Java FFI on Windows

2021-02-01 Thread Bernard Devlin via use-livecode
In the Extending Livecode part of the Dictionary it says:

*"Java bindings are currently supported on Android, Mac and Linux."*

I am at a loss to understand why Java extensions are supported on Mac and
Linux but *not* on Windows.

I've seen no explanation for this anywhere. Can someone enlighten me?  I'd
understand why Objective C extensions would not work on Windows, but Java?

Can extensions on Windows *only* be written in C?

Regards Bernard
___
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: open secure socket... using certificate

2021-01-29 Thread Bernard Devlin via use-livecode
Hi Richard

The idea of client certificates was why I was looking for this feature for
the past 15 years. I know that PKI is complex but it is important (which is
why browsers implemented it decades ago).  Even tsNet behaves very
differently on Windows and OS X when it comes to server certificates (looks
to me that Microsoft is mis-implementing their libcurl/schannel interface,
but I guess it stung them 20 years ago when they hadn't implemented
Certificate Revocation Lists).

Without LC having client certificates the only option I face is moving to
another tool. Even the idea of implementing real secure sockets in a FFI
external would be a huge amount of work for any of us users (as there is
not even a single lower level language which we could rely on in each
platform, so we'd be implementing a complex security layer and learning how
to do it in several different languages).

I'm going to look at moving to Xojo. Their docs say they implemented secure
sockets with certificates in 2006.  Their Android deployment platform is
close to completion. It really would have been better for LC to have
removed those claims from the Dictionary back in 2014, as then I'd have
seen in recent years it still hadn't been implemented and would have looked
at another tool months ago rather than get to this point and realise I'd
wasted a lot of time. I've spent an entire week now just pondering on the
options on how I can move forward.

Regards, Bernard


On Fri, Jan 29, 2021 at 4:12 PM Richard Gaskin via use-livecode <
use-livecode@lists.runrev.com> wrote:

>
>
> Thank you, Bernard.  In bringing myself up to speed on this, once I
> stumbled across the phrase "two-way SSL" I was able to find much good
> reading about it.
>
> I just added Comment #11 to the bug report on this:
>
>  The range of services requiring two-way SSL is increasing,
>  and with IoT the scope of use cases is multiplying.
>
>  Where required, I don't see an opportunity for a scripted
>  workaround, so in those cases the only alternative is to
>  leave LC for a tool that supports two-way SSL.
>
>  This would seem a good time to bite the bullet on this
>  implementation. What would be required to make it happen?
>
> https://quality.livecode.com/show_bug.cgi?id=13410#c11
>
>
___
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: open secure socket... using certificate

2021-01-29 Thread Bernard Devlin via use-livecode
Hi Richard, it's client side.

The project is a messaging app, principally for mobile.  SSL/TLS
certificates issued with client software enable servers to know that the
connection is allowed at a network level before any user authentication.

In the early stages of development it was simple enough for a LC messaging
server to dismiss connections which didn't come from an accepted client and
didn't meet the specification.  Once I moved to an industry standard server
then the connections have to be encrypted with certificates at both ends.

Regards, Bernard

On Thu, Jan 28, 2021 at 4:44 PM Richard Gaskin via use-livecode <
use-livecode@lists.runrev.com> wrote:

> Bernard Devlin wrote:
>
>  > I feel like a sucker. I started on this project a few months ago on
>  > the assumption that socket certificates now worked as they'd been
>  > included in the Dictionary for years. I was away from development for
>  > a few years and hadn't noticed the complaints that the Dictionary was
>  > misrepresenting the situation.
>
> Is this specific to the server "accept" not supporting SSL, or something
> client-side?
>
> --
>   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
>
___
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: open secure socket... using certificate

2021-01-27 Thread Bernard Devlin via use-livecode
Hi Tom

You shouldn't get any hopes up. I'd commented in the bug report in 2014
that this was something that we'd been told was coming back in the days of
LC version 2.


I think when a development environment has failed to deliver a feature from
version 2 to version 10 that thing is never going to appear.


On Tue, Jan 26, 2021 at 9:22 PM Tom Glod via use-livecode <
use-livecode@lists.runrev.com> wrote:

> super happy to see this, hopefully it will when i need it.
> Going to look this up.
> Wondering if it will be super hard to create certificate and make it work
> on localhost sockets.
>
>
> On Tue, Jan 26, 2021 at 2:26 PM Brian Milby via use-livecode <
> use-livecode@lists.runrev.com> wrote:
>
> > “Using tCertificate and tKey” compiles but I don’t know how to test.  It
> > does seem like the parser doesn’t recognize those keywords.
> >
> > Sent from my iPhone
> >
> > > On Jan 26, 2021, at 2:13 PM, Bernard Devlin via use-livecode <
> > use-livecode@lists.runrev.com> wrote:
> > >
> > > Thanks for the suggestion Erik, but I don't see from that how one
> > specifies
> > > the certificate.
> > >
> > > Regards, Bernard
> > >
> > >> On Tue, Jan 26, 2021 at 6:03 PM Erik Beugelaar via use-livecode <
> > >> use-livecode@lists.runrev.com> wrote:
> > >>
> > >> Maybe this:
> > >>
> > >> secure socket "livecode.com:443"
> > >>
> > >> Examples:
> > >>
> > >> https://livecode.fandom.com/wiki/Secure_socket
> > >>
> > >> -Original Message-
> > >> From: use-livecode  On Behalf
> Of
> > >> Bernard Devlin via use-livecode
> > >> Sent: dinsdag 26 januari 2021 16:40
> > >> To: How to use LiveCode 
> > >> Cc: Bernard Devlin 
> > >> Subject: Re: open secure socket... using certificate
> > >>
> > >> I did. I tried these too:
> > >>
> > >> *open* *secure* socket to "localhost:443"  using certificate tc and
> key
> > tk
> > >>
> > >> *open* *secure* socket to "localhost:443" without verification using
> > >> certificate tc and key tk
> > >>
> > >> When the above lines are entered in the script editor they are flagged
> > as
> > >> being syntax errors. In both cases it is what comes after
> "certificate"
> > >> that is flagged as a syntax error (flagged as: missing "," near "tc").
> > >> There seems to be no combination of command options that works with
> > >> certificates.
> > >>
> > >> The fact that the Dictionary has zero information about what is
> expected
> > >> for certificate/key was not a good sign, which is why I searched the
> > >> archive.  I just went to have a look at the code on Github and I can
> > seen
> > >> nothing to suggest that "using certificate and key" is implemented.
> > >>
> > >> The server and client certificate are working in a browser, so the
> > problem
> > >> is definitely on the LC side.
> > >>
> > >> On Tue, Jan 26, 2021 at 2:34 PM Brian Milby via use-livecode <
> > >> use-livecode@lists.runrev.com> wrote:
> > >>
> > >>> Did you try with “and key tKey”... it does not look like that part is
> > >>> optional.
> > >>>
> > >>> Sent from my iPhone
> > >>>
> > >>>> On Jan 26, 2021, at 9:07 AM, Bernard Devlin via use-livecode <
> > >>> use-livecode@lists.runrev.com> wrote:
> > >>>>
> > >>>> According to the Dictionary in LC 9.5.1 there is this command:
> > >>>>
> > >>>> open secure socket [from [localHostName][:localPort]] [to] socketID
> > >>>> [with message callbackMessage] [without verification] *[using
> > >>>> certificate certificate and key key]*
> > >>>>
> > >>>> However I can't get it to work.
> > >>>>
> > >>>> open secure socket to "localhost:443" using certificate
> > >>>>
> > >>>> throws a runtime error "no handler: using"
> > >>>>
> > >>>> If I use
> > >>>>
> > >>>> open secure socket to "localhost:443"
> > >>>>
> > >>>> I get a socket connect

Re: open secure socket... using certificate

2021-01-27 Thread Bernard Devlin via use-livecode
Thanks for confirming it Mark.

I feel like a sucker. I started on this project a few months ago on the
assumption that socket certificates now worked as they'd been included in
the Dictionary for years. I was away from development for a few years and
hadn't noticed the complaints that the Dictionary was misrepresenting the
situation.  Depressing to see my own name in that bug report from 7 years
ago, pointing out that secure sockets with certificates was something that
we'd been told was on the horizon in LC version 2, which must have been
around 2004.

https://quality.livecode.com/show_bug.cgi?id=13410

Regards, Bernard


On Wed, Jan 27, 2021 at 12:57 AM Mark Wieder via use-livecode <
use-livecode@lists.runrev.com> wrote:

> On 1/26/21 6:05 AM, Bernard Devlin via use-livecode wrote:
>
> > Is it really the case that for the past 6 years LC documentation has been
> > misleading people concerning the implementation of certificates for
> secure
> > socket connections?
>
> Sadly, yes.
>
> >
> > I notice in the Dictionary the entry for "open socket" in the table of
> > options for this command has entries for "certificate" and "key", but
> these
> > are both empty.  As if these features were meant to be implemented but
> were
> > never implemented and the Dictionary was never updated to remove this
> > misleading information.
> >
> > I just find that hard to believe.
>
> https://quality.livecode.com/show_bug.cgi?id=13410
>
> although the earliest bug report seems to be from 2006:
>
> https://quality.livecode.com/show_bug.cgi?id=3737
>
> --
>   Mark Wieder
>
>
___
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: open secure socket... using certificate

2021-01-26 Thread Bernard Devlin via use-livecode
Thanks for the suggestion Erik, but I don't see from that how one specifies
the certificate.

