Hi Idris,

You have a cascade of errors there. I have no suggestions about the "error
while opening ctl", but if the runtime's trying to open User_table.dat, it
must be finding that table name somewhere. Export the data repository and
search for that string; you may find that one of your tables has the wrong
SQL name.

Andy J

-----Original Message-----
From: magicu-l@yahoogroups.com [mailto:magicu-l@yahoogroups.com] On Behalf
Of Idris
Sent: Monday, November 09, 2009 5:39 PM
To: magicu-l@yahoogroups.com
Subject: [magicu-l] Re: Saga continues Pervasive SQL 9x Magic SQL Command.
;(

Here is an output from the mgerror.log file.  I still cannot make sense of
it.  Error while opening ctl file, give me a break, I'm modifying the
program and I have no issues with that?  There is no table as User_table.dat
in the repository. ? but such a named table does exist in the pervasive data
folder.

17:24:14.096     - >> ERROR >> Error while opening ctl.
17:24:21.346     -  >> ERROR >> Cannot modify Read Only, table:
User_Table.dat  >> Application : D:\Web_Test_Pro\Control\T1CTL.MCF, program
: Navigation_Menu
17:25:00.252     -  >> ERROR >> Failed to open, table:  >> Application :
D:\Web_Test_Pro\Control\T1CTL.MCF, program : Order List.Order Display - Sort
SQL
17:25:00.408     -  >> ERROR >> MRGSendResponse(): Failed to open,
table:<BR>[Pervasive][ODBC Client Interface][Pervasive][ODBC Engine
Interface]Syntax Error: Select



h.Order_H_PO ,

h.Order_H_Purchase_Req,

h.Order_H_Ordered_By_C,

h.Order_H_Order_Status,

h.Order_H_Order_Type,

h.Order_H_Order_SubTyp,

h.Order_H_Customer_No,

h.Order_H_Customer_Div,

h.Order_H<BR> : "order_header_list" ("Web_Test_Procurement")  (msgid 10),
program : Main Program


Idris....

I'm still digging , I know I'm in hole but no choice :) One thing I realized
that the control file needs to be closed before running the web program, is
that right or does it not make a diff?


--- In magicu-l@yahoogroups.com, Idris Bengali <idrisb59@...> wrote:
>
> Yes, I tried that, did not give me a log file.
>  
> I am now shifting over to the mreq.ini file change the log level to "C"
and what results I get
>  
> Hang in there everyone.
>  
> Thanks
>  
> Idris
>
> --- On Mon, 11/9/09, Andy Jerison <ajerison@...> wrote:
>
>
> From: Andy Jerison <ajerison@...>
> Subject: RE: [magicu-l] Re: Saga continues Pervasive SQL 9x Magic SQL
Command. ;(
> To: magicu-l@yahoogroups.com
> Date: Monday, November 9, 2009, 3:22 PM
>
>
>  
>
>
>
> >> you have to file a log file name <<
>
> ... should be "you have to *enter* a log file name"
>
> AJ
>
> -----Original Message-----
> From: magicu-l@yahoogroup s.com [mailto:magicu-l@yahoogroup s.com] On
Behalf
> Of Andy Jerison
> Sent: Monday, November 09, 2009 4:02 PM
> To: magicu-l@yahoogroup s.com
> Subject: RE: [magicu-l] Re: Saga continues Pervasive SQL 9x Magic SQL
> Command. ;(
>
> Hi Idris,
>
> In eDev 9.4, IIRR, DBMS logging settings take effect only after you close
> and re-start eDev. Also, you have to file a log file name, not just a
path.
> Given your previous setting, you might want to look for a file named
"temp"
> on the root of your C drive.
>
> If you check the "log synch" box, performance will suffer even more than
> usual with logging -- but your log file will be updated even if eDev
> crashes.
>
> Good luck,
>
> Andy J

Here is an output from the mgerror.log file.  I still cannot make sense of it. 
Error while opening ctl file, give me a break, I'm modifying the program and I
have no issues with that?  There is no table as User_table.dat in the
repository. ? but such a named table does exist in the pervasive data folder.

17:24:14.096     - >> ERROR >> Error while opening ctl.
17:24:21.346     -  >> ERROR >> Cannot modify Read Only, table: User_Table.dat 
>> Application : D:\Web_Test_Pro\Control\T1CTL.MCF, program : Navigation_Menu
17:25:00.252     -  >> ERROR >> Failed to open, table:  >> Application :
D:\Web_Test_Pro\Control\T1CTL.MCF, program : Order List.Order Display - Sort SQL
17:25:00.408     -  >> ERROR >> MRGSendResponse(): Failed to open,
table:<BR>[Pervasive][ODBC Client Interface][Pervasive][ODBC Engine
Interface]Syntax Error: Select



h.Order_H_PO ,

h.Order_H_Purchase_Req,

h.Order_H_Ordered_By_C,

h.Order_H_Order_Status,

h.Order_H_Order_Type,

h.Order_H_Order_SubTyp,

h.Order_H_Customer_No,

h.Order_H_Customer_Div,

h.Order_H<BR> : "order_header_list" ("Web_Test_Procurement")  (msgid 10),
program : Main Program


Idris....

I'm still digging , I know I'm in hole but no choice :) One thing I realized
that the control file needs to be closed before running the web program, is that
right or does it not make a diff?


--- In magicu-l@yahoogroups.com, Idris Bengali <idrisb59@...> wrote:
>
> Yes, I tried that, did not give me a log file.
>  
> I am now shifting over to the mreq.ini file change the log level to "C" and
what results I get
>  
> Hang in there everyone.
>  
> Thanks
>  
> Idris
>
> --- On Mon, 11/9/09, Andy Jerison <ajerison@...> wrote:
>
>
> From: Andy Jerison <ajerison@...>
> Subject: RE: [magicu-l] Re: Saga continues Pervasive SQL 9x Magic SQL Command.
;(
> To: magicu-l@yahoogroups.com
> Date: Monday, November 9, 2009, 3:22 PM
>
>
>  
>
>
>
> >> you have to file a log file name <<
>
> ... should be "you have to *enter* a log file name"
>
> AJ
>
> -----Original Message-----
> From: magicu-l@yahoogroup s.com [mailto:magicu-l@yahoogroup s.com] On Behalf
> Of Andy Jerison
> Sent: Monday, November 09, 2009 4:02 PM
> To: magicu-l@yahoogroup s.com
> Subject: RE: [magicu-l] Re: Saga continues Pervasive SQL 9x Magic SQL
> Command. ;(
>
> Hi Idris,
>
> In eDev 9.4, IIRR, DBMS logging settings take effect only after you close
> and re-start eDev. Also, you have to file a log file name, not just a path.
> Given your previous setting, you might want to look for a file named "temp"
> on the root of your C drive.
>
> If you check the "log synch" box, performance will suffer even more than
> usual with logging -- but your log file will be updated even if eDev
> crashes.
>
> Good luck,
>
> Andy J
>
>
>
>
>
>
>
>
>
>
>
> [Non-text portions of this message have been removed]
>

What's written in the help is about calling a component program, but in our case
it's when the component program is calling a program from the a potential host
app, and in this case it's a call by name.

I did this in v9.4 and i think that there was no global function at the point,
and anyways, the reason i had to use the bridge, was jus to fool magic with the
recursive limitation, I don't have the time now to test it with a global user
function, but it would nice to test it, and i'll do so when i have some extra
time.

Zvi


--- In magicu-l@yahoogroups.com, "Andy Jerison" <ajerison@...> wrote:
>
> Hi Zvi,
>
> Wouldn't that depend on how the component's Load Immediate property is set?
> According to the help:
>
> "Load Immediate - Determines whether or not to load the components together
> with the project load. If this is selected, then the component will load
> when the project starts. Otherwise, it will load when it is first used."
>
> But even so, I imagine a bridge program of any sort would impose non-trivial
> overhead, because initializing programs (even when they're resident) is a
> relatively slow task.
>
> You might get better performance if you replaced the bridge program with a
> global user function. The user function would receive a set of parameters
> and would invoke the selected program by name.
>
> Andy J
>
> -----Original Message-----
> From: magicu-l@yahoogroups.com [mailto:magicu-l@yahoogroups.com] On Behalf
> Of zvidev
> Sent: Monday, November 09, 2009 11:13 AM
> To: magicu-l@yahoogroups.com
> Subject: [magicu-l] Re: Call By Name Color and Font issues
>
> I'm using it for online calls, and what i see is, that the first time the
> user is calling something through call by name (same results with and
> without the tiny bridge app) it's taking like 6-8 seconds, but then any call
> thereafter is like it's a call to a local program (at least that's the
> feeling i got in online calls - but it's possible that for multiple calls in
> a batch task there will be a significant difference). I remember seeing in
> the magic helpfiles, something about the first time you call byname to
> another app, it's loading that application in memory, therefore the results
> isee makes sense, since after the first call it's loaded in memory.
>
> Zvi
>
> --- In magicu-l@yahoogroups.com, "Andy Jerison" <ajerison@> wrote:
> >
> > Hi Zvi,
> >
> > If you get a chance, it'd be really interesting to see whether the bridge
> > component technique has a significant effect on performance. A simple
> > benchmark would be to execute some large number of Invoke Program
> operations
> > on the same program, once going through the bridge, once going directly to
> > the component -- and of course, once calling a local copy of the program.
> >
> > Andy J
> >
> > -----Original Message-----
> > From: magicu-l@yahoogroups.com [mailto:magicu-l@yahoogroups.com] On Behalf
> > Of zvidev
> > Sent: Monday, November 09, 2009 10:17 AM
> > To: magicu-l@yahoogroups.com
> > Subject: [magicu-l] Re: Call By Name Color and Font issues
> >
> > I see your point.
> >
> > I happened to encounter the issue of recursive components many times from
> > non generalized calls, and I had to struggle with it when my issue was not
> > at all the source code, until I found a solution which took me a couple of
> > full days to figure out.
> > It's basically by involving a 3rd component application which I just use
> as
> > a bridge to call back anything from a host application and depending on
> the
> > circumstances I also have a generic exposed event in all host
> applications,
> > with a parameter of the program public name to call and some extra alpha
> > parms. Now this structure can be generalized with passing public names and
> > parms from a table the way you described it. Although I did not do it,
> > because it was not my purpose, but I'm sure it can be worked out, the only
> > drawback is that you need an extra component app as a bridge.
> >
> > I did not elaborate enough on the exact structure, if someone is really in
> a
> > need of recursive usage of components I can send him a demo privately, or
> > describe it step by step when I'm in my office.
> >
> > In any case we all agree that the recursive & color/font issue should be a
> > high priority for MSE to resolve natively.
> >
> > Zvi
> >
> >
> > --- In magicu-l@yahoogroups.com, "ola_ip" <omar.lamin@> wrote:
> > >
> > > Hi Zvi,
> > >
> > > What I mean is that a generic component can have generic 'call by Name'
> > operations already in place at pertinent places in its code. On these, the
> > 'Public Program Name' and 'Cabinet file Name' properties would reference
> (by
> > expressions) columns of a generic component table that would hold the
> values
> > of the program to call.
> > >
> > > Source code of the component would not be required. Just an online
> > configuration program that would enable maintaining programs public names
> > and ecf(s) locations.
> > >
> > > BTW, that concept already works. Except for the recursive limitation
> > everything is in place for it to work. A protocol also needs to be
> > established when it comes to arguments, but that is acceptable at this
> > point.
> > >
> > > The idea is that the component process could have generic hooks to
> > processes that would reside in the host project where the owner of this
> host
> > application would have the ability to control further the execution of the
> > combined solution at this point.
> > >
> > > It would open a door that is already there on a new set of generic
> > possibilities. However, currently there is that useless lock (recursion
> > limitation) on it.
> > >
> > > Does that makes more sense to you at this point. Do you see that for
> these
> > generic hooks, the source code of the component would not be required?
> > >
> > > Omar,
> > >
> > >
> > > --- In magicu-l@yahoogroups.com, "zvidev" <magicusergroup@> wrote:
> > > >
> > > > I agree with your concept that to benefit at it's max from components,
> > those obstacles exist, i just have a little comment about the paragragreph
> > below.
> > > >
> > > > "Furthermore, it is also so when it comes to the recursion restriction
> > that MSE imposes on components. Again, the work around for this
> restriction
> > implies that one has to modify the source code of the component to be able
> > to make it work with its host application defeating again the sole purpose
> > of being able to generically share the component where recursion would be
> > applicable."
> > > >
> > > > To use a component recursively - meaning to call back from the
> component
> > application a host application program etc.- you will always need the
> source
> > code and the ability to modify the source code. or maybe my understanding
> of
> > recursive is incorrect..
> >
> > > >
> > > > Zvi
> > > >
> > > >
> > > > --- In magicu-l@yahoogroups.com, "ola_ip" <omar.lamin@> wrote:
> > > > >
> > > > > Just to emphasize what Thomas is saying.
> > > > >
> > > > > The fact that a work around for a component issue implies that one
> > needs the source code to be able to modify both the component and the host
> > application means that the sole purpose of what components are for is
> broken
> > and as such the work around is not acceptable even if it is a work around.
> > > > >
> > > > > Using components to divide your application into several components
> is
> > good. But the true purpose should be that we can share components among us
> > in the community where each component is a black box that does a
> pre-defined
> > numbers of things. To take advantage of someone else's component, on
> should
> > not have to have access to this component source code and have to fix it
> for
> > it to work with his own application. Currently it is very likely to be so
> > when it comes to font and colors handling and as such is a deal breaker.
> > > > >
> > > > > Furthermore, it is also so when it comes to the recursion
> restriction
> > that MSE imposes on components. Again, the work around for this
> restriction
> > implies that one has to modify the source code of the component to be able
> > to make it work with its host application defeating again the sole purpose
> > of being able to generically share the component where recursion would be
> > applicable.
> > > > >
> > > > > When, but only when MSE will come to realize the severity of these
> > deficiencies and fix them will components be able to significantly achieve
> > their true potential by enabling more and more component base add-on.
> > Meanwhile, with the source code or in circumstances that avoid colors and
> > fonts issue (batch) as well as the recursion limitation it is still
> possible
> > to do things like for those products I plug here and there in my posts
> here.
> > But it really limits the scope of what can be done to always have to keep
> > such limitations in mind.
> > > > >
> > > > > Omar,
> > > > >
> > > > > --- In magicu-l@yahoogroups.com, "THOMAS" <ttitus_aerocontrols_com@>
> > wrote:
> > > > > >
> > > > > >
> > > > > >
> > > > > > Thanks Ariel,
> > > > > >
> > > > > > I believe, the settings in the external files tab of the project
> > properties must always override the INI settings, but this works only if
> the
> > ecf is opened as main application.
> > > > > >
> > > > > > When the component was introduced by MSE as I recall, they are
> > referred as easily "distributable libraries".
> > > > > >
> > > > > > If I understand the endeavor correctly, to use a color/font
> > sensitive distributed component, in practice, you have to get the source
> > code also. Then you incorporate colors/fonts of the component to the
> > colors/fonts of your main project, change the color/font pointers inside
> the
> > component's project accordingly, rebuild the component cabinet file and
> then
> > add the component to your main project.
> > > > > >
> > > > > > The interesting part is, when you design form screens for the
> > purpose of running as main application, tool or component, the color/fonts
> > files being in effect are files specified in the project properties. When
> > you run the application as a component (or call by name), the engine
> > completely ignores these files. There is no way to associate these files
> > back to the application at runtime.
> > > > > >
> > > > > > Simply from the point of project management and considering what
> is
> > in stake, this should have been corrected soon after the introduction of
> > components. This high maintenance work around also makes the distribution
> of
> > version updates of components with colors, fonts, key board mapping etc
> very
> > difficult.
> > > > > >
> > > > > > Thanks
> > > > > >
> > > > > > Thomas Titus
> > > > > >
> > > > > >
> > > > > > --- In magicu-l@yahoogroups.com, "Ariel Kotler" <prosto4u@> wrote:
> > > > > > >
> > > > > > > Hi ,
> > > > > > >
> > > > > > > Yes , it's true , my wish to have separated font/colors for each
> > applications opened in the "Wish List" since they released v9 with
> > components .
> > > > > > > Till now we can not share our components between us .
> > > > > > >
> > > > > > > Ariel Kotler
> > > > > > >
> > > > > > > --- In magicu-l@yahoogroups.com, "THOMAS"
> > <ttitus_aerocontrols_com@> wrote:
> > > > > > > >
> > > > > > > > Hi Folks,
> > > > > > > > I have two projects; each uses its own application color file
> > and application font file (in the application properties).
> > > > > > > > When I call Task2 by name of ecf2 from ecf1, it executes with
> > color and font file of ecf1 which displays a very tasteless screen. I
> > contacted MSE support. They advised me to insert eval
> > INIPUT(RuntimeApplicationColorDefinitionFile, color2.eng) and
> > INIPUT(RuntimeApplicationFontDefinitionFile, font2.eng) in the called
> > program. I used the evals  before the call, after the call, task prefix of
> > task2, task prefix of main2  etc and none worked. I send my sample
> projects
> > to MSE and now MSE agrees with me, saying "There is a wish opened for this
> > to have it changed however..".
> > > > > > > > I am assuming anyone who worked with cross called ecfs should
> > have experienced this before and I like to know how you handled it.
> > > > > > > > Thomas Titus
>

Yes, I tried that, did not give me a log file.
 
