Yes, I use stored procs for heavy lifting that I don't want to bring
across the wire. "Just give me the final (aggregated) dataset."
On 7/24/2020 9:23 AM, Stephen Russell wrote:
I'll bite. If you don't use sprocs for standard CRUD operations, why do you
need them? Do you work with a great dea
gt; environment with your data.
>
> --
>
> rk
>
> -Original Message-
> From: ProfoxTech On Behalf Of MB Software
> Solutions, LLC
> Sent: Friday, July 24, 2020 9:30 AM
> To: profoxt...@leafe.com
> Subject: Re: Unit tests (was Re: [NF] I will .)
>
&
st in your
environment with your data.
--
rk
-Original Message-
From: ProfoxTech On Behalf Of MB Software
Solutions, LLC
Sent: Friday, July 24, 2020 9:30 AM
To: profoxt...@leafe.com
Subject: Re: Unit tests (was Re: [NF] I will .)
I always thought stored procedures were faster??
On
This is an actual test in my code:
[TestMethod]
public void custItemView()
{
String item = "351000287";
String VenItem = EFData.GetCustomerPartFromView(item);
Assert.IsTrue(VenItem.Length > 0);
}
My function has a view of all items in ou
: profoxt...@leafe.com
Subject: Re: Unit tests (was Re: [NF] I will .)
On 7/23/2020 8:07 PM, Stephen Russell wrote:
Pretty much for anything that is repetitive in nature. All apps yes.
I have the instances set to not allow homemade sql except for some of
our new stuff in R, or my certificate
I'll bite. If you don't use sprocs for standard CRUD operations, why do you
need them? Do you work with a great deal of 4th normal data that the
joining of the same tables over and over is easier in a sproc? Or do you
have a process like EOM where you are doing repetitive things to data
tables?
> I still have trouble conceptualizing how to write unit tests for data
> access.
That's an integration test, really, not a unit test.
What I do is set up data sets of VFP tables or SQL Server and then zip\unzip
the one that applies to the test, as part of the test.
--
Alan Bourke
alanpb
There are performance implications to using SPs, at least in MSSQL.
--
rk
-Original Message-
From: ProfoxTech On Behalf Of MB Software
Solutions, LLC
Sent: Thursday, July 23, 2020 10:27 PM
To: profoxt...@leafe.com
Subject: Re: Unit tests (was Re: [NF] I will .)
On 7/23/2020 8:07
What I have seen from TDD is the following:
- Lazy behavior - developer types (or the 'testing group') run off to
write a lot of 'test code' instead of asking the hard questions of
stakeholders - pushing on things that are not solid, etc (TDD generally
has it's own 'syntax' like an additional
On 7/23/2020 8:07 PM, Stephen Russell wrote:
Pretty much for anything that is repetitive in nature. All apps yes. I
have the instances set to not allow homemade sql except for some of our new
stuff in R, or my certificate report that passes SQL to include all of the
Lot Numbers needed. It is o
Pretty much for anything that is repetitive in nature. All apps yes. I
have the instances set to not allow homemade sql except for some of our new
stuff in R, or my certificate report that passes SQL to include all of the
Lot Numbers needed. It is on a separate sql server and operates very
slowl
Stephen -- Are you still using stored procedures for all your CRUD?
On 7/23/2020 5:54 PM, Stephen Russell wrote:
I only see it as testing a connection to a defined data source. You
have creds that work for a positive test. You have creds that fail to
notify you that no connection was made.
I
I only see it as testing a connection to a defined data source. You
have creds that work for a positive test. You have creds that fail to
notify you that no connection was made.
If your function receives params for the actual source you are just
changing the params you pass back, or where they a
On 7/23/2020 2:12 PM, Christof Wollenhaupt wrote:
For projects that use a more object oriented approach I use foxmock to
simulate data. Most times what I really need isn't the actual data
access, because I know that these classes work, rather I have to test
the code that accesses data. foxmock
> I still have trouble conceptualizing how to write unit tests for data
> access.
I have a function that reads a table and recreates the same structure as an
empty cursor with all indexes, and such. Optionally, it appends the existing
content of the table. In my unit test I can then make changes
On Jul 23, 2020, at 12:17, Garrett Fitzgerald wrote:
>
> I still have trouble conceptualizing how to write unit tests for data
> access.
