PDA

View Full Version : Re: Verify Question



Stephen W. Meeley
19-Oct-2005, 07:57 AM
Phil,

I would also recommend that you check out the Crystal KBase and/or
forums. The feature you are looking for feedback on is not DataFlex
specific, so there is no need to limit your sample size to this NG.


Best regards,

-SWM-


-----Original Message-----
From: Phil Kirby [mailto:phil@ksoft.com.au]
Posted At: Tuesday, October 18, 2005 7:48 PM
Posted To: crystal-reports
Conversation: Verify Question
Subject: Re: Verify Question

Man, it never ceases to amaze me the number of frustrating issues that
we
come across in software developement that one would expect dozens of
others
to have encountered, only to find you are "the first", or as near as
dammit
!

As Professor Julius Summner-Miller would have said between chocolate
bars
"Why is it so ?"

Regards
Phil



"Phil Kirby" <phil@ksoft.com.au> wrote in message
news:5j0pMr60FHA.5020@dacmail.dataaccess.com...
> Greetings from Downunder,
>
> We have always been in the habit of verifying our Crystal reports when
our
> data file structure changes. On a few disasterous ocassions, an RPT
has
> been shipped that has not been verified after a file structure change.

> This inevitably causes intermittent DFRUN crashes. This behaviour in
> itself is rather poor, but the underlying problem of data verification
can
> also be handled in Crystal by setting the report to "verify on each
> print". We are wondering if we should change all our reports to this
> method of verification, but are not prepared to take any further shots
in
> the dark due to the extremely delicate nature of the interface between
VDF
> and Crystal.
>
> I would be most interested in any comments from developers who have
used
> the "verify on each print" approach. Speed, reliability, problems,
etc
>
> Regards
> Phil Kirby
>
>

Phil Kirby
19-Oct-2005, 09:46 PM
True, Stephen,
But my issue is with verification against Dataflex files, so this NG was a
good place to start, given I would have expected hundreds of developers to
have had to face the same problem. (oh well, time for another chocolate bar
!)
Phil


"Stephen Meeley" <stephen-m@dataaccess.com> wrote in message
news:xnfnpyK1FHA.3232@dacmail.dataaccess.com...
Phil,

I would also recommend that you check out the Crystal KBase and/or
forums. The feature you are looking for feedback on is not DataFlex
specific, so there is no need to limit your sample size to this NG.


Best regards,

-SWM-


-----Original Message-----
From: Phil Kirby [mailto:phil@ksoft.com.au]
Posted At: Tuesday, October 18, 2005 7:48 PM
Posted To: crystal-reports
Conversation: Verify Question
Subject: Re: Verify Question

Man, it never ceases to amaze me the number of frustrating issues that
we
come across in software developement that one would expect dozens of
others
to have encountered, only to find you are "the first", or as near as
dammit
!

As Professor Julius Summner-Miller would have said between chocolate
bars
"Why is it so ?"

Regards
Phil



"Phil Kirby" <phil@ksoft.com.au> wrote in message
news:5j0pMr60FHA.5020@dacmail.dataaccess.com...
> Greetings from Downunder,
>
> We have always been in the habit of verifying our Crystal reports when
our
> data file structure changes. On a few disasterous ocassions, an RPT
has
> been shipped that has not been verified after a file structure change.

> This inevitably causes intermittent DFRUN crashes. This behaviour in
> itself is rather poor, but the underlying problem of data verification
can
> also be handled in Crystal by setting the report to "verify on each
> print". We are wondering if we should change all our reports to this
> method of verification, but are not prepared to take any further shots
in
> the dark due to the extremely delicate nature of the interface between
VDF
> and Crystal.
>
> I would be most interested in any comments from developers who have
used
> the "verify on each print" approach. Speed, reliability, problems,
etc
>
> Regards
> Phil Kirby
>
>

Baz
20-Oct-2005, 03:42 PM
If you've shipped an app with wobbly def's so that Crystal goes poop, mebbe
such apps have other 'features' that slipped thru your sanity testing.
Thing is, what does the client do when Crystal says it has found a verify
error?
One can hardly take a debugger on site in this current climate.

Doing every report after structure change is a sledge, but a needed chore.

Baz.
(yumm, more choc Tim Tams)

David Martinko
20-Oct-2005, 09:36 PM
BTW, Verify before every print means that it will verify before every
print... it doesn't save after verifying. So for every report you _forgot_
to verify, the user will have to hit the OK on a popup box "Customer file
changed"... [OK] "Inventory file changed"... [OK]... Database is up to date
[OK].