I am now shifting over to the mreq.ini file change the log level to "C" and what
results I get
 
Hang in there everyone.
 
Thanks
 
Idris

--- On Mon, 11/9/09, Andy Jerison <ajerison@...> wrote:


From: Andy Jerison <ajerison@...>
Subject: RE: [magicu-l] Re: Saga continues Pervasive SQL 9x Magic SQL Command.
;(
To: magicu-l@yahoogroups.com
Date: Monday, November 9, 2009, 3:22 PM


 



>> you have to file a log file name <<

... should be "you have to *enter* a log file name"

AJ

-----Original Message-----
From: magicu-l@yahoogroup s.com [mailto:magicu-l@yahoogroup s.com] On Behalf
Of Andy Jerison
Sent: Monday, November 09, 2009 4:02 PM
To: magicu-l@yahoogroup s.com
Subject: RE: [magicu-l] Re: Saga continues Pervasive SQL 9x Magic SQL
Command. ;(

Hi Idris,

In eDev 9.4, IIRR, DBMS logging settings take effect only after you close
and re-start eDev. Also, you have to file a log file name, not just a path.
Given your previous setting, you might want to look for a file named "temp"
on the root of your C drive.

If you check the "log synch" box, performance will suffer even more than
usual with logging -- but your log file will be updated even if eDev
crashes.

Good luck,

Andy J











[Non-text portions of this message have been removed]

>> you have to file a log file name <<

... should be "you have to *enter* a log file name"

AJ

-----Original Message-----
From: magicu-l@yahoogroups.com [mailto:magicu-l@yahoogroups.com] On Behalf
Of Andy Jerison
Sent: Monday, November 09, 2009 4:02 PM
To: magicu-l@yahoogroups.com
Subject: RE: [magicu-l] Re: Saga continues Pervasive SQL 9x Magic SQL
Command. ;(

Hi Idris,

In eDev 9.4, IIRR, DBMS logging settings take effect only after you close
and re-start eDev. Also, you have to file a log file name, not just a path.
Given your previous setting, you might want to look for a file named "temp"
on the root of your C drive.

If you check the "log synch" box, performance will suffer even more than
usual with logging -- but your log file will be updated even if eDev
crashes.


Good luck,

Andy J

Hi Zvi,

Wouldn't that depend on how the component's Load Immediate property is set?
According to the help:

"Load Immediate - Determines whether or not to load the components together
with the project load. If this is selected, then the component will load
when the project starts. Otherwise, it will load when it is first used."

But even so, I imagine a bridge program of any sort would impose non-trivial
overhead, because initializing programs (even when they're resident) is a
relatively slow task.

You might get better performance if you replaced the bridge program with a
global user function. The user function would receive a set of parameters
and would invoke the selected program by name.

Andy J

-----Original Message-----
From: magicu-l@yahoogroups.com [mailto:magicu-l@yahoogroups.com] On Behalf
Of zvidev
Sent: Monday, November 09, 2009 11:13 AM
To: magicu-l@yahoogroups.com
Subject: [magicu-l] Re: Call By Name Color and Font issues

I'm using it for online calls, and what i see is, that the first time the
user is calling something through call by name (same results with and
without the tiny bridge app) it's taking like 6-8 seconds, but then any call
thereafter is like it's a call to a local program (at least that's the
feeling i got in online calls - but it's possible that for multiple calls in
a batch task there will be a significant difference). I remember seeing in
the magic helpfiles, something about the first time you call byname to
another app, it's loading that application in memory, therefore the results
isee makes sense, since after the first call it's loaded in memory.

Zvi

--- In magicu-l@yahoogroups.com, "Andy Jerison" <ajerison@...> wrote:
>
> Hi Zvi,
>
> If you get a chance, it'd be really interesting to see whether the bridge
> component technique has a significant effect on performance. A simple
> benchmark would be to execute some large number of Invoke Program
operations
> on the same program, once going through the bridge, once going directly to
> the component -- and of course, once calling a local copy of the program.
>
> Andy J
>
> -----Original Message-----
> From: magicu-l@yahoogroups.com [mailto:magicu-l@yahoogroups.com] On Behalf
> Of zvidev
> Sent: Monday, November 09, 2009 10:17 AM
> To: magicu-l@yahoogroups.com
> Subject: [magicu-l] Re: Call By Name Color and Font issues
>
> I see your point.
>
> I happened to encounter the issue of recursive components many times from
> non generalized calls, and I had to struggle with it when my issue was not
> at all the source code, until I found a solution which took me a couple of
> full days to figure out.
> It's basically by involving a 3rd component application which I just use
as
> a bridge to call back anything from a host application and depending on
the
> circumstances I also have a generic exposed event in all host
applications,
> with a parameter of the program public name to call and some extra alpha
> parms. Now this structure can be generalized with passing public names and
> parms from a table the way you described it. Although I did not do it,
> because it was not my purpose, but I'm sure it can be worked out, the only
> drawback is that you need an extra component app as a bridge.
>
> I did not elaborate enough on the exact structure, if someone is really in
a
> need of recursive usage of components I can send him a demo privately, or
> describe it step by step when I'm in my office.
>
> In any case we all agree that the recursive & color/font issue should be a
> high priority for MSE to resolve natively.
>
> Zvi
>
>
> --- In magicu-l@yahoogroups.com, "ola_ip" <omar.lamin@> wrote:
> >
> > Hi Zvi,
> >
> > What I mean is that a generic component can have generic 'call by Name'
> operations already in place at pertinent places in its code. On these, the
> 'Public Program Name' and 'Cabinet file Name' properties would reference
(by
> expressions) columns of a generic component table that would hold the
values
> of the program to call.
> >
> > Source code of the component would not be required. Just an online
> configuration program that would enable maintaining programs public names
> and ecf(s) locations.
> >
> > BTW, that concept already works. Except for the recursive limitation
> everything is in place for it to work. A protocol also needs to be
> established when it comes to arguments, but that is acceptable at this
> point.
> >
> > The idea is that the component process could have generic hooks to
> processes that would reside in the host project where the owner of this
host
> application would have the ability to control further the execution of the
> combined solution at this point.
> >
> > It would open a door that is already there on a new set of generic
> possibilities. However, currently there is that useless lock (recursion
> limitation) on it.
> >
> > Does that makes more sense to you at this point. Do you see that for
these
> generic hooks, the source code of the component would not be required?
> >
> > Omar,
> >
> >
> > --- In magicu-l@yahoogroups.com, "zvidev" <magicusergroup@> wrote:
> > >
> > > I agree with your concept that to benefit at it's max from components,
> those obstacles exist, i just have a little comment about the paragragreph
> below.
> > >
> > > "Furthermore, it is also so when it comes to the recursion restriction
> that MSE imposes on components. Again, the work around for this
restriction
> implies that one has to modify the source code of the component to be able
> to make it work with its host application defeating again the sole purpose
> of being able to generically share the component where recursion would be
> applicable."
> > >
> > > To use a component recursively - meaning to call back from the
component
> application a host application program etc.- you will always need the
source
> code and the ability to modify the source code. or maybe my understanding
of
> recursive is incorrect..
>
> > >
> > > Zvi
> > >
> > >
> > > --- In magicu-l@yahoogroups.com, "ola_ip" <omar.lamin@> wrote:
> > > >
> > > > Just to emphasize what Thomas is saying.
> > > >
> > > > The fact that a work around for a component issue implies that one
> needs the source code to be able to modify both the component and the host
> application means that the sole purpose of what components are for is
broken
> and as such the work around is not acceptable even if it is a work around.
> > > >
> > > > Using components to divide your application into several components
is
> good. But the true purpose should be that we can share components among us
> in the community where each component is a black box that does a
pre-defined
> numbers of things. To take advantage of someone else's component, on
should
> not have to have access to this component source code and have to fix it
for
> it to work with his own application. Currently it is very likely to be so
> when it comes to font and colors handling and as such is a deal breaker.
> > > >
> > > > Furthermore, it is also so when it comes to the recursion
restriction
> that MSE imposes on components. Again, the work around for this
restriction
> implies that one has to modify the source code of the component to be able
> to make it work with its host application defeating again the sole purpose
> of being able to generically share the component where recursion would be
> applicable.
> > > >
> > > > When, but only when MSE will come to realize the severity of these
> deficiencies and fix them will components be able to significantly achieve
> their true potential by enabling more and more component base add-on.
> Meanwhile, with the source code or in circumstances that avoid colors and
> fonts issue (batch) as well as the recursion limitation it is still
possible
> to do things like for those products I plug here and there in my posts
here.
> But it really limits the scope of what can be done to always have to keep
> such limitations in mind.
> > > >
> > > > Omar,
> > > >
> > > > --- In magicu-l@yahoogroups.com, "THOMAS" <ttitus_aerocontrols_com@>
> wrote:
> > > > >
> > > > >
> > > > >
> > > > > Thanks Ariel,
> > > > >
> > > > > I believe, the settings in the external files tab of the project
> properties must always override the INI settings, but this works only if
the
> ecf is opened as main application.
> > > > >
> > > > > When the component was introduced by MSE as I recall, they are
> referred as easily "distributable libraries".
> > > > >
> > > > > If I understand the endeavor correctly, to use a color/font
> sensitive distributed component, in practice, you have to get the source
> code also. Then you incorporate colors/fonts of the component to the
> colors/fonts of your main project, change the color/font pointers inside
the
> component's project accordingly, rebuild the component cabinet file and
then
> add the component to your main project.
> > > > >
> > > > > The interesting part is, when you design form screens for the
> purpose of running as main application, tool or component, the color/fonts
> files being in effect are files specified in the project properties. When
> you run the application as a component (or call by name), the engine
> completely ignores these files. There is no way to associate these files
> back to the application at runtime.
> > > > >
> > > > > Simply from the point of project management and considering what
is
> in stake, this should have been corrected soon after the introduction of
> components. This high maintenance work around also makes the distribution
of
> version updates of components with colors, fonts, key board mapping etc
very
> difficult.
> > > > >
> > > > > Thanks
> > > > >
> > > > > Thomas Titus
> > > > >
> > > > >
> > > > > --- In magicu-l@yahoogroups.com, "Ariel Kotler" <prosto4u@> wrote:
> > > > > >
> > > > > > Hi ,
> > > > > >
> > > > > > Yes , it's true , my wish to have separated font/colors for each
> applications opened in the "Wish List" since they released v9 with
> components .
> > > > > > Till now we can not share our components between us .
> > > > > >
> > > > > > Ariel Kotler
> > > > > >
> > > > > > --- In magicu-l@yahoogroups.com, "THOMAS"
> <ttitus_aerocontrols_com@> wrote:
> > > > > > >
> > > > > > > Hi Folks,
> > > > > > > I have two projects; each uses its own application color file
> and application font file (in the application properties).
> > > > > > > When I call Task2 by name of ecf2 from ecf1, it executes with
> color and font file of ecf1 which displays a very tasteless screen. I
> contacted MSE support. They advised me to insert eval
> INIPUT(RuntimeApplicationColorDefinitionFile, color2.eng) and
> INIPUT(RuntimeApplicationFontDefinitionFile, font2.eng) in the called
> program. I used the evals  before the call, after the call, task prefix of
> task2, task prefix of main2  etc and none worked. I send my sample
projects
> to MSE and now MSE agrees with me, saying "There is a wish opened for this
> to have it changed however..".
> > > > > > > I am assuming anyone who worked with cross called ecfs should
> have experienced this before and I like to know how you handled it.
> > > > > > > Thomas Titus

speaking recently with a few of our component software customers who just happen
to have a passion for the Magic uniPaaS IDE, each has inquired if DBI would be
interested in having a more direct interaction with the Magic uniPaaS developer
community.  (I won't name names just yet)

This is kind of picking up where we (DBI and Magic) left off a few years ago...
DBI is interested in exploring ways to offer uniPaaS developers snap-in,
reuseable, royalty-free software components for creating Windows-standard user
interfaces, application navigation, data input, data presentation and scheduling
functionality.

Some of you might recall seeing DBI components bundled with Magic Software a few
years ago...

Here are a couple of examples of what we've done with other developer
communities - LabVIEW  http://www.labviewcomponents.com  and FoxPro
http://dbi-tech.com/VisualFoxProSedna.aspx, (the links are avilable on
http://www.dbi-tech.com too)

... we work with one or two constituents from a community to create an offering
that includes common samples for the IDE and a collection of DBI components and
then work to build on the community interaction for everyone's benefit.

If you're interested in helping build a Magic uniPaaS component-related
community site I would be happy to hear from you.

I look forward to hearing from you.

Best...
Rod

Hi Idris,

In eDev 9.4, IIRR, DBMS logging settings take effect only after you close
and re-start eDev. Also, you have to file a log file name, not just a path.
Given your previous setting, you might want to look for a file named "temp"
on the root of your C drive.

If you check the "log synch" box, performance will suffer even more than
usual with logging -- but your log file will be updated even if eDev
crashes.


Good luck,

Andy J

-----Original Message-----
From: magicu-l@yahoogroups.com [mailto:magicu-l@yahoogroups.com] On Behalf
Of Idris
Sent: Monday, November 09, 2009 3:52 PM
To: magicu-l@yahoogroups.com
Subject: [magicu-l] Re: Saga continues Pervasive SQL 9x Magic SQL Command.
;(

Andy,

Magic eDev 9.4 SP7c

Pervasive 9.5

I have access to the Pervasive control center and not able to locate (yet)
anything to do with SQL capture / logging.