Regards, Bernard

On Tue, Jan 26, 2021 at 6:03 PM Erik Beugelaar via use-livecode <
use-livecode@lists.runrev.com> wrote:

> Maybe this:
>
> secure socket "livecode.com:443"
>
> Examples:
>
> https://livecode.fandom.com/wiki/Secure_socket
>
> -Original Message-
> From: use-livecode  On Behalf Of
> Bernard Devlin via use-livecode
> Sent: dinsdag 26 januari 2021 16:40
> To: How to use LiveCode 
> Cc: Bernard Devlin 
> Subject: Re: open secure socket... using certificate
>
> I did. I tried these too:
>
> *open* *secure* socket to "localhost:443"  using certificate tc and key tk
>
> *open* *secure* socket to "localhost:443" without verification using
> certificate tc and key tk
>
> When the above lines are entered in the script editor they are flagged as
> being syntax errors. In both cases it is what comes after "certificate"
> that is flagged as a syntax error (flagged as: missing "," near "tc").
> There seems to be no combination of command options that works with
> certificates.
>
> The fact that the Dictionary has zero information about what is expected
> for certificate/key was not a good sign, which is why I searched the
> archive.  I just went to have a look at the code on Github and I can seen
> nothing to suggest that "using certificate and key" is implemented.
>
> The server and client certificate are working in a browser, so the problem
> is definitely on the LC side.
>
> On Tue, Jan 26, 2021 at 2:34 PM Brian Milby via use-livecode <
> use-livecode@lists.runrev.com> wrote:
>
> > Did you try with “and key tKey”... it does not look like that part is
> > optional.
> >
> > Sent from my iPhone
> >
> > > On Jan 26, 2021, at 9:07 AM, Bernard Devlin via use-livecode <
> > use-livecode@lists.runrev.com> wrote:
> > >
> > > According to the Dictionary in LC 9.5.1 there is this command:
> > >
> > > open secure socket [from [localHostName][:localPort]] [to] socketID
> > > [with message callbackMessage] [without verification] *[using
> > > certificate certificate and key key]*
> > >
> > > However I can't get it to work.
> > >
> > > open secure socket to "localhost:443" using certificate
> > >
> > > throws a runtime error "no handler: using"
> > >
> > > If I use
> > >
> > >  open secure socket to "localhost:443"
> > >
> > > I get a socket connection, but all the security of a client
> > > certificate does not work
> > >
> > >
> > > This causes a syntax error in the IDE:
> > >
> > >  open secure socket to "localhost:443" using certificate tName
> > >
> > > Looking through the archives I see that a couple of discussions
> > > where people were asking about this variant of the "open socket"
> > > command 5 to 6 years ago, *saying that the "certificate" part has
> > > not been implemented*, regardless of what the Dictionary says.
> > >
> > > Is it really the case that for the past 6 years LC documentation has
> > > been misleading people concerning the implementation of certificates
> > > for
> > secure
> > > socket connections?
> > >
> > > I notice in the Dictionary the entry for "open socket" in the table
> > > of options for this command has entries for "certificate" and "key",
> > > but
> > these
> > > are both empty.  As if these features were meant to be implemented
> > > but
> > were
> > > never implemented and the Dictionary was never updated to remove
> > > this misleading information.
> > >
> > > I just find that hard to believe.
> > >
> > > Regards
> > >
> > > Bernard
> > > ___
> > > 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
>
>
> ___
> 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: open secure socket... using certificate

2021-01-26 Thread Bernard Devlin via use-livecode
I did. I tried these too:

*open* *secure* socket to "localhost:443"  using certificate tc and key tk

*open* *secure* socket to "localhost:443" without verification using
certificate tc and key tk

When the above lines are entered in the script editor they are flagged as
being syntax errors. In both cases it is what comes after "certificate"
that is flagged as a syntax error (flagged as: missing "," near "tc").
There seems to be no combination of command options that works with
certificates.

The fact that the Dictionary has zero information about what is expected
for certificate/key was not a good sign, which is why I searched the
archive.  I just went to have a look at the code on Github and I can seen
nothing to suggest that "using certificate and key" is implemented.

The server and client certificate are working in a browser, so the problem
is definitely on the LC side.

On Tue, Jan 26, 2021 at 2:34 PM Brian Milby via use-livecode <
use-livecode@lists.runrev.com> wrote:

> Did you try with “and key tKey”... it does not look like that part is
> optional.
>
> Sent from my iPhone
>
> > On Jan 26, 2021, at 9:07 AM, Bernard Devlin via use-livecode <
> use-livecode@lists.runrev.com> wrote:
> >
> > According to the Dictionary in LC 9.5.1 there is this command:
> >
> > open secure socket [from [localHostName][:localPort]] [to] socketID [with
> > message callbackMessage] [without verification] *[using certificate
> > certificate and key key]*
> >
> > However I can't get it to work.
> >
> > open secure socket to "localhost:443" using certificate
> >
> > throws a runtime error "no handler: using"
> >
> > If I use
> >
> >  open secure socket to "localhost:443"
> >
> > I get a socket connection, but all the security of a client certificate
> > does not work
> >
> >
> > This causes a syntax error in the IDE:
> >
> >  open secure socket to "localhost:443" using certificate tName
> >
> > Looking through the archives I see that a couple of discussions where
> > people were asking about this variant of the "open socket" command 5 to 6
> > years ago, *saying that the "certificate" part has not been implemented*,
> > regardless of what the Dictionary says.
> >
> > Is it really the case that for the past 6 years LC documentation has been
> > misleading people concerning the implementation of certificates for
> secure
> > socket connections?
> >
> > I notice in the Dictionary the entry for "open socket" in the table of
> > options for this command has entries for "certificate" and "key", but
> these
> > are both empty.  As if these features were meant to be implemented but
> were
> > never implemented and the Dictionary was never updated to remove this
> > misleading information.
> >
> > I just find that hard to believe.
> >
> > Regards
> >
> > Bernard
> > ___
> > 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


open secure socket... using certificate

2021-01-26 Thread Bernard Devlin via use-livecode
According to the Dictionary in LC 9.5.1 there is this command:

open secure socket [from [localHostName][:localPort]] [to] socketID [with
message callbackMessage] [without verification] *[using certificate
certificate and key key]*

However I can't get it to work.

 open secure socket to "localhost:443" using certificate

throws a runtime error "no handler: using"

If I use

  open secure socket to "localhost:443"

I get a socket connection, but all the security of a client certificate
does not work


This causes a syntax error in the IDE:

  open secure socket to "localhost:443" using certificate tName

Looking through the archives I see that a couple of discussions where
people were asking about this variant of the "open socket" command 5 to 6
years ago, *saying that the "certificate" part has not been implemented*,
regardless of what the Dictionary says.

Is it really the case that for the past 6 years LC documentation has been
misleading people concerning the implementation of certificates for secure
socket connections?

I notice in the Dictionary the entry for "open socket" in the table of
options for this command has entries for "certificate" and "key", but these
are both empty.  As if these features were meant to be implemented but were
never implemented and the Dictionary was never updated to remove this
misleading information.

I just find that hard to believe.

Regards

Bernard
___
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: opened socket in IDE prevents script changes

2021-01-16 Thread Bernard Devlin via use-livecode
Hi Bernd

Thanks for the suggestion. But I think the evidence points away from this.

One thing I've done is
- change the script (could be just addition of a comment)
- copy entire script
- apply & close
- re-open script editor --> *change is missing*

- close IDE
- re-open LC
- open stack script in editor
- delete entire script
- paste script from clipboard
- apply & close
- re-open script editor --> *change is now there*

As soon as I open a socket, back to the situation that where script
changed, apply etc. does not take. And closing all sockets does not work
(although there is always one socket left open, which I believe is a socket
the IDE creates in the background).

I'm working on the assumption that it's something to do with the way I'm
reading/writing to the socket.

It's definitely strange.  But if I can get to the bottom of this it will
help others. The weird thing is that everything is working with the socket
communications, no script or runtime errors thrown by LC, yet this working
app breaks the IDE in the sense that LC is no longer a dynamic scripting
environment (for me anyway).

It's important to get to the bottom of this as it makes LC look bad,
especially the numerous crashes, which all seem to be related to setting
breakpoints in the script editor to try and get to the bottom of what was
not working. At least on OSX I'm getting crash logs in the console.
Strangely the crash errors are not all the same.

Regards
Bernard

On Fri, Jan 15, 2021 at 5:28 PM Niggemann, Bernd via use-livecode <
use-livecode@lists.runrev.com> wrote:

> Hi Bernard,
>
> I have seen similar behavior of the apply button indicating "clean"  when
> the script was "dirty"/had compile errors. That was without sockets.
>
> https://quality.livecode.com/show_bug.cgi?id=22555
>
> maybe your problem is related
>
> Kind regards
> Bernd
> ___
> 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: opened socket in IDE prevents script changes

2021-01-15 Thread Bernard Devlin via use-livecode
Hi Alex

- no plugins using sockets
- Business 9.6.1
- no other debuggers/editors in use or used (I have crashed/restarted LC a
few dozen times today)
- no remote debugging

This is just a plain old stack, only a mouseup in a button, everything else
in the stack script.

I'm going to have to look back at my previous socket project to see what I
was doing there that was different.

It seems to me that this is what is going on: LC loads the script & the
socket interaaction makes it impossible for any script change to be
applied. Feels like the engine is simply not "letting go" in order to
reload the script.  Very peculiar, as I would expect that closing each of
the open sockets would release whatever was bound and stopping the script
changes from being applied.

Thanks for testing & showing that it's something idiosyncratic to me.

