Re: What exactly is the scope of the On err call between 4D server & Client?
Hi Ben, On Fri, Jul 7, 2017 at 5:15 PM, Ben Kershaw via 4D_Tech < 4d_tech@lists.4d.com> wrote: > When a method with this property checked is executed, it runs in the > context of the servers “twin” process (which isn’t really a twin, since it > doesn’t really share anything with the client process). If you want your > OEC method called for an error in that method, you have to set that in the > server processes context. > > Do you have some sort of generic “init process” method that you call > whenever you start a new process? If so, create a new method with the > “Execute on Server” attribute checked, and then call THAT method from your > “init process” method; this will initialize the server processes context. > Well I do now... > I have a method called “InitServerProcess_sv” with the EoS property > checked, and I call it from my InitProcess method. It sets up the OEC > method and does a few other things. Has worked like a charm for years. > That is a great idea. I just recently wrote a method to run like that when I start processes to capture the process startup data Joaquin clued me into last week. That is a good place to put such a call. -- Kirk Brooks San Francisco, CA === *The only thing necessary for the triumph of evil is for good men to do nothing.* *- Edmund Burke* ** 4D Internet Users Group (4D iNUG) FAQ: http://lists.4d.com/faqnug.html Archive: http://lists.4d.com/archives.html Options: http://lists.4d.com/mailman/options/4d_tech Unsub: mailto:4d_tech-unsubscr...@lists.4d.com **
Re: What exactly is the scope of the On err call between 4D server & Client?
We always install new On err call on triggers. Alan Chan 4D iNug Technical <4d_tech@lists.4d.com> writes: >I don't remember...triggers? ** 4D Internet Users Group (4D iNUG) FAQ: http://lists.4d.com/faqnug.html Archive: http://lists.4d.com/archives.html Options: http://lists.4d.com/mailman/options/4d_tech Unsub: mailto:4d_tech-unsubscr...@lists.4d.com **
Re: What exactly is the scope of the On err call between 4D server & Client?
Ben, Thanks for these details, that's a huge help. Does anyone know if this is documented anywhere? For that matter, are the "twin" processes documented anywhere? I know that they were discussed at some Summits in the past but, well, I haven't been to a Summit in over a decade. Too. Darn. Far. Thanks. ** 4D Internet Users Group (4D iNUG) FAQ: http://lists.4d.com/faqnug.html Archive: http://lists.4d.com/archives.html Options: http://lists.4d.com/mailman/options/4d_tech Unsub: mailto:4d_tech-unsubscr...@lists.4d.com **
Re: What exactly is the scope of the On err call between 4D server & Client?
Kirk Brooks wrote: >Yeah, so I'm talking about a method with the Execute on server attribute (or >property - whatever) checked. When a method with this property checked is executed, it runs in the context of the servers “twin” process (which isn’t really a twin, since it doesn’t really share anything with the client process). If you want your OEC method called for an error in that method, you have to set that in the server processes context. Do you have some sort of generic “init process” method that you call whenever you start a new process? If so, create a new method with the “Execute on Server” attribute checked, and then call THAT method from your “init process” method; this will initialize the server processes context. I have a method called “InitServerProcess_sv” with the EoS property checked, and I call it from my InitProcess method. It sets up the OEC method and does a few other things. Has worked like a charm for years. Keep in mind, though, that unless you pass something back in a parameter (or handle it some other way), the method running on the client doesn’t know that there was an error during execution on the server - the error method was called on the server, so anything that happened during your error handler happened on the server and doesn’t change anything on the client. Regards, Ben ** 4D Internet Users Group (4D iNUG) FAQ: http://lists.4d.com/faqnug.html Archive: http://lists.4d.com/archives.html Options: http://lists.4d.com/mailman/options/4d_tech Unsub: mailto:4d_tech-unsubscr...@lists.4d.com **
Any issues with running v13.6 server and 13.5 clients?
Just wondering. Thanks -- Kirk Brooks San Francisco, CA === *The only thing necessary for the triumph of evil is for good men to do nothing.* *- Edmund Burke* ** 4D Internet Users Group (4D iNUG) FAQ: http://lists.4d.com/faqnug.html Archive: http://lists.4d.com/archives.html Options: http://lists.4d.com/mailman/options/4d_tech Unsub: mailto:4d_tech-unsubscr...@lists.4d.com **
Re: Request for tech note: Dave Terry's note on 2D Arrays
Speaking of 2D arrays, a couple of things to note:
* APPEND TO ARRAY doesn't work when you're trying to append an array. You
need to insert an element at the end and then use COPY ARRAY. I'm not sure
that the 2D array commands have been updated since V2 apart from type
compatability, and APPEND TO ARRAY is a more recent command.