Thanks

Idris


--- In magicu-l@yahoogroups.com, "Andy Jerison" <ajerison@...> wrote:
>
> Idris,
>
> What version of Magic/eDev/UP?
>
> Andy J
>
> -----Original Message-----
> From: magicu-l@yahoogroups.com [mailto:magicu-l@yahoogroups.com] On Behalf
> Of Idris
> Sent: Monday, November 09, 2009 3:32 PM
> To: magicu-l@yahoogroups.com
> Subject: [magicu-l] Re: Saga continues Pervasive SQL 9x Magic SQL Command.
> ;(
>
> Hi Andy,
>
> Thanks for the details.
>
> Here is what I did.
>
> Closed the application.
>
> Went to DBMS, positioned at Pervasive, then ^P and Log Level= Customer and
> Log name C:\temp
>
> executed the main screen from the web and when this program is called, I
got
> the same error but no log file was created in C:\temp.
>
> The program is called from the web, in this case will it still create a
log
> file ?
>
> Still hoping.
>
> Idris
>
> --- In magicu-l@yahoogroups.com, "Andy Jerison" <ajerison@> wrote:
> >
> > Hi Idris,
> >
> > The best way to troubleshoot direct SQL problems is usually to capture
the
> > SQL that the gateway is issuing. Enable gateway logging at the Customer
> > level. Remember to disable it as soon as you've gotten the information
you
> > need!
> >
> > If you log only to the activity monitor, start your application, pause
the
> > monitor after it has some contents, and erase the contents. Then,
> > immediately before triggering the problem query, resume monitoring. (A
bug
> > in the monitor makes it fail to begin logging under certain
circumstances
> if
> > it wasn't running when you start an application.) You'll find your query
> in
> > the monitor.
> >
> > Your query will appear in the log more than once. Choose the last
instance
> > of it and examine the SQL closely, comparing it to the version you
entered
> > into the query editor. If the problem doesn't jump right out at you,
paste
> > it into PSQL's query tool, execute it, and try to find the error. It's
> often
> > challenging!
> >
> > BTW, SQL Server's query editor includes a great feature that made
> > troubleshooting one of my queries trivially easy. Come to think of it I
> > think PSQL's editor may have the same feature: it displays SQL commands,
> > strings, and comments in different colors. In my case, the problem was
> that
> > one version of a dynamic direct SQL query had an extra single-quote
> > character in it. When I pasted the query into SQL Server's query editor,
> > everything after the extra quote was red; the extra character had turned
> the
> > rest of my very long query into an unusually long string.
> >
> > SQL Server 2005+ has another nice troubleshooting feature. It displays
its
> > not-very-useful error messages in a message pane at the bottom of the
> > screen. When you double-click the error, the editor positions its
> insertion
> > point on the line of code that triggered the error.
> >
> > Andy J
> >
> > -----Original Message-----
> > From: magicu-l@yahoogroups.com [mailto:magicu-l@yahoogroups.com] On
Behalf
> > Of Idris
> > Sent: Monday, November 09, 2009 2:52 PM
> > To: magicu-l@yahoogroups.com
> > Subject: [magicu-l] Saga continues Pervasive SQL 9x Magic SQL Command.
;(
> >
> > I fnally completed the SQL and tested it in a standalone program. After
> > incoporating this same query in a batch program , which is called from
the
> > web I get this error.  Gosh Syntax Error , when the same query works
fine
> >
> > Failed to open, table:
> > [Pervasive][ODBC Client Interface][Pervasive][ODBC Engine
Interface]Syntax
> > Error: Select h.Order_H_PO , h.Order_H_Purchase_Req,
> h.Order_H_Ordered_By_C,
> > h.Order_H_Order_Status, h.Order_H_Order_Type, h.Order_H_Order_SubTyp,
> > h.Order_H_Customer_No, h.Order_H_Customer_Div, h.Order_H
> >
> > I'll paste the query at the bottom.
> >
> > So in this case how does one debug the Syntax error. ?
> >
> > Thanks
> >
> > Idris
> >
> > Sorry for the longish query , but here it is.. repeat it works fine by
> > itself in a test program, run directly from the program repository.
> >
> > SELECT
> >
> > h.Order_H_PO ,
> > h.Order_H_Purchase_Req,
> > h.Order_H_Ordered_By_C,
> > h.Order_H_Order_Status,
> > h.Order_H_Order_Type,
> > h.Order_H_Order_SubTyp,
> > h.Order_H_Invoice_No
> >
> > FROm Order_Header h
> >
> > WHERE
> > ( h.Order_H_PO LIKE ':1' OR ':1' = '')
> > AND ( position(RTRIM(':2'),h.Order_H_Purchase_Req) > 0 OR ':2' = '' )
> > AND (
> > (:3 < 500 and h.Order_H_Ordered_By_C= RTRIM(':4')) OR
> > (:3 > 499 and ( (h.Order_H_Ordered_By_C >= RTRIM(':4') and
> >                   h.Order_H_Ordered_By_C <= REPLICATE ('Z',30)) OR ':4'
=
> > ''))
> >     )
> > AND ( h.Order_H_Order_Status = ':5' OR ':5' = '')
> > AND ( h.Order_H_Order_Type = ':6' OR ':6' = '')
> > AND ( h.Order_H_Order_SubTyp = ':7' OR ':7' = '')
> > AND ( (':8' = 'BI' AND ( h.Order_H_Order_Type = 'PO' OR
> h.Order_H_Order_Type
> > = 'SO' ) AND
> >       h.Order_H_Order_Status <> 'QS' AND h.Order_H_Order_Status <> 'QC'
> AND
> >       h.Order_H_Order_Status <> 'RJ' AND h.Order_H_Order_Status <> 'CA'
> AND
> >       ( ( UPPER(':9') =  'TRUE' AND h.Order_H_Order_Status <> 'PD') OR
> >         ( UPPER(':9') <> 'TRUE')
> >       )
> >       ) OR ':8' <> 'BI'
> >     )
> > AND ( (:10 = 99999 AND h.Order_H_Customer_No = RTRIM(':11')) OR :10 <>
> 99999
> > )
> > AND ( h.Order_H_Customer_Div = RTRIM(':12') OR ':12' = '' )
> > AND ( (:10 < 1000 AND h.Order_H_Customer_Dep = RTRIM(':13')) OR :10 >=
> 1000
> > )
> > AND ( (':14' = 'OR' AND h.Order_H_Date BETWEEN ':15' AND ':16' ) OR
> >        ':14' <> 'OR')
> > AND ( (':14' = 'IP' AND h.Order_H_Confirm_Date BETWEEN ':15' AND ':16' )
> OR
> >        ':14' <> 'IP')
> > AND ( (':14' = 'SH' AND h.Order_H_Shipped_Comp BETWEEN ':15' AND ':16' )
> OR
> >        ':14' <> 'SH')
> > AND ( (':14' = 'AP' AND h.Order_H_Date_Appr_Rj BETWEEN ':15' AND ':16' )
> OR
> >        ':14' <> 'AP')
> > AND ( (':14' = 'IN' AND h.Order_H_Invoice_Date BETWEEN ':15' AND ':16' )
> OR
> >        ':14' <> 'IN')
> > AND ( h.Order_H_Invoice_No LIKE ':17' OR ':17' = '')
> > AND ( (':14' = 'PA' AND h.Order_H_Confirm_Date BETWEEN ':15' AND ':16' )
> OR
> >        ':14' <> 'PA')
> > AND ( (':14' = 'RC' AND h.Order_H_Shipped_Comp BETWEEN ':15' AND ':16' )
> OR
> >        ':14' <> 'RC')
> > AND ( (':14' = 'ID' AND h.Order_H_Date_Appr_Rj BETWEEN ':15' AND ':16' )
> OR
> >        ':14' <> 'ID')
> > AND ( (':14' = 'PI' AND h.Order_H_Invoice_Date BETWEEN ':15' AND ':16' )
> OR
> >        ':14' <> 'PI')
> > AND ( ((UPPER(':18') = 'TRUE' OR UPPER(':19') = 'TRUE' ) AND
> > h.Order_H_Credit_Card_ = 1) OR
> >       (UPPER(':18') <> 'TRUE' AND UPPER(':19') <> 'TRUE') )
> > AND ( (UPPER(':18') = 'TRUE' AND h.Order_H_Credit_Car_1 = 'I')  OR
> >       (UPPER(':19') = 'TRUE' AND h.Order_H_Credit_Car_1 = 'E')  OR
> >       (UPPER(':18') <> 'TRUE' AND UPPER(':19') <> 'TRUE'))
> > AND ( (:20 > 0 AND ( h.Order_H_Vendor_ID = 0 OR h.Order_H_Vendor_ID =
:20
> ))
> > OR :20 = 0 ) ORDER BY :21

Andy,

Magic eDev 9.4 SP7c

Pervasive 9.5

I have access to the Pervasive control center and not able to locate (yet)
anything to do with SQL capture / logging.


Thanks

Idris


--- In magicu-l@yahoogroups.com, "Andy Jerison" <ajerison@...> wrote:
>
> Idris,
>
> What version of Magic/eDev/UP?
>
> Andy J
>
> -----Original Message-----
> From: magicu-l@yahoogroups.com [mailto:magicu-l@yahoogroups.com] On Behalf
> Of Idris
> Sent: Monday, November 09, 2009 3:32 PM
> To: magicu-l@yahoogroups.com
> Subject: [magicu-l] Re: Saga continues Pervasive SQL 9x Magic SQL Command.
> ;(
>
> Hi Andy,
>
> Thanks for the details.
>
> Here is what I did.
>
> Closed the application.
>
> Went to DBMS, positioned at Pervasive, then ^P and Log Level= Customer and
> Log name C:\temp
>
> executed the main screen from the web and when this program is called, I got
> the same error but no log file was created in C:\temp.
>
> The program is called from the web, in this case will it still create a log
> file ?
>
> Still hoping.
>
> Idris
>
> --- In magicu-l@yahoogroups.com, "Andy Jerison" <ajerison@> wrote:
> >
> > Hi Idris,
> >
> > The best way to troubleshoot direct SQL problems is usually to capture the
> > SQL that the gateway is issuing. Enable gateway logging at the Customer
> > level. Remember to disable it as soon as you've gotten the information you
> > need!
> >
> > If you log only to the activity monitor, start your application, pause the
> > monitor after it has some contents, and erase the contents. Then,
> > immediately before triggering the problem query, resume monitoring. (A bug
> > in the monitor makes it fail to begin logging under certain circumstances
> if
> > it wasn't running when you start an application.) You'll find your query
> in
> > the monitor.
> >
> > Your query will appear in the log more than once. Choose the last instance
> > of it and examine the SQL closely, comparing it to the version you entered
> > into the query editor. If the problem doesn't jump right out at you, paste
> > it into PSQL's query tool, execute it, and try to find the error. It's
> often
> > challenging!
> >
> > BTW, SQL Server's query editor includes a great feature that made
> > troubleshooting one of my queries trivially easy. Come to think of it I
> > think PSQL's editor may have the same feature: it displays SQL commands,
> > strings, and comments in different colors. In my case, the problem was
> that
> > one version of a dynamic direct SQL query had an extra single-quote
> > character in it. When I pasted the query into SQL Server's query editor,
> > everything after the extra quote was red; the extra character had turned
> the
> > rest of my very long query into an unusually long string.
> >
> > SQL Server 2005+ has another nice troubleshooting feature. It displays its
> > not-very-useful error messages in a message pane at the bottom of the
> > screen. When you double-click the error, the editor positions its
> insertion
> > point on the line of code that triggered the error.
> >
> > Andy J
> >
> > -----Original Message-----
> > From: magicu-l@yahoogroups.com [mailto:magicu-l@yahoogroups.com] On Behalf
> > Of Idris
> > Sent: Monday, November 09, 2009 2:52 PM
> > To: magicu-l@yahoogroups.com
> > Subject: [magicu-l] Saga continues Pervasive SQL 9x Magic SQL Command. ;(
> >
> > I fnally completed the SQL and tested it in a standalone program. After
> > incoporating this same query in a batch program , which is called from the
> > web I get this error.  Gosh Syntax Error , when the same query works fine
> >
> > Failed to open, table:
> > [Pervasive][ODBC Client Interface][Pervasive][ODBC Engine Interface]Syntax
> > Error: Select h.Order_H_PO , h.Order_H_Purchase_Req,
> h.Order_H_Ordered_By_C,
> > h.Order_H_Order_Status, h.Order_H_Order_Type, h.Order_H_Order_SubTyp,
> > h.Order_H_Customer_No, h.Order_H_Customer_Div, h.Order_H
> >
> > I'll paste the query at the bottom.
> >
> > So in this case how does one debug the Syntax error. ?
> >
> > Thanks
> >
> > Idris
> >
> > Sorry for the longish query , but here it is.. repeat it works fine by
> > itself in a test program, run directly from the program repository.
> >
> > SELECT
> >
> > h.Order_H_PO ,
> > h.Order_H_Purchase_Req,
> > h.Order_H_Ordered_By_C,
> > h.Order_H_Order_Status,
> > h.Order_H_Order_Type,
> > h.Order_H_Order_SubTyp,
> > h.Order_H_Invoice_No
> >
> > FROm Order_Header h
> >
> > WHERE
> > ( h.Order_H_PO LIKE ':1' OR ':1' = '')
> > AND ( position(RTRIM(':2'),h.Order_H_Purchase_Req) > 0 OR ':2' = '' )
> > AND (
> > (:3 < 500 and h.Order_H_Ordered_By_C= RTRIM(':4')) OR
> > (:3 > 499 and ( (h.Order_H_Ordered_By_C >= RTRIM(':4') and
> >                   h.Order_H_Ordered_By_C <= REPLICATE ('Z',30)) OR ':4' =
> > ''))
> >     )
> > AND ( h.Order_H_Order_Status = ':5' OR ':5' = '')
> > AND ( h.Order_H_Order_Type = ':6' OR ':6' = '')
> > AND ( h.Order_H_Order_SubTyp = ':7' OR ':7' = '')
> > AND ( (':8' = 'BI' AND ( h.Order_H_Order_Type = 'PO' OR
> h.Order_H_Order_Type
> > = 'SO' ) AND
> >       h.Order_H_Order_Status <> 'QS' AND h.Order_H_Order_Status <> 'QC'
> AND
> >       h.Order_H_Order_Status <> 'RJ' AND h.Order_H_Order_Status <> 'CA'
> AND
> >       ( ( UPPER(':9') =  'TRUE' AND h.Order_H_Order_Status <> 'PD') OR
> >         ( UPPER(':9') <> 'TRUE')
> >       )
> >       ) OR ':8' <> 'BI'
> >     )
> > AND ( (:10 = 99999 AND h.Order_H_Customer_No = RTRIM(':11')) OR :10 <>
> 99999
> > )
> > AND ( h.Order_H_Customer_Div = RTRIM(':12') OR ':12' = '' )
> > AND ( (:10 < 1000 AND h.Order_H_Customer_Dep = RTRIM(':13')) OR :10 >=
> 1000
> > )
> > AND ( (':14' = 'OR' AND h.Order_H_Date BETWEEN ':15' AND ':16' ) OR
> >        ':14' <> 'OR')
> > AND ( (':14' = 'IP' AND h.Order_H_Confirm_Date BETWEEN ':15' AND ':16' )
> OR
> >        ':14' <> 'IP')
> > AND ( (':14' = 'SH' AND h.Order_H_Shipped_Comp BETWEEN ':15' AND ':16' )
> OR
> >        ':14' <> 'SH')
> > AND ( (':14' = 'AP' AND h.Order_H_Date_Appr_Rj BETWEEN ':15' AND ':16' )
> OR
> >        ':14' <> 'AP')
> > AND ( (':14' = 'IN' AND h.Order_H_Invoice_Date BETWEEN ':15' AND ':16' )
> OR
> >        ':14' <> 'IN')
> > AND ( h.Order_H_Invoice_No LIKE ':17' OR ':17' = '')
> > AND ( (':14' = 'PA' AND h.Order_H_Confirm_Date BETWEEN ':15' AND ':16' )
> OR
> >        ':14' <> 'PA')
> > AND ( (':14' = 'RC' AND h.Order_H_Shipped_Comp BETWEEN ':15' AND ':16' )
> OR
> >        ':14' <> 'RC')
> > AND ( (':14' = 'ID' AND h.Order_H_Date_Appr_Rj BETWEEN ':15' AND ':16' )
> OR
> >        ':14' <> 'ID')
> > AND ( (':14' = 'PI' AND h.Order_H_Invoice_Date BETWEEN ':15' AND ':16' )
> OR
> >        ':14' <> 'PI')
> > AND ( ((UPPER(':18') = 'TRUE' OR UPPER(':19') = 'TRUE' ) AND
> > h.Order_H_Credit_Card_ = 1) OR
> >       (UPPER(':18') <> 'TRUE' AND UPPER(':19') <> 'TRUE') )
> > AND ( (UPPER(':18') = 'TRUE' AND h.Order_H_Credit_Car_1 = 'I')  OR
> >       (UPPER(':19') = 'TRUE' AND h.Order_H_Credit_Car_1 = 'E')  OR
> >       (UPPER(':18') <> 'TRUE' AND UPPER(':19') <> 'TRUE'))
> > AND ( (:20 > 0 AND ( h.Order_H_Vendor_ID = 0 OR h.Order_H_Vendor_ID = :20
> ))
> > OR :20 = 0 ) ORDER BY :21
>

Idris,

What version of Magic/eDev/UP?

Andy J

-----Original Message-----
From: magicu-l@yahoogroups.com [mailto:magicu-l@yahoogroups.com] On Behalf
Of Idris
Sent: Monday, November 09, 2009 3:32 PM
To: magicu-l@yahoogroups.com
Subject: [magicu-l] Re: Saga continues Pervasive SQL 9x Magic SQL Command.
;(

Hi Andy,

Thanks for the details.

Here is what I did.

Closed the application.

Went to DBMS, positioned at Pervasive, then ^P and Log Level= Customer and
Log name C:\temp

executed the main screen from the web and when this program is called, I got
the same error but no log file was created in C:\temp.

The program is called from the web, in this case will it still create a log
file ?

Still hoping.

