Re: Why is everything in the IDE so tiny?

[J. Landman Gay via use-livecode]

On 11/1/17 12:04 AM, AcidJazz via use-livecode wrote:

I'm using a*much*  larger screen (34"
diameter), and find that virtually everything in the IDE, but especially the
icons in the object/card/stack inspectors to be freakishly small.


At the top right of the inspector, tool palette, and project browser, 
there is an icon that looks like a sunburst (or maybe a gear with a lot 
of points.) That opens a menu. Choose "Header/Footer Size" and choose 
"Largest". The icons still aren't very big on a high-resolution display 
but it's much better than the default.



why is everything in the object inspector still buried in tabs?


While you were away, the tabs became a popdown menu. It was that way for 
years. People complained it took too many clicks to change categories so 
it was recently changed back to tabs. You missed your opening I guess.



It would be so much easier to
quickly check and make changes if the inspectors were a long column with few
if any tabs.  You'd just scroll through it, but with sections that you could
expand or contract.


An accordian control would be a nice compromise. I like that idea.

--
Jacqueline Landman Gay | jac...@hyperactivesw.com
HyperActive Software   | http://www.hyperactivesw.com

___
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription 
preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode

Re: OT: ...and everybody hates Perl...

[Kaveh Bazargan via use-livecode]

And nobody's heard of livecode. :-(

On Wed, 1 Nov 2017 at 04:26, Mark Wieder via use-livecode <
use-livecode@lists.runrev.com> wrote:

> https://boingboing.net/2017/10/31/perl-is-the-most-hated-program.html
> https://stackoverflow.blog/2017/10/31/disliked-programming-languages/?cb=1
>
> --
>   Mark Wieder
>   ahsoftw...@gmail.com
>
> ___
> use-livecode mailing list
> use-livecode@lists.runrev.com
> Please visit this url to subscribe, unsubscribe and manage your
> subscription preferences:
> http://lists.runrev.com/mailman/listinfo/use-livecode
>
-- 
Sent from MetroMail
___
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription 
preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode

Why is everything in the IDE so tiny?

[AcidJazz via use-livecode]

I used Livecode a lot about 6 years ago, and have recently started playing
around again.  I know a lot of features have been added under the hood, but
the IDE is really disappointing.  I'm using a *much* larger screen (34"
diameter), and find that virtually everything in the IDE, but especially the
icons in the object/card/stack inspectors to be freakishly small.  Yes, my
screen has a high resolution, but every other application is perfectly
readable on my screen.  Why not LiveCode?   I do understand that some people
use laptops with little screen real estate, but why can't this be adjusted
for folks with big screens and high resolution?  And while I'm kvetching,
why is everything in the object inspector still buried in tabs?   There
don't appear to be any obvious key commands to quickly switch between them. 
We all have scroll wheels and track pads.  It would be so much easier to
quickly check and make changes if the inspectors were a long column with few
if any tabs.  You'd just scroll through it, but with sections that you could
expand or contract.  I bought a monster screen so I could see everything at
once, but Livecode is damned if it's going to let me do that.  I'm not
normally negative, but it's 2017, and the IDE looks like it was created in
the 1980s and is actually less usable than when I used  it 6 years ago.  Am
I missing a magic button somewhere that makes this thing less clunky?  
Surely, I'm not the only one wondering these things.   Am I? 

- Mark

   



--
Sent from: 
http://runtime-revolution.278305.n4.nabble.com/Revolution-User-f278306.html

___
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription 
preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode

OT: ...and everybody hates Perl...

[Mark Wieder via use-livecode]

https://boingboing.net/2017/10/31/perl-is-the-most-hated-program.html
https://stackoverflow.blog/2017/10/31/disliked-programming-languages/?cb=1

--
 Mark Wieder
 ahsoftw...@gmail.com

___
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription 
preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode

Re: Remote Debugger Frustrations

[Monte Goulding via use-livecode]

> On 1 Nov 2017, at 1:45 pm, J. Landman Gay via use-livecode 
>  wrote:
> 
>> Once you hit a breakpoint you should be able to change breakpoints in the 
>> open debugger.
> 
> This I didn't know, I wasn't able to do that. To verify: I have to be stopped 
> at a breakpoint first before changing or adding one? I didn't do that when I 
> last tried.

Yes, you should be able to manipulate red dot breakpoints once the debugger is 
open. If you aren’t stopped at a breakpoint or execution error then you are 
probably just looking at your local script and therefore any changes to 
breakpoints won’t go over the wire.

Cheers

Monte
___
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription 
preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode

Re: Remote Debugger Frustrations

[J. Landman Gay via use-livecode]

On 10/31/17 4:14 PM, Monte Goulding via use-livecode wrote:

Once you hit a breakpoint you should be able to change breakpoints in the open 
debugger.


This I didn't know, I wasn't able to do that. To verify: I have to be 
stopped at a breakpoint first before changing or adding one? I didn't do 
that when I last tried.