That’s good, because once you exercise anything outside of the “unit”, it’s no
longer a unit test.
What you generally do is mock the database call, returning
I still have trouble conceptualizing how to write unit tests for data
access.
On Thu, Jul 23, 2020, 11:37 MB Software Solutions, LLC <
mbsoftwaresoluti...@mbsoftwaresolutions.com> wrote:
> I have done it that way before, and it works well. I wrote the DataObj
> and BizObj first before the UI for
I have done it that way before, and it works well. I wrote the DataObj
and BizObj first before the UI for those purposes. But alas, those are
mere validation and data access classes; those are not Unit Tests, by my
definition anyway. Ever since I did this n-tier design after watching
Bob Lee
Whenever you go back into it for one reason or another, add a new test for
that area you are working in.
If all of your code is in screens or forms, you might be doing it wrong.
On Wed, Jul 22, 2020 at 10:55 AM Eric Selje wrote:
> /It doesn't make sense to retrospectively create unit test for a
/It doesn't make sense to retrospectively create unit test for an existing
application/
Useful for when you're going to make changes and you want to ensure they
don't break your existing app, especially for code that you inherited and
may not fully understand. That was the impetus behind that feat
> My UI rarely does any voodoo for making things work. Been that way for
> almost 20 years now.
Plus with Fox of course the more you keep in PRG files the easier it is to work
with source control.
--
Alan Bourke
alanpbourke (at) fastmail (dot) fm
On Wed, 22 Jul 2020, at 3:21 PM, Stephen R
DBT. Design by Test is how I work.
My UI rarely does any voodoo for making things work. Been that way for
almost 20 years now.
On Wed, Jul 22, 2020 at 7:59 AM Eric Selje wrote:
> I may be biased here, having given the Unit Test talk at SW Fox and now
> maintaining FoxUnit on GitHub, but to me
You test success and expected failure.
Your code lives in the business layer/object. In one prg you call all of
that functionality one at a time to make sure it works, as expected. If a
function receives a date as a parameter, pass in a bad date to validate it
doesn't work.
Over the life of you
On Wed, 22 Jul 2020, at 7:35 AM, Christof Wollenhaupt wrote:
> It's even quicker if you use FoxUnit and foxmock.
I couldn't do without FoxUnit, and NUnit in C# now.
--
Alan Bourke
alanpbourke (at) fastmail (dot) fm
___
Post Messages to: ProFo
> Having said that, TDD does work "best" for new development, and many VFP
> devs are not doing a ton of apps from scratch which is why it may not have
> caught on. It also works best with logic that's not buried in the UI, and
> unfortunately a lot of us buried it there (forgive me father).
We h
FoxPro let’s you put code in the UI pieces?
I’m gonna check that out. Sounds neat.
Matt Slay
> On Jul 22, 2020, at 7:58 AM, Eric Selje wrote:
>
> I may be biased here, having given the Unit Test talk at SW Fox and now
> maintaining FoxUnit on GitHub, but to me Test-Driven Development is
I may be biased here, having given the Unit Test talk at SW Fox and now
maintaining FoxUnit on GitHub, but to me Test-Driven Development is one of
those things where I "saw the light". As Stephen says, writing the tests
first forces you to think about design first, which always leads to better
prog
>
> To make an unit test in VFP is as easy as anything:
> just write your code, highlight it, activate the shortcut (right mouse
> click), select the option 'Execute the selection'. A very neat and quick
> solution.
It's even quicker if you use FoxUnit and foxmock. Both are free on VFPX. I've
h
To make an unit test in VFP is as easy as anything:
just write your code, highlight it, activate the shortcut (right mouse
click), select the option 'Execute the selection'. A very neat and quick
solution.
Stay healthy,
Koen
Op wo 22 jul. 2020 om 02:09 schreef Stephen Russell :
> Why not?
>
> De
Why not?
Design by Test. You create your functionality in TESTS and then use those
functions in a UI.
On Tue, Jul 21, 2020 at 5:53 PM MB Software Solutions, LLC <
mbsoftwaresoluti...@mbsoftwaresolutions.com> wrote:
> I would wager that less than 5% of devs (in VFP anyway) have unit tests
> in t
LMAO!!! "Tabs vs. spaces!"