Idris

--- In magicu-l@yahoogroups.com, "Andy Jerison" <ajerison@...> wrote:
>
> Hi Idris,
>
> The best way to troubleshoot direct SQL problems is usually to capture the
> SQL that the gateway is issuing. Enable gateway logging at the Customer
> level. Remember to disable it as soon as you've gotten the information you
> need!
>
> If you log only to the activity monitor, start your application, pause the
> monitor after it has some contents, and erase the contents. Then,
> immediately before triggering the problem query, resume monitoring. (A bug
> in the monitor makes it fail to begin logging under certain circumstances
if
> it wasn't running when you start an application.) You'll find your query
in
> the monitor.
>
> Your query will appear in the log more than once. Choose the last instance
> of it and examine the SQL closely, comparing it to the version you entered
> into the query editor. If the problem doesn't jump right out at you, paste
> it into PSQL's query tool, execute it, and try to find the error. It's
often
> challenging!
>
> BTW, SQL Server's query editor includes a great feature that made
> troubleshooting one of my queries trivially easy. Come to think of it I
> think PSQL's editor may have the same feature: it displays SQL commands,
> strings, and comments in different colors. In my case, the problem was
that
> one version of a dynamic direct SQL query had an extra single-quote
> character in it. When I pasted the query into SQL Server's query editor,
> everything after the extra quote was red; the extra character had turned
the
> rest of my very long query into an unusually long string.
>
> SQL Server 2005+ has another nice troubleshooting feature. It displays its
> not-very-useful error messages in a message pane at the bottom of the
> screen. When you double-click the error, the editor positions its
insertion
> point on the line of code that triggered the error.
>
> Andy J
>
> -----Original Message-----
> From: magicu-l@yahoogroups.com [mailto:magicu-l@yahoogroups.com] On Behalf
> Of Idris
> Sent: Monday, November 09, 2009 2:52 PM
> To: magicu-l@yahoogroups.com
> Subject: [magicu-l] Saga continues Pervasive SQL 9x Magic SQL Command. ;(
>
> I fnally completed the SQL and tested it in a standalone program. After
> incoporating this same query in a batch program , which is called from the
> web I get this error.  Gosh Syntax Error , when the same query works fine
>
> Failed to open, table:
> [Pervasive][ODBC Client Interface][Pervasive][ODBC Engine Interface]Syntax
> Error: Select h.Order_H_PO , h.Order_H_Purchase_Req,
h.Order_H_Ordered_By_C,
> h.Order_H_Order_Status, h.Order_H_Order_Type, h.Order_H_Order_SubTyp,
> h.Order_H_Customer_No, h.Order_H_Customer_Div, h.Order_H
>
> I'll paste the query at the bottom.
>
> So in this case how does one debug the Syntax error. ?
>
> Thanks
>
> Idris
>
> Sorry for the longish query , but here it is.. repeat it works fine by
> itself in a test program, run directly from the program repository.
>
> SELECT
>
> h.Order_H_PO ,
> h.Order_H_Purchase_Req,
> h.Order_H_Ordered_By_C,
> h.Order_H_Order_Status,
> h.Order_H_Order_Type,
> h.Order_H_Order_SubTyp,
> h.Order_H_Invoice_No
>
> FROm Order_Header h
>
> WHERE
> ( h.Order_H_PO LIKE ':1' OR ':1' = '')
> AND ( position(RTRIM(':2'),h.Order_H_Purchase_Req) > 0 OR ':2' = '' )
> AND (
> (:3 < 500 and h.Order_H_Ordered_By_C= RTRIM(':4')) OR
> (:3 > 499 and ( (h.Order_H_Ordered_By_C >= RTRIM(':4') and
>                   h.Order_H_Ordered_By_C <= REPLICATE ('Z',30)) OR ':4' =
> ''))
>     )
> AND ( h.Order_H_Order_Status = ':5' OR ':5' = '')
> AND ( h.Order_H_Order_Type = ':6' OR ':6' = '')
> AND ( h.Order_H_Order_SubTyp = ':7' OR ':7' = '')
> AND ( (':8' = 'BI' AND ( h.Order_H_Order_Type = 'PO' OR
h.Order_H_Order_Type
> = 'SO' ) AND
>       h.Order_H_Order_Status <> 'QS' AND h.Order_H_Order_Status <> 'QC'
AND
>       h.Order_H_Order_Status <> 'RJ' AND h.Order_H_Order_Status <> 'CA'
AND
>       ( ( UPPER(':9') =  'TRUE' AND h.Order_H_Order_Status <> 'PD') OR
>         ( UPPER(':9') <> 'TRUE')
>       )
>       ) OR ':8' <> 'BI'
>     )
> AND ( (:10 = 99999 AND h.Order_H_Customer_No = RTRIM(':11')) OR :10 <>
99999
> )
> AND ( h.Order_H_Customer_Div = RTRIM(':12') OR ':12' = '' )
> AND ( (:10 < 1000 AND h.Order_H_Customer_Dep = RTRIM(':13')) OR :10 >=
1000
> )
> AND ( (':14' = 'OR' AND h.Order_H_Date BETWEEN ':15' AND ':16' ) OR
>        ':14' <> 'OR')
> AND ( (':14' = 'IP' AND h.Order_H_Confirm_Date BETWEEN ':15' AND ':16' )
OR
>        ':14' <> 'IP')
> AND ( (':14' = 'SH' AND h.Order_H_Shipped_Comp BETWEEN ':15' AND ':16' )
OR
>        ':14' <> 'SH')
> AND ( (':14' = 'AP' AND h.Order_H_Date_Appr_Rj BETWEEN ':15' AND ':16' )
OR
>        ':14' <> 'AP')
> AND ( (':14' = 'IN' AND h.Order_H_Invoice_Date BETWEEN ':15' AND ':16' )
OR
>        ':14' <> 'IN')
> AND ( h.Order_H_Invoice_No LIKE ':17' OR ':17' = '')
> AND ( (':14' = 'PA' AND h.Order_H_Confirm_Date BETWEEN ':15' AND ':16' )
OR
>        ':14' <> 'PA')
> AND ( (':14' = 'RC' AND h.Order_H_Shipped_Comp BETWEEN ':15' AND ':16' )
OR
>        ':14' <> 'RC')
> AND ( (':14' = 'ID' AND h.Order_H_Date_Appr_Rj BETWEEN ':15' AND ':16' )
OR
>        ':14' <> 'ID')
> AND ( (':14' = 'PI' AND h.Order_H_Invoice_Date BETWEEN ':15' AND ':16' )
OR
>        ':14' <> 'PI')
> AND ( ((UPPER(':18') = 'TRUE' OR UPPER(':19') = 'TRUE' ) AND
> h.Order_H_Credit_Card_ = 1) OR
>       (UPPER(':18') <> 'TRUE' AND UPPER(':19') <> 'TRUE') )
> AND ( (UPPER(':18') = 'TRUE' AND h.Order_H_Credit_Car_1 = 'I')  OR
>       (UPPER(':19') = 'TRUE' AND h.Order_H_Credit_Car_1 = 'E')  OR
>       (UPPER(':18') <> 'TRUE' AND UPPER(':19') <> 'TRUE'))
> AND ( (:20 > 0 AND ( h.Order_H_Vendor_ID = 0 OR h.Order_H_Vendor_ID = :20
))
> OR :20 = 0 ) ORDER BY :21

Hi Andy,

Thanks for the details.

Here is what I did.

Closed the application.

Went to DBMS, positioned at Pervasive, then ^P and Log Level= Customer and Log
name C:\temp

executed the main screen from the web and when this program is called, I got the
same error but no log file was created in C:\temp.

The program is called from the web, in this case will it still create a log file
?

Still hoping.

Idris

--- In magicu-l@yahoogroups.com, "Andy Jerison" <ajerison@...> wrote:
>
> Hi Idris,
>
> The best way to troubleshoot direct SQL problems is usually to capture the
> SQL that the gateway is issuing. Enable gateway logging at the Customer
> level. Remember to disable it as soon as you've gotten the information you
> need!
>
> If you log only to the activity monitor, start your application, pause the
> monitor after it has some contents, and erase the contents. Then,
> immediately before triggering the problem query, resume monitoring. (A bug
> in the monitor makes it fail to begin logging under certain circumstances if
> it wasn't running when you start an application.) You'll find your query in
> the monitor.
>
> Your query will appear in the log more than once. Choose the last instance
> of it and examine the SQL closely, comparing it to the version you entered
> into the query editor. If the problem doesn't jump right out at you, paste
> it into PSQL's query tool, execute it, and try to find the error. It's often
> challenging!
>
> BTW, SQL Server's query editor includes a great feature that made
> troubleshooting one of my queries trivially easy. Come to think of it I
> think PSQL's editor may have the same feature: it displays SQL commands,
> strings, and comments in different colors. In my case, the problem was that
> one version of a dynamic direct SQL query had an extra single-quote
> character in it. When I pasted the query into SQL Server's query editor,
> everything after the extra quote was red; the extra character had turned the
> rest of my very long query into an unusually long string.
>
> SQL Server 2005+ has another nice troubleshooting feature. It displays its
> not-very-useful error messages in a message pane at the bottom of the
> screen. When you double-click the error, the editor positions its insertion
> point on the line of code that triggered the error.
>
> Andy J
>
> -----Original Message-----
> From: magicu-l@yahoogroups.com [mailto:magicu-l@yahoogroups.com] On Behalf
> Of Idris
> Sent: Monday, November 09, 2009 2:52 PM
> To: magicu-l@yahoogroups.com
> Subject: [magicu-l] Saga continues Pervasive SQL 9x Magic SQL Command. ;(
>
> I fnally completed the SQL and tested it in a standalone program. After
> incoporating this same query in a batch program , which is called from the
> web I get this error.  Gosh Syntax Error , when the same query works fine
>
> Failed to open, table:
> [Pervasive][ODBC Client Interface][Pervasive][ODBC Engine Interface]Syntax
> Error: Select h.Order_H_PO , h.Order_H_Purchase_Req, h.Order_H_Ordered_By_C,
> h.Order_H_Order_Status, h.Order_H_Order_Type, h.Order_H_Order_SubTyp,
> h.Order_H_Customer_No, h.Order_H_Customer_Div, h.Order_H
>
> I'll paste the query at the bottom.
>
> So in this case how does one debug the Syntax error. ?
>
> Thanks
>
> Idris
>
> Sorry for the longish query , but here it is.. repeat it works fine by
> itself in a test program, run directly from the program repository.
>
> SELECT
>
> h.Order_H_PO ,
> h.Order_H_Purchase_Req,
> h.Order_H_Ordered_By_C,
> h.Order_H_Order_Status,
> h.Order_H_Order_Type,
> h.Order_H_Order_SubTyp,
> h.Order_H_Invoice_No
>
> FROm Order_Header h
>
> WHERE
> ( h.Order_H_PO LIKE ':1' OR ':1' = '')
> AND ( position(RTRIM(':2'),h.Order_H_Purchase_Req) > 0 OR ':2' = '' )
> AND (
> (:3 < 500 and h.Order_H_Ordered_By_C= RTRIM(':4')) OR
> (:3 > 499 and ( (h.Order_H_Ordered_By_C >= RTRIM(':4') and
>                   h.Order_H_Ordered_By_C <= REPLICATE ('Z',30)) OR ':4' =
> ''))
>     )
> AND ( h.Order_H_Order_Status = ':5' OR ':5' = '')
> AND ( h.Order_H_Order_Type = ':6' OR ':6' = '')
> AND ( h.Order_H_Order_SubTyp = ':7' OR ':7' = '')
> AND ( (':8' = 'BI' AND ( h.Order_H_Order_Type = 'PO' OR h.Order_H_Order_Type
> = 'SO' ) AND
>       h.Order_H_Order_Status <> 'QS' AND h.Order_H_Order_Status <> 'QC' AND
>       h.Order_H_Order_Status <> 'RJ' AND h.Order_H_Order_Status <> 'CA' AND
>       ( ( UPPER(':9') =  'TRUE' AND h.Order_H_Order_Status <> 'PD') OR
>         ( UPPER(':9') <> 'TRUE')
>       )
>       ) OR ':8' <> 'BI'
>     )
> AND ( (:10 = 99999 AND h.Order_H_Customer_No = RTRIM(':11')) OR :10 <> 99999
> )
> AND ( h.Order_H_Customer_Div = RTRIM(':12') OR ':12' = '' )
> AND ( (:10 < 1000 AND h.Order_H_Customer_Dep = RTRIM(':13')) OR :10 >= 1000
> )
> AND ( (':14' = 'OR' AND h.Order_H_Date BETWEEN ':15' AND ':16' ) OR
>        ':14' <> 'OR')
> AND ( (':14' = 'IP' AND h.Order_H_Confirm_Date BETWEEN ':15' AND ':16' ) OR
>        ':14' <> 'IP')
> AND ( (':14' = 'SH' AND h.Order_H_Shipped_Comp BETWEEN ':15' AND ':16' ) OR
>        ':14' <> 'SH')
> AND ( (':14' = 'AP' AND h.Order_H_Date_Appr_Rj BETWEEN ':15' AND ':16' ) OR
>        ':14' <> 'AP')
> AND ( (':14' = 'IN' AND h.Order_H_Invoice_Date BETWEEN ':15' AND ':16' ) OR
>        ':14' <> 'IN')
> AND ( h.Order_H_Invoice_No LIKE ':17' OR ':17' = '')
> AND ( (':14' = 'PA' AND h.Order_H_Confirm_Date BETWEEN ':15' AND ':16' ) OR
>        ':14' <> 'PA')
> AND ( (':14' = 'RC' AND h.Order_H_Shipped_Comp BETWEEN ':15' AND ':16' ) OR
>        ':14' <> 'RC')
> AND ( (':14' = 'ID' AND h.Order_H_Date_Appr_Rj BETWEEN ':15' AND ':16' ) OR
>        ':14' <> 'ID')
> AND ( (':14' = 'PI' AND h.Order_H_Invoice_Date BETWEEN ':15' AND ':16' ) OR
>        ':14' <> 'PI')
> AND ( ((UPPER(':18') = 'TRUE' OR UPPER(':19') = 'TRUE' ) AND
> h.Order_H_Credit_Card_ = 1) OR
>       (UPPER(':18') <> 'TRUE' AND UPPER(':19') <> 'TRUE') )
> AND ( (UPPER(':18') = 'TRUE' AND h.Order_H_Credit_Car_1 = 'I')  OR
>       (UPPER(':19') = 'TRUE' AND h.Order_H_Credit_Car_1 = 'E')  OR
>       (UPPER(':18') <> 'TRUE' AND UPPER(':19') <> 'TRUE'))
> AND ( (:20 > 0 AND ( h.Order_H_Vendor_ID = 0 OR h.Order_H_Vendor_ID = :20 ))
> OR :20 = 0 ) ORDER BY :21
>

In the MGA the messages of October 2009.

You can find it at http://www.ng-sw.de/mg-wikka/MagicListMessages

I have added dates to see easily which are new and need a download.

Check out also the new Video-section on the wiki.

Regards,
Frank Van Herreweghe





[Non-text portions of this message have been removed]

Hi Idris,

The best way to troubleshoot direct SQL problems is usually to capture the
SQL that the gateway is issuing. Enable gateway logging at the Customer
level. Remember to disable it as soon as you've gotten the information you
need!

If you log only to the activity monitor, start your application, pause the
monitor after it has some contents, and erase the contents. Then,
immediately before triggering the problem query, resume monitoring. (A bug
in the monitor makes it fail to begin logging under certain circumstances if
it wasn't running when you start an application.) You'll find your query in
the monitor.

Your query will appear in the log more than once. Choose the last instance
of it and examine the SQL closely, comparing it to the version you entered
into the query editor. If the problem doesn't jump right out at you, paste
it into PSQL's query tool, execute it, and try to find the error. It's often
challenging!

BTW, SQL Server's query editor includes a great feature that made
troubleshooting one of my queries trivially easy. Come to think of it I
think PSQL's editor may have the same feature: it displays SQL commands,
strings, and comments in different colors. In my case, the problem was that
one version of a dynamic direct SQL query had an extra single-quote
character in it. When I pasted the query into SQL Server's query editor,
everything after the extra quote was red; the extra character had turned the
rest of my very long query into an unusually long string.

SQL Server 2005+ has another nice troubleshooting feature. It displays its
not-very-useful error messages in a message pane at the bottom of the
screen. When you double-click the error, the editor positions its insertion
point on the line of code that triggered the error.