--
Jacqueline Landman Gay | jac...@hyperactivesw.com
HyperActive Software   | http://www.hyperactivesw.com

___
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription 
preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode

Re: Strange permissions issues calling 'shell' in Windows

[Monte Goulding via use-livecode]

> On 31 Oct 2017, at 9:40 pm, Ben Rubinstein via use-livecode 
>  wrote:
> 
> Is there some way that calling 'shell' enters a context with less permission 
> than the app which calls it? And is there something I can do about this?

LiveCode’s shell function on windows is basically telling cmd.exe to run the 
command. I have run into permissions issues before on school networks that have 
policies that block the use of the command prompt. In that case moving to open 
process with a read from process loop worked around the policy. You might try 
that. It could be in the scheduled task environment there is something missing 
that cmd.exe needs.

Cheers

Monte
___
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription 
preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode

Re: Remote Debugger Frustrations

[Monte Goulding via use-livecode]

> On 31 Oct 2017, at 9:15 pm, Peter Reid via use-livecode 
>  wrote:
> 
> How are people getting on with the remote debugger facility? I'm finding it 
> very useful but also frustrating for the following reasons:
> 
> 1. It's SLOW. I'm running LC 9 on an iMac with a 4GHz quad i7 processor and 
> 24GB RAM connected by USB to an Amazon Fire 7 tablet, yet each step through 
> the script in the debugger takes several seconds.

The remote debugger diverts calls to the debugger library to go over a socket 
to the remote process. It will always be much slower than debugging in the 
engine. If you have lots of variables or a lot of data in them then all of the 
values need to be requested and updated for each step.
> 
> 2. It's difficult to stop to make changes. Clicking the stop debugging button 
> often doesn't work or takes a long time to react and even when apparently 
> stopped it's still making everything slow.

If you have a good recipe for this then it would be good to open a report about 
it. It shouldn’t take more than a few seconds to stop the debugger.
> 
> 3. Often it steps through the script as it was just before the latest change, 
> i.e. it doesn't take into account the latest script edits.

The script of being stepped through is copied over from the remote process so 
it will only show you the script as it was when you clicked the Test button.
> 
> 4. It can be difficult to change break points during remote debugging.

Can you give more detail? Once you hit a breakpoint you should be able to 
change breakpoints in the open debugger.
> 
> 5. The only way that guarantees a clean run is to quit LC and start again.

What do you mean by clean run here?
> 
> 6. A partial reset can be achieved by the following:
>   a. having clicked 'stop debugging'...
>   b. open & close Standalone Application Settings
>   c. click Test, which prompts for Save
>   d. click Save and the remote build & install starts
>   e. however, the above doesn't always recognise script changes.

I think the above is related to #3. Perhaps a misunderstanding about what 
happens once you edit the local stack scripts during a remote debugger session?

Cheers

Monte
___
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription 
preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode

Re: Go card "xyz"

[Bob Sneidar via use-livecode]

No, and it can't work that way because as you have foreseen, that would wreak 
havoc on the message hierarchy. Some messages, a very few get processed as a 
high priority to resolve some issue they would have had if they went through 
the regular hierarchy (someone correct me if I am wrong) but these are 
typically not the ones you would read about in the dictionary. 

Bob S


> On Oct 31, 2017, at 13:18 , Pyyhtiä Christer via use-livecode 
>  wrote:
> 
> "When in a script you issue a command
> 
>   go card "xyz"
> 
> does the next command line of the script get into execution before the "go" 
> has been completed?

___
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription 
preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode

Re: Relative Paths in Property Inspector

[Bob Sneidar via use-livecode]

This is called a Watch Folder approach and any cloud based system that uses it 
(nextcloud does sort of, icloud does as well) will work the same way. Pretty 
slick. 

Bob S


> On Oct 31, 2017, at 11:43 , J. Landman Gay via use-livecode 
>  wrote:
> 
> On 10/31/17 11:57 AM, Mike Kerner via use-livecode wrote:
>> no I'm referring to the usernames on the machine.  I'm using google drive
>> (backup and sync and file stream).  I use dropbox for app data storage and
>> transfer.
> 
> Here is my setup: I created a folder in Dropbox named "Application Support". 
> I put the LC folder in there as well as BBEdit's preferences and a few other 
> things I want to share among machines.
> 
> On Mac #1, the path in the LC prefs is:
> /Users/Username1/Dropbox/Application Support/My LiveCode
> 
> Mac #2:
> /Users/username2/Dropbox/Application Support/My LiveCode
> 
> And so forth. Dropbox keeps the folders in sync and because it's a local 
> folder on each drive, LC has no problem accessing it.
> 
> -- 
> Jacqueline Landman Gay | jac...@hyperactivesw.com
> HyperActive Software   | http://www.hyperactivesw.com


___
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription 
preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode

Re: Go card "xyz"

[Pyyhtiä Christer via use-livecode]

Thank you for a good set of answers!

Now, here is the question asked, the heck I could not express it correctly the 
first time:

"When in a script you issue a command

go card "xyz"

does the next command line of the script get into execution before the "go" has 
been completed?

Is it comparable to the "move" command, where this is true, with the additional 
statement "without waiting"?

Is it possible to find a delay in the "go" command if you for example go from 
card 2 to card 28 and there are many objects in the cards in between?