* I misinterpreted this part of the docs on SORT ARRAY:
Note: You can sort the elements of a two-dimensional array (i.e.,
a2DArray{$vlThisElem}) but you cannot sort the two-dimensional array itself
(i.e., a2DArray).
I find myself thinking, "Bummer! I can't have a group of 1D and 2D arrays
and keep them together :(" Like a 1D array of item numbers and then 2D
arrays where you store how many of each item was ordered on the 1st, 2nd,
3rd, etc. Well, no, 4D works better than that. You can sort the 2D arrays
in the way you would hope, along with parallel 1D arrays. I just confirmed
this in a new database with some simple code:
ARRAY TEXT($check_1D_at;3)
ARRAY TEXT($check_2D_at;3;5)
$check_1D_at{1}:="Jan"
$check_1D_at{2}:="Feb"
$check_1D_at{3}:="Mar"
$check_2D_at{1}{1}:="1"
$check_2D_at{1}{2}:="1"
$check_2D_at{1}{3}:="1"
$check_2D_at{1}{4}:="1"
$check_2D_at{1}{5}:="1"
$check_2D_at{2}{1}:="2"
$check_2D_at{2}{2}:="2"
$check_2D_at{2}{3}:="2"
$check_2D_at{2}{4}:="2"
$check_2D_at{2}{5}:="2"
$check_2D_at{2}{1}:="3"
$check_2D_at{2}{2}:="3"
$check_2D_at{2}{3}:="3"
$check_2D_at{2}{4}:="3"
$check_2D_at{2}{5}:="3"
ALERT($check_1D_at{1}+" : "+$check_2D_at{1}{1})
SORT ARRAY($check_1D_at;$check_2D_at)
ALERT($check_1D_at{1}+" : "+$check_2D_at{1}{1})
On Sat, Jul 8, 2017 at 8:20 AM, David Adams wrote:
> Thanks everyone! It's great having a community with a long history. So,
> not only did I get what I was looking for, I got more. Just to put things
> in perspective, I was after a tech note from *before the PDF format even
> existed.* Back in the days when that second MB of RAM was going to solve
> all of your problems
>
> Anyway, the note that I remembered is "Two-Dimensional Arrays" by Dave
> Terry, TN #92-14. This note is from 25 years ago and it remains the best
> documentation on the subject. I have no idea why it's never been folded
> into the main docs, it should be. Bart Davis said that he found it on one
> of the 4D FTP sites, so it's out there somewhere.
>
> Mike Goldsmith sent me a couple of articles that he wrote on cross-tabs
> some time back. Cross-tabs are great! They're also a sensible use of 2D
> arrays. I used to love the cross-tab editor in the original 4D Calc. (I
> think cross-tabs are often called 'pivot reports' or 'summarizing across'
> in other parts of the world.)
>
> Dennis Gallagher sent me something totally unexpected. I think that it is
> a V1 tech note, TN #92 from 1988 titled “Multi-Dimensional Arrays in 4th
> Dimension” by Djundi Karjadi & Joshua Wachs and later modified by Dave
> Terry & Ron Dell’Aquila. These guys were *emulating* 2D arrays using 1D
> arrays. I'm not worthy! I'm going to check it out as it's probably got some
> good ideas about packing.
>
> Thanks everyone!
>
**
4D Internet Users Group (4D iNUG)
FAQ: http://lists.4d.com/faqnug.html
Archive: http://lists.4d.com/archives.html
Options: http://lists.4d.com/mailman/options/4d_tech
Unsub: mailto:4d_tech-unsubscr...@lists.4d.com
**
Re: What exactly is the scope of the On err call between 4D server & Client?
> And this is really my question - is this a bug or am I missing something? Try it compiled and interpreted and see what you find - it might work compiled. Either way, if the behavior is undefined it would be great to get it sorted out. Error dialogs popping up on the 4D Server machine? Not good. I guess you can file a case on-line as a US Partner, or there are the forums in France? I think that you know the procedures better than I do. Thanks for pointing this out, it's important! Please let us know what you learn. P.S. If someone actually knows or if it's in the docs and we've missed it, please chime in. P.P.S. Kirk, I guess if you can reproduce it in the latest R release, you could report it as a bug in that release... ** 4D Internet Users Group (4D iNUG) FAQ: http://lists.4d.com/faqnug.html Archive: http://lists.4d.com/archives.html Options: http://lists.4d.com/mailman/options/4d_tech Unsub: mailto:4d_tech-unsubscr...@lists.4d.com **
Re: What exactly is the scope of the On err call between 4D server & Client?