Regards,
Bernard

On Fri, Jan 15, 2021 at 4:59 PM Alex Tweedly via use-livecode <
use-livecode@lists.runrev.com> wrote:

>
> On 15/01/2021 15:30, Bernard Devlin via use-livecode wrote:
> > I'm seeing exactly the same range of behaviours I described earlier for
> > 9.5.1 on Win 10 but with 9.6.1 on OS X.
> >
> > It's very strange that Alex is also editing scripts with sockets
> > opened/closed but  not losing changes made in the script editor. I also
> had
> > 3 crashes on OS X within 30 minutes of working on this stack.
> >
> > Alex, are you using the IDE's script editor or some external editor?  If
> > you are using that script editor, once you've opened a socket could you
> add
> > a comment to your stack script, apply, close the script editor, then edit
> > the script again to see if the comment has persisted?  I have to quit LC
> to
> > make any changes to a stack once I've opened a socket.
>
> Yes, I'm using standard IDE script editor.
>
> The added comment persists, both through  "apply - close - reopen" and
> "apply - save stack - close editor - reopen", and to stack and card
> scripts.
>
> > It seems to me that once a socket has been opened that the IDE is no
> longer
> > persisting changes.  I'm pretty sure I didn't see this behaviour on my
> > previous project using sockets.
>
> Have to ask
>
>   - do you have any plugins that might use sockets ?
>
>   - I'm using Indy version - let me know if you're on Community /
> Business and I can try that too
>I think :-)
>
>   - do you, or have you, used other editors / debuggers that might
> affect this ?
>
>   - do you, or have you, used remote debugging ?
>
> That's all the straws I can think of for now.
>
> Alex.
>
> > Regards
> > Bernard
> >
> > On Fri, Jan 15, 2021 at 12:21 PM Bernard Devlin 
> wrote:
> >
> >> Thanks Alex for your quick response.
> >>
> >> I'll try with 9.6.1 on Mac.  The only other difference between us is
> that
> >> my sockets are interacting with a process outside of LC. But I can't see
> >> why that would make a difference (particularly once I've closed all
> >> sockets, I wouldn't expect sockets to interfere with scripts being
> >> changed).
> >>
> >> Regards,
> >> Bernard
> >>
> >> On Fri, Jan 15, 2021 at 12:07 PM Alex Tweedly via use-livecode <
> >> use-livecode@lists.runrev.com> wrote:
> >>
> >>> I've not seen any such problems - though with 9.6.1 / Mac.
> >>>
> >>> I've spent most of the last week in the IDE, with two or more stacks
> >>> open (one server, one client - occasionally multiple client stacks) so
> I
> >>> have multiple open sockets (both 'accept'ing and 'open'ed)  almost all
> >>> the time, and had no problems with IDE or script editor.
> >>>
> >>> (I know - that's no consolation, but I thought I'd mention it in case
> >>> 9.6.1 was an option for you and might just help:-)
> >>>
> >>>
> > ___
> > 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: opened socket in IDE prevents script changes

2021-01-15 Thread Bernard Devlin via use-livecode
I'm seeing exactly the same range of behaviours I described earlier for
9.5.1 on Win 10 but with 9.6.1 on OS X.

It's very strange that Alex is also editing scripts with sockets
opened/closed but  not losing changes made in the script editor. I also had
3 crashes on OS X within 30 minutes of working on this stack.

Alex, are you using the IDE's script editor or some external editor?  If
you are using that script editor, once you've opened a socket could you add
a comment to your stack script, apply, close the script editor, then edit
the script again to see if the comment has persisted?  I have to quit LC to
make any changes to a stack once I've opened a socket.

It seems to me that once a socket has been opened that the IDE is no longer
persisting changes.  I'm pretty sure I didn't see this behaviour on my
previous project using sockets.

Regards
Bernard

On Fri, Jan 15, 2021 at 12:21 PM Bernard Devlin  wrote:

> Thanks Alex for your quick response.
>
> I'll try with 9.6.1 on Mac.  The only other difference between us is that
> my sockets are interacting with a process outside of LC. But I can't see
> why that would make a difference (particularly once I've closed all
> sockets, I wouldn't expect sockets to interfere with scripts being
> changed).
>

> Regards,
> Bernard
>
> On Fri, Jan 15, 2021 at 12:07 PM Alex Tweedly via use-livecode <
> use-livecode@lists.runrev.com> wrote:
>
>> I've not seen any such problems - though with 9.6.1 / Mac.
>>
>> I've spent most of the last week in the IDE, with two or more stacks
>> open (one server, one client - occasionally multiple client stacks) so I
>> have multiple open sockets (both 'accept'ing and 'open'ed)  almost all
>> the time, and had no problems with IDE or script editor.
>>
>> (I know - that's no consolation, but I thought I'd mention it in case
>> 9.6.1 was an option for you and might just help:-)
>>
>>
___
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: opened socket in IDE prevents script changes

2021-01-15 Thread Bernard Devlin via use-livecode
"do i understand it correctly that both, server and client stack, can
communicate with each other using sockets in one instance of the LC IDE"

Your understanding is correct. On the other project (where I didn't have
this current problem) I had half a dozen clients and 1 "server" running in
the IDE.

On Fri, Jan 15, 2021 at 2:47 PM matthias rebbe via use-livecode <
use-livecode@lists.runrev.com> wrote:

> Alex,
> for a new project which should include socket communication  i have to
> prototype a short "demo" and i was already wondering if i need to have
> running 2 instances of the LC IDE for development.
>
> But now you wrote that you had one server and even multiple client stacks
> opened in LC.
> So do i understand it correctly that both, server and client stack, can
> communicate with each other using sockets in one instance of the LC IDE?
>
>
___
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: opened socket in IDE prevents script changes

2021-01-15 Thread Bernard Devlin via use-livecode
Thanks Alex for your quick response.

I'll try with 9.6.1 on Mac.  The only other difference between us is that
my sockets are interacting with a process outside of LC. But I can't see
why that would make a difference (particularly once I've closed all
sockets, I wouldn't expect sockets to interfere with scripts being changed).

Regards,
Bernard

On Fri, Jan 15, 2021 at 12:07 PM Alex Tweedly via use-livecode <
use-livecode@lists.runrev.com> wrote:

> I've not seen any such problems - though with 9.6.1 / Mac.
>
> I've spent most of the last week in the IDE, with two or more stacks
> open (one server, one client - occasionally multiple client stacks) so I
> have multiple open sockets (both 'accept'ing and 'open'ed)  almost all
> the time, and had no problems with IDE or script editor.
>
> (I know - that's no consolation, but I thought I'd mention it in case
> 9.6.1 was an option for you and might just help:-)
>
>
___
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


opened socket in IDE prevents script changes

2021-01-15 Thread Bernard Devlin via use-livecode
This is a strange phenomenon.I wonder if anyone else has seen this.

Livecode 9.5.1 Windows 10.

If a socket is opened then changes to scripts in the IDE are not
applied/saved.  There's no error message. It *appears* from the IDE that
the changes are applied (i.e. the changes are visible in the script editor,
the "apply" button indicates the script is no longer dirty).

*Even if the socket is then closed and changes made this problem occurs.*
There are also many crashes of the IDE in working on scripts involving
sockets.  It's because of these crashes that I realised changes are not
applied.

Once a socket has been opened/closed then in the IDE the stack cannot be
closed/reopened without closing/reopening the entire IDE. Saving the stack
from the message box also results in the (not really applied) changes being
saved.

Regards
Bernard
___
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: Some sad news

2020-12-04 Thread Bernard Devlin via use-livecode
That's very sad news. I'd noticed he had been quiet on the forum. HH made
some truly extraordinary contributions. Just a few days ago I was just
looking at his Jigsaw HTML5 app, and was astonished at what he had
achieved.  He might not have been happy with the limitations in the HTML5
standalone, but he pushed what could be done to the max, inspiring many of
us to look at what was possible rather than on what was not possible.

I've never been part of an online community for so long that I've seen half
a dozen great contributors pass away. HH always went the extra mile with
any problem in our community where he got involved.  He will be missed :(

On Fri, Dec 4, 2020 at 12:07 PM Heather Laine via use-livecode <
use-livecode@lists.runrev.com> wrote:

> Dear List Folks,
>
> I have today received some sad news. I am told that Hermann Hoch passed
> away in April this year, from a stroke. I know many of you knew him either
> remotely or having met him in person at a conference. He will be missed. I
> met him myself and found him to be a wonderfully warm and kind person. He
> could be volatile, but I always felt it was because he cared passionately
> about things.
>
> I will take what steps I can to preserve his legacy of HTML5 examples. As
> these are hosted on an on-rev account we should be able to maintain it in
> perpetuity for the benefit of the LiveCode Community.
>
> LiveCode HTML5 didn't always meet his expectations or progress as fast as
> he wanted it to and I do wish he could have seen our next planned release,
> where HTML5 support will be a focus. He was of great benefit to this
> community and shared his expertise freely.
>
> I'm not sure what else I can say. Rest in peace hh, you will be missed :(
>
> Heather
>
>
> Heather Laine
> Customer Services Manager
> LiveCode Ltd
> www.livecode.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: Naive XML questions

2020-10-25 Thread Bernard Devlin via use-livecode
In his reply to you I think Alex is suggesting you use the handlers that
come with the  Datagrid.  These can convert xml to a LC array.  You might
find it easier just to deal with a familair data structure rather than
learn XML.

I found them here: http://revonline2.runrev.com/stack/571/XMLAndArrays

But the handlers are also to be found inside the datagrid library in the
IDE (I suspect they are more up to date).