Andy J

-----Original Message-----
From: magicu-l@yahoogroups.com [mailto:magicu-l@yahoogroups.com] On Behalf
Of Idris
Sent: Monday, November 09, 2009 2:52 PM
To: magicu-l@yahoogroups.com
Subject: [magicu-l] Saga continues Pervasive SQL 9x Magic SQL Command. ;(

I fnally completed the SQL and tested it in a standalone program. After
incoporating this same query in a batch program , which is called from the
web I get this error.  Gosh Syntax Error , when the same query works fine

Failed to open, table:
[Pervasive][ODBC Client Interface][Pervasive][ODBC Engine Interface]Syntax
Error: Select h.Order_H_PO , h.Order_H_Purchase_Req, h.Order_H_Ordered_By_C,
h.Order_H_Order_Status, h.Order_H_Order_Type, h.Order_H_Order_SubTyp,
h.Order_H_Customer_No, h.Order_H_Customer_Div, h.Order_H

I'll paste the query at the bottom.

So in this case how does one debug the Syntax error. ?

Thanks

Idris

Sorry for the longish query , but here it is.. repeat it works fine by
itself in a test program, run directly from the program repository.

SELECT

h.Order_H_PO ,
h.Order_H_Purchase_Req,
h.Order_H_Ordered_By_C,
h.Order_H_Order_Status,
h.Order_H_Order_Type,
h.Order_H_Order_SubTyp,
h.Order_H_Invoice_No

FROm Order_Header h

WHERE
( h.Order_H_PO LIKE ':1' OR ':1' = '')
AND ( position(RTRIM(':2'),h.Order_H_Purchase_Req) > 0 OR ':2' = '' )
AND (
(:3 < 500 and h.Order_H_Ordered_By_C= RTRIM(':4')) OR
(:3 > 499 and ( (h.Order_H_Ordered_By_C >= RTRIM(':4') and
                   h.Order_H_Ordered_By_C <= REPLICATE ('Z',30)) OR ':4' =
''))
     )
AND ( h.Order_H_Order_Status = ':5' OR ':5' = '')
AND ( h.Order_H_Order_Type = ':6' OR ':6' = '')
AND ( h.Order_H_Order_SubTyp = ':7' OR ':7' = '')
AND ( (':8' = 'BI' AND ( h.Order_H_Order_Type = 'PO' OR h.Order_H_Order_Type
= 'SO' ) AND
       h.Order_H_Order_Status <> 'QS' AND h.Order_H_Order_Status <> 'QC' AND
       h.Order_H_Order_Status <> 'RJ' AND h.Order_H_Order_Status <> 'CA' AND
       ( ( UPPER(':9') =  'TRUE' AND h.Order_H_Order_Status <> 'PD') OR
         ( UPPER(':9') <> 'TRUE')
       )
       ) OR ':8' <> 'BI'
     )
AND ( (:10 = 99999 AND h.Order_H_Customer_No = RTRIM(':11')) OR :10 <> 99999
)
AND ( h.Order_H_Customer_Div = RTRIM(':12') OR ':12' = '' )
AND ( (:10 < 1000 AND h.Order_H_Customer_Dep = RTRIM(':13')) OR :10 >= 1000
)
AND ( (':14' = 'OR' AND h.Order_H_Date BETWEEN ':15' AND ':16' ) OR
        ':14' <> 'OR')
AND ( (':14' = 'IP' AND h.Order_H_Confirm_Date BETWEEN ':15' AND ':16' ) OR
        ':14' <> 'IP')
AND ( (':14' = 'SH' AND h.Order_H_Shipped_Comp BETWEEN ':15' AND ':16' ) OR
        ':14' <> 'SH')
AND ( (':14' = 'AP' AND h.Order_H_Date_Appr_Rj BETWEEN ':15' AND ':16' ) OR
        ':14' <> 'AP')
AND ( (':14' = 'IN' AND h.Order_H_Invoice_Date BETWEEN ':15' AND ':16' ) OR
        ':14' <> 'IN')
AND ( h.Order_H_Invoice_No LIKE ':17' OR ':17' = '')
AND ( (':14' = 'PA' AND h.Order_H_Confirm_Date BETWEEN ':15' AND ':16' ) OR
        ':14' <> 'PA')
AND ( (':14' = 'RC' AND h.Order_H_Shipped_Comp BETWEEN ':15' AND ':16' ) OR
        ':14' <> 'RC')
AND ( (':14' = 'ID' AND h.Order_H_Date_Appr_Rj BETWEEN ':15' AND ':16' ) OR
        ':14' <> 'ID')
AND ( (':14' = 'PI' AND h.Order_H_Invoice_Date BETWEEN ':15' AND ':16' ) OR
        ':14' <> 'PI')
AND ( ((UPPER(':18') = 'TRUE' OR UPPER(':19') = 'TRUE' ) AND
h.Order_H_Credit_Card_ = 1) OR
       (UPPER(':18') <> 'TRUE' AND UPPER(':19') <> 'TRUE') )
AND ( (UPPER(':18') = 'TRUE' AND h.Order_H_Credit_Car_1 = 'I')  OR
       (UPPER(':19') = 'TRUE' AND h.Order_H_Credit_Car_1 = 'E')  OR
       (UPPER(':18') <> 'TRUE' AND UPPER(':19') <> 'TRUE'))
AND ( (:20 > 0 AND ( h.Order_H_Vendor_ID = 0 OR h.Order_H_Vendor_ID = :20 ))
OR :20 = 0 ) ORDER BY :21

Yes you can. all the webservices I am hosting is in Document literal and in
v9.4



Create your batch program to have a return Value which is the XML Blob you
will be returning for your webservice and one parameter as input which is
the XML Blob your program will receive. Each will need to have a valid XSD
that you will need in the builder program to generate the WSDL file.



When you choose the WSDL builder in v9.4,  make sure there is only one
Argument for Public Program. with Direction In. This is your xml blob the
batch program is receiving and whose schema is defined in the "Schema"
Button. In the Optional Schema-Based  Document. this is where you define how
the XML you are receiving and XML you are sending back is defined. ( you can
use XMLSPy or any other tool to generate this file)



I am in the process of moving it to UP1.8 but is having some trouble. in
generating the required files trhough the builder that came with UP1.8



HTH

C









From: magicu-l@yahoogroups.com [mailto:magicu-l@yahoogroups.com] On Behalf
Of Jackson, Adam
Sent: November-06-09 11:47 AM
To: magicu-l@yahoogroups.com
Subject: [magicu-l] document literal vs RPC encoding





HI,



I have created a web service for our Sharepoint developer to invoke a
transaction in our ERP app. He is trying create his form in MS Infopath
but that says it does not support RPC encoding only document literal. Is
there a way in Magic 9.4 to creat document literal WSDL files or is it
strictly RPC only?



Thanks



Adam






The contents of this email are confidential and for the exclusive use of the
intended recipient. If you receive this email in error you should not copy
it, retransmit it, use it or disclose its contents but should return it to
the sender immediately and delete your copy.

[Non-text portions of this message have been removed]





[Non-text portions of this message have been removed]

I fnally completed the SQL and tested it in a standalone program. After
incoporating this same query in a batch program , which is called from the web I
get this error.  Gosh Syntax Error , when the same query works fine

Failed to open, table:
[Pervasive][ODBC Client Interface][Pervasive][ODBC Engine Interface]Syntax
Error: Select h.Order_H_PO , h.Order_H_Purchase_Req, h.Order_H_Ordered_By_C,
h.Order_H_Order_Status, h.Order_H_Order_Type, h.Order_H_Order_SubTyp,
h.Order_H_Customer_No, h.Order_H_Customer_Div, h.Order_H

I'll paste the query at the bottom.

So in this case how does one debug the Syntax error. ?

Thanks

Idris

Sorry for the longish query , but here it is.. repeat it works fine by itself in
a test program, run directly from the program repository.

SELECT

h.Order_H_PO ,
h.Order_H_Purchase_Req,
h.Order_H_Ordered_By_C,
h.Order_H_Order_Status,
h.Order_H_Order_Type,
h.Order_H_Order_SubTyp,
h.Order_H_Invoice_No

FROm Order_Header h

WHERE
( h.Order_H_PO LIKE ':1' OR ':1' = '')
AND ( position(RTRIM(':2'),h.Order_H_Purchase_Req) > 0 OR ':2' = '' )
AND (
(:3 < 500 and h.Order_H_Ordered_By_C= RTRIM(':4')) OR
(:3 > 499 and ( (h.Order_H_Ordered_By_C >= RTRIM(':4') and
                   h.Order_H_Ordered_By_C <= REPLICATE ('Z',30)) OR ':4' = ''))
     )
AND ( h.Order_H_Order_Status = ':5' OR ':5' = '')
AND ( h.Order_H_Order_Type = ':6' OR ':6' = '')
AND ( h.Order_H_Order_SubTyp = ':7' OR ':7' = '')
AND ( (':8' = 'BI' AND ( h.Order_H_Order_Type = 'PO' OR h.Order_H_Order_Type =
'SO' ) AND
       h.Order_H_Order_Status <> 'QS' AND h.Order_H_Order_Status <> 'QC' AND
       h.Order_H_Order_Status <> 'RJ' AND h.Order_H_Order_Status <> 'CA' AND
       ( ( UPPER(':9') =  'TRUE' AND h.Order_H_Order_Status <> 'PD') OR
         ( UPPER(':9') <> 'TRUE')
       )
       ) OR ':8' <> 'BI'
     )
AND ( (:10 = 99999 AND h.Order_H_Customer_No = RTRIM(':11')) OR :10 <> 99999 )
AND ( h.Order_H_Customer_Div = RTRIM(':12') OR ':12' = '' )
AND ( (:10 < 1000 AND h.Order_H_Customer_Dep = RTRIM(':13')) OR :10 >= 1000 )
AND ( (':14' = 'OR' AND h.Order_H_Date BETWEEN ':15' AND ':16' ) OR
        ':14' <> 'OR')
AND ( (':14' = 'IP' AND h.Order_H_Confirm_Date BETWEEN ':15' AND ':16' ) OR
        ':14' <> 'IP')
AND ( (':14' = 'SH' AND h.Order_H_Shipped_Comp BETWEEN ':15' AND ':16' ) OR
        ':14' <> 'SH')
AND ( (':14' = 'AP' AND h.Order_H_Date_Appr_Rj BETWEEN ':15' AND ':16' ) OR
        ':14' <> 'AP')
AND ( (':14' = 'IN' AND h.Order_H_Invoice_Date BETWEEN ':15' AND ':16' ) OR
        ':14' <> 'IN')
AND ( h.Order_H_Invoice_No LIKE ':17' OR ':17' = '')
AND ( (':14' = 'PA' AND h.Order_H_Confirm_Date BETWEEN ':15' AND ':16' ) OR
        ':14' <> 'PA')
AND ( (':14' = 'RC' AND h.Order_H_Shipped_Comp BETWEEN ':15' AND ':16' ) OR
        ':14' <> 'RC')
AND ( (':14' = 'ID' AND h.Order_H_Date_Appr_Rj BETWEEN ':15' AND ':16' ) OR
        ':14' <> 'ID')
AND ( (':14' = 'PI' AND h.Order_H_Invoice_Date BETWEEN ':15' AND ':16' ) OR
        ':14' <> 'PI')
AND ( ((UPPER(':18') = 'TRUE' OR UPPER(':19') = 'TRUE' ) AND
h.Order_H_Credit_Card_ = 1) OR
       (UPPER(':18') <> 'TRUE' AND UPPER(':19') <> 'TRUE') )
AND ( (UPPER(':18') = 'TRUE' AND h.Order_H_Credit_Car_1 = 'I')  OR
       (UPPER(':19') = 'TRUE' AND h.Order_H_Credit_Car_1 = 'E')  OR
       (UPPER(':18') <> 'TRUE' AND UPPER(':19') <> 'TRUE'))
AND ( (:20 > 0 AND ( h.Order_H_Vendor_ID = 0 OR h.Order_H_Vendor_ID = :20 )) OR
:20 = 0 ) ORDER BY :21

Frank,
Your comment emphasizes the seriousness of the issue.

To distribute a colors/fonts/Key Mapping/Print Attributes/HTML_Style sensitive
component, practically you have to distribute the source code also!  Let us talk
about the lost business opportunity of the world of redistributable components
since its conception with Ver 9.x.

Thanks.

Thomas Titus


--- In magicu-l@yahoogroups.com, "Frank Van Herreweghe"
<Frank.Van.Herreweghe@...> wrote:
>
> Same here, I never released my component framework because of this serious
> problem.
>
> It was on the planning in 2006 (as yuval said on the miug) but it fell off
> the table.
>
> Last month on a Magic Café day in NL, I asked Yuval again and I had to send
> him a reminder which I did.
>
> … so let's hope this time the table is big enough :-)
>
>
>
> Regards,
>
> Frank Van Herreweghe
>
>
>
>   _____
>
> From: magicu-l@yahoogroups.com [mailto:magicu-l@yahoogroups.com] On Behalf
> Of Ariel Kotler
> Sent: vrijdag 6 november 2009 14:43
> To: magicu-l@yahoogroups.com
> Subject: [magicu-l] Re: Call By Name Color and Font issues
>
>
>
>
>
> Hi ,
>
> Yes , it's true , my wish to have separated font/colors for each
> applications opened in the "Wish List" since they released v9 with
> components .
> Till now we can not share our components between us .
>
> Ariel Kotler
>
> --- In magicu-l@yahoogroup <mailto:magicu-l%40yahoogroups.com> s.com,
> "THOMAS" <ttitus_aerocontrols_com@> wrote:
> >
> > Hi Folks,
> > I have two projects; each uses its own application color file and
> application font file (in the application properties).
> > When I call Task2 by name of ecf2 from ecf1, it executes with color and
> font file of ecf1 which displays a very tasteless screen. I contacted MSE
> support. They advised me to insert eval
> INIPUT(RuntimeApplicationColorDefinitionFile, color2.eng) and
> INIPUT(RuntimeApplicationFontDefinitionFile, font2.eng) in the called
> program. I used the evals before the call, after the call, task prefix of
> task2, task prefix of main2 etc and none worked. I send my sample projects
> to MSE and now MSE agrees with me, saying "There is a wish opened for this
> to have it changed however….".
> > I am assuming anyone who worked with cross called ecfs should have
> experienced this before and I like to know how you handled it.
> > Thomas Titus
> >
>
>
>
>
>
> [Non-text portions of this message have been removed]
>

I'm using it for online calls, and what i see is, that the first time the user
is calling something through call by name (same results with and without the
tiny bridge app) it's taking like 6-8 seconds, but then any call thereafter is
like it's a call to a local program (at least that's the feeling i got in online
calls - but it's possible that for multiple calls in a batch task there will be
a significant difference). I remember seeing in the magic helpfiles, something
about the first time you call byname to another app, it's loading that
application in memory, therefore the results isee makes sense, since after the
first call it's loaded in memory.

