| Re: Wt/Motif Interface (was: Real Number on PV-WAVE) [message #18097] |
Tue, 30 November 1999 00:00  |
rkj
Messages: 66
Registered: February 1996
|
Member |
|
|
Mark D. Williams (markw@resource-eng.com) wrote:
: "R. Kyle Justice" wrote:
: > > Then you must not be developing "elegant" user-interfaces through
: > > direct calls to the Motif toolkit via the Wt functions. This is
: > > the epitome of "obscure."
: > > But even when you have found the little-known resource that does just
: > > what you want, there is only a 50/50 chance that setting it will have
: > > any effect.
: > > It's too bad IDL users don't have the "joy" of going through this process
: > since they
: > > don't have the illusion of so many interface design choices :-)
: Just as you note, it is a tradeoff between obscurity/difficulty and flexibility.
: For the
: additional flexibility that the Wt layer affords, I don't mind putting up with
: some
: obscurity and having to deal with the O'Reilly Motif "tomes". What the Ww layer
: gives you in cross-platform portability (it is the only interface design layer
: that
: works consistently on Windows) you give up in interface flexibility.
: There are some historical reasons behind Ww being the only supported layer
: on both Motif and Windows. When VNI first developed PV-WAVE on the
: Windows platform, they ported a subset of their Wt functions from Motif to
: the Microsoft Foundation Classes. How's that for obscure? I don't envy the
: programmers that accomplished that their jobs. The subset of Wt/Motif that
: VNI chose to port to Windows was exactly the functions necessary to get
: Ww to work properly. So it is a good bet that if VNI is using
: a Wt function in the Ww layer functions, that particular Wt function is
: supported on Windows, too. At least, that Wt function, in the specific way
: that it is called from the Ww function is supported :-)
Actually, I primarily develop in a UNIX environement (I would not recommend
Wt if you want cross-platform independence. Definitely stick with the Ww
functions). This is my gripe -- many of the Motif resources appear to have
no effect when accessed through the Wt layer. For instance, why can I set
the "no resize" resources for bulletin board and form layouts just fine, but
I can't for row-column layouts? One good side effect (pun intended) is that
I have learned a lot about Motif. Just staring at the "tomes" makes one a
better programmer ;-)
: Now, I suppose this is all fine and good for someone coming to the task
: with a good familiarity with UNIX/Motif. It is admittedly problematic and
: not quite desirable for the programmer coming to PV-WAVE from either
: a Visual Basic or MSVC++ background.
I don't know if you meant to imply that I am coming from the perspective/
background of a Windows programmer (*shudder* :-P ). Anyway, this stuff
would be pretty mind-boggling for a brilliant Xlib guru (I refuse to
consult the even more ancient tomes -- volumes 1,2,4,and 5. As an engineer
I don't feel worthy :-O). But then again, a brilliant Xlib guru would not
be writing code in PV-Wave, or would they???
: Regards,
: Mark Williams
: Resource Engineering, Inc.
Kyle
PS And I have not even mentioned "secret" keywords known only by VNI tech
support and the few that gain access to this "gnosis" (i.e. /Raise for
WtSet and /Fast for SPAWN. If you know others, please e-mail me!).
|
|
|
|
| Re: Wt/Motif Interface (was: Real Number on PV-WAVE) [message #18099 is a reply to message #18097] |
Mon, 29 November 1999 00:00  |
Mark D. Williams
Messages: 15
Registered: November 1999
|
Junior Member |
|
|
"R. Kyle Justice" wrote:
>> Then you must not be developing "elegant" user-interfaces through
>> direct calls to the Motif toolkit via the Wt functions. This is
>> the epitome of "obscure."
>> But even when you have found the little-known resource that does just
>> what you want, there is only a 50/50 chance that setting it will have
>> any effect.
>> It's too bad IDL users don't have the "joy" of going through this process
> since they
>> don't have the illusion of so many interface design choices :-)
Just as you note, it is a tradeoff between obscurity/difficulty and flexibility.
For the
additional flexibility that the Wt layer affords, I don't mind putting up with
some
obscurity and having to deal with the O'Reilly Motif "tomes". What the Ww layer
gives you in cross-platform portability (it is the only interface design layer
that
works consistently on Windows) you give up in interface flexibility.
There are some historical reasons behind Ww being the only supported layer
on both Motif and Windows. When VNI first developed PV-WAVE on the
Windows platform, they ported a subset of their Wt functions from Motif to
the Microsoft Foundation Classes. How's that for obscure? I don't envy the
programmers that accomplished that their jobs. The subset of Wt/Motif that
VNI chose to port to Windows was exactly the functions necessary to get
Ww to work properly. So it is a good bet that if VNI is using
a Wt function in the Ww layer functions, that particular Wt function is
supported on Windows, too. At least, that Wt function, in the specific way
that it is called from the Ww function is supported :-)
Now, I suppose this is all fine and good for someone coming to the task
with a good familiarity with UNIX/Motif. It is admittedly problematic and
not quite desirable for the programmer coming to PV-WAVE from either
a Visual Basic or MSVC++ background.
Regards,
Mark Williams
Resource Engineering, Inc.
|
|
|
|
| Re: Wt/Motif Interface (was: Real Number on PV-WAVE) [message #18100 is a reply to message #18099] |
Mon, 29 November 1999 00:00 |
Mark D. Williams
Messages: 15
Registered: November 1999
|
Junior Member |
|
|
"R. Kyle Justice" wrote:
>> Then you must not be developing "elegant" user-interfaces through
>> direct calls to the Motif toolkit via the Wt functions. This is
>> the epitome of "obscure."
