Groups > Superbase > Superbase NG > Re: I still don't think these things are really being tested at all




I still don't think these things are really being tested at
all

I still don't think these things are really being tested at all
Thu, 14 Feb 2008 13:11:13 -050
The code generated by Desktop when save a dataform1 program gives you
this:

fc =@ f.addcontrol(dataform1text, 50, 101, 96, 19, text="DateOpened",

alignment="right", name="text5"", page=page, error=e)

SHOULD BE

fc =@ f.addcontrol(dataform1text, 50, 101, 96, 19, text="DateOpened",

alignment="right", name="text5", page=page, error=e)

NOTE

name="text5",

NOT

name="text5"",


Surely, if you tested by created a form with one of each object on a 
form and then by running the code, this would have been caught.

Post Reply
Re: I still don't think these things are really being tested at all
Thu, 14 Feb 2008 13:21:59 -050
And what really cheeses me off was that I reported this same error
in an 10/15/2007 post "Example needed"

How hard could it be to just do a simple test to see if what being 
generated actually works?

JDK


Kromkowski wrote:
> 
> The code generated by Desktop when save a dataform1 program gives you
> this:
> 
> fc =@ f.addcontrol(dataform1text, 50, 101, 96, 19,
text="DateOpened", 
> alignment="right", name="text5"", page=page,
error=e)
> 
> SHOULD BE
> 
> fc =@ f.addcontrol(dataform1text, 50, 101, 96, 19,
text="DateOpened", 
> alignment="right", name="text5", page=page, error=e)
> 
> NOTE
> 
> name="text5",
> 
> NOT
> 
> name="text5"",
> 
> 
> Surely, if you tested by created a form with one of each object on a 
> form and then by running the code, this would have been caught.
> 
Post Reply
Re: I still don't think these things are really being tested at all
Thu, 14 Feb 2008 13:28:52 -050
WOW.

Having fixed that "textnn"" code, I now get problem in this line

  dtable =@ f.addtable(table, src, error=e)

Object not found.

SOUND FAMILIAR.  Because that was also reported in the same doggone thread!

The frustration for me is that it is very hard for me to determine
whether problems result from bugs or my lack of understanding. 
Heretofore, I've presumed as a default that my lack of understanding is 
the source.  No longer, will that be the case.

JDK


Kromkowski wrote:
> And what really cheeses me off was that I reported this same error
> in an 10/15/2007 post "Example needed"
> 
> How hard could it be to just do a simple test to see if what being 
> generated actually works?
> 
> JDK
> 
> 
> Kromkowski wrote:
>>
>> The code generated by Desktop when save a dataform1 program gives you
>> this:
>>
>> fc =@ f.addcontrol(dataform1text, 50, 101, 96, 19,
text="DateOpened", 
>> alignment="right", name="text5"", page=page,
error=e)
>>
>> SHOULD BE
>>
>> fc =@ f.addcontrol(dataform1text, 50, 101, 96, 19,
text="DateOpened", 
>> alignment="right", name="text5", page=page,
error=e)
>>
>> NOTE
>>
>> name="text5",
>>
>> NOT
>>
>> name="text5"",
>>
>>
>> Surely, if you tested by created a form with one of each object on a 
>> form and then by running the code, this would have been caught.
>>
Post Reply
Re: I still don't think these things are really being tested at all
Fri, 15 Feb 2008 09:02:51 +000
Kromkowski wrote:
> WOW.
> 
> Having fixed that "textnn"" code, I now get problem in this
line
> 
>  dtable =@ f.addtable(table, src, error=e)
> 
> Object not found.
> 
> SOUND FAMILIAR.  Because that was also reported in the same doggone
thread!
> 
> The frustration for me is that it is very hard for me to determine
> whether problems result from bugs or my lack of understanding. 
> Heretofore, I've presumed as a default that my lack of understanding is 
> the source.  No longer, will that be the case.

JD,

I am sorry that this chunk of code hasn't gotten heavy scrutiny yet. 
There was a lot of work on my plate to get other things done and in. If 
you checked the release notes, you would have seen all of the reported 
bugs that *were* fixed. This includes most of yours. I simply did not 
have time to check the dataform1 *.smp generation code yet. I assure 
you, this will be fixed in the next release.