On Sun, Oct 25, 2020 at 5:22 PM Graham Samuel via use-livecode <
use-livecode@lists.runrev.com> wrote:

> I feel stupid, but even with Sarah Reichelt’s help, there are very very
> simple things I can’t do with the LC XML functions. My current problem is
> this:
>
> I have a series of XML files which are in fact .gpx files - a collection
> of GPS coordinates representing a route on the Earth’s surface. There are
> thousands of files around which I can use as test data, so I don’t have to
> roll my own. Taking just one of these examples, I can do the following in LC
>
> 1. Get the user to find the file and download it
>
> 2. Check it’s a .gpx file
>
> 3. Turn it into an LC XML data structure with an integer ID
>
> 4. Find out how many child nodes there are.
>
> Then it all goes wrong, because I don’t seem to be able to address the
> individual nodes. This is obviously needed for further progress, in
> functions like revXMLFirstChild  and revXMLNodeContents.
>
> I just don’t know what format to use. Sarah’s tutorial doesn’t help
> because her data structure is too simplified.
>
> If you’re still with me, this is what the beginning of a typical file
> looks like:
>
> > 
> > http://www.topografix.com/GPX/1/1; creator="Digital Crow"
> version="1.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance;
> xsi:schemaLocation="http://www.topografix.com/GPX/1/1
> http://www.topografix.com/GPX/1/1/gpx.xsd;>
> >  
> >  
> >
>
> > 
> > 169.623637 
>
> > 
> > 169.744722 
>
> > 
> > 169.744722 
>
> > 
> > 169.266626 
> >
> etc - there may sometimes be thousands of these “trkpt” elements.
>
> So, suppose I want to get the first instance of the “lon” attribute into a
> local variable. How do I address the first  node etc in order to do
> it? I have attempted various common-sense methods, but none work.
>
> This must be child’s play to a lot of people, just not me.
>
> Hoping for some guidance.
>
> Graham
> ___
> 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: Script Editor freezes in Windows 10 (was Can't edit scripts in DataGrid group in Windows 10)

2020-10-23 Thread Bernard Devlin via use-livecode
Hi Martin

In the Script Editor try unchecking the various entries under Edit/Options
e.g. variable checking, live errors, etc.

I've recently moved to using LC on OS X and was surprised how snappy the
editor was i.e. I didn't need to switch off most of those things as I did
on Windows 10.

I don't have any advice about Proj Browser, as I almost never used it.

Regards, Bernard

On Fri, Oct 23, 2020 at 1:31 AM Martin Koob via use-livecode <
use-livecode@lists.runrev.com> wrote:

> Another bit of info regarding the length of script where this happens.
>
> I have been going through the project browser trying to open small scripts
> then opening those with more and more lines.   I found that a script with
> 53 lines opens, a script with 62 lines freezes.
>
> anyway.  Down with this for tonight.  Hope someone has some ideas as to
> what might be going on and how to solve it.
>
> Martin
>
>
___
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: LC iOS Command 'mergPopActivity'

2020-10-21 Thread Bernard Devlin via use-livecode
Hi Graham

I wondered if this information is of use.  URL calls are not limited to
"http://;.

Having a LC app respond to a URL call:
http://lessons.livecode.com/m/4069/l/58672-using-custom-url-schemes

Apps and their URL schemas:
https://ios.gadgethacks.com/news/always-updated-list-ios-app-url-scheme-names-0184033/

Regards
Bernard

On Wed, Oct 21, 2020 at 4:33 PM Graham Samuel via use-livecode <
use-livecode@lists.runrev.com> wrote:

So far I have not found out how to invoke another app directly from inside
> an iOS app, although I think I can invoke a browser using a URL. Perhaps
> this is easier than I thought. Will report.
>
> Graham
>
>
>
___
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: Livecode server UNIX version (not Linux).

2020-10-18 Thread Bernard Devlin via use-livecode
Which is why my sentence finishes with "... to be in a situation to take on
this level of complexity." :-)

Apple first bought the domain iphone.org in 1999. That the iPhone was under
development was even being discussed by mainstream media such as the New
York Times in 2002. The public availability of the iPhone was announced by
Apple at the start of January 2007. Those with an ear to the ground would
have been considering their future options between 2002 and 2007. I'm glad
they had the foresight I didn't have.

Kind regards,
Bernard


On Sat, Oct 17, 2020 at 7:45 PM Richmond via use-livecode <
use-livecode@lists.runrev.com> wrote:

> That sounds super if it were true, but I don't think it is as RunRev (as
> they then were) dropped support for SPARC, UNIX and so
> forth a long time before they began work on mobile platforms.
>
> Richmond.
>
>
___
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: Livecode server UNIX version (not Linux).

2020-10-16 Thread Bernard Devlin via use-livecode
 Hi Heriberto

Back in the day (20 years ago) the engine/IDE ran on FreeBSD and various
proprietary unixes.

The Linux server version has been seen to work on FreeBSD back in 2011
(after installing Linux compatibility layer).

http://runtime-revolution.278305.n4.nabble.com/Yay-Victory-RevServer-runs-on-FreeBSD-with-Linux-Compat-installed-td3445454.html

You _might_ be able to get that to work now.  I doubt it would be supported
by Livecode.  What amazes me nowadays is just how much more complex things
are than they were 20 years ago - looking at the compatibility matrix for
Livecode dependencies on OS version, XCode version, device version -- all
just to produce apps that run on iOS:

https://livecode.com/docs/9-5-0/faq/faq/

If someone had said 20 years ago that a small company in Scotland could
manage that kind of complexity people would have laughed in disbelief.  But
rightly LC saw where the future was headed with mobile computing and they
obviously had to make sacrifices along the way (e.g. FreeBSD, etc) to be in
a situation to take on this level of complexity.

HTH Bernard

On Thu, Oct 15, 2020 at 5:08 PM Heriberto Torrado via use-livecode <
use-livecode@lists.runrev.com> wrote:

> Dear all,
>
> I have a question.   It's not a very important question and I don’t want
> to bother you y 'all, so if you think is off-topic, feel free to not to
> respond,  I'm just curious.
>
> I've been working with Livecode for almost five years, and I never saw a
> LiveCode server  "UNIX" version.
>
> The LiveCode Wikipedia’s article says this: LiveCode runs on iOS, Android,
> OS X, Windows 95 through Windows 10, Raspberry Pi and "several variations
> of Unix (I think is just means Linux)".
>
> I'm thinking about tinkering with a FreeBSD server and LiveCode server,
> but I didn't see a "UNIX" version, so I suppose that I have to compile it.
> Have any of you installed LiveCode server on FreeBSD (or Solaris)?
>
> I'm just thinking, but maybe there's a small niche working  with LiveCode
> server on BSD or Solaris (still many companies use them and not many people
> develop software for this platforms nowadays).
> We have a few companies in Spain (my country) still using Solaris or BSD
> servers (mainly in the Graphic arts business).
>
> PS, I currently work with LiveCode Server on Centos, but it could be
> interesting  to test it in FreeBSD.
>
> Best regards/ Saludos cordiales/ Cordialement
>
> Heriberto Torrado
> ​Chief Technology Officer (CTO)
> ​Director de informática
> Directeur informatique
>
> https://networkdreams.net
>
>
>
>
> ___
> 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: Fun with the templateimage

2020-10-15 Thread Bernard Devlin via use-livecode
This works:

export snapshot from rect tRect of window tWinID at size tDim1,tDim2 to
file tFile as GIF

I saw a report in the forum saying that this scaled incorrectly if the
snapshot was of a text field, but in my tests this was not a problem.

On Thu, Oct 15, 2020 at 11:34 AM Bernard Devlin  wrote:

> Richard, just FYI
>
> The dictionary (LC 9.5.1) does say that "export snapshot ... at size
> x,x..." is supported since version 6.0, but I can't get it to change the
> size of the exported image.  I'll add a note here if I get it to work.
>
> On Thu, Oct 31, 2019 at 4:38 PM Richard Gaskin via use-livecode <
> use-livecode@lists.runrev.com> wrote:
>
>> The "import snapshot" command had an "at size" option added several
>> versions ago to facilitate some scaling tasks:
>>
>> import snapshot from the selectedObject at size 100,100
>>
>> But oddly, no such option has been added to the "export snapshot" command.
>>
>> --
>>   Richard Gaskin
>>   Fourth World Systems
>>
>>
>>
___
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: Fun with the templateimage

2020-10-15 Thread Bernard Devlin via use-livecode
Richard, just FYI

The dictionary (LC 9.5.1) does say that "export snapshot ... at size
x,x..." is supported since version 6.0, but I can't get it to change the
size of the exported image.  I'll add a note here if I get it to work.

On Thu, Oct 31, 2019 at 4:38 PM Richard Gaskin via use-livecode <
use-livecode@lists.runrev.com> wrote:

> The "import snapshot" command had an "at size" option added several
> versions ago to facilitate some scaling tasks:
>
> import snapshot from the selectedObject at size 100,100
>
> But oddly, no such option has been added to the "export snapshot" command.
>
> --
>   Richard Gaskin
>   Fourth World Systems
>
>
>
___
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: Getting URL to Work in specialFolderDocuments

2020-10-05 Thread Bernard Devlin via use-livecode
 Hi

I get a headache looking at concatenations like this.  I don't use them
unless I'm constructing a simple string.

  put URL (tHttpsServerRoot & sStoryTitle & "/" & sStoryTitle &
".livecode") into tLocalDocumentsStories & sStoryTitle & "/" & sStoryTitle
& ".livecode"

It will be a cold day in hell when I don't use Merge() to construct
non-simple strings.

*put* "blurb1" into sStoryTitle

*put* "https://dev.himalayanacademy.com/media/stories/; into
tHttpsServerRoot

*put* specialFolderPath("Documents") & "/SivaSivaApp/media/stories/" into
tLocalDocumentsStories

*if* char -1 of tHttpsServerRoot is "/" *then* *delete* char -1 of
tHttpsServerRoot ## deleting "/" to make merge() template clearer

*if* char -1 of tLocalDocumentsStories is "/" *then* *delete* char -1 of
 tLocalDocumentsStories

*put* merge(  "[[tHttpsServerRoot
]]/[[sStoryTitle]]/[[sStoryTitle]].livecode" ) into tSourceURL

*put* merge( "binfile:[[tLocalDocumentsStories]]/[[
sStoryTitle]]/[[sStoryTitle]].livecode" ) into tDestURL
put URL  tSourceURL into URL tDestURL


On each of the above lines one can use breakpoints to check the values of
the variables and at the last line check that the URLs to use are correct.