David, I was hoping you might weigh in here. Yeah, so I'm talking about a method with the Execute on server attribute (or property - whatever) checked. So the situation is a client process is churning away and passes some arrays off to the EOS method for processing on the server (looking things up, grinding away). It's not running on the client (that's the point) but it's not entirely native to the server either. I'm hearing the guy from the Princess Bride here... "If he's not completely dead that means he's a little bit alive." And this is really my question - is this a bug or am I missing something? On Fri, Jul 7, 2017 at 3:43 PM, David Adams via 4D_Tech < 4d_tech@lists.4d.com> wrote: > I'm not sure I'm following the basic setup because 'execute on server' can > mean two things in 4D, can you clarify? > > If you're running a method with the 'execute on server' flag set then, huh. > What a great question! I hope this is what you mean because, yeah, that's a > nice detail. I don't know the answer and I won't be shocked to learn that > this is a gap in the ON ERR CALL coverage in 4D. (4D seems to approach > error handling as something tacked on top of things - there isn't a > predictable or systemic scheme for it. At least it doesn't look like there > is.) Does anyone know? I've got it on my list to do a deep dive on the > Execute on server flag, but, well, I've got a long list. > > -- Kirk Brooks San Francisco, CA === *The only thing necessary for the triumph of evil is for good men to do nothing.* *- Edmund Burke* ** 4D Internet Users Group (4D iNUG) FAQ: http://lists.4d.com/faqnug.html Archive: http://lists.4d.com/archives.html Options: http://lists.4d.com/mailman/options/4d_tech Unsub: mailto:4d_tech-unsubscr...@lists.4d.com **
Re: What exactly is the scope of the On err call between 4D server & Client?
I'm not sure I'm following the basic setup because 'execute on server' can
mean two things in 4D, can you clarify?
Just to clarify my question...
$id:=Execute on server ("method name;etc)
If you're calling the Execute on server *command*, then you're spawning a
new process on 4D Server that runs there. No, it won't get the ON ERR CALL
you set up startup. The 'global' ON ERR CALL on 4D Server is for the
application process and..I don't remember...triggers? It used to depend on
compiled/interpreted and may depend on version. I don't remember off-hand
and would want to verify with your version anyway.
If you're running a method with the 'execute on server' flag set then, huh.
What a great question! I hope this is what you mean because, yeah, that's a
nice detail. I don't know the answer and I won't be shocked to learn that
this is a gap in the ON ERR CALL coverage in 4D. (4D seems to approach
error handling as something tacked on top of things - there isn't a
predictable or systemic scheme for it. At least it doesn't look like there
is.) Does anyone know? I've got it on my list to do a deep dive on the
Execute on server flag, but, well, I've got a long list.
**
4D Internet Users Group (4D iNUG)
FAQ: http://lists.4d.com/faqnug.html
Archive: http://lists.4d.com/archives.html
Options: http://lists.4d.com/mailman/options/4d_tech
Unsub: mailto:4d_tech-unsubscr...@lists.4d.com
**
Re: Request for tech note: Dave Terry's note on 2D Arrays
Thanks everyone! It's great having a community with a long history. So, not only did I get what I was looking for, I got more. Just to put things in perspective, I was after a tech note from *before the PDF format even existed.* Back in the days when that second MB of RAM was going to solve all of your problems Anyway, the note that I remembered is "Two-Dimensional Arrays" by Dave Terry, TN #92-14. This note is from 25 years ago and it remains the best documentation on the subject. I have no idea why it's never been folded into the main docs, it should be. Bart Davis said that he found it on one of the 4D FTP sites, so it's out there somewhere. Mike Goldsmith sent me a couple of articles that he wrote on cross-tabs some time back. Cross-tabs are great! They're also a sensible use of 2D arrays. I used to love the cross-tab editor in the original 4D Calc. (I think cross-tabs are often called 'pivot reports' or 'summarizing across' in other parts of the world.) Dennis Gallagher sent me something totally unexpected. I think that it is a V1 tech note, TN #92 from 1988 titled “Multi-Dimensional Arrays in 4th Dimension” by Djundi Karjadi & Joshua Wachs and later modified by Dave Terry & Ron Dell’Aquila. These guys were *emulating* 2D arrays using 1D arrays. I'm not worthy! I'm going to check it out as it's probably got some good ideas about packing. Thanks everyone! ** 4D Internet Users Group (4D iNUG) FAQ: http://lists.4d.com/faqnug.html Archive: http://lists.4d.com/archives.html Options: http://lists.4d.com/mailman/options/4d_tech Unsub: mailto:4d_tech-unsubscr...@lists.4d.com **
Re: What exactly is the scope of the On err call between 4D server & Client?