Never assume things don't work only because of your lack of 
understanding. I appreciate that what you wanted to do, and the way that 
you wanted to do it have not been the highest priority to date. Most 
people approach building a product in Superbase by first creating the 
tables interactively, and then either creating a form or by using record 
view they enter data. The next steps, which are lacking, are viewing the 
data using table view and reporting on the data using a report engine. 
Finally, they start to automate what they have using the language.

In our case, it is possible to build some great things, just by using 
the language, but that takes a little time (though it is getting 
faster). The tools are getting there, but they also take some time to 
build and get right. It is gradually becoming possible to speed up the 
development process by using the tools provided, specifically the 
creation of database tables, modification of them, and the creation of 
forms and dialogs. The final pieces of the forms puzzle are coming 
(data-aware grids and detail blocks). Following that will be a 
wizard-based ad-hoc reporting tool that can save and load reports and 
the underlying engine which can also be used from the language to load 
and run reports.

Post Reply
Re: I still don't think these things are really being tested at all
Fri, 15 Feb 2008 14:59:33 -050
> I am sorry that this chunk of code hasn't gotten heavy scrutiny yet. 
> There was a lot of work on my plate to get other things done and in. If 
> you checked the release notes, you would have seen all of the reported 
> bugs that *were* fixed. This includes most of yours. I simply did not 
> have time to check the dataform1 *.smp generation code yet. I assure 
> you, this will be fixed in the next release.

I had checked the notes.

> Never assume things don't work only because of your lack of 
> understanding. I appreciate that what you wanted to do, and the way that 
> you wanted to do it have not been the highest priority to date. Most 
> people approach building a product in Superbase by first creating the 
> tables interactively, and then either creating a form or by using record 
> view they enter data. 

But that is not really not building application.


>The next steps, which are lacking, are viewing the 
>data using table view 

This is not so much a big deal for me.

First, if you go back to the first simpol thing I wrote and posted it 
was a grid like display using an array.

Now that there is a grid, all you really need to do to have a table 
display is running through the table and stuff the information into a 
grid object.

Second, I'm only looking at stuff in a table view if I'm quickly trying 
to find some nonsense in the data.  Mostly, we look at data a record at 
a time plus associated detail blocks.

Third, if looking at data in a table view is that important export the 
data to one of the many types that Excel or Quattro Pro or Google 
Spreadsheets can open.  Or just use MySql and any of the free interfaces.

> and reporting on the data using a report engine. 

Ok, I agree the lack here is a problem.

> In our case, it is possible to build some great things, just by using 
> the language, but that takes a little time (though it is getting 
> faster). The tools are getting there, but they also take some time to 
> build and get right. It is gradually becoming possible to speed up the 
> development process by using the tools provided, specifically the 
> creation of database tables, modification of them, and the creation of 
> forms and dialogs. The final pieces of the forms puzzle are coming 
> (data-aware grids and detail blocks). Following that will be a 
> wizard-based ad-hoc reporting tool that can save and load reports and 
> the underlying engine which can also be used from the language to load 
> and run reports.

I must be approaching things a different way.  I can create and load 
forms to show data.  The issue is hooking up buttons and onchanges, etc. 
to appropriate functions.

I guess I'm not that clear how to do that if I just load the form using 
opendataform1

f =@ opendataform1(filename, error=e, errortext=errtext)

Let's say I have a button on a data form and presuming that I can 
remember the name of the function correctly (e.g. TestOnClick)that I've 
associate for example with it say.

Is this the kind of function that I add to my application

function TestOnClick(dataform1button me)
// do stuff here
I think that somehow  f=@ me.form will let me do stuff to the form but I 
am not sure because I can't seem to look at variables within this 
function or any of the variations below.
end function

OR is it simply

function TestOnClick()
// do stuff here
end function

or

function TestOnClick(wxformbutton me)
//
end function

or maybe

function TestOnClick(dataform1 me)
//
end function

My confusion is why I want to pragramtically create the form so that I 
could see what was going on and what could be (or is being) passed.

JDK








Post Reply
about | contact