Zvi

--- In magicu-l@yahoogroups.com, "Andy Jerison" <ajerison@...> wrote:
>
> Hi Zvi,
>
> If you get a chance, it'd be really interesting to see whether the bridge
> component technique has a significant effect on performance. A simple
> benchmark would be to execute some large number of Invoke Program operations
> on the same program, once going through the bridge, once going directly to
> the component -- and of course, once calling a local copy of the program.
>
> Andy J
>
> -----Original Message-----
> From: magicu-l@yahoogroups.com [mailto:magicu-l@yahoogroups.com] On Behalf
> Of zvidev
> Sent: Monday, November 09, 2009 10:17 AM
> To: magicu-l@yahoogroups.com
> Subject: [magicu-l] Re: Call By Name Color and Font issues
>
> I see your point.
>
> I happened to encounter the issue of recursive components many times from
> non generalized calls, and I had to struggle with it when my issue was not
> at all the source code, until I found a solution which took me a couple of
> full days to figure out.
> It's basically by involving a 3rd component application which I just use as
> a bridge to call back anything from a host application and depending on the
> circumstances I also have a generic exposed event in all host applications,
> with a parameter of the program public name to call and some extra alpha
> parms. Now this structure can be generalized with passing public names and
> parms from a table the way you described it. Although I did not do it,
> because it was not my purpose, but I'm sure it can be worked out, the only
> drawback is that you need an extra component app as a bridge.
>
> I did not elaborate enough on the exact structure, if someone is really in a
> need of recursive usage of components I can send him a demo privately, or
> describe it step by step when I'm in my office.
>
> In any case we all agree that the recursive & color/font issue should be a
> high priority for MSE to resolve natively.
>
> Zvi
>
>
> --- In magicu-l@yahoogroups.com, "ola_ip" <omar.lamin@> wrote:
> >
> > Hi Zvi,
> >
> > What I mean is that a generic component can have generic 'call by Name'
> operations already in place at pertinent places in its code. On these, the
> 'Public Program Name' and 'Cabinet file Name' properties would reference (by
> expressions) columns of a generic component table that would hold the values
> of the program to call.
> >
> > Source code of the component would not be required. Just an online
> configuration program that would enable maintaining programs public names
> and ecf(s) locations.
> >
> > BTW, that concept already works. Except for the recursive limitation
> everything is in place for it to work. A protocol also needs to be
> established when it comes to arguments, but that is acceptable at this
> point.
> >
> > The idea is that the component process could have generic hooks to
> processes that would reside in the host project where the owner of this host
> application would have the ability to control further the execution of the
> combined solution at this point.
> >
> > It would open a door that is already there on a new set of generic
> possibilities. However, currently there is that useless lock (recursion
> limitation) on it.
> >
> > Does that makes more sense to you at this point. Do you see that for these
> generic hooks, the source code of the component would not be required?
> >
> > Omar,
> >
> >
> > --- In magicu-l@yahoogroups.com, "zvidev" <magicusergroup@> wrote:
> > >
> > > I agree with your concept that to benefit at it's max from components,
> those obstacles exist, i just have a little comment about the paragragreph
> below.
> > >
> > > "Furthermore, it is also so when it comes to the recursion restriction
> that MSE imposes on components. Again, the work around for this restriction
> implies that one has to modify the source code of the component to be able
> to make it work with its host application defeating again the sole purpose
> of being able to generically share the component where recursion would be
> applicable."
> > >
> > > To use a component recursively - meaning to call back from the component
> application a host application program etc.- you will always need the source
> code and the ability to modify the source code. or maybe my understanding of
> recursive is incorrect..
>
> > >
> > > Zvi
> > >
> > >
> > > --- In magicu-l@yahoogroups.com, "ola_ip" <omar.lamin@> wrote:
> > > >
> > > > Just to emphasize what Thomas is saying.
> > > >
> > > > The fact that a work around for a component issue implies that one
> needs the source code to be able to modify both the component and the host
> application means that the sole purpose of what components are for is broken
> and as such the work around is not acceptable even if it is a work around.
> > > >
> > > > Using components to divide your application into several components is
> good. But the true purpose should be that we can share components among us
> in the community where each component is a black box that does a pre-defined
> numbers of things. To take advantage of someone else's component, on should
> not have to have access to this component source code and have to fix it for
> it to work with his own application. Currently it is very likely to be so
> when it comes to font and colors handling and as such is a deal breaker.
> > > >
> > > > Furthermore, it is also so when it comes to the recursion restriction
> that MSE imposes on components. Again, the work around for this restriction
> implies that one has to modify the source code of the component to be able
> to make it work with its host application defeating again the sole purpose
> of being able to generically share the component where recursion would be
> applicable.
> > > >
> > > > When, but only when MSE will come to realize the severity of these
> deficiencies and fix them will components be able to significantly achieve
> their true potential by enabling more and more component base add-on.
> Meanwhile, with the source code or in circumstances that avoid colors and
> fonts issue (batch) as well as the recursion limitation it is still possible
> to do things like for those products I plug here and there in my posts here.
> But it really limits the scope of what can be done to always have to keep
> such limitations in mind.
> > > >
> > > > Omar,
> > > >
> > > > --- In magicu-l@yahoogroups.com, "THOMAS" <ttitus_aerocontrols_com@>
> wrote:
> > > > >
> > > > >
> > > > >
> > > > > Thanks Ariel,
> > > > >
> > > > > I believe, the settings in the external files tab of the project
> properties must always override the INI settings, but this works only if the
> ecf is opened as main application.
> > > > >
> > > > > When the component was introduced by MSE as I recall, they are
> referred as easily "distributable libraries".
> > > > >
> > > > > If I understand the endeavor correctly, to use a color/font
> sensitive distributed component, in practice, you have to get the source
> code also. Then you incorporate colors/fonts of the component to the
> colors/fonts of your main project, change the color/font pointers inside the
> component's project accordingly, rebuild the component cabinet file and then
> add the component to your main project.
> > > > >
> > > > > The interesting part is, when you design form screens for the
> purpose of running as main application, tool or component, the color/fonts
> files being in effect are files specified in the project properties. When
> you run the application as a component (or call by name), the engine
> completely ignores these files. There is no way to associate these files
> back to the application at runtime.
> > > > >
> > > > > Simply from the point of project management and considering what is
> in stake, this should have been corrected soon after the introduction of
> components. This high maintenance work around also makes the distribution of
> version updates of components with colors, fonts, key board mapping etc very
> difficult.
> > > > >
> > > > > Thanks
> > > > >
> > > > > Thomas Titus
> > > > >
> > > > >
> > > > > --- In magicu-l@yahoogroups.com, "Ariel Kotler" <prosto4u@> wrote:
> > > > > >
> > > > > > Hi ,
> > > > > >
> > > > > > Yes , it's true , my wish to have separated font/colors for each
> applications opened in the "Wish List" since they released v9 with
> components .
> > > > > > Till now we can not share our components between us .
> > > > > >
> > > > > > Ariel Kotler
> > > > > >
> > > > > > --- In magicu-l@yahoogroups.com, "THOMAS"
> <ttitus_aerocontrols_com@> wrote:
> > > > > > >
> > > > > > > Hi Folks,
> > > > > > > I have two projects; each uses its own application color file
> and application font file (in the application properties).
> > > > > > > When I call Task2 by name of ecf2 from ecf1, it executes with
> color and font file of ecf1 which displays a very tasteless screen. I
> contacted MSE support. They advised me to insert eval
> INIPUT(RuntimeApplicationColorDefinitionFile, color2.eng) and
> INIPUT(RuntimeApplicationFontDefinitionFile, font2.eng) in the called
> program. I used the evals  before the call, after the call, task prefix of
> task2, task prefix of main2  etc and none worked. I send my sample projects
> to MSE and now MSE agrees with me, saying "There is a wish opened for this
> to have it changed however..".
> > > > > > > I am assuming anyone who worked with cross called ecfs should
> have experienced this before and I like to know how you handled it.
> > > > > > > Thomas Titus
>

Thanks again, Zvi.

It's not in the release notes but I'll watch out for it.




Alan


-----Original Message-----
From: magicu-l@yahoogroups.com [mailto:magicu-l@yahoogroups.com] On
Behalf Of zvidev
Sent: Monday, November 09, 2009 3:41 PM
To: magicu-l@yahoogroups.com
Subject: 1.8SP1a

Just in case it matters to you, in SP1 there is a bug where  there are
calls to .bat file using invokeOS and the Wait Setting = Yes, that you
will get the error "File not found". The way to overcome this error was
to use 'CMD /C '&[your batch file], instead of a straight call to the
batch file. But it's possible that it was resolved in SP1a, i didn't
check the release notes yet.

HTH
Zvi

--- In magicu-l@yahoogroups.com, "Brookes, Alan" <alan.j.brookes@...>
wrote:
>
> Thanks for that Zvi.
>
> Time to upgrade my Studio environment then, I guess!
>
>
>
>
> Alan
>
>
> -----Original Message-----
> From: magicu-l@yahoogroups.com [mailto:magicu-l@yahoogroups.com] On
> Behalf Of zvidev
> Sent: Monday, November 09, 2009 3:20 PM
> To: magicu-l@yahoogroups.com
> Subject: [magicu-l] Re: 1.8SP1a
>
> Alan
>
> I don't know about 1.8SP1a, but ecf filles created with 1.8SP1 worked
> for in deployments of 1.8b. I assume SP1a wouldn't be any different.
>
> Zvi
>
> --- In magicu-l@yahoogroups.com, "Brookes, Alan" <alan.j.brookes@>
> wrote:
> >
> > Does anybody know if the new, better-compressed .ecf files produced
by
> > UP 1.8 SP1 can be read by 1.8b?
> >
> > My Live environment is 1.8b but I am anxious to test SP1a...
> >
> >
> >
> >
> > Alan Brookes
> > Service Desk Analyst
> > IS Service Delivery, UK
> >
> >
> > e2v
> > 106 Waterhouse Lane, Chelmsford, Essex, CM1 2QU, England
> > Tel:  +44 (0)1245 453698
> >  www.e2v.com
> >
> > Consider the environment: do you really need to print this email?
> >
> >
> > -----Original Message-----
> > From: magicu-l@yahoogroups.com [mailto:magicu-l@yahoogroups.com] On
> > Behalf Of Rosie Coller
> > Sent: Monday, November 09, 2009 11:24 AM
> > To: magicu-l@yahoogroups.com
> > Subject: 1.8SP1a
> >
> > You mean there is an SP1a already!
> >
> > We've only just started using SP1. Will have to take a look at that.
> >
> > Rosie
> >
> > 2009/11/7 Andy Jerison <ajerison@>
> >
> > >
> > >
> > > They did? I actually interrupted the download after I read the
> release
> > > notes
> > > and didn't find that one!
> > >
> > > Oh well....
> > >
> > > Andy J
> > >
> > > -----Original Message-----
> > > From: magicu-l@yahoogroups.com <magicu-l%40yahoogroups.com>
[mailto:
> > > magicu-l@yahoogroups.com <magicu-l%40yahoogroups.com>] On Behalf
> > > Of Wes
> > > Sent: Friday, November 06, 2009 4:29 PM
> > > To: magicu-l@yahoogroups.com <magicu-l%40yahoogroups.com>
> > > Subject: [magicu-l] SP1a
> > >
> > > They fixed the transparent radio button issue!
> > >
> > > Wes

Sent by a member of the e2v group of companies. The parent company, e2v
technologies plc, is registered in England and Wales. Company number; 04439718.
Registered address; 106 Waterhouse Lane, Chelmsford, Essex, CM1 2QU, UK.  This
email and any attachments are confidential and meant solely for the use of the
intended recipient. If you are not the intended recipient and have received this
email in error, please notify us immediately by replying to the sender and then
deleting this copy and the reply from your system without further disclosing,
copying, distributing or using the e-mail or any attachment. Thank you for your
cooperation.
______________________________________________________
________________
This email has been scanned by the MessageLabs Email Security System.
For more information please visit http://www.messagelabs.com/email
______________________________________________________________________

Hi Zvi,

If you get a chance, it'd be really interesting to see whether the bridge
component technique has a significant effect on performance. A simple
benchmark would be to execute some large number of Invoke Program operations
on the same program, once going through the bridge, once going directly to
the component -- and of course, once calling a local copy of the program.

Andy J

-----Original Message-----
From: magicu-l@yahoogroups.com [mailto:magicu-l@yahoogroups.com] On Behalf
Of zvidev
Sent: Monday, November 09, 2009 10:17 AM
To: magicu-l@yahoogroups.com
Subject: [magicu-l] Re: Call By Name Color and Font issues

I see your point.

I happened to encounter the issue of recursive components many times from
non generalized calls, and I had to struggle with it when my issue was not
at all the source code, until I found a solution which took me a couple of
full days to figure out.
It's basically by involving a 3rd component application which I just use as
a bridge to call back anything from a host application and depending on the
circumstances I also have a generic exposed event in all host applications,
with a parameter of the program public name to call and some extra alpha
parms. Now this structure can be generalized with passing public names and
parms from a table the way you described it. Although I did not do it,
because it was not my purpose, but I'm sure it can be worked out, the only
drawback is that you need an extra component app as a bridge.

I did not elaborate enough on the exact structure, if someone is really in a
need of recursive usage of components I can send him a demo privately, or
describe it step by step when I'm in my office.

In any case we all agree that the recursive & color/font issue should be a
high priority for MSE to resolve natively.

Zvi


--- In magicu-l@yahoogroups.com, "ola_ip" <omar.lamin@...> wrote:
>
> Hi Zvi,
>
> What I mean is that a generic component can have generic 'call by Name'
operations already in place at pertinent places in its code. On these, the
'Public Program Name' and 'Cabinet file Name' properties would reference (by
expressions) columns of a generic component table that would hold the values
of the program to call.
>
> Source code of the component would not be required. Just an online
configuration program that would enable maintaining programs public names
and ecf(s) locations.
>
> BTW, that concept already works. Except for the recursive limitation
everything is in place for it to work. A protocol also needs to be
established when it comes to arguments, but that is acceptable at this
point.
>
> The idea is that the component process could have generic hooks to
processes that would reside in the host project where the owner of this host
application would have the ability to control further the execution of the
combined solution at this point.
>
> It would open a door that is already there on a new set of generic
possibilities. However, currently there is that useless lock (recursion
limitation) on it.
>
> Does that makes more sense to you at this point. Do you see that for these
generic hooks, the source code of the component would not be required?
>
> Omar,
>
>
> --- In magicu-l@yahoogroups.com, "zvidev" <magicusergroup@> wrote:
> >
> > I agree with your concept that to benefit at it's max from components,
those obstacles exist, i just have a little comment about the paragragreph
below.
> >
> > "Furthermore, it is also so when it comes to the recursion restriction
that MSE imposes on components. Again, the work around for this restriction
implies that one has to modify the source code of the component to be able
to make it work with its host application defeating again the sole purpose
of being able to generically share the component where recursion would be
applicable."
> >
> > To use a component recursively - meaning to call back from the component
application a host application program etc.- you will always need the source
code and the ability to modify the source code. or maybe my understanding of
recursive is incorrect..

> >
> > Zvi
> >
> >
> > --- In magicu-l@yahoogroups.com, "ola_ip" <omar.lamin@> wrote:
> > >
> > > Just to emphasize what Thomas is saying.
> > >
> > > The fact that a work around for a component issue implies that one
needs the source code to be able to modify both the component and the host
application means that the sole purpose of what components are for is broken
and as such the work around is not acceptable even if it is a work around.
> > >
> > > Using components to divide your application into several components is
good. But the true purpose should be that we can share components among us
in the community where each component is a black box that does a pre-defined
numbers of things. To take advantage of someone else's component, on should
not have to have access to this component source code and have to fix it for
it to work with his own application. Currently it is very likely to be so
when it comes to font and colors handling and as such is a deal breaker.
> > >
> > > Furthermore, it is also so when it comes to the recursion restriction
that MSE imposes on components. Again, the work around for this restriction
implies that one has to modify the source code of the component to be able
to make it work with its host application defeating again the sole purpose
of being able to generically share the component where recursion would be
applicable.
> > >
> > > When, but only when MSE will come to realize the severity of these
deficiencies and fix them will components be able to significantly achieve
their true potential by enabling more and more component base add-on.
Meanwhile, with the source code or in circumstances that avoid colors and
fonts issue (batch) as well as the recursion limitation it is still possible
to do things like for those products I plug here and there in my posts here.
But it really limits the scope of what can be done to always have to keep
such limitations in mind.
> > >
> > > Omar,
> > >
> > > --- In magicu-l@yahoogroups.com, "THOMAS" <ttitus_aerocontrols_com@>
wrote:
> > > >
> > > >
> > > >
> > > > Thanks Ariel,
> > > >
> > > > I believe, the settings in the external files tab of the project
properties must always override the INI settings, but this works only if the
ecf is opened as main application.
> > > >
> > > > When the component was introduced by MSE as I recall, they are
referred as easily "distributable libraries".
> > > >
> > > > If I understand the endeavor correctly, to use a color/font
sensitive distributed component, in practice, you have to get the source
code also. Then you incorporate colors/fonts of the component to the
colors/fonts of your main project, change the color/font pointers inside the
component's project accordingly, rebuild the component cabinet file and then
add the component to your main project.
> > > >
> > > > The interesting part is, when you design form screens for the
purpose of running as main application, tool or component, the color/fonts
files being in effect are files specified in the project properties. When
you run the application as a component (or call by name), the engine
completely ignores these files. There is no way to associate these files
back to the application at runtime.
> > > >
> > > > Simply from the point of project management and considering what is
in stake, this should have been corrected soon after the introduction of
components. This high maintenance work around also makes the distribution of
version updates of components with colors, fonts, key board mapping etc very
difficult.
> > > >
> > > > Thanks
> > > >
> > > > Thomas Titus
> > > >
> > > >
> > > > --- In magicu-l@yahoogroups.com, "Ariel Kotler" <prosto4u@> wrote:
> > > > >
> > > > > Hi ,
> > > > >
> > > > > Yes , it's true , my wish to have separated font/colors for each
applications opened in the "Wish List" since they released v9 with
components .
> > > > > Till now we can not share our components between us .
> > > > >
> > > > > Ariel Kotler
> > > > >
> > > > > --- In magicu-l@yahoogroups.com, "THOMAS"
<ttitus_aerocontrols_com@> wrote:
> > > > > >
> > > > > > Hi Folks,
> > > > > > I have two projects; each uses its own application color file
and application font file (in the application properties).
> > > > > > When I call Task2 by name of ecf2 from ecf1, it executes with
color and font file of ecf1 which displays a very tasteless screen. I
contacted MSE support. They advised me to insert eval
INIPUT(RuntimeApplicationColorDefinitionFile, color2.eng) and
INIPUT(RuntimeApplicationFontDefinitionFile, font2.eng) in the called
program. I used the evals  before the call, after the call, task prefix of
task2, task prefix of main2  etc and none worked. I send my sample projects
to MSE and now MSE agrees with me, saying "There is a wish opened for this
to have it changed however..".
> > > > > > I am assuming anyone who worked with cross called ecfs should
have experienced this before and I like to know how you handled it.
> > > > > > Thomas Titus

Just in case it matters to you, in SP1 there is a bug where  there are calls to
.bat file using invokeOS and the Wait Setting = Yes, that you will get the error
"File not found". The way to overcome this error was to use 'CMD /C '&[your
batch file], instead of a straight call to the batch file. But it's possible
that it was resolved in SP1a, i didn't check the release notes yet.