The fact is that the behaviour varies, same code at different times in some 
cases looks like an action is tried to be done for the card going out."

-

I should have added that in the development / IDE environmen,t yes it behaves 
as nicely as Jeremy Clarkson - no, much more nicely.   But.

But this is the question of LC (8.1.7) running in the target device (good old 
Samsung Galaxy S4 with 1.6GHz octa-core CPU (still beats some of the devices, 
which have come out after it)).  There isn't really a chance of debugging in 
real time how the sequences of messages vs the code execution will sequence / 
synch.  If I add any commands on-top-of-real-time-execution to debug or ask to 
process logging message timings the real-time execution does not really exist 
any more.

The question really is looking into the device execution environment.  

The interpretation is that if a 'go to card "xyz"' is issued, it seems to be 
that the execution continues before all these messages closing the card & 
objects and opening the new card takes place, and possible references to the 
expected new card are causing a stop of execution of the handler asking for the 
go and shoveling new commands into the CPUs..

Thus, my interpretation is that the guiding how to code with LiveCode (or any 
other event-driven middleware) is missing a key point.

There are the following possibilities to manage the problem.

1.  wait x ticks / milliSeconds after the go card command; the problem is that 
if you target to devices, you need to be able to measure what is the best value 
or your app misbehaves or is a laggard;  somewhat ? ...

2.  LC to enhance the "go card" with something like "wait until done";  does 
not sound right, either;

3.  Instead of in a handler issuing 'go card "xyz"' use 'send "go card xyz of 
this stack" (probably need to do it by building a do statement) and put any 
action behind that new card script handler "openCard", which initiates any 
action you want to perform with any objects on that card.  This is probably the 
safest path.  Also, with such a way you would avoid the continuous path of 
handlers chained by continuous handler calls, and break the chain of returns 
handler by handler.

Is there any sense behind this thinking?


Christer Pyyhtiä
MindCrea Ltd
Mobile: +358-400-410216
Skype:  christerp1
chris...@mindcrea.com

___
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription 
preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode

Re: Go card "xyz"

[dunbarx via use-livecode]

"When you go to another card, lots of things happen before and after the
go...

What Bob said.

I assumed you knew that lots of messages are generated in the case you
described, all the comings and goings, and were only asking about the order
of events between the two lines of code in the initial handler, the one that
sent you off, and the next line.

Craig



--
Sent from: 
http://runtime-revolution.278305.n4.nabble.com/Revolution-User-f278306.html

___
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription 
preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode

Re: Go card "xyz"

[Richard Gaskin via use-livecode]

Bob Sneidar wrote:

> The best way to tell what is going on is to turn on message watching

Agreed, and if you find LC's Message Watcher as unwieldy as I do you may 
enjoy this more flexible, tailorable tool that produces more readable 
and useful logs of messages during execution:




--
 Richard Gaskin
 Fourth World Systems
 Software Design and Development for the Desktop, Mobile, and the Web
 
 ambassa...@fourthworld.comhttp://www.FourthWorld.com

___
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription 
preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode

Re: Remote Debugger Frustrations

[J. Landman Gay via use-livecode]

On 10/31/17 5:15 AM, Peter Reid via use-livecode wrote:

3. Often it steps through the script as it was just before the latest change, 
i.e. it doesn't take into account the latest script edits.

4. It can be difficult to change break points during remote debugging.


You need to recompile/rebuild the app every time you make a change. 
Built apps can't be modified, and that includes the location of the 
breakpoints. That does slow down debuging because you can't move or 
create breakpoints without doing a rebuild, and the scripts are of 
course permanent until the app is recreated.


--
Jacqueline Landman Gay | jac...@hyperactivesw.com
HyperActive Software   | http://www.hyperactivesw.com

___
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription 
preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode

Re: Relative Paths in Property Inspector

[J. Landman Gay via use-livecode]

On 10/31/17 11:57 AM, Mike Kerner via use-livecode wrote:

no I'm referring to the usernames on the machine.  I'm using google drive
(backup and sync and file stream).  I use dropbox for app data storage and
transfer.


Here is my setup: I created a folder in Dropbox named "Application 
Support". I put the LC folder in there as well as BBEdit's preferences 
and a few other things I want to share among machines.


On Mac #1, the path in the LC prefs is:
/Users/Username1/Dropbox/Application Support/My LiveCode

Mac #2:
/Users/username2/Dropbox/Application Support/My LiveCode

And so forth. Dropbox keeps the folders in sync and because it's a local 
folder on each drive, LC has no problem accessing it.


--
Jacqueline Landman Gay | jac...@hyperactivesw.com
HyperActive Software   | http://www.hyperactivesw.com

___
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription 
preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode

Re: Go card "xyz"

[Bob Sneidar via use-livecode]

When you go to another card, lots of things happen before and after the go. By 
way of example, Closecard gets sent to the curent card, but before that close 
field gets sent to any field that you are in that has been edited, etc. Before 
the card opens, it gets a preOpenCard. So don't think that by going to a card, 
that command gets executed first, then all the other stuff happens. That isn't 
how it works. 