--
David Martinko
Redeemed Software
248-535-7495
RedeemedSoftware(SHIFT+2)Hotmail(PERIOD)com
www.redeemedsoftware.com

"Baz" <baz@nospamstuff.doodle> wrote in message
news:G3A2%23bb1FHA.4292@dacmail.dataaccess.com...
> If you've shipped an app with wobbly def's so that Crystal goes poop,
> mebbe
> such apps have other 'features' that slipped thru your sanity testing.
> Thing is, what does the client do when Crystal says it has found a verify
> error?
> One can hardly take a debugger on site in this current climate.
>
> Doing every report after structure change is a sledge, but a needed chore.
>
> Baz.
> (yumm, more choc Tim Tams)
>
>

Phil Kirby
21-Oct-2005, 02:21 AM
Hmmmm thats not the behaviour we experienced David.

We have our usual "Crystal Crash Fest" happening randomly around our client
sites, and a couple of problems of this nature were resolved by issuing them
with a modified Crystal report that verified before print. Miraculously
this stopped the crashing, but there was no pop-up confirmation activity at
all.

Kind Regards,
Phil Kirby
MediFlex P/L
phil@mediflex.net.au
www.mediflex.net.au



"David Martinko" <RedeemedSoftware@Hotmail.com> wrote in message
news:nSPwAhe1FHA.2260@dacmail.dataaccess.com...
> BTW, Verify before every print means that it will verify before every
> print... it doesn't save after verifying. So for every report you _forgot_
> to verify, the user will have to hit the OK on a popup box "Customer file
> changed"... [OK] "Inventory file changed"... [OK]... Database is up to
> date [OK].
>
> --
> David Martinko
> Redeemed Software
> 248-535-7495
> RedeemedSoftware(SHIFT+2)Hotmail(PERIOD)com
> www.redeemedsoftware.com
>
> "Baz" <baz@nospamstuff.doodle> wrote in message
> news:G3A2%23bb1FHA.4292@dacmail.dataaccess.com...
>> If you've shipped an app with wobbly def's so that Crystal goes poop,
>> mebbe
>> such apps have other 'features' that slipped thru your sanity testing.
>> Thing is, what does the client do when Crystal says it has found a verify
>> error?
>> One can hardly take a debugger on site in this current climate.
>>
>> Doing every report after structure change is a sledge, but a needed
>> chore.
>>
>> Baz.
>> (yumm, more choc Tim Tams)
>>
>>
>
>

Ken Claflin
12-Nov-2005, 10:34 PM
The "verify on every print" goes away by version 10, but the problem is
still there. If the user is running from a runtime (ie from a vdf
front-end) and not from Crystal Reports, you will have lots of problems.

This is a very real problem. David M. has the solution, a batch verify
utiltity. The last time I checked softwareforces there product did not
support the dataflex database driver.

From my experience, there are three choices:
1) Open every report and verify it or
2) contact David and get his utility or
3) eventually be sorry or embarrassed.

I have 270 reports in my app and it has been painful.

:-(

-Ken


"Phil Kirby" <phil@ksoft.com.au> wrote in message
news:4vPK38g1FHA.3636@dacmail.dataaccess.com...
> Hmmmm thats not the behaviour we experienced David.
>
> We have our usual "Crystal Crash Fest" happening randomly around our
> client sites, and a couple of problems of this nature were resolved by
> issuing them with a modified Crystal report that verified before print.
> Miraculously this stopped the crashing, but there was no pop-up
> confirmation activity at all.
>
> Kind Regards,
> Phil Kirby
> MediFlex P/L
> phil@mediflex.net.au
> www.mediflex.net.au
>
>
>
> "David Martinko" <RedeemedSoftware@Hotmail.com> wrote in message
> news:nSPwAhe1FHA.2260@dacmail.dataaccess.com...
>> BTW, Verify before every print means that it will verify before every
>> print... it doesn't save after verifying. So for every report you
>> _forgot_ to verify, the user will have to hit the OK on a popup box
>> "Customer file changed"... [OK] "Inventory file changed"... [OK]...
>> Database is up to date [OK].
>>
>> --
>> David Martinko
>> Redeemed Software
>> 248-535-7495
>> RedeemedSoftware(SHIFT+2)Hotmail(PERIOD)com
>> www.redeemedsoftware.com
>>
>> "Baz" <baz@nospamstuff.doodle> wrote in message
>> news:G3A2%23bb1FHA.4292@dacmail.dataaccess.com...
>>> If you've shipped an app with wobbly def's so that Crystal goes poop,
>>> mebbe
>>> such apps have other 'features' that slipped thru your sanity testing.
>>> Thing is, what does the client do when Crystal says it has found a
>>> verify
>>> error?
>>> One can hardly take a debugger on site in this current climate.
>>>
>>> Doing every report after structure change is a sledge, but a needed
>>> chore.
>>>
>>> Baz.
>>> (yumm, more choc Tim Tams)
>>>
>>>
>>
>>
>
>