Merge() will ignore spaces within [[]] and outside of the quotes around the
template string, so one can line up placeholders to check that all the
template parts match visually.

HTH
Bernard

On Fri, Oct 2, 2020 at 11:00 PM Sannyasin Brahmanathaswami via use-livecode
 wrote:

> on dataServerStack
>local tHttpsServerRoot,tLocalDocumentsStories
>put "https://dev.himalayanacademy.com/media/stories/; --into
> tHttpsServerRoot
>put specialFolderPath("Documents") & "/SivaSivaApp/media/stories/" into
> tLocalDocumentsStories
>put URL (tHttpsServerRoot & sStoryTitle & "/" & sStoryTitle &
> ".livecode") into tLocalDocumentsStories & sStoryTitle & "/" & sStoryTitle
> & ".livecode"
> end dataServerStack
>
> it goes badly to documents folder,
>
> on dataServerStack
>local tHttpsServerRoot,tLocalDocumentsStories
>put "https://dev.himalayanacademy.com/media/stories/; into
> tHttpsServerRoot
>put specialFolderPath("Documents") & "/SivaSivaApp/media/stories/" into
> tLocalDocumentsStories
>put URL (tHttpsServerRoot & sStoryTitle & "/" & sStoryTitle &
> ".livecode") into tLocalDocumentsStories & sStoryTitle & "/" & sStoryTitle
> & ".livecode"
> end dataServerStack
>
> it goes badly to documents folder,
>
> ?
> put URL (tHttpsServerRoot & sStoryTitle & "/" & sStoryTitle & ".livecode")
> into tLocalDocumentsStories & sStoryTitle & "/" & sStoryTitle & ".livecode"
>
> Compile error line 119 (Handler: bad command) near "&", char 84
>
>
> BR
>
>
> ___
> 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: Accessing files on a local network file server - BEST PRACTICE?

2020-09-25 Thread Bernard Devlin via use-livecode
>>
So that in case (1) , we regenerate the file and in case (2) we present
a warning that the server is disconnected or down?
<<

For (2) you could save an empty file (as a flag of availability) and set
it's attribute to hidden.

If the file is not there when you get the files, then the folder is
unavailable.  Because it's a hidden file, users won't see it if they open
the folder using (most) other applications (Explorer only shows such files
if the 'show hidden' checkbox is ticked).

This file can then be filtered out based on file name when you get the
files of that folder to present to the user. It's just your internal check
for folder availability.  I guess there are some "net use" commands that
would show which shares are available/unavailable, but it's a long time
since I did that kind of system admin.

To programmatically set a file attribute to hidden on Windows use the shell
command to issue this "attrib +h _ServerAvailable.txt"

Hope that helps.

On Thu, Sep 24, 2020 at 9:54 PM Paul Dupuis via use-livecode <
use-livecode@lists.runrev.com> wrote:

> This question may have an obvious answer, but I am brain dead today, so
> here goes:
>
> I have a app installed with some customers in a Windows network
> environment. They have all their laptops set so that their Home
> directories are on a local network server.
>
> So, where as on a local laptop, specialFolderPath("documents") might
> return "C:/Users//Documents" as the path, for them
> specialFolderPath("documents") returns something like
> "//s1.somedomain.com/mountPoint//Documents"
>
> We save a number of folders and files in the user's Documents folder. In
> some cases, if the file is not present (i.e. if there is a file tFile is
> false) we regenerate the file from a default.
>
> Sometime, the server is offline or the user is disconnected from the
> network.
>
> What is a good way to differentiate between:
>
> 1) one of these files is not present so must be regenerated, i.e
>
> if there is not a file (specialFolderPath("documents"))
> then
>-- regenerate the file tFile at location specialFolderPath("documents")
>
> Versus
>
> 2) The Server (//s1.somedomain.com/ is disconnected to down?
>
> So that in case (1) , we regenerate the file and in case (2) we present
> a warning that the server is disconnected or down?
>
> Mu gut says this should be simple, but I just can not wrap my brain
> around it today.
>
> ___
> 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: Adding items to a group

2020-08-26 Thread Bernard Devlin via use-livecode
Whilst Mark provided what Phil needed, I too cannot believe how easy and
reliable Geoff's Navigator is.  I use it for everything. When I need an
Inspector a double-click on a line in Navigator brings it up.  I love it.

On Wed, Aug 26, 2020 at 7:14 PM Bob Sneidar via use-livecode <
use-livecode@lists.runrev.com> wrote:

> Also, Geoff Canyon wrote a GREAT utility called Navigator, which among
> other things, allows you to drag objects around in a hierarchical list. If
> you drag an object in a group to somewhere outside that group, into another
> group let’s say, it does all the relaying for you.
>
>
___
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: Cross Platform Font Layout - current workarounds

2020-08-25 Thread Bernard Devlin via use-livecode
I wondered how Java dealt with the issue of rendering fonts across
platforms. It seems they have (had) the same problems you are experiencing.

https://stackoverflow.com/questions/12202758/how-to-make-java-swing-application-has-a-consistent-font-in-windows-and-linux

https://stackoverflow.com/questions/11994407/using-java-to-develop-cross-platform-fonts-scaling-differently-on-different-pla

Those posts are 8 years old (that they've been viewed 1000s of times
suggests this is pain that was felt by many). Maybe things have changed for
the better in Javaland. But it's more likely that Java is less used for
desktop apps than it was 8 years ago. I know that I only use one Java app
across my computers.

Regards
Bernard

On Tue, Aug 25, 2020 at 5:03 AM Sean Cole (Pi) via use-livecode <
use-livecode@lists.runrev.com> wrote:

> Yet again, everyone happy to make excuses in accepting their fate to have
> to do all the work twice. I might as well code it in Java, C# and Swift for
> all the hassle I get in LC. It's laughable how excuses are made for them.
>
> 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: breakpoint causes 3 second delay then exit to top

2020-07-27 Thread Bernard Devlin via use-livecode
Here's another data point for this oddity.

Once the IDE starts tripping out from breakpoints and I restart Livecode to
fix the problem, the behavior in which the breakpoints stop working has
"detached" from the object pointing to it.  I say "detached", as the object
still shows in the Inspector that it is pointing to the behavior-containing
control and choosing to edit the behavior script from the object's
Inspector opens up the relevant behavior in the Script Editor.  Breakpoints
work once again when I once more point the object's Inspector at the object
it knows contains the behavior script

Somewhere inside LC this connection is being lost.  At the moment the only
variable I've been able to isolate as a cause is: how many hours I've had
LC open working on this project.

On Fri, Jul 17, 2020 at 7:35 PM Bernard Devlin  wrote:

> I found later that the same thing was happening away from this code block,
> and with just a red breakpoint.
>
> As I don't restart my PC for weeks on end, I quit LC and restarted.  Both
> of these strange error situations disappeared.  Which suggests that it is
> some complex condition (either within my app or within the IDE) that leads
> to this situation.
>
>
___
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: breakpoint causes 3 second delay then exit to top

2020-07-17 Thread Bernard Devlin via use-livecode
I found later that the same thing was happening away from this code block,
and with just a red breakpoint.

As I don't restart my PC for weeks on end, I quit LC and restarted.  Both
of these strange error situations disappeared.  Which suggests that it is
some complex condition (either within my app or within the IDE) that leads
to this situation.

On Fri, Jul 17, 2020 at 5:09 PM Ludovic THEBAULT via use-livecode <
use-livecode@lists.runrev.com> wrote:

>
>
> > Le 17 juil. 2020 à 13:52, Bernard Devlin via use-livecode <
> use-livecode@lists.runrev.com> a écrit :
> >
> > This is mystifying in so many ways.  I've seen this a few times and I'm
> at
> > a loss to explain it. It's LC 9.5.1 on Windows 10.
> >
> > *When the code gets to the above breakpoint, the debugger exits to top.
> It
> > takes about 3 seconds with the debugger sitting at the breakpoint before
> > the engine exits to top.*
> >
>
> Hello,
>
> I've already had the same behavior but I noticed that it happens when
> there is an error in the executed code (but an error that doesn't block the
> execution of the script !). By fixing the script, the breakpoint then works
> as it should.
>
> Maybe there is an error in your script (before the code you displayed
> here).
>
> Ludovic
> ___
> 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


breakpoint causes 3 second delay then exit to top

2020-07-17 Thread Bernard Devlin via use-livecode
 This is mystifying in so many ways.  I've seen this a few times and I'm at
a loss to explain it. It's LC 9.5.1 on Windows 10.

I have a simple conditional, where I want to stop execution and debug some
following actions. Here's the code:

*try*

  *if* "15474" is in tControlLongId *then*

   *put* the long time && tControlLongId

   *breakpoint*

  *end* *if*

*catch* tErr

  *put* "error:" && tErr

*end* *try*



*When the code gets to the above breakpoint, the debugger exits to top. It
takes about 3 seconds with the debugger sitting at the breakpoint before
the engine exits to top.*


* A long id containing "15474" is put into the message box. *


*If I comment out that condition, then the code runs to completion.  *

*The Catch clause never catches any error.*

* If I try to use a red breakpoint in the gutter, the debugger exits to top
(again, after the IDE has sat at the breakpoint for about 3  seconds). *

*So the condition triggers.  The variable has the tested value. The value
of that variable can be expressed.  But it looks like the very act of
invoking the debugger causes the code to fail, and in a way where even a
try/catch block cannot intercept.*


*Mystifying. And a significant bug IMO, rendering interactive debugging
impossible.*


*Kind regards, Bernard *
___
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: DataGrid question...