The best way to tell what is going on is to turn on message watching, then set 
a breakpoint right at the first and then step through every line of code. Then 
you can review the message watcher and you can see all kinds of messages are 
flying around that you were unaware of. 

Bob S


> On Oct 31, 2017, at 06:09 , Pyyhtiä Christer via use-livecode 
>  wrote:
> 
> When in a script you issue a command
> 
>   go card "xyz"
> 
> does the next command line of the script get into execution before the "go" 
> has been completed?
> 
> Is it comparable to the "move" command, where this is true, with the 
> additional statement "without waiting"?
> 
> Is it possible to find a delay in the "go" command if you for example go from 
> card 2 to card 28 and there are many objects in the cards in between?
> 
> The fact is that the behaviour varies, same code at different times in some 
> cases looks like an action is tried to be done for the card going out.
> 
> 
> Christer Pyyhtiä
> chris...@mindcrea.com
> 
> 
> ___
> use-livecode mailing list
> use-livecode@lists.runrev.com
> Please visit this url to subscribe, unsubscribe and manage your subscription 
> preferences:
> http://lists.runrev.com/mailman/listinfo/use-livecode

___
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription 
preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode

Re: I Hate Full Screen!

[Brian Milby via use-livecode]

And I’m guessing that you do want to allow the stack to be manually resized
(since turning that off will hide the green button).
On Tue, Oct 31, 2017 at 12:08 PM Roger Guay via use-livecode <
use-livecode@lists.runrev.com> wrote:

> Well, that helps but I still hate Full Screen!
>
>
>
> > On Oct 31, 2017, at 7:45 AM, Bob Sneidar via use-livecode <
> use-livecode@lists.runrev.com> wrote:
> >
> > If you option-click the zoom on a Mac, it uses the old zoom where the
> menus are still visible.
> >
> > Bob S
> >
> >
> >> On Oct 30, 2017, at 18:03 , Brian Milby via use-livecode <
> use-livecode@lists.runrev.com> wrote:
> >>
> >> I figured out how to disable full screen and replace it with the old
> zoom,
> >> but that isn’t really that much of an improvement. Something in the SDK
> has
> >> changed since the code didn’t change between 7 and 8 (at least the
> engine
> >> code that sets up the window).
> >
> > ___
> > use-livecode mailing list
> > use-livecode@lists.runrev.com
> > Please visit this url to subscribe, unsubscribe and manage your
> subscription preferences:
> > http://lists.runrev.com/mailman/listinfo/use-livecode
>
> ___
> use-livecode mailing list
> use-livecode@lists.runrev.com
> Please visit this url to subscribe, unsubscribe and manage your
> subscription preferences:
> http://lists.runrev.com/mailman/listinfo/use-livecode
___
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription 
preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode

Re: I Hate Full Screen!

[Roger Guay via use-livecode]

Well, that helps but I still hate Full Screen!



> On Oct 31, 2017, at 7:45 AM, Bob Sneidar via use-livecode 
>  wrote:
> 
> If you option-click the zoom on a Mac, it uses the old zoom where the menus 
> are still visible. 
> 
> Bob S
> 
> 
>> On Oct 30, 2017, at 18:03 , Brian Milby via use-livecode 
>>  wrote:
>> 
>> I figured out how to disable full screen and replace it with the old zoom,
>> but that isn’t really that much of an improvement. Something in the SDK has
>> changed since the code didn’t change between 7 and 8 (at least the engine
>> code that sets up the window).
> 
> ___
> use-livecode mailing list
> use-livecode@lists.runrev.com
> Please visit this url to subscribe, unsubscribe and manage your subscription 
> preferences:
> http://lists.runrev.com/mailman/listinfo/use-livecode

___
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription 
preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode

Re: Relative Paths in Property Inspector

[Mike Kerner via use-livecode]

no I'm referring to the usernames on the machine.  I'm using google drive
(backup and sync and file stream).  I use dropbox for app data storage and
transfer.

On Tue, Oct 31, 2017 at 11:17 AM, Mark Wieder via use-livecode <
use-livecode@lists.runrev.com> wrote:

> On 10/30/2017 07:55 PM, Mike Kerner via use-livecode wrote:
>
>> So what do you have for your paths?  Are your usernames the same?  For me,
>> I run into the issue because my usernames aren't the same on every box, so
>> the absolute path varies.
>>
>
> Are you referring to the *dropbox* usernames being different? If that's
> the case, you can share a folder between dropbox users and still use a
> common repository.
>
>
> --
>  Mark Wieder
>  ahsoftw...@gmail.com
>
> ___
> use-livecode mailing list
> use-livecode@lists.runrev.com
> Please visit this url to subscribe, unsubscribe and manage your
> subscription preferences:
> http://lists.runrev.com/mailman/listinfo/use-livecode
>



-- 
On the first day, God created the heavens and the Earth
On the second day, God created the oceans.
On the third day, God put the animals on hold for a few hours,
   and did a little diving.
And God said, "This is good."
___
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription 
preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode

Re: Strange permissions issues calling 'shell' in Windows

[Mike Bonner via use-livecode]

nvm my response, my earlier message poofed due to size. *sigh* here goes
again...

one thing you might try is to look at the properties for the scheduled
job.  You obviously have it set to "run whether user is logged in or not"
but i'm wondering if "do not store password" box is checked. If it is,
might test with it unchecked.  there is also a box for "use highest
permissions" that you might try if the first idea fails.

disclaimer:  I have no idea how this would affect machine security.