HTH
Zvi

--- In magicu-l@yahoogroups.com, "Brookes, Alan" <alan.j.brookes@...> wrote:
>
> Thanks for that Zvi.
>
> Time to upgrade my Studio environment then, I guess!
>
>
>
>
> Alan
>
>
> -----Original Message-----
> From: magicu-l@yahoogroups.com [mailto:magicu-l@yahoogroups.com] On
> Behalf Of zvidev
> Sent: Monday, November 09, 2009 3:20 PM
> To: magicu-l@yahoogroups.com
> Subject: [magicu-l] Re: 1.8SP1a
>
> Alan
>
> I don't know about 1.8SP1a, but ecf filles created with 1.8SP1 worked
> for in deployments of 1.8b. I assume SP1a wouldn't be any different.
>
> Zvi
>
> --- In magicu-l@yahoogroups.com, "Brookes, Alan" <alan.j.brookes@>
> wrote:
> >
> > Does anybody know if the new, better-compressed .ecf files produced by
> > UP 1.8 SP1 can be read by 1.8b?
> >
> > My Live environment is 1.8b but I am anxious to test SP1a...
> >
> >
> >
> >
> > Alan Brookes
> > Service Desk Analyst
> > IS Service Delivery, UK
> >
> >
> > e2v
> > 106 Waterhouse Lane, Chelmsford, Essex, CM1 2QU, England
> > Tel:  +44 (0)1245 453698
> >  www.e2v.com
> >
> > Consider the environment: do you really need to print this email?
> >
> >
> > -----Original Message-----
> > From: magicu-l@yahoogroups.com [mailto:magicu-l@yahoogroups.com] On
> > Behalf Of Rosie Coller
> > Sent: Monday, November 09, 2009 11:24 AM
> > To: magicu-l@yahoogroups.com
> > Subject: 1.8SP1a
> >
> > You mean there is an SP1a already!
> >
> > We've only just started using SP1. Will have to take a look at that.
> >
> > Rosie
> >
> > 2009/11/7 Andy Jerison <ajerison@>
> >
> > >
> > >
> > > They did? I actually interrupted the download after I read the
> release
> > > notes
> > > and didn't find that one!
> > >
> > > Oh well....
> > >
> > > Andy J
> > >
> > > -----Original Message-----
> > > From: magicu-l@yahoogroups.com <magicu-l%40yahoogroups.com> [mailto:
> > > magicu-l@yahoogroups.com <magicu-l%40yahoogroups.com>] On Behalf
> > > Of Wes
> > > Sent: Friday, November 06, 2009 4:29 PM
> > > To: magicu-l@yahoogroups.com <magicu-l%40yahoogroups.com>
> > > Subject: [magicu-l] SP1a
> > >
> > > They fixed the transparent radio button issue!
> > >
> > > Wes
>
> Sent by a member of the e2v group of companies. The parent company, e2v
technologies plc, is registered in England and Wales. Company number; 04439718.
Registered address; 106 Waterhouse Lane, Chelmsford, Essex, CM1 2QU, UK.  This
email and any attachments are confidential and meant solely for the use of the
intended recipient. If you are not the intended recipient and have received this
email in error, please notify us immediately by replying to the sender and then
deleting this copy and the reply from your system without further disclosing,
copying, distributing or using the e-mail or any attachment. Thank you for your
cooperation.
> ______________________________________________________
> ________________
> This email has been scanned by the MessageLabs Email Security System.
> For more information please visit http://www.messagelabs.com/email
> ______________________________________________________________________
>

Thanks for that Zvi.

Time to upgrade my Studio environment then, I guess!




Alan


-----Original Message-----
From: magicu-l@yahoogroups.com [mailto:magicu-l@yahoogroups.com] On
Behalf Of zvidev
Sent: Monday, November 09, 2009 3:20 PM
To: magicu-l@yahoogroups.com
Subject: [magicu-l] Re: 1.8SP1a

Alan

I don't know about 1.8SP1a, but ecf filles created with 1.8SP1 worked
for in deployments of 1.8b. I assume SP1a wouldn't be any different.

Zvi

--- In magicu-l@yahoogroups.com, "Brookes, Alan" <alan.j.brookes@...>
wrote:
>
> Does anybody know if the new, better-compressed .ecf files produced by
> UP 1.8 SP1 can be read by 1.8b?
>
> My Live environment is 1.8b but I am anxious to test SP1a...
>
>
>
>
> Alan Brookes
> Service Desk Analyst
> IS Service Delivery, UK
>
>
> e2v
> 106 Waterhouse Lane, Chelmsford, Essex, CM1 2QU, England
> Tel:  +44 (0)1245 453698
>  www.e2v.com
>
> Consider the environment: do you really need to print this email?
>
>
> -----Original Message-----
> From: magicu-l@yahoogroups.com [mailto:magicu-l@yahoogroups.com] On
> Behalf Of Rosie Coller
> Sent: Monday, November 09, 2009 11:24 AM
> To: magicu-l@yahoogroups.com
> Subject: 1.8SP1a
>
> You mean there is an SP1a already!
>
> We've only just started using SP1. Will have to take a look at that.
>
> Rosie
>
> 2009/11/7 Andy Jerison <ajerison@...>
>
> >
> >
> > They did? I actually interrupted the download after I read the
release
> > notes
> > and didn't find that one!
> >
> > Oh well....
> >
> > Andy J
> >
> > -----Original Message-----
> > From: magicu-l@yahoogroups.com <magicu-l%40yahoogroups.com> [mailto:
> > magicu-l@yahoogroups.com <magicu-l%40yahoogroups.com>] On Behalf
> > Of Wes
> > Sent: Friday, November 06, 2009 4:29 PM
> > To: magicu-l@yahoogroups.com <magicu-l%40yahoogroups.com>
> > Subject: [magicu-l] SP1a
> >
> > They fixed the transparent radio button issue!
> >
> > Wes

Sent by a member of the e2v group of companies. The parent company, e2v
technologies plc, is registered in England and Wales. Company number; 04439718.
Registered address; 106 Waterhouse Lane, Chelmsford, Essex, CM1 2QU, UK.  This
email and any attachments are confidential and meant solely for the use of the
intended recipient. If you are not the intended recipient and have received this
email in error, please notify us immediately by replying to the sender and then
deleting this copy and the reply from your system without further disclosing,
copying, distributing or using the e-mail or any attachment. Thank you for your
cooperation.
______________________________________________________
________________
This email has been scanned by the MessageLabs Email Security System.
For more information please visit http://www.messagelabs.com/email
______________________________________________________________________

Hi Andy.

That was my gut feeling too - so I though I'd ask the group.

I felt sure _someone_ would have tested it, if only for devilment!




Alan


-----Original Message-----
From: magicu-l@yahoogroups.com [mailto:magicu-l@yahoogroups.com] On
Behalf Of Andy Jerison
Sent: Monday, November 09, 2009 2:05 PM
To: magicu-l@yahoogroups.com
Subject: 1.8SP1a

Hi Alan,

I'd bet a few beers that 1.8sp1 .ecf's are not usable by 1.8b. It's a
bet
I'd be happy to lose, of course!

Andy J

-----Original Message-----
From: magicu-l@yahoogroups.com [mailto:magicu-l@yahoogroups.com] On
Behalf
Of Brookes, Alan
Sent: Monday, November 09, 2009 7:57 AM
To: magicu-l@yahoogroups.com
Subject: [magicu-l] RE: 1.8SP1a

Does anybody know if the new, better-compressed .ecf files produced by
UP 1.8 SP1 can be read by 1.8b?

My Live environment is 1.8b but I am anxious to test SP1a...




Alan Brookes
Service Desk Analyst
IS Service Delivery, UK


e2v
106 Waterhouse Lane, Chelmsford, Essex, CM1 2QU, England
Tel:  +44 (0)1245 453698
  www.e2v.com

Consider the environment: do you really need to print this email?


-----Original Message-----
From: magicu-l@yahoogroups.com [mailto:magicu-l@yahoogroups.com] On
Behalf Of Rosie Coller
Sent: Monday, November 09, 2009 11:24 AM
To: magicu-l@yahoogroups.com
Subject: 1.8SP1a

You mean there is an SP1a already!

We've only just started using SP1. Will have to take a look at that.

Rosie

2009/11/7 Andy Jerison <ajerison@...>

>
>
> They did? I actually interrupted the download after I read the release
> notes
> and didn't find that one!
>
> Oh well....
>
> Andy J
>
> -----Original Message-----
> From: magicu-l@yahoogroups.com <magicu-l%40yahoogroups.com> [mailto:
> magicu-l@yahoogroups.com <magicu-l%40yahoogroups.com>] On Behalf
> Of Wes
> Sent: Friday, November 06, 2009 4:29 PM
> To: magicu-l@yahoogroups.com <magicu-l%40yahoogroups.com>
> Subject: [magicu-l] SP1a
>
> They fixed the transparent radio button issue!
>
> Wes

Sent by a member of the e2v group of companies. The parent company, e2v
technologies plc, is registered in England and Wales. Company number; 04439718.
Registered address; 106 Waterhouse Lane, Chelmsford, Essex, CM1 2QU, UK.  This
email and any attachments are confidential and meant solely for the use of the
intended recipient. If you are not the intended recipient and have received this
email in error, please notify us immediately by replying to the sender and then
deleting this copy and the reply from your system without further disclosing,
copying, distributing or using the e-mail or any attachment. Thank you for your
cooperation.
______________________________________________________
________________
This email has been scanned by the MessageLabs Email Security System.
For more information please visit http://www.messagelabs.com/email
______________________________________________________________________

Alan

I don't know about 1.8SP1a, but ecf filles created with 1.8SP1 worked for in
deployments of 1.8b. I assume SP1a wouldn't be any different.

Zvi

--- In magicu-l@yahoogroups.com, "Brookes, Alan" <alan.j.brookes@...> wrote:
>
> Does anybody know if the new, better-compressed .ecf files produced by
> UP 1.8 SP1 can be read by 1.8b?
>
> My Live environment is 1.8b but I am anxious to test SP1a...
>
>
>
>
> Alan Brookes
> Service Desk Analyst
> IS Service Delivery, UK
>
>
> e2v
> 106 Waterhouse Lane, Chelmsford, Essex, CM1 2QU, England
> Tel:  +44 (0)1245 453698
>  www.e2v.com
>
> Consider the environment: do you really need to print this email?
>
>
> -----Original Message-----
> From: magicu-l@yahoogroups.com [mailto:magicu-l@yahoogroups.com] On
> Behalf Of Rosie Coller
> Sent: Monday, November 09, 2009 11:24 AM
> To: magicu-l@yahoogroups.com
> Subject: 1.8SP1a
>
> You mean there is an SP1a already!
>
> We've only just started using SP1. Will have to take a look at that.
>
> Rosie
>
> 2009/11/7 Andy Jerison <ajerison@...>
>
> >
> >
> > They did? I actually interrupted the download after I read the release
> > notes
> > and didn't find that one!
> >
> > Oh well....
> >
> > Andy J
> >
> > -----Original Message-----
> > From: magicu-l@yahoogroups.com <magicu-l%40yahoogroups.com> [mailto:
> > magicu-l@yahoogroups.com <magicu-l%40yahoogroups.com>] On Behalf
> > Of Wes
> > Sent: Friday, November 06, 2009 4:29 PM
> > To: magicu-l@yahoogroups.com <magicu-l%40yahoogroups.com>
> > Subject: [magicu-l] SP1a
> >
> > They fixed the transparent radio button issue!
> >
> > Wes
>
> Sent by a member of the e2v group of companies. The parent company, e2v
technologies plc, is registered in England and Wales. Company number; 04439718.
Registered address; 106 Waterhouse Lane, Chelmsford, Essex, CM1 2QU, UK.  This
email and any attachments are confidential and meant solely for the use of the
intended recipient. If you are not the intended recipient and have received this
email in error, please notify us immediately by replying to the sender and then
deleting this copy and the reply from your system without further disclosing,
copying, distributing or using the e-mail or any attachment. Thank you for your
cooperation.
> ______________________________________________________
> ________________
> This email has been scanned by the MessageLabs Email Security System.
> For more information please visit http://www.messagelabs.com/email
> ______________________________________________________________________
>

I see your point.

I happened to encounter the issue of recursive components many times from non
generalized calls, and I had to struggle with it when my issue was not at all
the source code, until I found a solution which took me a couple of full days to
figure out.
It's basically by involving a 3rd component application which I just use as a
bridge to call back anything from a host application and depending on the
circumstances I also have a generic exposed event in all host applications, with
a parameter of the program public name to call and some extra alpha parms. Now
this structure can be generalized with passing public names and parms from a
table the way you described it. Although I did not do it, because it was not my
purpose, but I'm sure it can be worked out, the only drawback is that you need
an extra component app as a bridge.

I did not elaborate enough on the exact structure, if someone is really in a
need of recursive usage of components I can send him a demo privately, or
describe it step by step when I'm in my office.

In any case we all agree that the recursive & color/font issue should be a high
priority for MSE to resolve natively.

