroutine [/color]TEST(Type, Indata, Branch, Ret_Val) 
[/color] 
Status()=[/color]"" 
[/color]If [/color]Len(Indata) ] [/color]3 [/color]Then  
  [/color]Status()=[/color]3 
[/color]End   
[/color]Ret_Val=Indata 
[/color]Return[/color][/color][/size]         
               

]]I think the following code fixes the problem

Status should always be cleared prior, when the status is going to be checked. AFAIK that is an OI/Revelation stipulation

-Barry

Jim,

I used your original code in 8.03 and could not get the behavior you mention. I did not check status so it is possible it was fine in my instance. I was not running the test as part of an overall app.

John

Interesting.

I was running the code on a single test form with two editlines (not in an larger application).

If I open 8.03 and immediately run the following code via the editor I get test=2

test=Status()

Debug

Something has changed from 8.02 (at least the way 8.03 upgraded my 8.01 install).

If the solution is to always set

Status()="

at the start of all my validation subroutines that's OK.

However it makes me a tad nervous that this is a symptom of some maybe larger problem; especially given the complete lack of any open beta test for this version (or 8.02).

Thats weird… I receive 0

I tried your code and my code:

function test_status(void)

return status()

]] status should always be cleared prior, when the status is going to be checked. AFAIK that is an OI/Revelation stipulation «

If that is then case then maybe that is the problem.

However in my code Status is not being checked and the Status() function is not even being called. It is in the code but the number of characters in the string is less than the value I was checking for and hence the Status() function is never called or checked. As a result I would expect nothing to happen.

However the dialog (not mine but one built into OI) still pops up.

Therefor the bug (it seems to me is in OI) and it's probably because they are not clearing Status() prior to checking it.

It is indeed.

Something must be different.

What O/S? What network driver?

I am using Vista / All 2.1

I am using XP SP2

Lets try that again… XP SP2, All Networks 2.1

Does anyone from Revelation have any ideas what might be going on here?

Tried that , I get 0 ; 8.0.3

F5

run test_status

Do you have any "start up" ssp

Do you have background indexing on; if so turn it off and retry

What about creating a LH3.LOG file in the OI dir, open OI, close OI.

See what is at the end of the log (I get some scary FILE ERROR's in the middle of mine, so dont freak out)

Rename the LH3.log so OI is not slowed down.

]

That might work - but if any of your validation routines call others, it may not (they won't pass on the status from one to another in some circumstances).

But, you could have the startup for your app set status() one time (or just try to figure out what is failing on startup and fix it?)

If I switch to UD4 the problem goes away. Status() is zero on startup of OI.

If I use All Networks driver 2.1; I see the problem and the errors in the LH2.LOG.

Is OI8 accessing a UD file on startup that is causing these errors?

or is something wrong with my install?

Could someone from Revelation respond?

Re:

OP REV30223 10 OK

RD 10 ERROR:1003 FILE:1 GROUP:0

I am only guessing, but, file is being open with id of 10.

File id 10 is getting the errors.

find system file REV30223 in DB Manager - LH File info (LISTMEDIA REVBOOT if you have it). display the lh file info and see if it has file type 3, if it does it means it was created using LH 3 driver - if 4, the LH 4 driver(i suppose).

If so, as it will be a system file, revsoft will need to help to fix.

The table in question is MD, one that belongs to REVBOOT. The file version is 3.

I am using the All Networks Driver.

Could someone from Revelation tell me what I need to do to fix this?

Jim

This appears to be a localised problem. If you recreated any SYS files (SYSREPOSIX) while running the UD the system will not run if the AN 2.1 driver is selected. The error you are seeing in the LH2log is FS1003, a GFE in the table.

Run the ZZ_RevertLHFiles utility here to convert any UD formated files to All Network 2.1 formated files.

The Sprezzatura Group

World leaders in all things RevSoft

]] If you recreated any SYS files (SYSREPOSIX) while running the UD the system will not run if the AN 2.1 driver is selected. The error you are seeing in the LH2log is FS1003, a GFE in the table. ] Run the ZZ_RevertLHFiles utility here to convert any UD formated files to All Network 2.1 formated files. «

I will backup my install and give this a try.

Running ZZ_RevertLHFiles("c:\oi32"), seems to have fixed the problem.

Many thanks to everyone who helped, it's appreciated.

I don't think btw that I just caused these files to change to UD files. Could

(a) The install of 8.03 changed this files to UD.

or

(b) does OI 8.03 use these files differently at startup. Maybe they have been UD files for a long time and I only just saw the problem.

We changed how they're attached initially (looks like), but I'm not sure how/why this would have affected your system in this way.

Mike

In a normal install are these files UD files?

If so how can that be, if they also have to work with the All networks driver?

View this thread on the Works forum...