Also, there is a button that allows you to change user or group, I don't
know how to do so, but if you can give a group permission to run pslist,
and then set the task to that group, that might be a way also.

Here is the screen i'm looking at for all this...
https://www.dropbox.com/s/0chbhqcn8hi4cu1/image%20%281%29.png?dl=0

On Tue, Oct 31, 2017 at 10:32 AM, Mike Bonner  wrote:

> i'm on 10, no clue what earlier versions would look like.
>
> On Tue, Oct 31, 2017 at 10:23 AM, Bob Sneidar via use-livecode <
> use-livecode@lists.runrev.com> wrote:
>
>> Windows 10?
>>
>> Bob S
>>
>>
>>
>> ___
>> use-livecode mailing list
>> use-livecode@lists.runrev.com
>> Please visit this url to subscribe, unsubscribe and manage your
>> subscription preferences:
>> http://lists.runrev.com/mailman/listinfo/use-livecode
>>
>
>
___
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription 
preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode

Re: Strange permissions issues calling 'shell' in Windows

[Mike Bonner via use-livecode]

i'm on 10, no clue what earlier versions would look like.

On Tue, Oct 31, 2017 at 10:23 AM, Bob Sneidar via use-livecode <
use-livecode@lists.runrev.com> wrote:

> Windows 10?
>
> Bob S
>
>
>
> ___
> use-livecode mailing list
> use-livecode@lists.runrev.com
> Please visit this url to subscribe, unsubscribe and manage your
> subscription preferences:
> http://lists.runrev.com/mailman/listinfo/use-livecode
>
___
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription 
preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode

Re: Strange permissions issues calling 'shell' in Windows

[Bob Sneidar via use-livecode]

Windows 10?

Bob S



___
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription 
preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode

Re: Strange permissions issues calling 'shell' in Windows

[Ben Rubinstein via use-livecode]

Hi Bob,

Thanks for taking the trouble to consider this.


The first thing that comes to my mind is, where are you putting the log
file? If you are using a specialfolderpath that only has a context when
logged in, then that might be an issue. 


It isn't that, the file is being created, the output from pslist.exe is being 
written to it - but the output is the report from pslist.exe saying that it 
didn't have permission.



Now if you are logged off, I am not sure what account is being used to
access the registry (and am not sure why it's accessing it or why you would
want to in the first place).


The scheduled task is set to use the same account as I use to log in.

I thought that the issue was the difference between being logged in as this 
account, and the task running with the same account. But it isn't (just) that, 
because when I made the scheduled task a batch script to run the command, it 
worked.


So there is an additional difference imposed by using an LC app using 'shell' 
to run the command, rather than using a batch script. Both work fine when 
logged in; only one when it is a scheduled task. The command works fine 
invoked from a batch script which itself is run as a scheduled task. But the 
command reports the permission problem when it is invoked from the app running 
as a scheduled task.


Ben


On 31/10/2017 14:57, Bob Sneidar via use-livecode wrote:

The first thing that comes to my mind is, where are you putting the log file? 
If you are using a specialfolderpath that only has a context when logged in, 
then that might be an issue. As far as the firewall goes, unless you or your IT 
team has intentionally blocked RPC traffic, it should have an exception by 
default, as a lot of things use this service. Now if you are logged off, I am 
not sure what account is being used to access the registry (and am not sure why 
it's accessing it or why you would want to in the first place).

Bob S



On Oct 31, 2017, at 03:40 , Ben Rubinstein via use-livecode 
 wrote:

I have an app in LC, which normally runs as a nightly scheduled task on a VM in 
the client's network.

It's been having memory problems, and to help diagnose this I wanted to track 
memory usage through the job and over time (it runs every night, takes about 40 
minutes crunching lots of data, and emails me a log after each run).

Unfortunately LC doesn't have any reliable ways to inspect its memory usage, so 
I've been trying to use a combination of an LC function that estimates the size 
of all global variables, with a shell call to the 'pslist' sysinternal tool.

Run 'live', this works fine. But when run as a scheduled task, logged out of 
the machine, instead of the expected process information the result of the 
shell call was
Failed to take process snapshot on ORGAPP01.
Make sure that the Remote Registry service is running on the remote
system, that you have firewall ports allow RPC access, and your
account has read access the following key on the remote system:
HKLM\Software\Microsoft\Windows NT\CurrentVersion\Perflib

I don't have that kind of access (nor the Windows knowledge to use it safely if 
I did!) so I appealed to the client's IT dept, who said some words I didn't 
understand to the affect that it should work. When I said again that it worked 
logged-in, but not as a scheduled task, they asked for details of the batch 
file; so I thought the simplest thing would be to set up a simple batch file to 
demonstrate the issue.