>> But even when you have found the little-known resource that does just
>> what you want, there is only a 50/50 chance that setting it will have
>> any effect.
>> It's too bad IDL users don't have the "joy" of going through this process
> since they
>> don't have the illusion of so many interface design choices :-)
Just as you note, it is a tradeoff between obscurity/difficulty and
flexibility.
For the
additional flexibility that the Wt layer affords, I don't mind putting
up with
some
obscurity and having to deal with the O'Reilly Motif "tomes". What the
Ww layer
gives you in cross-platform portability (it is the only interface design
layer
that
works consistently on Windows) you give up in interface flexibility.
There are some historical reasons behind Ww being the only supported
layer
on both Motif and Windows. When VNI first developed PV-WAVE on the
Windows platform, they ported a subset of their Wt functions from Motif
to
the Microsoft Foundation Classes. How's that for obscure? I don't envy
the
programmers that accomplished that their jobs. The subset of Wt/Motif
that
VNI chose to port to Windows was exactly the functions necessary to get
Ww to work properly. So it is a good bet that if VNI is using
a Wt function in the Ww layer functions, that particular Wt function is
supported on Windows, too. At least, that Wt function, in the specific
way
that it is called from the Ww function is supported :-)
Now, I suppose this is all fine and good for someone coming to the task
with a good familiarity with UNIX/Motif. It is admittedly problematic
and
not quite desirable for the programmer coming to PV-WAVE from either
a Visual Basic or MSVC++ background.
Regards,
Mark Williams
Resource Engineering, Inc.
|
|
|
|
|
|
| Re: Wt/Motif Interface (was: Real Number on PV-WAVE) [message #18173 is a reply to message #18099] |
Wed, 01 December 1999 00:00 |
Mark D. Williams
Messages: 15
Registered: November 1999
|
Junior Member |
|
|
"R. Kyle Justice" wrote:
> Actually, I primarily develop in a UNIX environement (I would not recommend
> Wt if you want cross-platform independence. Definitely stick with the Ww
> functions).
> : Now, I suppose this is all fine and good for someone coming to the task
> : with a good familiarity with UNIX/Motif. It is admittedly problematic and
> : not quite desirable for the programmer coming to PV-WAVE from either
> : a Visual Basic or MSVC++ background.
> I don't know if you meant to imply that I am coming from the perspective/
> background of a Windows programmer (*shudder* :-P ).
No, I didn't mean to imply this. My post more reflects my current bias
as one whose primary development background is on a UNIX platform, but
in my
current job, what development work I do is almost all on Windoze, both
in PV-WAVE and Visual Basic. PV-WAVE on Windows requires some shifts in
thinking, and a different bag of tricks to get the sometimes-quirky
Ww layer into behaving on Windows--like spinning a WwTimer off to
allow event processing to catch up, or the helpful WtProcessEvent(...,
/Drain).
> PS And I have not even mentioned "secret" keywords known only by VNI tech
> support and the few that gain access to this "gnosis" (i.e. /Raise for
> WtSet and /Fast for SPAWN. If you know others, please e-mail me!).
These are good ones to know, for sure! About the only other "secret"
WtSet keywords I have played with in the past involved using /Manage and
/UnManage along with /Child for a WwOptionMenu to explicitly set the
selected item.
> Anyway, this stuff would be pretty mind-boggling for a brilliant Xlib guru
Ah, sometimes I long for good ol' UNIX.
Regards,
Mark Williams
Resource Engineering, Inc.
|
|
|
|
| Re: Wt/Motif Interface (was: Real Number on PV-WAVE) [message #18184 is a reply to message #18099] |
Tue, 14 December 1999 00:00 |
nittler
Messages: 2
Registered: December 1999
|
Junior Member |
|
|
First of all, thanks Mark for your answer to my question about setting
the scroll position in wwtext. I have been programming in PV-WAVE for
eight years now without really being aware of the Tips database. I'm not
sure why.
>> Wt if you want cross-platform independence. Definitely stick with
the Ww
>> functions).
I always have, mainly because I have no clue about motif programming.
However, now that I have been doing widget programming in IDL there is a
lot of things annoyingly missing from the ww layer in Pv-WAVE.
Of course, I was hoping for a method of setting the WWTEXT scroll
position that would work on windows as well. I am in a weird position
of only ever having used PV-WAVE on Unix, but trying to port some of my
code to windows (for an overseas colleague), but not actually having a
Wave license on Windows. Thus, I am sending my colleague programs,
asking him to test them and then debugging by trial and error.
>
>> : Now, I suppose this is all fine and good for someone coming to the
task
>> : with a good familiarity with UNIX/Motif. It is admittedly
problematic and
>> : not quite desirable for the programmer coming to PV-WAVE from
either
>> : a Visual Basic or MSVC++ background.
How about someone who learned to program using PV-WAVE?
>> thinking, and a different bag of tricks to get the sometimes-quirky
> Ww layer into behaving on Windows--like spinning a WwTimer off to
> allow event processing to catch up, or the helpful WtProcessEvent(...,
> /Drain).
>
This sounds interesting; what the hell are you talking about?
Cheers,
Larry
------------------------------------------------------------ ------------------
Larry R. Nittler Laboratory for Extraterrestrial Physics,
Code 691
Interstellar Dust Buster NASA Goddard Spaceflight Center
Greenbelt MD 20771
phone: 301-286-4572 fax:301-286-0212
nittler@lepvax.gsfc.nasa.gov http://www.ciw.edu/lrn/
------------------------------------------------------------ ------------------
Sent via Deja.com http://www.deja.com/
Before you buy.
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