Zvi


--- In magicu-l@yahoogroups.com, "ola_ip" <omar.lamin@...> wrote:
>
> Hi Zvi,
>
> What I mean is that a generic component can have generic 'call by Name'
operations already in place at pertinent places in its code. On these, the
'Public Program Name' and 'Cabinet file Name' properties would reference (by
expressions) columns of a generic component table that would hold the values of
the program to call.
>
> Source code of the component would not be required. Just an online
configuration program that would enable maintaining programs public names and
ecf(s) locations.
>
> BTW, that concept already works. Except for the recursive limitation
everything is in place for it to work. A protocol also needs to be established
when it comes to arguments, but that is acceptable at this point.
>
> The idea is that the component process could have generic hooks to processes
that would reside in the host project where the owner of this host application
would have the ability to control further the execution of the combined solution
at this point.
>
> It would open a door that is already there on a new set of generic
possibilities. However, currently there is that useless lock (recursion
limitation) on it.
>
> Does that makes more sense to you at this point. Do you see that for these
generic hooks, the source code of the component would not be required?
>
> Omar,
>
>
> --- In magicu-l@yahoogroups.com, "zvidev" <magicusergroup@> wrote:
> >
> > I agree with your concept that to benefit at it's max from components, those
obstacles exist, i just have a little comment about the paragragreph below.
> >
> > "Furthermore, it is also so when it comes to the recursion restriction that
MSE imposes on components. Again, the work around for this restriction implies
that one has to modify the source code of the component to be able to make it
work with its host application defeating again the sole purpose of being able to
generically share the component where recursion would be applicable."
> >
> > To use a component recursively - meaning to call back from the component
application a host application program etc.- you will always need the source
code and the ability to modify the source code. or maybe my understanding of
recursive is incorrect..
> >
> > Zvi
> >
> >
> > --- In magicu-l@yahoogroups.com, "ola_ip" <omar.lamin@> wrote:
> > >
> > > Just to emphasize what Thomas is saying.
> > >
> > > The fact that a work around for a component issue implies that one needs
the source code to be able to modify both the component and the host application
means that the sole purpose of what components are for is broken and as such the
work around is not acceptable even if it is a work around.
> > >
> > > Using components to divide your application into several components is
good. But the true purpose should be that we can share components among us in
the community where each component is a black box that does a pre-defined
numbers of things. To take advantage of someone else's component, on should not
have to have access to this component source code and have to fix it for it to
work with his own application. Currently it is very likely to be so when it
comes to font and colors handling and as such is a deal breaker.
> > >
> > > Furthermore, it is also so when it comes to the recursion restriction that
MSE imposes on components. Again, the work around for this restriction implies
that one has to modify the source code of the component to be able to make it
work with its host application defeating again the sole purpose of being able to
generically share the component where recursion would be applicable.
> > >
> > > When, but only when MSE will come to realize the severity of these
deficiencies and fix them will components be able to significantly achieve their
true potential by enabling more and more component base add-on. Meanwhile, with
the source code or in circumstances that avoid colors and fonts issue (batch) as
well as the recursion limitation it is still possible to do things like for
those products I plug here and there in my posts here. But it really limits the
scope of what can be done to always have to keep such limitations in mind.
> > >
> > > Omar,
> > >
> > > --- In magicu-l@yahoogroups.com, "THOMAS" <ttitus_aerocontrols_com@>
wrote:
> > > >
> > > >
> > > >
> > > > Thanks Ariel,
> > > >
> > > > I believe, the settings in the external files tab of the project
properties must always override the INI settings, but this works only if the ecf
is opened as main application.
> > > >
> > > > When the component was introduced by MSE as I recall, they are referred
as easily "distributable libraries".
> > > >
> > > > If I understand the endeavor correctly, to use a color/font sensitive
distributed component, in practice, you have to get the source code also. Then
you incorporate colors/fonts of the component to the colors/fonts of your main
project, change the color/font pointers inside the component's project
accordingly, rebuild the component cabinet file and then add the component to
your main project.
> > > >
> > > > The interesting part is, when you design form screens for the purpose of
running as main application, tool or component, the color/fonts files being in
effect are files specified in the project properties. When you run the
application as a component (or call by name), the engine completely ignores
these files. There is no way to associate these files back to the application at
runtime.
> > > >
> > > > Simply from the point of project management and considering what is in
stake, this should have been corrected soon after the introduction of
components. This high maintenance work around also makes the distribution of
version updates of components with colors, fonts, key board mapping etc very
difficult.
> > > >
> > > > Thanks
> > > >
> > > > Thomas Titus
> > > >
> > > >
> > > > --- In magicu-l@yahoogroups.com, "Ariel Kotler" <prosto4u@> wrote:
> > > > >
> > > > > Hi ,
> > > > >
> > > > > Yes , it's true , my wish to have separated font/colors for each
applications opened in the "Wish List" since they released v9 with components .
> > > > > Till now we can not share our components between us .
> > > > >
> > > > > Ariel Kotler
> > > > >
> > > > > --- In magicu-l@yahoogroups.com, "THOMAS" <ttitus_aerocontrols_com@>
wrote:
> > > > > >
> > > > > > Hi Folks,
> > > > > > I have two projects; each uses its own application color file and
application font file (in the application properties).
> > > > > > When I call Task2 by name of ecf2 from ecf1, it executes with color
and font file of ecf1 which displays a very tasteless screen. I contacted MSE
support. They advised me to insert eval
INIPUT(RuntimeApplicationColorDefinitionFile, color2.eng) and
INIPUT(RuntimeApplicationFontDefinitionFile, font2.eng) in the called program. I
used the evals  before the call, after the call, task prefix of task2, task
prefix of main2  etc and none worked. I send my sample projects to MSE and now
MSE agrees with me, saying "There is a wish opened for this to have it changed
however….".
> > > > > > I am assuming anyone who worked with cross called ecfs should have
experienced this before and I like to know how you handled it.
> > > > > > Thomas Titus
> > > > > >
> > > > >
> > > >
> > >
> >
>

Hi Alan,

I'd bet a few beers that 1.8sp1 .ecf's are not usable by 1.8b. It's a bet
I'd be happy to lose, of course!

Andy J

-----Original Message-----
From: magicu-l@yahoogroups.com [mailto:magicu-l@yahoogroups.com] On Behalf
Of Brookes, Alan
Sent: Monday, November 09, 2009 7:57 AM
To: magicu-l@yahoogroups.com
Subject: [magicu-l] RE: 1.8SP1a

Does anybody know if the new, better-compressed .ecf files produced by
UP 1.8 SP1 can be read by 1.8b?

My Live environment is 1.8b but I am anxious to test SP1a...




Alan Brookes
Service Desk Analyst
IS Service Delivery, UK


e2v
106 Waterhouse Lane, Chelmsford, Essex, CM1 2QU, England
Tel:  +44 (0)1245 453698
  www.e2v.com

Consider the environment: do you really need to print this email?


-----Original Message-----
From: magicu-l@yahoogroups.com [mailto:magicu-l@yahoogroups.com] On
Behalf Of Rosie Coller
Sent: Monday, November 09, 2009 11:24 AM
To: magicu-l@yahoogroups.com
Subject: 1.8SP1a

You mean there is an SP1a already!

We've only just started using SP1. Will have to take a look at that.

Rosie

2009/11/7 Andy Jerison <ajerison@...>

>
>
> They did? I actually interrupted the download after I read the release
> notes
> and didn't find that one!
>
> Oh well....
>
> Andy J
>
> -----Original Message-----
> From: magicu-l@yahoogroups.com <magicu-l%40yahoogroups.com> [mailto:
> magicu-l@yahoogroups.com <magicu-l%40yahoogroups.com>] On Behalf
> Of Wes
> Sent: Friday, November 06, 2009 4:29 PM
> To: magicu-l@yahoogroups.com <magicu-l%40yahoogroups.com>
> Subject: [magicu-l] SP1a
>
> They fixed the transparent radio button issue!
>
> Wes

Hello Omar,

I hope someone having field experience will answer your question.

I would be interested too.

Does anyone use MySql with UP for serious tasks , succesfully ?


Meanwhile I dug interesting thread about MySql & UP & Unicode :
http://tech.groups.yahoo.com/group/magicu-l/message/128796

Best regards
Przemek



--- In magicu-l@yahoogroups.com, "ola_ip" <omar.lamin@...> wrote:
>
> Hi,
>
> Just went over Craig's tutorial to work with MySQL.
>
> It suggests working with MySql 5 and odbc 3.5
>
> I vaguely remember a discussion here about possibly 2 ways or 2 versions sets
considerations. I try to look for it here, but can not find anything. I might be
out to lunch again.
>
> Can someone confirm what is the most correct way for what ever reason to setup
MySql to work with Magic or if what Craig proposes is this correct and only way.
>
> Omar
>

Does anybody know if the new, better-compressed .ecf files produced by
UP 1.8 SP1 can be read by 1.8b?

My Live environment is 1.8b but I am anxious to test SP1a...




Alan Brookes
Service Desk Analyst
IS Service Delivery, UK


e2v
106 Waterhouse Lane, Chelmsford, Essex, CM1 2QU, England
Tel:  +44 (0)1245 453698
  www.e2v.com

Consider the environment: do you really need to print this email?


-----Original Message-----
From: magicu-l@yahoogroups.com [mailto:magicu-l@yahoogroups.com] On
Behalf Of Rosie Coller
Sent: Monday, November 09, 2009 11:24 AM
To: magicu-l@yahoogroups.com
Subject: 1.8SP1a

You mean there is an SP1a already!

We've only just started using SP1. Will have to take a look at that.

Rosie

2009/11/7 Andy Jerison <ajerison@...>

>
>
> They did? I actually interrupted the download after I read the release
> notes
> and didn't find that one!
>
> Oh well....
>
> Andy J
>
> -----Original Message-----
> From: magicu-l@yahoogroups.com <magicu-l%40yahoogroups.com> [mailto:
> magicu-l@yahoogroups.com <magicu-l%40yahoogroups.com>] On Behalf
> Of Wes
> Sent: Friday, November 06, 2009 4:29 PM
> To: magicu-l@yahoogroups.com <magicu-l%40yahoogroups.com>
> Subject: [magicu-l] SP1a
>
> They fixed the transparent radio button issue!
>
> Wes

Sent by a member of the e2v group of companies. The parent company, e2v
technologies plc, is registered in England and Wales. Company number; 04439718.
Registered address; 106 Waterhouse Lane, Chelmsford, Essex, CM1 2QU, UK.  This
email and any attachments are confidential and meant solely for the use of the
intended recipient. If you are not the intended recipient and have received this
email in error, please notify us immediately by replying to the sender and then
deleting this copy and the reply from your system without further disclosing,
copying, distributing or using the e-mail or any attachment. Thank you for your
cooperation.
______________________________________________________
________________
This email has been scanned by the MessageLabs Email Security System.
For more information please visit http://www.messagelabs.com/email
______________________________________________________________________

You mean there is an SP1a already!

We've only just started using SP1. Will have to take a look at that.

Rosie

2009/11/7 Andy Jerison <ajerison@...>

>
>
> They did? I actually interrupted the download after I read the release
> notes
> and didn't find that one!
>
> Oh well....
>
> Andy J
>
> -----Original Message-----
> From: magicu-l@yahoogroups.com <magicu-l%40yahoogroups.com> [mailto:
> magicu-l@yahoogroups.com <magicu-l%40yahoogroups.com>] On Behalf
> Of Wes
> Sent: Friday, November 06, 2009 4:29 PM
> To: magicu-l@yahoogroups.com <magicu-l%40yahoogroups.com>
> Subject: [magicu-l] SP1a
>
> They fixed the transparent radio button issue!
>
> Wes
>
>
>


[Non-text portions of this message have been removed]

Charismatic Infotech has around 50+ man years of Magic Software design,
development, implementation and maintenance experience. Have extensively worked
on client/server, web based, distributed architecture.

Databases used with Magic: SQL server, Oracle, DB2, Informix, Sybase, Btrieve,
Pervasive SQL.

Magic implemented on Platforms: MS-DOS, SCO-UNIX, Windows NT 3.51, Windows 95,
Windows 3.11, Sun-Solaris, HP-900, AS/400

Here is the brief of the services we offer.
• Design
• Development
• Implementation
• Maintenance
• Redesign and development for different architecture. (Client/Server, Web,
Distributed).

Specializes in migration to
• Higher version
• Database
• Platform
• Architecture

Some of the possible models are:
• Onsite
• Onsite/Offsite (Offshore)
• Offsite (Offshore)
• Remote Access (High speed bandwidth is available inhouse)

For onsite/offsite model the ratio will be 10% Onsite & 90% Offsite.

Work model may be tailored to suit customer needs including onsite, offsite
services.

niranjan@...
niranjan.phadke@...

Hi Zvi,

What I mean is that a generic component can have generic 'call by Name'
operations already in place at pertinent places in its code. On these, the
'Public Program Name' and 'Cabinet file Name' properties would reference (by
expressions) columns of a generic component table that would hold the values of
the program to call.

Source code of the component would not be required. Just an online configuration
program that would enable maintaining programs public names and ecf(s)
locations.

BTW, that concept already works. Except for the recursive limitation everything
is in place for it to work. A protocol also needs to be established when it
comes to arguments, but that is acceptable at this point.

The idea is that the component process could have generic hooks to processes
that would reside in the host project where the owner of this host application
would have the ability to control further the execution of the combined solution
at this point.

It would open a door that is already there on a new set of generic
possibilities. However, currently there is that useless lock (recursion
limitation) on it.

Does that makes more sense to you at this point. Do you see that for these
generic hooks, the source code of the component would not be required?

Omar,


--- In magicu-l@yahoogroups.com, "zvidev" <magicusergroup@...> wrote:
>
> I agree with your concept that to benefit at it's max from components, those
obstacles exist, i just have a little comment about the paragragreph below.
>
> "Furthermore, it is also so when it comes to the recursion restriction that
MSE imposes on components. Again, the work around for this restriction implies
that one has to modify the source code of the component to be able to make it
work with its host application defeating again the sole purpose of being able to
generically share the component where recursion would be applicable."
>
> To use a component recursively - meaning to call back from the component
application a host application program etc.- you will always need the source
code and the ability to modify the source code. or maybe my understanding of
recursive is incorrect..
>
> Zvi
>
>
> --- In magicu-l@yahoogroups.com, "ola_ip" <omar.lamin@> wrote:
> >
> > Just to emphasize what Thomas is saying.
> >
> > The fact that a work around for a component issue implies that one needs the
source code to be able to modify both the component and the host application
means that the sole purpose of what components are for is broken and as such the
work around is not acceptable even if it is a work around.
> >
> > Using components to divide your application into several components is good.
But the true purpose should be that we can share components among us in the
community where each component is a black box that does a pre-defined numbers of
things. To take advantage of someone else's component, on should not have to
have access to this component source code and have to fix it for it to work with
his own application. Currently it is very likely to be so when it comes to font
and colors handling and as such is a deal breaker.
> >
> > Furthermore, it is also so when it comes to the recursion restriction that
MSE imposes on components. Again, the work around for this restriction implies
that one has to modify the source code of the component to be able to make it
work with its host application defeating again the sole purpose of being able to
generically share the component where recursion would be applicable.
> >
> > When, but only when MSE will come to realize the severity of these
deficiencies and fix them will components be able to significantly achieve their
true potential by enabling more and more component base add-on. Meanwhile, with
the source code or in circumstances that avoid colors and fonts issue (batch) as
well as the recursion limitation it is still possible to do things like for
those products I plug here and there in my posts here. But it really limits the
scope of what can be done to always have to keep such limitations in mind.
> >
> > Omar,
> >
> > --- In magicu-l@yahoogroups.com, "THOMAS" <ttitus_aerocontrols_com@> wrote:
> > >
> > >
> > >
> > > Thanks Ariel,
> > >
> > > I believe, the settings in the external files tab of the project
properties must always override the INI settings, but this works only if the ecf
is opened as main application.
> > >
> > > When the component was introduced by MSE as I recall, they are referred as
easily "distributable libraries".
> > >
> > > If I understand the endeavor correctly, to use a color/font sensitive
distributed component, in practice, you have to get the source code also. Then
you incorporate colors/fonts of the component to the colors/fonts of your main
project, change the color/font pointers inside the component's project
accordingly, rebuild the component cabinet file and then add the component to
your main project.
> > >
> > > The interesting part is, when you design form screens for the purpose of
running as main application, tool or component, the color/fonts files being in
effect are files specified in the project properties. When you run the
application as a component (or call by name), the engine completely ignores
these files. There is no way to associate these files back to the application at
runtime.
> > >
> > > Simply from the point of project management and considering what is in
stake, this should have been corrected soon after the introduction of
components. This high maintenance work around also makes the distribution of
version updates of components with colors, fonts, key board mapping etc very
difficult.
> > >
> > > Thanks
> > >
> > > Thomas Titus
> > >
> > >
> > > --- In magicu-l@yahoogroups.com, "Ariel Kotler" <prosto4u@> wrote:
> > > >
> > > > Hi ,
> > > >
> > > > Yes , it's true , my wish to have separated font/colors for each
applications opened in the "Wish List" since they released v9 with components .
> > > > Till now we can not share our components between us .
> > > >
> > > > Ariel Kotler
> > > >
> > > > --- In magicu-l@yahoogroups.com, "THOMAS" <ttitus_aerocontrols_com@>
wrote:
> > > > >
> > > > > Hi Folks,
> > > > > I have two projects; each uses its own application color file and
application font file (in the application properties).
> > > > > When I call Task2 by name of ecf2 from ecf1, it executes with color
and font file of ecf1 which displays a very tasteless screen. I contacted MSE
support. They advised me to insert eval
INIPUT(RuntimeApplicationColorDefinitionFile, color2.eng) and
INIPUT(RuntimeApplicationFontDefinitionFile, font2.eng) in the called program. I
used the evals  before the call, after the call, task prefix of task2, task
prefix of main2  etc and none worked. I send my sample projects to MSE and now
MSE agrees with me, saying "There is a wish opened for this to have it changed
however….".
> > > > > I am assuming anyone who worked with cross called ecfs should have
experienced this before and I like to know how you handled it.
> > > > > Thomas Titus
> > > > >
> > > >
> > >
> >
>