2020-04-03 Thread Bernard Devlin via use-livecode
Hi Bernd

Your modTableField is very impressive.  Reminds me of the work of the late,
great Eric Chatonet (author of many works on Hypercard).  I see only one
slight problem: after I click into a cell a trace of the I beam is left
behind on leaving the cell (this becomes visible on selecting another
row).  I guess this is a LC bug.  If I click away (e.g. on the "examples"
tabbed button) the trace vanishes.  If I click on the column header to sort
and re-sort, the I beam trace is still there in the field in which I
clicked. If I click in the output field that also does not remove the I
beam vestige.

But apart from that visual vestige, in 20 minutes of playing with mTF I was
very impressed.  BTW, in case you have not seen this yourself, this vestige
was visible for me with LC 9.5.1 on Windows 10.

Regards, Bernard

On Mon, Mar 30, 2020 at 10:44 PM Niggemann, Bernd via use-livecode <
use-livecode@lists.runrev.com> wrote:

>
> >Peter Bogdanoff wrote:
>
> >Does anyone have a link to obtain Bernd's modTableField ?
>
> Here is a link to the latest version of modifiedTableField
>
> berndniggemann.on-rev.com/mtf/modTableField.zip<
> http://berndniggemann.on-rev.com/mtf/modTableField.zip>
>
> there is a zip of a demo stack with a helper stack and a .rtf document
> with the API
>
> Use the demo stack to explore modTableField. Have a little patience to
> explore. It shows many options. Please have a look at the API document to
> get a feeling how to address and query modTableField.
>
> If you want to use modTableField in a project copy the group
> "modTableField 1" from the demo stack to your stack and use the API to work
> with it.
>
> This version is HTML5 safe. There was a superfluous wait in former
> versions. Hermann used tinyDict with his HTML5 demo of an IDE.
>
> tinyDictionary uses modTableField
>
> There are a couple of users that seem quite content with modTableField
> (thanks Jerry for mentioning it)
>
> I would be happy to know if anybody uses modTableField in a project.
> (aside from the people that mailed me before).
>
> If you have questions please feel free to email me.
>
> Kind regards
> Bernd
>
>
> ___
> 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: The Roadmap

2019-10-30 Thread Bernard Devlin via use-livecode
On Tue, Oct 29, 2019 at 7:00 PM Richard Gaskin via use-livecode <
use-livecode@lists.runrev.com> wrote:

>
> If it were up to me I'd ditch it altogether.
>

 Whatever frustrations I have with Livecode they are vastly outweighed by
the things that it can do and the things it can do now that it couldn't do
a decade ago.

I don't think people appreciate what an incredible achievement it is.  As I
used to be a Lotus Notes developer, I can see the contrast.  IBM recently
sold their Lotus technologies to an Indian company for $1.8bn. Over the
last ten years a company the size of IBM could not successfully manage the
development of a cross-platform application development environment.   For
an entire decade IBM only added the v.9 release and fixpacks (and Notes 9.0
was actually Notes 8.5.4 but re-branded).

Notes used to exist in native clients for OS/2, Windows, MacOS.  IBM
jettisoned the native clients and moved to running the entire thing inside
the JVM.  It took IBM about 15 years to move from Notes 7 to Notes 10.
When I recently downloaded Notes 10 to do some programming on thousands of
emails (the very thing where Notes should shine) the client crashed
repeatedly on trying to import the emails.  Bear in mind that Notes even
has a menu action to import emails from a folder, so importing emails to
process the text is not even something that requires any programming, it's
that basic to the product.  Having moved from programming in C with all the
possible problems with pointers and memory allocation, Notes was moved to a
language with garbage collection and memory management and still the client
crashes doing basic things it could do 15 years ago.

By comparison in the last 10 to 15 years Livecode has added app delivery
for iOS, Android, Linux, Windows64 and HTML5 to Windows32 and OSX.  That's
more than IBM ever managed to do and a movement in the opposite direction
to that taken by IBM.  And as well as doing this Livecode re-architected
the engine.   IBM's major new feature over the last decade (XPages) looks
like it is now almost certainly going to be thrown away.  By moving to the
JVM for the development environment IBM was banking on delivering most
Notes apps as web apps.

Obviously the Notes servers do a whole range of things that Livecode
doesn't do.  But these server-side technologies were mostly in the Notes
product 15 years ago. Going back 15 years ago and Livecode was at v2.5.
Remember back then?  The clipboarddata was a new thing.  We didn't have a
built-in web browser, nor multi-dimensional arrays, nor unicode, nor the
datagrid, nor behaviors, nor widgets, nor xslt,  nor the enhanced liburl.
That's just off the top of my head.

IBM with all their resources and access to capital markets couldn't hold
themselves to their own limited roadmap.  Livecode has successfully moved
forward with a cross-platform solution when IBM couldn't do so.  And
without having $billions to play with.  They shouldn't attempt to hold to a
roadmap when the elephant in the room can't do it.  And IBM never allowed
businesses to use Notes for free nor did they open the source code.
Livecode did that and has a free edition which contains probably 99% of the
functionality of the most expensive license.

I think many of us lose sight of this small company's achievements.

Regards, Bernard
___
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: Thank you for the 9.0.5 update

2019-10-10 Thread Bernard Devlin via use-livecode
>>
I created a play/test stack yesterday - 1 button 2 functions and a mousup
of a couple of lines no more than 30 lines of code. Did some debugging and
got a crash
in less than 15 minutes.
<<

I was getting multiple random crashes with a test stack in 9.5 that were
apparently totally unrelated to debugging.   Having not seen such crashes
for years, I assumed that it was the introduction of the 64bit build for
Windows. So I switched to using the 32bit build for Windows and the random
crashes continued.  I haven't got anywhere with finding a reproducible
pattern.

Regards, Bernard
___
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: livecode connect framework

2019-04-19 Thread Bernard Devlin via use-livecode
Does anyone know if LiveCode Connect ever come out of beta?  I have a
project coming up which uses OpenAPI and it would save me a lot of work if
Connect is available.

Regards, Bernard

On Fri, Jun 9, 2017 at 9:03 PM Tom Glod via use-livecode <
use-livecode@lists.runrev.com> wrote:

> if the only hint is that its available to all attendees of the conference,
> we can't be sure i guess we will find out...what a great feature for LC
> and its bright future.  In a way i am indifferent, since I can build my own
> handlers to talk to any api i want and i do want livecode inc to
> prosper ALOT, so if they decide its a non-community feature thats ok with
> me.
>
>
___
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


Enhancement: transparently persist/retrieve large amounts of array data as JSON

2019-02-04 Thread Bernard Devlin via use-livecode
As far as I can see, providing this enhancement is (in essence) no more
than the addition of one line to Livecode's compilation process.

Details here:

https://quality.livecode.com/show_bug.cgi?id=21821

Seems to me to be a huge gain for virtually no effort.  Array data could be
stored and indexed outside of a stack.

Regards
Bernard
___
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


HTML5 severely undocumented

2018-12-24 Thread Bernard Devlin via use-livecode
As I'm trying to work out why it is that the datagrid does not work in
HTML5, I've realised that there are many really basic features of Livecode
that do not work in HTML5.  For instance, in a field the backspace key does
not work.  Moreover, it appears that the rawKeyDown message is not
implemented, so I can't implement my own backspacekey handler.

Surely the Dictionary should at least indicate which features are not
implemented?  When I start to go through the Dictionary, I can'd even
discover if something is implemented without building an HTML5 app and then
seeing if that command works.

We can't even rely on the Dictionary NOT mentioning HTML5 in the
Platform/OS category of the Dictionary entry as being indicative of whether
or not something is supported.  For example, the Dictionary entry for
"Create [button]" does not say that this is implemented in HTML5 but it
works (I found out through trial and error). So a command that's been in
xCard for some decades is not indicated as being implemented for HTML5.
But rawKeyDown has probably been in Livecode/Metacard for 20 years, but
that too is not listed as being implemented in the HTML5 engine. This
vagueness about what should work and what doesn't is woefully inadequate.

It is over 4 years since the HTML5 standalone was first funded, and I think
3 years since it was first working (for some definition of working). Surely
it should by now at least be documented what works and what does not?
Considering that Runrev expects most customers to pay an additional license
for HTML5 deployment this is very disappointing.

Here's what Kevin said in 2014. "Anywhere you can run an HTML5 capable
browser, you will be able to run your app."  Now I'm sure that there are
limitations imposed by the emscripten engine that Livecode over which
Livecode have no control.  But at the very least, there should have been a
list of Livecode features that don't work in HMTL5.

I'm not expecting any reply to this until after christmas.  But I need to
get it off my chest.  After being away from Livecode for some years, I am
disappointed to find I can't even work out if I can use HTML5 except
through trial and error.

Regards
Bernard
___
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


Datagrids in HTML5

2018-12-24 Thread Bernard Devlin via use-livecode
Hi folks

Am I mistaken in thinking datagrids are now supposed to work in HTML5?  My
datagrid and its contents display, but clicking on column headers has no
effect.

I ensured that the datagrid library was being included in the standalone
settings. The datagrid template substack of the mainstack is listed in the
HTML5 app as a substack.  However, when a pertinent "do" command e.g.

set the dgProps["show header"] of grp "datagrid 1" to false

is issued in the HTML5 app, it has no effect. I built the same stack for
Windows and HTML5, and the datagrid works as expected in the Windows
executable.

If datagrids don't work for HTML5, then I'm surprised they display at all.
Is there a list somewhere of what is supposed to work in HTML5 and what
does not work?

Regards
Bernard
___
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: WordReport Registration

2018-12-10 Thread Bernard Devlin via use-livecode
Thanks Matthias.  I did try variations of what I thought were a likely
names for the registration handler (but that wasn't one of my guesses for
the likely handler name).

The file "WordReport Registered Uses.pdf" is not included in the zip file
that is downloaded from within our user area on livecode.com.  I saw it
mentioned somewhere in the stacks/pdfs.  If you could email me a copy of
that file to bdrunrev at gmail, that might help too.

The beauty of the WordReport concept is that it puts report-writing in the
hands of the user.

Regards
Bernard

On Mon, Dec 10, 2018 at 9:17 AM Matthias Rebbe via use-livecode <
use-livecode@lists.runrev.com> wrote:

> Bernard,
>
> this is done with registerWordReport.
>
> The following is an excerpt from the file "WordReport Registered Uses.pdf"
> which was included in the download file, at least when i downloaded it.
>
>
___
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


WordReport Registration

2018-12-09 Thread Bernard Devlin via use-livecode
Hi all

I've got a copy of this as part of one of my previous Livecode purchases.
How does one go about using the registration code?

The "Try WordReport" stack says either to buy a registration code *or* to
simply use the code if one has it. Neither of the PDF files which come with
the stack explains how to use the registration code.

I'd contact the provider of the software but I've seen numerous messages
saying he doesn't respond to emails that I thought I'd ask an existing user.

Regards
Bernard
___
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: LiveCoders from London, lets meet!

2018-11-13 Thread Bernard Devlin via use-livecode
I live outside London, but easy enough to get in to meet up.

On Sat, Nov 10, 2018 at 2:14 PM Keith Martin via use-livecode <
use-livecode@lists.runrev.com> wrote:

> Hi Andre,
>
> I'm based in London – I live in Mitcham/Tooting and I work in Elephant
> & Castle. I'm game!
>
> k
>
>
> On 9 Nov 2018, at 17:04, Andre Alves Garzia via use-livecode wrote:
>
> > Hey Friends,
> >
> > Who here is from London or nearby and would be interested in regular
> > meetups? We could meet once a month or so in a pub or quieter setting.
> > I am thinking of informal meet & drink, chatting.
> >
> > Cheers
> >
> > andre
> >
> >
___
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: LC 9.0 Windows - tooltip broken for new datagrid?

2018-10-17 Thread Bernard Devlin via use-livecode
Thanks.
If it's not supposed to work then best to remove it from the property
inspector so the product doesn't appear buggy to new users.

I have Livecode 5.5 on another machine.  I'll see if the tooltip works
there.  I see no reason why a datagrid (even if it is a fancy group)
shouldn't have a tooltip the way other UI controls have one.  I guess these
features are not even used on mobile interfaces, so maybe they are "legacy".

Either way I'll enter this as a bug for the 9.x series.
Bernard

On Tue, Oct 16, 2018 at 4:18 PM Bob Sneidar via use-livecode <
use-livecode@lists.runrev.com> wrote:

> Actually, everything in a datagrid is an object of some sort with it's own
> tooltip (empty by default). There would have to be special code to display
> the tooltip for a datagrid instead of the tooltip for the object you are
> pointing at.
>
> I tested in a new stack and the tooltip for a datagrid NEVER displays. Not
> even sure why it's there.
>
> 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


LC 9.0 Windows - tooltip broken for new datagrid?

2018-10-16 Thread Bernard Devlin via use-livecode
Am I right that this is broken?

A new stack with a single button will show the button's tooltip when the
pointer hovers over it.
Add a datagrid, and set the tooltip for the datagrid through the inspector.
The tooltip does not display no matter where the pointer is in the datagrid.

Regards
Bernard
___
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


sqlite config/compilation files on github

2018-08-12 Thread Bernard Devlin via use-livecode
Hi all

Could anyone point me to the file(s) on github determining how the sqlite
external is configured and built.  I've had a look round, but I can't seem
to find the relevant file.  I'm looking for the location at which the
full-text search option is configured.

Regards, Bernard
___
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: LC on Windows - slow for others or just me?

2018-05-01 Thread Bernard Devlin via use-livecode
I've been using the final version of LC9 on Windows 7.  I don't have these
kinds of problems with any other program.  I didn't notice it with earlier
DP versions on the same machine (although admittedly I would leap-frog over
many DP point versions, so I can't say at what point this problem was
introduced).

In the default install it is so slow it is unusable.  I switched off many
of the recent auto-complete, live error, etc. features, and it became more
usable.

On Wed, May 2, 2018 at 12:12 AM, Andre Garzia via use-livecode <
use-livecode@lists.runrev.com> wrote:

> I use LC 9 on a Surface Pro 4 Machine, it is almost unusable. Everything is
> as slow as molasses. Typing a script sometimes takes many seconds to
> respond and it even goes as far as missing keystrokes. Trying to drag and
> drop stuff onto the stack is also horrible. I don't have any plugins
> anymore, this is a fresh install.
>
> I've been using this machine as my main machine for more than a year. LC is
> so slow for me (on a fast machine) that I go for my old mac sometimes to
> work. I have the impression that I am the only windows user here.
>
> On Tue, May 1, 2018 at 3:06 PM, Richard Gaskin via use-livecode <
> use-livecode@lists.runrev.com> wrote:
>
> > Ralph DiMola wrote:
> >
> > > I have not reported it. I wasn't sure that my setup (VMs/bloated Adobe
> > > Products...) was causing it. It's time to open a QCC report on this.
> > > Do you want to report it or should I?
> >
> > I had put off reporting it since I run Win only in VMs right now, and had
> > been thinking it was just me.
> >
> > If you have a native install it may carry more weight.  Even better if
> you
> > have anything more actionably specific than my current whingeing
> > abstraction, "it's slow".
> >
> > I'm trying to get some time for comparative unit tests, but this week I'm
> > up against some hairy deadlines so not sure when I can get those
> together.
> >
> >
> > --
> >  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
> >
>
>
>
> --
> http://www.andregarzia.com -- All We Do Is Code.
> http://fon.nu -- minimalist url shortening service.
> ___
> 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: debugging a CEF browser instance

2017-07-08 Thread Bernard Devlin via use-livecode
Hi Mark,

I suspect this is the problem. Is there a list of things which the LC
embedded browser forbids? Is there an explanation of the rationale for the
embedded browser to behave differently from a default browser?

This should be noted in the Dictionary, perhaps in the top entry
of: com.livecode.widget.browser.  Those who try Livecode for the first time
and discover that the browser widget "doesn't work reliably" (which is how
it appeared to me) may simply give up on Livecode.  Indeed, it might be
better if a warning appears when a page attempts to do something that has
been restricted (assuming the browser widget throws an exception when such
things happen).

Regards
Bernard

On Fri, Jul 7, 2017 at 1:25 PM, Mark Waddingham via use-livecode <
use-livecode@lists.runrev.com> wrote:

>
>
> The fact it doesn't work in either Mac's WebView of CEF may suggest that
> the webpage
> is doing something which is 'not allowed' by default
___
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: debugging a CEF browser instance

2017-07-07 Thread Bernard Devlin via use-livecode
Hi Mark,

I've now found an app which has embedded the CEF browser within it.  This
embedded CEF browser has no trouble loading the page in question.

This makes it very clear that the problem relates to the way that LC is
embedding the CEF browser. This is also confirmed by the fact the page will
not load in LC browser widget on OS X.  Is there any way of debugging the
browser that is embedded in OS X? If not I'll have to put this project on
hold without any way to debug any browser instance.

Regards
Bernard

On Thu, Jul 6, 2017 at 8:28 AM, Mark Waddingham via use-livecode <
use-livecode@lists.runrev.com> wrote:

> On 2017-07-06 08:39, Jonathan Lynch via use-livecode wrote:
>
>> Why?
>>
>
> In general there is a cost to logging - particularly CEF's which is very
> verbose. On Windows you might not notice (as stdout/stderr output is
> generally dumped to the equivalent of /dev/null), but on Linux if you
> happen to be working from the command-line and running UI stuff using the
> browser widget from there then you'll find your terminal flooded with CEF
> logging (and I mean flooded!).
>
> I don't think anyone has asked specifically about having it configurable
> before now - although I noticed it would be useful last month (
> http://quality.livecode.com/show_bug.cgi?id=19862) whilst attempting to
> figure out why the browser widget only works on *some* linux installs
> (seems to be somewhat independent of distribution - the workarounds some
> people have found with regards the locale don't seem to work anymore).
>
> Unfortunately, we aren't any closer to solving the linux problem... After
> at least three of us spending more hours than I'd care to comment on trying
> to figure out what is happening there, we're working through a couple of
> tasks so that we can more easily update CEF to the latest version.
>
> In particular, making it so that we can build our 'prebuilts' (currently
> ICU, OpenSSL and Curl) on vulcan (doing it manually is arduous and
> intensely error prone). We can then move the building of the CEF library
> component to a prebuilt and automate its generation based on a version tag
> (we can thank Spotify for taking over the management of binary releases of
> CEF - http://opensource.spotify.com/cefbuilds/index.html - as they've
> made it much much easier).
>
> So we're currently involved in a (small) yak-shave in this regard...
> Although one which will also mean we can solve a couple of other issues -
> the size of ICU data (has anyone noticed that the 9 engines are somewhat
> bigger than 8? That's down in good part to the ICU data), and also the
> several minute increase per platform in build time due to the Skia update.
> I also hope that this means that over time we can eliminate the thirdparty
> submodule entirely - which would be one less point of friction in our
> source base.
>
> Incidentally, Bernard and Jonathon - I take it you are using the browser
> widget on Windows? (The reason I ask that is because CEF is only used on
> Windows and Linux - Mac/Android and iOS all use the built-in browser - all
> three are WebKit derived, like CEF).
>
> Warmest Regards,
>
> Mark.
>
> --
> Mark Waddingham ~ m...@livecode.com ~ http://www.livecode.com/
> LiveCode: Everyone can create apps
>
>
> ___
> 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: debugging a CEF browser instance

2017-07-06 Thread Bernard Devlin via use-livecode
Hi Mark,

I had no doubt that the debug log was disabled because enabling it by
default would be expected to cause some significant slowdown. However,
since this is a configurable option, I would have hoped that (at the very
least) it could be turned on for situations such as the situation in which
I'm in (a page that works in all other browsers I've tried but that doesn't
work in either Livecode 8.1.4 on Windows or OS X - with no visible errors
there is no hint as to why it is not working).  The page is not mine and
therefore it's not like there is only a trivial amount of javascript for me
to test line by line (it was easier for me to discover why LC's CEF browser
doesn't allow debugging than it would be to find out why the web page does
not work).

That the page does not work in LC on BOTH OS X or Windows, suggests that it
is something discrepant with the way the browser widget works in LC (and
not just some limitation in the CEF widget).  The page not working on two
different implementations of the browser widget in LC suggests there may be
a wider set of problems developers will encounter (i.e. there may be many
more cases where LC's browser will not work the way that other browsers
work).  Without debugging information, I don't see how such discrepancies
will ever be fixed.

Hopefully it will be quicker and cleaner to enable logging (at least in the
CEF browser) by setting an environment variable or checking for a command
line parameter.  I believe that the verbosity of CEF's logging can be
configured. That logging level might also be worth setting at startup (in
case the amount of logging is so excessive that the IDE is unresponsive).

That the CEF logging on Linux is flooding the terminal window should also
be fixable.
https://stackoverflow.com/questions/11844208/linux-terminal-that-is-currently-running-and-logging-to-stdout-how-do-you-silen

Regards
Bernard

On Thu, Jul 6, 2017 at 8:28 AM, Mark Waddingham via use-livecode <
use-livecode@lists.runrev.com> wrote:

> On 2017-07-06 08:39, Jonathan Lynch via use-livecode wrote:
>
>> Why?
>>
>
> In general there is a cost to logging - particularly CEF's which is very
> verbose. On Windows you might not notice (as stdout/stderr output is
> generally dumped to the equivalent of /dev/null), but on Linux if you
> happen to be working from the command-line and running UI stuff using the
> browser widget from there then you'll find your terminal flooded with CEF
> logging (and I mean flooded!).
>
> I don't think anyone has asked specifically about having it configurable
> before now - although I noticed it would be useful last month (
> http://quality.livecode.com/show_bug.cgi?id=19862) whilst attempting to
> figure out why the browser widget only works on *some* linux installs
> (seems to be somewhat independent of distribution - the workarounds some
> people have found with regards the locale don't seem to work anymore).
>
> Unfortunately, we aren't any closer to solving the linux problem... After
> at least three of us spending more hours than I'd care to comment on trying
> to figure out what is happening there, we're working through a couple of
> tasks so that we can more easily update CEF to the latest version.
>
> In particular, making it so that we can build our 'prebuilts' (currently
> ICU, OpenSSL and Curl) on vulcan (doing it manually is arduous and
> intensely error prone). We can then move the building of the CEF library
> component to a prebuilt and automate its generation based on a version tag
> (we can thank Spotify for taking over the management of binary releases of
> CEF - http://opensource.spotify.com/cefbuilds/index.html - as they've
> made it much much easier).
>
> So we're currently involved in a (small) yak-shave in this regard...
> Although one which will also mean we can solve a couple of other issues -
> the size of ICU data (has anyone noticed that the 9 engines are somewhat
> bigger than 8? That's down in good part to the ICU data), and also the
> several minute increase per platform in build time due to the Skia update.
> I also hope that this means that over time we can eliminate the thirdparty
> submodule entirely - which would be one less point of friction in our
> source base.
>
> Incidentally, Bernard and Jonathon - I take it you are using the browser
> widget on Windows? (The reason I ask that is because CEF is only used on
> Windows and Linux - Mac/Android and iOS all use the built-in browser - all
> three are WebKit derived, like CEF).
>
> Warmest Regards,
>
> Mark.
>
___
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: debugging a CEF browser instance

2017-07-05 Thread Bernard Devlin via use-livecode
Replying to myself, for anyone else who thinks it ought to be possible to
debug a browser widget.

Ability to debug has been deliberately turned off.
https://github.com/livecode/livecode/commit/ccfab94499425b6245f98b3328ebd043876f9be7
___
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: debugging a CEF browser instance

2017-07-05 Thread Bernard Devlin via use-livecode
Thanks Jonathan for confirming that there is no debug mechanism for the
browser widget.

But what my examples shows is that one cannot rely on the browser widget to
present a working web page, even one that works in multiple standard
browsers (including Chrome).  This is not a web page of my construction,
but is one that is in use by other people.  So it's not that it's my
javascript that is failing, and therefore I can't carefully test each line
of javascript without taking their website apart and building it back up to
find the error.

LC needs to have some way of allowing at least the developer to see why a
page is not working within the widget.  The debuglog option of the CEF
browser widget shows
a) that the debuglog is being written to a system folder (and therefore
unlikely to be writeable)
b) that the browser widget has logging disabled anyway
c) that these options are hard-coded into the LC startup process

Anyone got suggestions about how this should be put forward as an
enhancement request/bug report?  It seems that a CEF executable ought to be
launchable with a supplied remote-debugging port. But that also does not
work with the libbrowser-cefprocess.exe which comes with Livecode.
http://magpcss.org/ceforum/viewtopic.php?f=6=11952#p22106

Regards
Bernard



On Wed, Jul 5, 2017 at 12:40 PM, Jonathan Lynch via use-livecode <
use-livecode@lists.runrev.com> wrote:

> Hi Bernard,
>
> I work with JS in the browser widget quite a bit.
>
> I have just gotten used to testing JS commands very carefully, step by
> step, because it does not really provide any debugging information.
>
> It would be nice to have that.
>
> Sent from my iPhone
>
>
___
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: debugging a CEF browser instance

2017-07-05 Thread Bernard Devlin via use-livecode
I tried with 9.0 dp7.

Still no luck. The webpage opens and works with both Chrome and Firefox.
But not with the browser widget in either LC 8 or LC 9.  I even tried with
LC8.1.4 on OS X and it fails there too.

I'm surprised no-one else has run into the need to access the debug log,
browser console or browser developer tools.

Regards
Bernard

On Tue, Jul 4, 2017 at 11:33 AM, Bernard Devlin  wrote:

> After a very long time away from Livecode, I'm trying to get back into it.
>
> On 8.1.4 in the Windows IDE I have a browser instance where a standalone
> web page with some Javascript code is not working.  The same page works if
> I load it into Chrome.
>
> So, I thought there might be a way to display the console of the browser
> instance, but I can't find a "chrome://..." command to do that. The most
> information I can get back from the browser instance is this:
>
> "C:/Program Files (x86)/RunRev/LiveCode Business 8.1.4/LiveCode Business"
> --browser-subprocess-path="C:\Program Files (x86)\RunRev\LiveCode
> Business 8.1.4\Externals\CEF\libbrowser-cefprocess.exe" --no-sandbox
> --lang=en-US --log-file="C:\Program Files (x86)\RunRev\LiveCode Business
> 8.1.4\debug.log" --log-severity=disable --locales-dir-path="C:\Program
> Files (x86)\RunRev\LiveCode Business 8.1.4\Externals\CEF\locales"
>
> So, I created a .cmd file where I launched the livecode .exe, passing in
> different values for
>
> --log-severity=disable
>
> but no alternative parameters in the .cmd file appear to be taken up (i.e.
> it looks like these CEF browser parameters might be hard-coded into the LC
> exe).
>
> I'm thinking there must be a way to debug CEF browser instances, but a
> search through the mail list didn't turn up anything that appeared relevant.
>
> Any suggestions?
>
> Regards
> Bernard
>
>
>
___
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


debugging a CEF browser instance

2017-07-04 Thread Bernard Devlin via use-livecode
After a very long time away from Livecode, I'm trying to get back into it.

On 8.1.4 in the Windows IDE I have a browser instance where a standalone
web page with some Javascript code is not working.  The same page works if
I load it into Chrome.

So, I thought there might be a way to display the console of the browser
instance, but I can't find a "chrome://..." command to do that. The most
information I can get back from the browser instance is this:

"C:/Program Files (x86)/RunRev/LiveCode Business 8.1.4/LiveCode Business"
--browser-subprocess-path="C:\Program Files (x86)\RunRev\LiveCode Business
8.1.4\Externals\CEF\libbrowser-cefprocess.exe" --no-sandbox --lang=en-US
--log-file="C:\Program Files (x86)\RunRev\LiveCode Business
8.1.4\debug.log" --log-severity=disable --locales-dir-path="C:\Program
Files (x86)\RunRev\LiveCode Business 8.1.4\Externals\CEF\locales"

So, I created a .cmd file where I launched the livecode .exe, passing in
different values for

--log-severity=disable

but no alternative parameters in the .cmd file appear to be taken up (i.e.
it looks like these CEF browser parameters might be hard-coded into the LC
exe).

I'm thinking there must be a way to debug CEF browser instances, but a
search through the mail list didn't turn up anything that appeared relevant.

Any suggestions?

Regards
Bernard
___
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