I made a batch file to invoke pslist and put the result in a text file; it 
worked. Then I made a scheduled task to invoke this batch file, expecting it to 
fail with the above issue thus proving my point and giving the IT guy a test 
case to work with - but in fact when the scheduled task ran, the output from 
pslist was perfectly good.

So then I thought that this was a work-around: instead of my app using shell to 
invoke pslist and then log the result of the shell call; it could instead use 
shell to invoke the batch file, and then read back and log the contents of the 
output file generated by the batch file.

So I made the change in the app, moved it over, ran it with the expected and 
desired result; logged out and waited.

This morning however I got the same error message as before in the log.

So the situation is:
1) logged-in app uses shell to call pslist: OK
2) not logged-in app uses shell to call pslist: permissions error
3) logged-in batch file calls pslist: OK
4) not logged-in batch file calls pslist: OK
5) not logged-in app uses shell to call batch file which calls pslist: 
permissions error


If step (4) produced a permissions error, I'd be sort-of happy. If step (5) 
worked, I'd be even happier. But it's the distinction between (4) and (5) that 
puzzles me.

Is there some way that calling 'shell' enters a context with less permission 
than the app which calls it? And is there something I can do about t

Re: Relative Paths in Property Inspector

[Mark Wieder via use-livecode]

On 10/30/2017 07:55 PM, Mike Kerner via use-livecode wrote:

So what do you have for your paths?  Are your usernames the same?  For me,
I run into the issue because my usernames aren't the same on every box, so
the absolute path varies.


Are you referring to the *dropbox* usernames being different? If that's 
the case, you can share a folder between dropbox users and still use a 
common repository.


--
 Mark Wieder
 ahsoftw...@gmail.com

___
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription 
preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode

Re: Strange permissions issues calling 'shell' in Windows

[Bob Sneidar via use-livecode]

The first thing that comes to my mind is, where are you putting the log file? 
If you are using a specialfolderpath that only has a context when logged in, 
then that might be an issue. As far as the firewall goes, unless you or your IT 
team has intentionally blocked RPC traffic, it should have an exception by 
default, as a lot of things use this service. Now if you are logged off, I am 
not sure what account is being used to access the registry (and am not sure why 
it's accessing it or why you would want to in the first place). 

Bob S


> On Oct 31, 2017, at 03:40 , Ben Rubinstein via use-livecode 
>  wrote:
> 
> I have an app in LC, which normally runs as a nightly scheduled task on a VM 
> in the client's network.
> 
> It's been having memory problems, and to help diagnose this I wanted to track 
> memory usage through the job and over time (it runs every night, takes about 
> 40 minutes crunching lots of data, and emails me a log after each run).
> 
> Unfortunately LC doesn't have any reliable ways to inspect its memory usage, 
> so I've been trying to use a combination of an LC function that estimates the 
> size of all global variables, with a shell call to the 'pslist' sysinternal 
> tool.
> 
> Run 'live', this works fine. But when run as a scheduled task, logged out of 
> the machine, instead of the expected process information the result of the 
> shell call was
>   Failed to take process snapshot on ORGAPP01.
>   Make sure that the Remote Registry service is running on the remote
>   system, that you have firewall ports allow RPC access, and your
>   account has read access the following key on the remote system:
>   HKLM\Software\Microsoft\Windows NT\CurrentVersion\Perflib
> 
> I don't have that kind of access (nor the Windows knowledge to use it safely 
> if I did!) so I appealed to the client's IT dept, who said some words I 
> didn't understand to the affect that it should work. When I said again that 
> it worked logged-in, but not as a scheduled task, they asked for details of 
> the batch file; so I thought the simplest thing would be to set up a simple 
> batch file to demonstrate the issue.
> 
> I made a batch file to invoke pslist and put the result in a text file; it 
> worked. Then I made a scheduled task to invoke this batch file, expecting it 
> to fail with the above issue thus proving my point and giving the IT guy a 
> test case to work with - but in fact when the scheduled task ran, the output 
> from pslist was perfectly good.
> 
> So then I thought that this was a work-around: instead of my app using shell 
> to invoke pslist and then log the result of the shell call; it could instead 
> use shell to invoke the batch file, and then read back and log the contents 
> of the output file generated by the batch file.
> 
> So I made the change in the app, moved it over, ran it with the expected and 
> desired result; logged out and waited.
> 
> This morning however I got the same error message as before in the log.
> 
> So the situation is:
>   1) logged-in app uses shell to call pslist: OK
>   2) not logged-in app uses shell to call pslist: permissions error
>   3) logged-in batch file calls pslist: OK
>   4) not logged-in batch file calls pslist: OK
>   5) not logged-in app uses shell to call batch file which calls pslist: 
> permissions error
> 
> 
> If step (4) produced a permissions error, I'd be sort-of happy. If step (5) 
> worked, I'd be even happier. But it's the distinction between (4) and (5) 
> that puzzles me.
> 
> Is there some way that calling 'shell' enters a context with less permission 
> than the app which calls it? And is there something I can do about this?
> 
> TIA,
> 
> Ben
> 
> 
> ___
> use-livecode mailing list
> use-livecode@lists.runrev.com
> Please visit this url to subscribe, unsubscribe and manage your subscription 
> preferences:
> http://lists.runrev.com/mailman/listinfo/use-livecode


___
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription 
preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode

Re: I Hate Full Screen!

