browse problem - please help

[melmckee72]

working in 9.1a chui environment

working with updateable browse and having trouble with 2 things.

first - if on leave trigger finds available record can not leave the browse (F1 does not send me to the next block of code)
second - if on leave trigger does not find available record I get the error message, but can not get back to the browse.

This is the last piece of this puzzle that I'm working on, if anyone has any ideas, i would appreciate the input.

Thanks in advance.

code looks like this:

def query jobfunc for bempfunc SCROLLING.
define browse func
query jobfunc NO-WAIT
display bempfunc.bfcde bempfunc.bfjob enable all
with 10 down no-box.
define fram job
func
with title "Job Functions" row 9.

mfcode = bempfunc.bfcde:handle in browse func.
mfjob = bempfunc.bfjob:handle in browse func.
ON ENTRY OF BROWSE FUNC DO:
ON TAB EDITOR-TAB.
END.
define fram mm
fcode label "Function Type"
fjob label "Job"
with row 9.
ON leave of bempfunc.bfcde in browse func DO:
FIND FIRST BFUNC WHERE BFUNC.BFCDE =MFCODE:SCREEN-VALUE NO-LOCK NO-ERROR.
IF NOT AVAILABLE BFUNC THEN DO:
DISPLAY "Invalid Function Type".
RETURN NO-APPLY.
END.
do transaction:
get current jobfunc exclusive-lock no-wait.
assign input browse func bempfunc.bfcde bempfunc.bfjob.
end.
get current jobfunc no-lock.
end.
FORM
BEMP.BFSTNME SKIP
BEMP.BLSTNME SKIP
BEMP.BDEPT SKIP
BEMP.BPHONE SKIP
BEMP.BEXT SKIP
BEMP.BACTIVE SKIP(2)
WITH FRAM H NO-BOX.

Prompt-for "Add New Employee(A) or Modify Existing Employee(M)?" mchoice
with fram z no-labels.
assign mchoice = input fram z mchoice.
If mchoice = "M" then RUN MODIFY.
Else if mchoice = "A" then do:
.....(code to create new employee record)
END.
PROCEDURE MODIFY:
...(code to update existing employee department, ext, etc)
END PROCEDURE.

open query jobfunc
for each bempfunc where BEMPFUNC.blstnme = MLAST AND BEMPFUNC.BFSTNME = MFIRST.
enable func WITH fram job.
wait-for close of current-window.

 

[MHotovec]

First, let me preface this with the fact that what follows is merely an educated guess. With equal emphasis on 'guess' as well as 'educated'.

After your error statement when you can't get back to the browse, try adding
apply 'entry' to bempfunc.bfcde in frame jobfunc .
That would be before 'return no-apply'.

As for leaving the browser. I don't see anything that satisfies the wait-for statement. I suspect the program is still waiting for the close of current-window.
Perhaps something along the lines of;
on f1 of frame jobfunc
apply 'close' to current-window.

Now on to a pet peeve of mine, (stepping onto the soap-box now). In my department I discourage the use of 'wait-for close of current-window' in favor of 'wait-for close of frame {framename}'.
Two reasons for this.
1 (that I can't explain) I've had code that misbehaved with the wait-for current-window line, but as soon as I changed it to wait-for frame it worked correctly (I dunno...).
2 From a debugging point of view, it's MUCH easier to see what's being waited-for when a frame name is specified. And it's also easier to find when the wait-for is satisfied.
(now stepping off soap-box)

Hope any or all of this helps!

Mark

working in 9.1a chui environment

working with updateable browse and having trouble with 2 things.

first - if on leave trigger finds available record can not leave the browse (F1 does not send me to the next block of code)
second - if on leave trigger does not find available record I get the error message, but can not get back to the browse.

This is the last piece of this puzzle that I'm working on, if anyone has any ideas, i would appreciate the input.

Thanks in advance.

code looks like this:

def query jobfunc for bempfunc SCROLLING.
define browse func
query jobfunc NO-WAIT
display bempfunc.bfcde bempfunc.bfjob enable all
with 10 down no-box.
define fram job
func
with title "Job Functions" row 9.

mfcode = bempfunc.bfcde:handle in browse func.
mfjob = bempfunc.bfjob:handle in browse func.
ON ENTRY OF BROWSE FUNC DO:
ON TAB EDITOR-TAB.
END.
define fram mm
fcode label "Function Type"
fjob label "Job"
with row 9.
ON leave of bempfunc.bfcde in browse func DO:
FIND FIRST BFUNC WHERE BFUNC.BFCDE =MFCODE:SCREEN-VALUE NO-LOCK NO-ERROR.
IF NOT AVAILABLE BFUNC THEN DO:
DISPLAY "Invalid Function Type".
RETURN NO-APPLY.
END.
do transaction:
get current jobfunc exclusive-lock no-wait.
assign input browse func bempfunc.bfcde bempfunc.bfjob.
end.
get current jobfunc no-lock.
end.
FORM
BEMP.BFSTNME SKIP
BEMP.BLSTNME SKIP
BEMP.BDEPT SKIP
BEMP.BPHONE SKIP
BEMP.BEXT SKIP
BEMP.BACTIVE SKIP(2)
WITH FRAM H NO-BOX.

Prompt-for "Add New Employee(A) or Modify Existing Employee(M)?" mchoice
with fram z no-labels.
assign mchoice = input fram z mchoice.
If mchoice = "M" then RUN MODIFY.
Else if mchoice = "A" then do:
.....(code to create new employee record)
END.
PROCEDURE MODIFY:
...(code to update existing employee department, ext, etc)
END PROCEDURE.

open query jobfunc
for each bempfunc where BEMPFUNC.blstnme = MLAST AND BEMPFUNC.BFSTNME = MFIRST.
enable func WITH fram job.
wait-for close of current-window.

 

[melmckee72]

Thank you so much. These were exactly the problems, one last thing I had to do was add a view frame statment before the "return no-apply" and bingo it works.

First, let me preface this with the fact that what follows is merely an educated guess. With equal emphasis on 'guess' as well as 'educated'.

After your error statement when you can't get back to the browse, try adding
apply 'entry' to bempfunc.bfcde in frame jobfunc .
That would be before 'return no-apply'.

As for leaving the browser. I don't see anything that satisfies the wait-for statement. I suspect the program is still waiting for the close of current-window.
Perhaps something along the lines of;
on f1 of frame jobfunc
apply 'close' to current-window.

Now on to a pet peeve of mine, (stepping onto the soap-box now). In my department I discourage the use of 'wait-for close of current-window' in favor of 'wait-for close of frame {framename}'.
Two reasons for this.
1 (that I can't explain) I've had code that misbehaved with the wait-for current-window line, but as soon as I changed it to wait-for frame it worked correctly (I dunno...).
2 From a debugging point of view, it's MUCH easier to see what's being waited-for when a frame name is specified. And it's also easier to find when the wait-for is satisfied.
(now stepping off soap-box)

Hope any or all of this helps!

Mark