Jody, Right, but what I'm seeing is a client side process which has an error handler installed. Running a method with EOS an error is blocking the server with the 4D native error handler. Even though the server has an error handler installed. This is following the what's in the docs: "ON ERR CALL is usually placed in the startup database method of an application, in order to handle errors for this application." Seems to suggest that would be the method called absent some specific method and instead of the native 4D error handler. So that's why I expect the server to use the handler I install there. It seems like this is something specific to methods running EOS because I've installed an error handler on both the server and the client process but the 4D error handler is winding up with the call. On Fri, Jul 7, 2017 at 2:52 PM, Jody Bevan via 4D_Tech <4d_tech@lists.4d.com > wrote: > On Error Call is process based. Therefore each process needs to call it. > > -- Kirk Brooks San Francisco, CA === *The only thing necessary for the triumph of evil is for good men to do nothing.* *- Edmund Burke* ** 4D Internet Users Group (4D iNUG) FAQ: http://lists.4d.com/faqnug.html Archive: http://lists.4d.com/archives.html Options: http://lists.4d.com/mailman/options/4d_tech Unsub: mailto:4d_tech-unsubscr...@lists.4d.com **
Re: What exactly is the scope of the On err call between 4D server & Client?
On Error Call is process based. Therefore each process needs to call it.
Sincerely
Jody Bevan
> On Jul 7, 2017, at 16:45, Kirk Brooks via 4D_Tech <4d_tech@lists.4d.com>
> wrote:
>
> I think I know the answer to this but think I'll ask anyway.
>
> When my server starts up in installs an error handler, ON ERR
> CALL("err_log_only"). As the name implies it simply logs errors and keeps
> churning.
>
> It doesn't seem to be called when a method called by Execute on server is
> running. At least I'm occasionally seeing errors on the server from a
> couple of methods. The thing about this is the process calling these
> methods has its own error handler installed as well which is not being
> called.
>
> So, is this a bug or is it the case that methods running as EOS have to
> explicitly call their own error handler? And if that's the case, as I think
> about it, is this a situation like the rest of process vars and the 'twin'
> process where it's the same but different. Ie, I can have process vars with
> the same names on both the client and server side but these vars can have
> totally different contents. Is it therefore the case with the Error
> handler? Installing a given handler on the client process doesn't affect
> the server side. If that's the case it sounds like I should be adding an
> EOS method to my process init to install a handler on the server twin.
>
> Anyone else encountered this? What did you do?
>
> --
> Kirk Brooks
> San Francisco, CA
> ===
>
> *The only thing necessary for the triumph of evil is for good men to do
> nothing.*
>
> *- Edmund Burke*
> **
> 4D Internet Users Group (4D iNUG)
> FAQ: http://lists.4d.com/faqnug.html
> Archive: http://lists.4d.com/archives.html
> Options: http://lists.4d.com/mailman/options/4d_tech
> Unsub: mailto:4d_tech-unsubscr...@lists.4d.com
> **
**
4D Internet Users Group (4D iNUG)
FAQ: http://lists.4d.com/faqnug.html
Archive: http://lists.4d.com/archives.html
Options: http://lists.4d.com/mailman/options/4d_tech
Unsub: mailto:4d_tech-unsubscr...@lists.4d.com
**
What exactly is the scope of the On err call between 4D server & Client?
I think I know the answer to this but think I'll ask anyway.
When my server starts up in installs an error handler, ON ERR
CALL("err_log_only"). As the name implies it simply logs errors and keeps
churning.
It doesn't seem to be called when a method called by Execute on server is
running. At least I'm occasionally seeing errors on the server from a
couple of methods. The thing about this is the process calling these
methods has its own error handler installed as well which is not being
called.
So, is this a bug or is it the case that methods running as EOS have to
explicitly call their own error handler? And if that's the case, as I think
about it, is this a situation like the rest of process vars and the 'twin'
process where it's the same but different. Ie, I can have process vars with
the same names on both the client and server side but these vars can have
totally different contents. Is it therefore the case with the Error
handler? Installing a given handler on the client process doesn't affect
the server side. If that's the case it sounds like I should be adding an
EOS method to my process init to install a handler on the server twin.
Anyone else encountered this? What did you do?
--
Kirk Brooks
San Francisco, CA
===
*The only thing necessary for the triumph of evil is for good men to do
nothing.*
*- Edmund Burke*
**
4D Internet Users Group (4D iNUG)
FAQ: http://lists.4d.com/faqnug.html
Archive: http://lists.4d.com/archives.html
Options: http://lists.4d.com/mailman/options/4d_tech
Unsub: mailto:4d_tech-unsubscr...@lists.4d.com
**
How to clear the Menu Bar
What is the best way to Clear or reset the menu bar to prevent multiple menus ? I do not see a Clear or Reset Menu Bar option. I am getting multiple menus displayed when more than one form is open with different Associated Menu Bars. Case: The Form for Orders is open with Menu #3 and I open a modal dialog for printing which has Menu #6 When I close the modal dialog for printing I now have both Menu # 3 and Menu # 6 displayed. Thanks, Jim Medlen Computer & Information Systems Functional Devices, Inc. j.med...@functionaldevices.com phone (765) 883-5538 x 428 fax (765) 883-4262 http://www.functionaldevices.com This email was transmitted on 100 percent recycled electrons ** 4D Internet Users Group (4D iNUG) FAQ: http://lists.4d.com/faqnug.html Archive: http://lists.4d.com/archives.html Options: http://lists.4d.com/mailman/options/4d_tech Unsub: mailto:4d_tech-unsubscr...@lists.4d.com **
Request for tech note: Dave Terry's note on 2D Arrays
I'm working heavily with 2D arrays right now for the first time in ages, and (re)discovered taht the docs are thin. Years (decades) ago Dave Terry wrote a tech note called something like "2D Arrays' of around 4 pages. The old notes aren't in the KB and I don't seem to have a copy of this one in my own archives. Does anyone have a copy? If I remember correctly, it covered the mechanics really well. Thanks ** 4D Internet Users Group (4D iNUG) FAQ: http://lists.4d.com/faqnug.html Archive: http://lists.4d.com/archives.html Options: http://lists.4d.com/mailman/options/4d_tech Unsub: mailto:4d_tech-unsubscr...@lists.4d.com **
Re: revelation! - Choose
Choose — Caution IMO the documentation is slightly faulty in its example for the “integer” use of this command. The issue is that Choose returns the “null” value for the type being used if the integer value is not in the range covered. From the Documentation **************** ***** Example 2 Here is an example of the typical use of this command with a Number type criterion: vStatus:=Choose([Person]Status;"Single";"Married";"Widowed";"Divorced") This code is strictly equivalent to: Case of :([Person]Status=0) vStatus:="Single" :([Person]Status=1) vStatus:="Married" :([Person]Status=2) vStatus:="Widowed" :([Person]Status=3) vStatus:="Divorced" End case **************** *** This is not "strictly" correct. You need to add an Else to the Case statement to get complete equivalency The null value of the type (in this case "") is returned if the integer value is out of range (in this case the range is 0-3) Case of :([Person]Status=0) vStatus:="Single" :([Person]Status=1) vStatus:="Married" :([Person]Status=2) vStatus:="Widowed" :([Person]Status=3) vStatus:="Divorced" Else vStatus:="" End case So if vStatus had some other value going into the Choose statement, it is going to loose that value even if there was no matching criteria. > On Jul 6, 2017, at 9:45 AM, Chip Scheide via 4D_Tech <4d_tech@lists.4d.com> > wrote: > > I found out about a command I had not known about > Choose > > I love this command > if anyone does not know about it - read it is so nice to do in one > line of code what other wee takes at least5 ** 4D Internet Users Group (4D iNUG) FAQ: http://lists.4d.com/faqnug.html Archive: http://lists.4d.com/archives.html Options: http://lists.4d.com/mailman/options/4d_tech Unsub: mailto:4d_tech-unsubscr...@lists.4d.com **