On 7/21/2020 12:19 PM, Bill Anderson wrote:
As a developer once (dramatically, suspiciously looks left, looks right) I
tried using tabs as an indentation...
Please forgive me
Bill Anderson
On Tue, Jul 21, 2020 at 6:15 AM Eric Selje wrote:
Forgive me father,
I would wager that less than 5% of devs (in VFP anyway) have unit tests
in their solution setup.
On 7/21/2020 10:04 AM, Johan Nel wrote:
Ah my child, I have to confess I have never in my complete development
career had a plan or unit test done...
On 2020/07/21 15:13, Eric Selje wrote:
Forgiv
Apostate! Heathen!
On Tue, Jul 21, 2020 at 11:19 AM Bill Anderson wrote:
> As a developer once (dramatically, suspiciously looks left, looks right) I
> tried using tabs as an indentation...
>
> Please forgive me
>
> Bill Anderson
>
> On Tue, Jul 21, 2020 at 6:15 AM Eric Selje wrote:
>
> > Forg
I do lots of unit\integration tests, but test-driven development where you
write a bunch of tests first? As if.
--
Alan Bourke
alanpbourke (at) fastmail (dot) fm
___
Post Messages to: ProFox@leafe.com
Subscription Maintenance: https://mail.leafe.
As a developer once (dramatically, suspiciously looks left, looks right) I
tried using tabs as an indentation...
Please forgive me
Bill Anderson
On Tue, Jul 21, 2020 at 6:15 AM Eric Selje wrote:
> Forgive me father, for I have started coding without having a plan or even
> a unit test written
Actually, things are more like, "... there are evil-doers who have crept
in, teaching deceit and all manner of untruths..."
There have been very few advancements in 'development approaches'. Most are
repackaging and new terminology of things previously discovered and used.
The same is true of most
Ah my child, I have to confess I have never in my complete development
career had a plan or unit test done...
On 2020/07/21 15:13, Eric Selje wrote:
Forgive me father, for I have started coding without having a plan or even
a unit test written first. It's been 51 years since my last confession.
On 07/21/20 9:13 AM, Eric Selje wrote:
Forgive me father, for I have started coding without having a plan or even
a unit test written first. It's been 51 years since my last confession.
You are forgiven. Go, and sin no more.
(And we need that program yesterday)
--
This email has been checked
Forgive me father, for I have started coding without having a plan or even
a unit test written first. It's been 51 years since my last confession.
On Sat, Jul 18, 2020 at 1:57 PM Dennis Schuette wrote:
> That's such an easy change, I won't even bill you for it.
>
> 1 hour of programming, 5 hours
That's such an easy change, I won't even bill you for it.
1 hour of programming, 5 hours of debugging, and 3 customer requested
enhancements later, I'm almost done!
Customized Business Services, LLC (928) 580-6352
Dennis Schuette Primary:
I seem to remember that that was one of the deadly sins - but can't remember
why 😊
John Weller
01380 723235
07976 393631
>
> My 15 month development window got shrunk to 3 weeks, and in a panic I used
> customer Social Security Numbers as primary keys. for no real reason
> other than
> at
On Fri, Jul 17, 2020 at 9:07 AM Stephen Russell
wrote:
> Alright, I'm not a priest, but I am a programmer.
>
> Confess to me your programming sins, and I shall absolve you.
>
My 15 month development window got shrunk to 3 weeks, and in a panic I used
customer Social Security Numbers as primary k
I have created variables on the fly without declaring them at the beginning
of my procedures.
I've been attending THOR BeautifyX meetings regularly and I've been able to
avoid that particular sin for at least six months now.
But the guilt remains...
Paul H. Tarver
-Original Message-
Playing with the snake? Not good.
On Fri, Jul 17, 2020 at 2:22 PM John Weller wrote:
> I have been unfaithful to my true love - I have been dabbling with Python
> and SQLite. I wish you could get FoxPro on a Raspberry Pi.
>
> John Weller
> 01380 723235
> 079763 93631
> Sent from my iPad
>
> > O
I have been unfaithful to my true love - I have been dabbling with Python and
SQLite. I wish you could get FoxPro on a Raspberry Pi.
John Weller
01380 723235
079763 93631
Sent from my iPad
> On 17 Jul 2020, at 17:07, Stephen Russell wrote:
>
> Alright, I'm not a priest, but I am a programmer
45 matches
Mail list logo