[Bob Sneidar via use-livecode]

If you option-click the zoom on a Mac, it uses the old zoom where the menus are 
still visible. 

Bob S


> On Oct 30, 2017, at 18:03 , Brian Milby via use-livecode 
>  wrote:
> 
> I figured out how to disable full screen and replace it with the old zoom,
> but that isn’t really that much of an improvement. Something in the SDK has
> changed since the code didn’t change between 7 and 8 (at least the engine
> code that sets up the window).

___
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription 
preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode

Re: Go card "xyz"

[dunbarx via use-livecode]

If you have a button on a stack with:

on mouseUp
  go stack "otherStack"
  answer "XYZ"
end mouseUp

and in the stack script of that other stack:

on resumeStack
  answer "ABC"
end resumeStack

The "ABC" appears before the "XYZ".

Is that what you meant?

Craig Newman





--
Sent from: 
http://runtime-revolution.278305.n4.nabble.com/Revolution-User-f278306.html

___
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription 
preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode

Go card "xyz"

[Pyyhtiä Christer via use-livecode]

When in a script you issue a command

go card "xyz"

does the next command line of the script get into execution before the "go" has 
been completed?

Is it comparable to the "move" command, where this is true, with the additional 
statement "without waiting"?

Is it possible to find a delay in the "go" command if you for example go from 
card 2 to card 28 and there are many objects in the cards in between?

The fact is that the behaviour varies, same code at different times in some 
cases looks like an action is tried to be done for the card going out.


Christer Pyyhtiä
chris...@mindcrea.com


___
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription 
preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode

Strange permissions issues calling 'shell' in Windows

[Ben Rubinstein via use-livecode]

I have an app in LC, which normally runs as a nightly scheduled task on a VM 
in the client's network.


It's been having memory problems, and to help diagnose this I wanted to track 
memory usage through the job and over time (it runs every night, takes about 
40 minutes crunching lots of data, and emails me a log after each run).


Unfortunately LC doesn't have any reliable ways to inspect its memory usage, 
so I've been trying to use a combination of an LC function that estimates the 
size of all global variables, with a shell call to the 'pslist' sysinternal tool.


Run 'live', this works fine. But when run as a scheduled task, logged out of 
the machine, instead of the expected process information the result of the 
shell call was

Failed to take process snapshot on ORGAPP01.
Make sure that the Remote Registry service is running on the remote
system, that you have firewall ports allow RPC access, and your
account has read access the following key on the remote system:
HKLM\Software\Microsoft\Windows NT\CurrentVersion\Perflib

I don't have that kind of access (nor the Windows knowledge to use it safely 
if I did!) so I appealed to the client's IT dept, who said some words I didn't 
understand to the affect that it should work. When I said again that it worked 
logged-in, but not as a scheduled task, they asked for details of the batch 
file; so I thought the simplest thing would be to set up a simple batch file 
to demonstrate the issue.


I made a batch file to invoke pslist and put the result in a text file; it 
worked. Then I made a scheduled task to invoke this batch file, expecting it 
to fail with the above issue thus proving my point and giving the IT guy a 
test case to work with - but in fact when the scheduled task ran, the output 
from pslist was perfectly good.


So then I thought that this was a work-around: instead of my app using shell 
to invoke pslist and then log the result of the shell call; it could instead 
use shell to invoke the batch file, and then read back and log the contents of 
the output file generated by the batch file.


So I made the change in the app, moved it over, ran it with the expected and 
desired result; logged out and waited.


This morning however I got the same error message as before in the log.

So the situation is:
1) logged-in app uses shell to call pslist: OK
2) not logged-in app uses shell to call pslist: permissions error
3) logged-in batch file calls pslist: OK
4) not logged-in batch file calls pslist: OK
 	5) not logged-in app uses shell to call batch file which calls pslist: 
permissions error



If step (4) produced a permissions error, I'd be sort-of happy. If step (5) 
worked, I'd be even happier. But it's the distinction between (4) and (5) that 
puzzles me.


Is there some way that calling 'shell' enters a context with less permission 
than the app which calls it? And is there something I can do about this?


TIA,

Ben


___
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription 
preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode

Re: Relative Paths in Property Inspector

[Mike Kerner via use-livecode]

Hmm.  I wonder why I've run into trouble with this.  Now I don't remember.

On Mon, Oct 30, 2017 at 11:19 PM, J. Landman Gay via use-livecode <
use-livecode@lists.runrev.com> wrote:

> I have different user names on each machine too. Just set the appropriate
> path in the LC prefs on each machine. You only have to do it once, and
> going forward it just works for any other LC versions you install since all
> versions on a machine share the same prefs file.
>
>
> On 10/30/17 9:55 PM, Mike Kerner via use-livecode wrote:
>
>> So what do you have for your paths?  Are your usernames the same?  For me,
>> I run into the issue because my usernames aren't the same on every box, so
>> the absolute path varies.
>>
>> On Mon, Oct 30, 2017 at 6:49 PM, Mark Wieder via use-livecode <
>> use-livecode@lists.runrev.com> wrote:
>>
>> On 10/30/2017 01:51 PM, J. Landman Gay via use-livecode wrote:
>>>
>>> I've got my extensionos folder in Dropbox. All my machines have Dropbox
 installed and it syncs to all of them.

 On 10/30/17 3:20 PM, Mike Kerner via use-livecode wrote:

 All I'm trying to "Get" is a simpler way to store the extensions folder
> on
> a network/sync so I get the exact same environment on every box.
>
>
 Same here. Trick learned from Andre Garzia.
>>>
>>> +1 for 'extensionos'.
>>>
>>>
>>> --
>>>   Mark Wieder
>>>   ahsoftw...@gmail.com
>>>
>>>
>>> ___
>>> use-livecode mailing list
>>> use-livecode@lists.runrev.com
>>> Please visit this url to subscribe, unsubscribe and manage your
>>> subscription preferences:
>>> http://lists.runrev.com/mailman/listinfo/use-livecode
>>>
>>>
>>
>>
>>
>
> --
> Jacqueline Landman Gay | jac...@hyperactivesw.com
> HyperActive Software   | http://www.hyperactivesw.com
>
> ___
> use-livecode mailing list
> use-livecode@lists.runrev.com
> Please visit this url to subscribe, unsubscribe and manage your
> subscription preferences:
> http://lists.runrev.com/mailman/listinfo/use-livecode
>



-- 
On the first day, God created the heavens and the Earth
On the second day, God created the oceans.
On the third day, God put the animals on hold for a few hours,
   and did a little diving.
And God said, "This is good."
___
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription 
preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode

Remote Debugger Frustrations

[Peter Reid via use-livecode]

How are people getting on with the remote debugger facility? I'm finding it 
very useful but also frustrating for the following reasons:

1. It's SLOW. I'm running LC 9 on an iMac with a 4GHz quad i7 processor and 
24GB RAM connected by USB to an Amazon Fire 7 tablet, yet each step through the 
script in the debugger takes several seconds.

2. It's difficult to stop to make changes. Clicking the stop debugging button 
often doesn't work or takes a long time to react and even when apparently 
stopped it's still making everything slow.

3. Often it steps through the script as it was just before the latest change, 
i.e. it doesn't take into account the latest script edits.

4. It can be difficult to change break points during remote debugging.

5. The only way that guarantees a clean run is to quit LC and start again.

6. A partial reset can be achieved by the following:
a. having clicked 'stop debugging'...
b. open & close Standalone Application Settings
c. click Test, which prompts for Save
d. click Save and the remote build & install starts
e. however, the above doesn't always recognise script changes.

I find the Remote Debugger to be very helpful. BUT, if only there was a quick 
reliable way of stopping, editing the script and continuing/restarting.  
Currently the run-check-edit-rerun cycle is far too slow and error prone using 
the remote debugger.

Any thoughts or advice from other users of the Remote Debugger?

Peter
--
Peter Reid
Loughborough, UK


___
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription 
preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode

Re: The Shrinking size of animated gif

[Richmond Mathewson via use-livecode]

I'll "have a bash at this" tonight: currently in my lunch break.

HOWEVER . . .

I ALWAYS decompose an animated GIF into constituent frames (which I then 
export on PNG format)

using GIMP . . .

OK, it's long and tedious, but at least I have a considerable amount of 
control over the process.


Richmond.

On 10/31/17 4:54 am, Alejandro Tejada via use-livecode wrote:

Hi All,

Could you verify if always that we convert
an animaated gif in a series of png images
the resulting images are always smaller
than the original animated gif?

Here is the recipe:

1) Create a new stack

2) create a button and paste this script
into the button: (check for broken lines
in the mail list server)


on mouseUp

-- first, delete all previous images
repeat with i = the number of controls of this card down to 1
   if word 1 of the long  name of control i of this card is "image" then
delete control i of this card
end repeat

answer file "Choose an animated gif..."
import paint from file it
try
   put the framecount of img 1 into k
   repeat with m = 1 to k
  set the currentframe of img 1 to m
  create image
  set the width of it to the width of img 1
  set the height of it to the height of img 1
  set the imagedata of it to the imagedata of img 1
  set the alphadata of it to the alphadata of img 1
   end repeat

   repeat with i = 1 to the number of controls of this card
  put the long name of control i of this card into q
  if word 1 of q is "image" then
 put the size of q & comma after n
  end if
   end repeat

   delete last char of n -- a comma
   put item 1 of n into tAnimatedGifSize
   delete item 1 of n
   put sum(n) into tPngImagesSize

   if tAnimatedGifSize > tPngImagesSize then
  put "Animated gif is " & (tAnimatedGifSize - tPngImagesSize) & "
bytes larger than png images"
   else
  -- sum of sizes of png images is larger
  -- than size of animated gif
  put "Png images are " & (tPngImagesSize - tAnimatedGifSize) & "
bytes larger than animated gif"
   end if
end try
end mouseUp



3) Click the button and import an animated gif

Post your results in this thread,
Thanks a lot in advance!

Al
___
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription 
preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode


___
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription 
preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode