RE: Running MetaCard based CGIs

[Monte Goulding]

Is this up and running yet? I'd like to check it out if you can post to the
list when it is?

> -Original Message-
> From: [EMAIL PROTECTED]
> [mailto:[EMAIL PROTECTED]]On Behalf Of Simon Lord
> Sent: Friday, 12 January 2001 3:00 PM
> To: [EMAIL PROTECTED]
> Subject: RE: Running MetaCard based CGIs
>
>
> We'll be launching this type of service very soon ourselves.  Testing
> has been completed on an in-house server and we'll be moving it to
> the web server next week.  Testing pages will soon be made public
> where the list can see what mchttp can do.
>
> We have lots of *goodies* planned for this, and if Kevin offers the
> same service then we'll just have ourselves some good'ol competition!
>
>
> >I would be very interested but haven't got much money to spend.
> The problem
> >is I don't know how to do it yet so I really need some way to practice
> >before I go telling clients what I can do. It would be good if you could
> >host a practice CGI bin or a demo script that we could post to
> and recieve
> >email results from. Once I learn I will be able to get money for
> this kind
> >of service. Sorry for being overly demanding but I think there would be
> >others in my situation.
> >
> >-Original Message-
> >From: [EMAIL PROTECTED]
> >[mailto:[EMAIL PROTECTED]]On Behalf Of Kevin Miller
> >Sent: Friday, 12 January 2001 9:35 AM
> >To: [EMAIL PROTECTED]
> >Subject: Running MetaCard based CGIs
> >
> >
> >Hi,
> >
> >We are considering launching a service to allow hosting of MC based CGIs.
> >If anyone would be interested in such a service, please could
> they email me
> >off list with a brief list of the sort of requirments you might
> expect.  How
> >much would you expect to pay?  What would the CGI do?  Would you want to
> >host your entire site with a domain name pointing to our IP
> address or would
> >you host the site elsewhere and refer CGI requests to our site?
> What kind
> >of disk and processor requirements might be involved?
> >
> >The idea is still up in the air, just trying to nail down what kind of a
> >demand there might be out there.  If this is something you might be
> >interested in then please do send me an email - obviously no demand = no
> >service! :-) Thanks for any info anyone provides...
> >
> >Regards,
> >
> >Kevin
> >
> >Kevin Miller <[EMAIL PROTECTED]> 
> >Runtime Revolution Limited (formerly Cross Worlds Computing).
> >Tel: +44 (0)131 672 2909.  Fax: +44 (0)1639 830 707.
> >
> >
> >Archives: http://www.mail-archive.com/metacard@lists.runrev.com/
> >Info: http://www.xworlds.com/metacard/mailinglist.htm
> >Please send bug reports to <[EMAIL PROTECTED]>, not this list.
> >
> >
> >
> >Archives: http://www.mail-archive.com/metacard@lists.runrev.com/
> >Info: http://www.xworlds.com/metacard/mailinglist.htm
> >Please send bug reports to <[EMAIL PROTECTED]>, not this list.
>
> --
>
>
> Cheers,
> Simon
>
> All your .com .net and .org domains for only $14.50 each.
> Get them while they last... http://www.amigo-3.com/hosting/
>
> --
> "The great discoveries in science are not punctuated by 'Eureka! I've
> found it!' but rather "Hmmm,that's funny" Isaac Asimov
>
> Archives: http://www.mail-archive.com/metacard@lists.runrev.com/
> Info: http://www.xworlds.com/metacard/mailinglist.htm
> Please send bug reports to <[EMAIL PROTECTED]>, not this list.
>


Archives: http://www.mail-archive.com/metacard@lists.runrev.com/
Info: http://www.xworlds.com/metacard/mailinglist.htm
Please send bug reports to <[EMAIL PROTECTED]>, not this list.

QT audio problem

[Jacqueline Landman Gay]

Remind me again -- what would cause an aiff file played via a QT player
object to stutter or stop playing? I can't reproduce the problem on my
G4, but a client says on his Mac 7300 the audio stutters, and on his G3
powerbook it plays for a few seconds and then stops entirely. The same
files, played from the same CD, play fine in HyperCard on both machines
using the Movie XCMD.

He has plenty of RAM on both machines and the CD-ROM drives appear to be
fast enough, at least for HyperCard. The files play okay in QT Player
too, but I know that isn't a fair test since Player takes over most of
the system resources.

Stuttering and stopping problems are not consistent. A file will play
okay two out of three times, and then stall. The length of the audio
files doesn't matter; short files are just as apt to error as long ones.

-- 
Jacqueline Landman Gay | [EMAIL PROTECTED]
HyperActive Software   | 612-724-1596
Custom hypermedia solutions| http://www.hyperactivesw.com

Archives: http://www.mail-archive.com/metacard@lists.runrev.com/
Info: http://www.xworlds.com/metacard/mailinglist.htm
Please send bug reports to <[EMAIL PROTECTED]>, not this list.

Re: launch browser problem

[Tariel Gogoberidze]

>
> From: Leston Drake <[EMAIL PROTECTED]>
> Subject: launch browser problem
> Date: Wed, 24 Jan 2001 10:17:32 -0700
> MIME-Version: 1.0
> Content-Type: text/plain; charset="us-ascii"; format=flowed
>
> Hello,
>
> A while back Kevin Miller shared this script to launch the default Internet
> browser, and go to a specific URL (on Win32):
>
> on launchWin32Browser httpDoc
># httpDoc: URL
>local bApp
>local EXE_off
>put
> queryRegistry("hkey_local_machine\software\classes\http\shell\open\command\")
> into bApp
>if char 1 of bApp is quote then clear char 1 of bApp  # strip leading quote
>put offset("EXE",bApp) into EXE_off
>put char 1 to (EXE_off+2) of bApp into bApp
>launch httpDoc with bApp
> end launchWin32Browser
>
> It works nicely.
> I've found, however, some unexpected behavior. When you call the routine,
> it launches the browser and loads the specified URL. If you leave the
> browser open and then call the routine again (e.g., with a different URL),
> nothing happens. The browser remains open to the 1st URL.

> This only happens

> when the browser has been opened with LAUNCH. In other words, if you open
> the browser manually (not through MetaCard), then another instance of the
> browser will be opened with LAUNCH.
>
> Is this standard behavior for the LAUNCH command?
> Is there some way to check to see if the app that was launched previously
> is still open?
> Is it possible to Close the app?
>
> Bottom line: I need to be able to go to another URL (in the browser)
> whether or not it was launched previously.
>
> Hmmm. I'm stumped. Any help GREATLY appreciated!
>
> Thanks in advance.
>
> Leston Drake, President
> LetterPress Software, Inc.
>

After Launch command MC opens process and then keeps track of it

You can check for open processes by..

  put the number of lines in (the openprocesses) into nlines
  put return && Nlines && "open process(es) registered.

or just

put (the openprocesses)

In my understanding you can't launch the  process if it is already opened in MC

If you have browser already opened by MC you can "quit" it with

  kill  process bApp

and then launch it again with different URL

Best regards
Tariel Gogoberidze


Archives: http://www.mail-archive.com/metacard@lists.runrev.com/
Info: http://www.xworlds.com/metacard/mailinglist.htm
Please send bug reports to <[EMAIL PROTECTED]>, not this list.

RE: Announcing MetaCard 2.3.2 for Darwin/Mac OS X

[Monte Goulding]

Hi I've been reading the latest thread and some old archives about CGI in MC
and I have a quick question.

It seems to me that it is being suggested that the best, quickest and
easiest way to run sripts with MCHTTPd is to use the stacks-bin and thus not
startUp a new process. This may be a stipid question but what happens if
there are two clients wanting the same thing at the same time? As it's the
same process wouldn't MCHTTPd need to wait untill the last script is
finnished before it can do anything?


> -Original Message-
> From: [EMAIL PROTECTED]
> [mailto:[EMAIL PROTECTED]]On Behalf Of Scott Raney
> Sent: Thursday, 25 January 2001 9:44 AM
> To: [EMAIL PROTECTED]
> Subject: Re: Announcing MetaCard 2.3.2 for Darwin/Mac OS X
>
>
> On Wed, 24 Jan 2001 David Bovill <[EMAIL PROTECTED]> wrote:
> >
> > >> What are the concrete advantages of using this engine over
> using Metacard to
> > >> develop CGI solutions?
> > >
> > > Not sure I follow this: the only way to do CGI with the full,
> > > graphical, version of MetaCard is under MacOS, where CGI is done with
> > > AppleEvents.
> >
> > ??? What was I doing when I used the full graphical
> version/engine for MC on
> > Linux as the CGI engine for Apache? I was sure I got this working -:) In
> > other words presuming this wasn't a case of false memory
> syndrome brought
> > about by an over enthusiatic adoption of the incomprehensible power of
> > Metacard...
>
> On UNIX it's the same *engine*, it just runs in two different *modes*.
> Which it runs in depends on what it finds on the command line when it
> starts up, a stack, a .mt script, or nothing.  There are two different
> Win32 engines, though, because you can't produce a single executable
> that does both on Windows.  The Darwin engine works just like the
> other UNIX engines, though the graphical mode will only work if you
> install and run an X server first.  The Carbon engine only supports
> the graphical mode.  So the situation on Mac OS X will be similar to
> Windows (you'll use one of two different engines depending on what you
> need to do).
>
> > How does the MC/Linux/Apache solution differ from the console
> version CGI
> > solutions performance/implementation wise?
>
> These are the same thing.  The only other way to do it is with
> mchttpd, which will run scripts (CGIs) faster but serve image and text
> files slower.  There is no MetaCard equivalent of something like
> "mod_perl" where the language interpreter is actually loaded into the
> address space of the HTTP server, which kind of gets you the best of
> both worlds.  It'd be pretty straightforward to build something like
> that (much easier than a WWW browser plugin, for one), but since we
> don't know of anyone that's run into a performance problem in this
> area yet, we haven't scheduled building anything like that.
>   Regards,
> Scott
>
> 
> Scott Raney  [EMAIL PROTECTED]  http://www.metacard.com
> MetaCard: You know, there's an easier way to do that...
>
>
>
> Archives: http://www.mail-archive.com/metacard@lists.runrev.com/
> Info: http://www.xworlds.com/metacard/mailinglist.htm
> Please send bug reports to <[EMAIL PROTECTED]>, not this list.
>


Archives: http://www.mail-archive.com/metacard@lists.runrev.com/
Info: http://www.xworlds.com/metacard/mailinglist.htm
Please send bug reports to <[EMAIL PROTECTED]>, not this list.

RE: launch browser problem

[Monte Goulding]

Would it be possible for you to post the script to the list. I'm interested
to know how you have implemented this and I have a LOT to learn about DDE.

PS to anyone that's interested. I still am not 100% sure about those
registry entried I posted the other day but they are what is placed into the
registry if you follow Tuviah's instructions which are in the old archives.
I don't know if I was clear about my intention. I was hoping that a guru who
had done this may confirm these entries before i go and crash my computer.

PPS I think we need an MC list FAQ. If we suggested a number of common
questions we could each put up our hand to write the detailed description of
each thing. A lot of lists have this kind of thing and is very helpfull and
can save a lot of time searching the archives.

Thanks in advance

Monte

> -Original Message-
> From: [EMAIL PROTECTED]
> [mailto:[EMAIL PROTECTED]]On Behalf Of Scott Rossi
> Sent: Thursday, 25 January 2001 9:08 AM
> To: [EMAIL PROTECTED]
> Subject: Re: launch browser problem
>
>
> Recently, Monte Goulding wrote:
>
> >> I need to be able to go to another URL (in the browser)
> >> whether or not it was launched previously.
>
> > You can use DDE from the externals collection. The example
> stack contains
> > the info you need to write the code. You'll need to check if
> it's open and
> > then use DDE if it is.
>
> FWIW, I've found you have to do a lot more jumping through hoops than what
> is available in the sample stack due to quirky behavior of
> Internet Explorer
> and possibly DDE in general.  The problem is IE doesn't always make itself
> known to the appsopen external on certain systems, so in my case, I had to
> use a combination of the launch command and DDE to make sure URLs got
> launched appropriately in this browser.
>
> Also, this may be a Windows thing or MetaCard thing, but rapid successive
> attempts to launch a URL can fail, making the launch process useless and
> requiring a restart of the browser and/or MC (I can't recall which).  If
> there is any chance of a user repeatedly clicking on a launch control, you
> would be wise to prevent these multiple launch requests using a
> timer and/or
> busy cursor.  In my experience a delay of 2 to 3 seconds appeared to work
> consistently under testing on Win95, 98 and 2000 systems.
>
> Hope this helps.
>
> Scott
>
> _
> Scott Rossi   Tactile Media - Multimedia & Design
> Creative Director Email: [EMAIL PROTECTED]
>   Web: www.tactilemedia.com
>
>
> Archives: http://www.mail-archive.com/metacard@lists.runrev.com/
> Info: http://www.xworlds.com/metacard/mailinglist.htm
> Please send bug reports to <[EMAIL PROTECTED]>, not this list.
>
>


Archives: http://www.mail-archive.com/metacard@lists.runrev.com/
Info: http://www.xworlds.com/metacard/mailinglist.htm
Please send bug reports to <[EMAIL PROTECTED]>, not this list.

Re: Announcing MetaCard 2.3.2 for Darwin/Mac OS X

[Scott Raney]

On Wed, 24 Jan 2001 David Bovill <[EMAIL PROTECTED]> wrote:
> 
> >> What are the concrete advantages of using this engine over using Metacard to
> >> develop CGI solutions?
> > 
> > Not sure I follow this: the only way to do CGI with the full,
> > graphical, version of MetaCard is under MacOS, where CGI is done with
> > AppleEvents. 
> 
> ??? What was I doing when I used the full graphical version/engine for MC on
> Linux as the CGI engine for Apache? I was sure I got this working -:) In
> other words presuming this wasn't a case of false memory syndrome brought
> about by an over enthusiatic adoption of the incomprehensible power of
> Metacard...

On UNIX it's the same *engine*, it just runs in two different *modes*.
Which it runs in depends on what it finds on the command line when it
starts up, a stack, a .mt script, or nothing.  There are two different
Win32 engines, though, because you can't produce a single executable
that does both on Windows.  The Darwin engine works just like the
other UNIX engines, though the graphical mode will only work if you
install and run an X server first.  The Carbon engine only supports
the graphical mode.  So the situation on Mac OS X will be similar to
Windows (you'll use one of two different engines depending on what you
need to do).

> How does the MC/Linux/Apache solution differ from the console version CGI
> solutions performance/implementation wise?

These are the same thing.  The only other way to do it is with
mchttpd, which will run scripts (CGIs) faster but serve image and text
files slower.  There is no MetaCard equivalent of something like
"mod_perl" where the language interpreter is actually loaded into the
address space of the HTTP server, which kind of gets you the best of
both worlds.  It'd be pretty straightforward to build something like
that (much easier than a WWW browser plugin, for one), but since we
don't know of anyone that's run into a performance problem in this
area yet, we haven't scheduled building anything like that.
  Regards,
Scott


Scott Raney  [EMAIL PROTECTED]  http://www.metacard.com
MetaCard: You know, there's an easier way to do that...



Archives: http://www.mail-archive.com/metacard@lists.runrev.com/
Info: http://www.xworlds.com/metacard/mailinglist.htm
Please send bug reports to <[EMAIL PROTECTED]>, not this list.

Re: Group Owner

[Mark Luetzelschwab]

ThanksI looked and this computer was running b4.  Works great, now!

-ml

>
>I had this problem, and there have been a few emails back and forth about
>this about 2 months back. As far as I remember Scott fixed/altered this in
>2.3.2 so that the owner returns the long id rather than the name. I think
>this is the case now...
>
>What version are you using? Otherwise there are some workarounds like
>defining your own function "myowner" and getting the function to return the
>long id of the target and then calling this function from your groups...
>
>
>
> > From: Mark Luetzelschwab <[EMAIL PROTECTED]>
> > Reply-To: [EMAIL PROTECTED]
> > Date: Tue, 23 Jan 2001 17:09:07 -0600
> > To: [EMAIL PROTECTED]
> > Subject: Group Owner
> >
> > Hi Everyone,
> >
> > Quick Question:
> >
> > I have two groups (custom scrollbars) that have the same name as each
> > other on one card

Mark J. Luetzelschwab   [EMAIL PROTECTED]
Graduate Research Assistant (v) (512) 232 6034
Instructional Technology(f) (512) 232 2322
Reading and Language Arts:
http://www.texasreading.org

Archives: http://www.mail-archive.com/metacard@lists.runrev.com/
Info: http://www.xworlds.com/metacard/mailinglist.htm
Please send bug reports to <[EMAIL PROTECTED]>, not this list.

Re: launch browser problem

[Scott Rossi]

Recently, Monte Goulding wrote:

>> I need to be able to go to another URL (in the browser)
>> whether or not it was launched previously.

> You can use DDE from the externals collection. The example stack contains
> the info you need to write the code. You'll need to check if it's open and
> then use DDE if it is.

FWIW, I've found you have to do a lot more jumping through hoops than what
is available in the sample stack due to quirky behavior of Internet Explorer
and possibly DDE in general.  The problem is IE doesn't always make itself
known to the appsopen external on certain systems, so in my case, I had to
use a combination of the launch command and DDE to make sure URLs got
launched appropriately in this browser.

Also, this may be a Windows thing or MetaCard thing, but rapid successive
attempts to launch a URL can fail, making the launch process useless and
requiring a restart of the browser and/or MC (I can't recall which).  If
there is any chance of a user repeatedly clicking on a launch control, you
would be wise to prevent these multiple launch requests using a timer and/or
busy cursor.  In my experience a delay of 2 to 3 seconds appeared to work
consistently under testing on Win95, 98 and 2000 systems.

Hope this helps.

Scott

_
Scott Rossi   Tactile Media - Multimedia & Design
Creative Director Email: [EMAIL PROTECTED]
  Web: www.tactilemedia.com


Archives: http://www.mail-archive.com/metacard@lists.runrev.com/
Info: http://www.xworlds.com/metacard/mailinglist.htm
Please send bug reports to <[EMAIL PROTECTED]>, not this list.

RE: Subject: Backdrops

[Monte Goulding]

Thanks that's one to remember. FormatForPrinting overrides
windowBoundingRect.

> -Original Message-
> From: [EMAIL PROTECTED]
> [mailto:[EMAIL PROTECTED]]On Behalf Of Raymond E. Griffith
> Sent: Thursday, 25 January 2001 6:00 AM
> To: [EMAIL PROTECTED]
> Subject: RE: Subject: Backdrops
>
>
> on Wed, 24 Jan 2001 19:40:10 +1030 "Monte Goulding"
> <[EMAIL PROTECTED]> perplexedly asked:
>
> > I still don't get what the formatForPrinting has to do
> with the
> > windowBoundingRect. This may override the property but
> surely changing the
> > windowBoundingRect would be the most obvious soloution.
>
> It isn't obvious, is it? Well, it had to be explained to me
> a couple of times.
>
> The windowBoundingRect is to prevent stacks from being off
> screen. So it looks at the screenrect, takes off for menu
> bars, and resizes your stack accordingly whether you like it
> or not. You can't set the windowBoundingRect to be larger
> than the screenrect.
>
> You can overcome this by typing into the message box:
>
> "set the formatForPrinting of this stack to true"
>
> It only has to be done once -- provided you save the stack.
> (I think.)
>
> This allows MC to understand that you intend to print the
> card.
>
> Why is this important? Say I want to print a letter from a
> card onto regular letter paper (say 8.5 inches by 11
> inches -- with apologies to our metric neighbors). You can't
> do it if the card won't open with enough height. The card
> would need to be 9 inches tall (9*72 = 648 pixels) assuming
> a 1-inch margin at the top and bottom. Now I generally use
> an 800x600 screen because the larger screen size makes the
> type too small for my eyes. The windowBoundingRect (which
> would set the height to about 580) is prevented by the
> formatForPrinting property of the stack. So the stack opens,
> partly off screen. And I can print it.
>
> I can't see another way around it unless Scott makes it
> possible to print the full size of groups or scrolling
> fields, etc. regardless of the window size. I have the
> feeling that he is moving in that direction, at least
> partially. But in the mean time
>
> Hope this helps.
>
> Raymond
>
>
> Archives: http://www.mail-archive.com/metacard@lists.runrev.com/
> Info: http://www.xworlds.com/metacard/mailinglist.htm
> Please send bug reports to <[EMAIL PROTECTED]>, not this list.
>
>


Archives: http://www.mail-archive.com/metacard@lists.runrev.com/
Info: http://www.xworlds.com/metacard/mailinglist.htm
Please send bug reports to <[EMAIL PROTECTED]>, not this list.

RE: Custom Property Storage

[Monte Goulding]

Thanks Geoff

Yup it seems custom properties are the way to go. The program will be
standalone but I will create my own file type and use document stacks for
data storage. Hence the DDE questions. The 2 to 4 GB limit is huge.

> -Original Message-
> From: [EMAIL PROTECTED]
> [mailto:[EMAIL PROTECTED]]On Behalf Of Geoff Canyon
> Sent: Thursday, 25 January 2001 3:13 AM
> To: [EMAIL PROTECTED]
> Subject: Re: Custom Property Storage
>
>
> At 3:13 PM +1030 1/23/01, Monte Goulding wrote:
> >I posted a question to this list asking about storage capacity of custom
> >properties. Still interested by the way. I just have tested for
> double the
> >capacity that I need and the property held 80,000 characters
> whithout even a
> >whince. I have tested fields to 4,000,000,000 characters before
> which worked
> >whilst bloating the file. unless someone tells me otherwise then
> I'll assume
> >that CP's have the same bottomless nature.
> >
> >I just thought i'd let anyone know that hasn't tested this
> before that for
> >some reason the way the data is stored in a CP is cheaper in
> terms of disk
> >space than in a field. Here are the stats:
> >
> >Base file: 35 KB
> >80,000 chars in a test field: over 250 KB
> >the same 80,000 chars in a CP: 108 KB
>
> I believer there is no practical limit to custom properties. They
> live inside a theoretical limit along with everything else that if
> memory serves is either two gigabytes or four.
>
> I'd guess the difference is style information. Fields store the text
> style, etc., while custom properties store only the actual data.
>
> For your requirements, you should definitely go with custom
> properties -- unless you plan to make changes and store them, and to
> build this into a standalone. In that case it gets more complex,
> because you can't save into a running application. The way to solve
> this is to store the data in an external file. The file can either be
> a text file you read in, or a stack that you open. Which is better
> has been debated on this list several times, check the archive if
> your curious. (my preference is a stack except for _very_ simple
> purposes)
>
> gc
>
>
> Archives: http://www.mail-archive.com/metacard@lists.runrev.com/
> Info: http://www.xworlds.com/metacard/mailinglist.htm
> Please send bug reports to <[EMAIL PROTECTED]>, not this list.
>


Archives: http://www.mail-archive.com/metacard@lists.runrev.com/
Info: http://www.xworlds.com/metacard/mailinglist.htm
Please send bug reports to <[EMAIL PROTECTED]>, not this list.

RE: launch browser problem

[Monte Goulding]

You can use DDE from the externals collection. The example stack contains
the info you need to write the code. You'll need to check if it's open and
then use DDE if it is.

> -Original Message-
> From: [EMAIL PROTECTED]
> [mailto:[EMAIL PROTECTED]]On Behalf Of Leston Drake
> Sent: Thursday, 25 January 2001 3:48 AM
> To: [EMAIL PROTECTED]
> Subject: launch browser problem
>
>
> Hello,
>
> A while back Kevin Miller shared this script to launch the
> default Internet
> browser, and go to a specific URL (on Win32):
>
> on launchWin32Browser httpDoc
># httpDoc: URL
>local bApp
>local EXE_off
>put
> queryRegistry("hkey_local_machine\software\classes\http\shell\open
> \command\")
> into bApp
>if char 1 of bApp is quote then clear char 1 of bApp  # strip
> leading quote
>put offset("EXE",bApp) into EXE_off
>put char 1 to (EXE_off+2) of bApp into bApp
>launch httpDoc with bApp
> end launchWin32Browser
>
> It works nicely.
> I've found, however, some unexpected behavior. When you call the routine,
> it launches the browser and loads the specified URL. If you leave the
> browser open and then call the routine again (e.g., with a
> different URL),
> nothing happens. The browser remains open to the 1st URL. This
> only happens
> when the browser has been opened with LAUNCH. In other words, if you open
> the browser manually (not through MetaCard), then another instance of the
> browser will be opened with LAUNCH.
>
> Is this standard behavior for the LAUNCH command?
> Is there some way to check to see if the app that was launched previously
> is still open?
> Is it possible to Close the app?
>
> Bottom line: I need to be able to go to another URL (in the browser)
> whether or not it was launched previously.
>
> Hmmm. I'm stumped. Any help GREATLY appreciated!
>
> Thanks in advance.
>
> Leston Drake, President
> LetterPress Software, Inc.
>
>
> Archives: http://www.mail-archive.com/metacard@lists.runrev.com/
> Info: http://www.xworlds.com/metacard/mailinglist.htm
> Please send bug reports to <[EMAIL PROTECTED]>, not this list.
>


Archives: http://www.mail-archive.com/metacard@lists.runrev.com/
Info: http://www.xworlds.com/metacard/mailinglist.htm
Please send bug reports to <[EMAIL PROTECTED]>, not this list.

RE: Subject: Backdrops

[Raymond E. Griffith]

on Wed, 24 Jan 2001 19:40:10 +1030 "Monte Goulding"
<[EMAIL PROTECTED]> perplexedly asked:

> I still don't get what the formatForPrinting has to do
with the
> windowBoundingRect. This may override the property but
surely changing the
> windowBoundingRect would be the most obvious soloution.

It isn't obvious, is it? Well, it had to be explained to me
a couple of times.

The windowBoundingRect is to prevent stacks from being off
screen. So it looks at the screenrect, takes off for menu
bars, and resizes your stack accordingly whether you like it
or not. You can't set the windowBoundingRect to be larger
than the screenrect.

You can overcome this by typing into the message box:

"set the formatForPrinting of this stack to true"

It only has to be done once -- provided you save the stack.
(I think.)

This allows MC to understand that you intend to print the
card.

Why is this important? Say I want to print a letter from a
card onto regular letter paper (say 8.5 inches by 11
inches -- with apologies to our metric neighbors). You can't
do it if the card won't open with enough height. The card
would need to be 9 inches tall (9*72 = 648 pixels) assuming
a 1-inch margin at the top and bottom. Now I generally use
an 800x600 screen because the larger screen size makes the
type too small for my eyes. The windowBoundingRect (which
would set the height to about 580) is prevented by the
formatForPrinting property of the stack. So the stack opens,
partly off screen. And I can print it.

I can't see another way around it unless Scott makes it
possible to print the full size of groups or scrolling
fields, etc. regardless of the window size. I have the
feeling that he is moving in that direction, at least
partially. But in the mean time

Hope this helps.

Raymond


Archives: http://www.mail-archive.com/metacard@lists.runrev.com/
Info: http://www.xworlds.com/metacard/mailinglist.htm
Please send bug reports to <[EMAIL PROTECTED]>, not this list.

Re: launch browser problem

[Sjoerd Op 't Land]

Leston Drake wrote/ schreef:

> Hello,
> 
> A while back Kevin Miller shared this script to launch the default Internet
> browser, and go to a specific URL (on Win32):
> 
> on launchWin32Browser httpDoc
> # httpDoc: URL
> local bApp
> local EXE_off
> put 
> queryRegistry("hkey_local_machine\software\classes\http\shell\open\command\")
> into bApp
> if char 1 of bApp is quote then clear char 1 of bApp  # strip leading quote
> put offset("EXE",bApp) into EXE_off
> put char 1 to (EXE_off+2) of bApp into bApp
> launch httpDoc with bApp
Just curious... what is 'the result' here?
> end launchWin32Browser
> 
> It works nicely.
> I've found, however, some unexpected behavior. When you call the routine,
> it launches the browser and loads the specified URL. If you leave the
> browser open and then call the routine again (e.g., with a different URL),
> nothing happens. The browser remains open to the 1st URL. This only happens
> when the browser has been opened with LAUNCH. In other words, if you open
> the browser manually (not through MetaCard), then another instance of the
> browser will be opened with LAUNCH.
> 
> Is this standard behavior for the LAUNCH command?
> Is there some way to check to see if the app that was launched previously
> is still open?
> Is it possible to Close the app?
> 
> Bottom line: I need to be able to go to another URL (in the browser)
> whether or not it was launched previously.
> 
> Hmmm. I'm stumped. Any help GREATLY appreciated!
> 
> Thanks in advance.
> 
> Leston Drake, President
> LetterPress Software, Inc.
Regards,
Sjoerd


Archives: http://www.mail-archive.com/metacard@lists.runrev.com/
Info: http://www.xworlds.com/metacard/mailinglist.htm
Please send bug reports to <[EMAIL PROTECTED]>, not this list.

Re: Announcing MetaCard 2.3.2 for Darwin/Mac

[andu]

>How does the MC/Linux/Apache solution differ from the console version CGI
>solutions performance/implementation wise?
>
>Thanks in advance...

In console mode you don't have to keep MC running all the time (if you don't need to), 
it
executes the script and exits. Besides you don't have the overhead of the UI. Most
servers don't have X or a monitor, remember, Linux's X is still  a major source of 
trouble
and not its stronger part not to mention the amount of resources it eats up.

Regards, Andu 
___
[EMAIL PROTECTED]

Archives: http://www.mail-archive.com/metacard@lists.runrev.com/
Info: http://www.xworlds.com/metacard/mailinglist.htm
Please send bug reports to <[EMAIL PROTECTED]>, not this list.

Re: Announcing MetaCard 2.3.2 for Darwin/Mac OS X

[David Bovill]

>> What are the concrete advantages of using this engine over using Metacard to
>> develop CGI solutions?
> 
> Not sure I follow this: the only way to do CGI with the full,
> graphical, version of MetaCard is under MacOS, where CGI is done with
> AppleEvents. 

??? What was I doing when I used the full graphical version/engine for MC on
Linux as the CGI engine for Apache? I was sure I got this working -:) In
other words presuming this wasn't a case of false memory syndrome brought
about by an over enthusiatic adoption of the incomprehensible power of
Metacard...

How does the MC/Linux/Apache solution differ from the console version CGI
solutions performance/implementation wise?

Thanks in advance...

> But this engine is only for the UNIX part of Mac OS X
> (aka Darwin) which doesn't support AppleEvents and instead uses stdio
> (Standard I/O) for communication between the HTTP server and CGI
> application.  These things are *very* different from anything you may
> have done with the graphical version of MetaCard.  There are some
> examples on the MetaCard WWW site.
> 
>> 1) Memory?
> 
> Should be considerably less than a full graphical application.
> 
>> 2) Speed?
> 
> Should be much faster than a graphical application, or any comparable
> MacOS application because of UNIX's better architecture.
> 
>> 3) Script limits
> 
> There are none for MetaTalk scripts like this.  This is true for all
> of the UNIX engines and the console-mode Win32 engine too.
> 
>> Will the engine run Metacard stacks direct in "console-mode"? Does this
>> require saving the stack as a standalone with this new engine?
> 
> You can refer to stacks to get data out of them, but they won't be
> visible anywhere when you "go" to them.  You can't build standalones
> with .mt scripts.
> Regards,
> Scott
> 

And in retrospect -:)


Archives: http://www.mail-archive.com/metacard@lists.runrev.com/
Info: http://www.xworlds.com/metacard/mailinglist.htm
Please send bug reports to <[EMAIL PROTECTED]>, not this list.

Re: Group Owner

[David Bovill]

I had this problem, and there have been a few emails back and forth about
this about 2 months back. As far as I remember Scott fixed/altered this in
2.3.2 so that the owner returns the long id rather than the name. I think
this is the case now...

What version are you using? Otherwise there are some workarounds like
defining your own function "myowner" and getting the function to return the
long id of the target and then calling this function from your groups...



> From: Mark Luetzelschwab <[EMAIL PROTECTED]>
> Reply-To: [EMAIL PROTECTED]
> Date: Tue, 23 Jan 2001 17:09:07 -0600
> To: [EMAIL PROTECTED]
> Subject: Group Owner
> 
> Hi Everyone,
> 
> Quick Question:
> 
> I have two groups (custom scrollbars) that have the same name as each
> other on one card
> 
> Each group has a graphic which needs to access customProperties of
> the parent (owner) group.
> 
> If I get the OWNER of the graphic, I get the name...any operations on
> that affect only the first group with that name.
> 
> Can I get the OWNER ID (and no, "put the id of the owner of me" does
> not work..it returns the same value for both groups).
> 
> The workaround is to remove the NAME of the group...and the owner
> returns the correct id.  But I have to reference the group by name at
> some point, so its only a temporary solution.
> 
> Thanks.
> -ml
> Mark J. Luetzelschwab  [EMAIL PROTECTED]
> Graduate Research Assistant (v) (512) 232 6034
> Instructional Technology (f) (512) 232 2322
> Reading and Language Arts:
> http://www.texasreading.org
> 
> Archives: http://www.mail-archive.com/metacard@lists.runrev.com/
> Info: http://www.xworlds.com/metacard/mailinglist.htm
> Please send bug reports to <[EMAIL PROTECTED]>, not this list.
> 
> 


Archives: http://www.mail-archive.com/metacard@lists.runrev.com/
Info: http://www.xworlds.com/metacard/mailinglist.htm
Please send bug reports to <[EMAIL PROTECTED]>, not this list.

Re: Group Owner

[David Bovill]

I had this problem, and there have been a few emails back and forth about
this about 2 months back. As far as I remember Scott fixed/altered this in
2.3.2 so that the owner returns the long id rather than the name. I think
this is the case now...

What version are you using? Otherwise there are some workarounds like
defining your own function "myowner" and getting the function to return the
long id of the target and then calling this function from your groups...



> From: Mark Luetzelschwab <[EMAIL PROTECTED]>
> Reply-To: [EMAIL PROTECTED]
> Date: Tue, 23 Jan 2001 17:09:07 -0600
> To: [EMAIL PROTECTED]
> Subject: Group Owner
> 
> Hi Everyone,
> 
> Quick Question:
> 
> I have two groups (custom scrollbars) that have the same name as each
> other on one card
> 
> Each group has a graphic which needs to access customProperties of
> the parent (owner) group.
> 
> If I get the OWNER of the graphic, I get the name...any operations on
> that affect only the first group with that name.
> 
> Can I get the OWNER ID (and no, "put the id of the owner of me" does
> not work..it returns the same value for both groups).
> 
> The workaround is to remove the NAME of the group...and the owner
> returns the correct id.  But I have to reference the group by name at
> some point, so its only a temporary solution.
> 
> Thanks.
> -ml
> Mark J. Luetzelschwab  [EMAIL PROTECTED]
> Graduate Research Assistant (v) (512) 232 6034
> Instructional Technology (f) (512) 232 2322
> Reading and Language Arts:
> http://www.texasreading.org
> 
> Archives: http://www.mail-archive.com/metacard@lists.runrev.com/
> Info: http://www.xworlds.com/metacard/mailinglist.htm
> Please send bug reports to <[EMAIL PROTECTED]>, not this list.
> 
> 


Archives: http://www.mail-archive.com/metacard@lists.runrev.com/
Info: http://www.xworlds.com/metacard/mailinglist.htm
Please send bug reports to <[EMAIL PROTECTED]>, not this list.

launch browser problem

[Leston Drake]

Hello,

A while back Kevin Miller shared this script to launch the default Internet 
browser, and go to a specific URL (on Win32):

on launchWin32Browser httpDoc
   # httpDoc: URL
   local bApp
   local EXE_off
   put 
queryRegistry("hkey_local_machine\software\classes\http\shell\open\command\") 
into bApp
   if char 1 of bApp is quote then clear char 1 of bApp  # strip leading quote
   put offset("EXE",bApp) into EXE_off
   put char 1 to (EXE_off+2) of bApp into bApp
   launch httpDoc with bApp
end launchWin32Browser

It works nicely.
I've found, however, some unexpected behavior. When you call the routine, 
it launches the browser and loads the specified URL. If you leave the 
browser open and then call the routine again (e.g., with a different URL), 
nothing happens. The browser remains open to the 1st URL. This only happens 
when the browser has been opened with LAUNCH. In other words, if you open 
the browser manually (not through MetaCard), then another instance of the 
browser will be opened with LAUNCH.

Is this standard behavior for the LAUNCH command?
Is there some way to check to see if the app that was launched previously 
is still open?
Is it possible to Close the app?

Bottom line: I need to be able to go to another URL (in the browser) 
whether or not it was launched previously.

Hmmm. I'm stumped. Any help GREATLY appreciated!

Thanks in advance.

Leston Drake, President
LetterPress Software, Inc.


Archives: http://www.mail-archive.com/metacard@lists.runrev.com/
Info: http://www.xworlds.com/metacard/mailinglist.htm
Please send bug reports to <[EMAIL PROTECTED]>, not this list.

Re: Custom Property Storage

[David Bovill]

That's pretty interesting Monte...

> From: "Monte Goulding" <[EMAIL PROTECTED]>
> Reply-To: [EMAIL PROTECTED]
> Date: Tue, 23 Jan 2001 15:13:35 +1030
> To: [EMAIL PROTECTED]
> Subject: Custom Property Storage
> 
> Hi All
> 
> I posted a question to this list asking about storage capacity of custom
> properties. Still interested by the way. I just have tested for double the
> capacity that I need and the property held 80,000 characters whithout even a
> whince. I have tested fields to 4,000,000,000 characters before which worked
> whilst bloating the file. unless someone tells me otherwise then I'll assume
> that CP's have the same bottomless nature.
> 
> I just thought i'd let anyone know that hasn't tested this before that for
> some reason the way the data is stored in a CP is cheaper in terms of disk
> space than in a field. Here are the stats:
> 
> Base file: 35 KB
> 80,000 chars in a test field: over 250 KB
> the same 80,000 chars in a CP: 108 KB
> 
> Bargain ;-)
> 
> 

I believe it is a *lot* faster to access custom properties as well.


Archives: http://www.mail-archive.com/metacard@lists.runrev.com/
Info: http://www.xworlds.com/metacard/mailinglist.htm
Please send bug reports to <[EMAIL PROTECTED]>, not this list.

Re: Darwin on cd

[David Bovill]

> I think the bigger question is how the battle between Darwin and
> LinuxPPC will go.  Each has its advantages, but it's hard to see how
> both of them can thrive in what must be a pretty small pond.  And if
> they get that X server for OS X polished up some, *both* of them may
> go away because you'll have the best of the Mac and UNIX worlds
> without even having to reboot your Mac.  Throw in SoftWindows or VPC
> and you'll be able to develop for all platforms MetaCard supports at
> the same time on one screen!

Don't get carried away Scott - you'll get mistaken for a Mac enthusiast -:)


Archives: http://www.mail-archive.com/metacard@lists.runrev.com/
Info: http://www.xworlds.com/metacard/mailinglist.htm
Please send bug reports to <[EMAIL PROTECTED]>, not this list.

Re: Custom Property Storage

[Geoff Canyon]

At 3:13 PM +1030 1/23/01, Monte Goulding wrote:
>I posted a question to this list asking about storage capacity of custom
>properties. Still interested by the way. I just have tested for double the
>capacity that I need and the property held 80,000 characters whithout even a
>whince. I have tested fields to 4,000,000,000 characters before which worked
>whilst bloating the file. unless someone tells me otherwise then I'll assume
>that CP's have the same bottomless nature.
>
>I just thought i'd let anyone know that hasn't tested this before that for
>some reason the way the data is stored in a CP is cheaper in terms of disk
>space than in a field. Here are the stats:
>
>Base file: 35 KB
>80,000 chars in a test field: over 250 KB
>the same 80,000 chars in a CP: 108 KB

I believer there is no practical limit to custom properties. They 
live inside a theoretical limit along with everything else that if 
memory serves is either two gigabytes or four.

I'd guess the difference is style information. Fields store the text 
style, etc., while custom properties store only the actual data.

For your requirements, you should definitely go with custom 
properties -- unless you plan to make changes and store them, and to 
build this into a standalone. In that case it gets more complex, 
because you can't save into a running application. The way to solve 
this is to store the data in an external file. The file can either be 
a text file you read in, or a stack that you open. Which is better 
has been debated on this list several times, check the archive if 
your curious. (my preference is a stack except for _very_ simple 
purposes)

gc


Archives: http://www.mail-archive.com/metacard@lists.runrev.com/
Info: http://www.xworlds.com/metacard/mailinglist.htm
Please send bug reports to <[EMAIL PROTECTED]>, not this list.

Re: Subject: Backdrops

[Jacqueline Landman Gay]

Monte Goulding wrote:
> 
> This seems incredably strange. Are you sure you didn't leave MC open while
> changging the screen resoloution. This would cause the same problem because
> the screenRect stays whatever it is at startUp

When I queried the message box for the screenrect, it always returned
the correct size. Same for the windowBoundingRect. I was quitting
MetaCard between sessions.

-- 
Jacqueline Landman Gay | [EMAIL PROTECTED]
HyperActive Software   | 612-724-1596
Custom hypermedia solutions| http://www.hyperactivesw.com

Archives: http://www.mail-archive.com/metacard@lists.runrev.com/
Info: http://www.xworlds.com/metacard/mailinglist.htm
Please send bug reports to <[EMAIL PROTECTED]>, not this list.

RE: Subject: Backdrops

[Monte Goulding]

This seems incredably strange. Are you sure you didn't leave MC open while
changging the screen resoloution. This would cause the same problem because
the screenRect stays whatever it is at startUp

> -Original Message-
> From: [EMAIL PROTECTED]
> [mailto:[EMAIL PROTECTED]]On Behalf Of Jacqueline Landman
> Gay
> Sent: Wednesday, 24 January 2001 1:35 PM
> To: [EMAIL PROTECTED]
> Subject: Re: Subject: Backdrops
>
>
> Thanks, everyone, for the replies. The decorations on the stack were
> already set to none, so that wasn't it. When I set the monitor
> resolution to 640x480 and asked for the windowBoundingRect in the
> message box, it returned the screenrect, as it is supposed to, adjusted
> to accomodate the Mac menubar. So that wasn't it. What appears to have
> been happening is that the stack actually resized itself when running at
> a lower resolution. If I put "set the width of this stack to 640" into
> the message box, it expanded to fill the screen. At larger monitor
> resolutions, the stack displayed its full width and height
> automatically, but at the lower resolution it resized itself improperly.
>
> The stack was developed entirely in a larger monitor resolution, but had
> never been saved at a lower one. So I set the correct dimensions via the
> message box and saved the stack while running at the lower resolution.
> That seems to have fixed it. I can now open the stack in any resolution
> and it appears at its complete size.
>
> Looks like a glitch somewhere, but explicitly saving while running at
> the lower resolution seems to solve the problem. This was a converted
> HyperCard stack, so maybe that has something to do with it.
>
> >
> > > From: Jacqueline Landman Gay <[EMAIL PROTECTED]>
> > > Subject: Backdrops
> > > Date: Tue, 23 Jan 2001 14:38:44 -0600
> > > MIME-Version: 1.0
> > > Content-Type: text/plain; charset=us-ascii
> > > Content-Transfer-Encoding: 7bit
> > >
> > > I have a stack that displays a backdrop and hides the
> > menubar. It works
> > > fine except when the monitor is set to 640x480 (which is
> > the same size
> > > as the stack). In that case, the stack is clipped off on
> > the right and
> > > bottom, eliminating much of the card. The idea is to fill
> > the screen
> > > with the stack.
> > >
> > > I have tried using a script that checks the screenrect and
> > sets the
> > > backdrop to "none" if the screenrect is the same size as
> > the stack, but
> > > it doesn't solve the problem. At 640x480 resolution, the
> > stack gets cut
> > > off no matter what. There is room at the edges of the
> > monitor for the
> > > rest of the stack, it just isn't being drawn to the
> > screen. Instead I
> > > can see through to the Finder at these edges (or I see the
> > backdrop if I
> > > don't use the script that checks the size of the
> > screenrect.)
> > >
> > > Solutions?
> > > --
> > > Jacqueline Landman Gay |
> > [EMAIL PROTECTED]
> > > HyperActive Software   | 612-724-1596
> > > Custom hypermedia solutions|
> > http://www.hyperactivesw.com
> > >
> >
> > Archives: http://www.mail-archive.com/metacard@lists.runrev.com/
> > Info: http://www.xworlds.com/metacard/mailinglist.htm
> > Please send bug reports to <[EMAIL PROTECTED]>, not this list.
>
>
> --
> Jacqueline Landman Gay | [EMAIL PROTECTED]
> HyperActive Software   | 612-724-1596
> Custom hypermedia solutions| http://www.hyperactivesw.com
>
>
>
> Archives: http://www.mail-archive.com/metacard@lists.runrev.com/
> Info: http://www.xworlds.com/metacard/mailinglist.htm
> Please send bug reports to <[EMAIL PROTECTED]>, not this list.
>


Archives: http://www.mail-archive.com/metacard@lists.runrev.com/
Info: http://www.xworlds.com/metacard/mailinglist.htm
Please send bug reports to <[EMAIL PROTECTED]>, not this list.

RE: Subject: Backdrops

[Monte Goulding]

I still don't get what the formatForPrinting has to do with the
windowBoundingRect. This may override the property but surely changing the
windowBoundingRect would be the most obvious soloution.

> -Original Message-
> From: [EMAIL PROTECTED]
> [mailto:[EMAIL PROTECTED]]On Behalf Of Raymond E. Griffith
> Sent: Wednesday, 24 January 2001 9:06 AM
> To: [EMAIL PROTECTED]
> Subject: Subject: Backdrops
>
>
> Jacque,
>
> You may be running into the windowBoundingRect property. It
> has a nasty habit of cutting off stacks. Try setting the
> formatForPrinting of the stack to true. You may still have
> to recenter your stack, but it won't cut things off.
>
> Cheers,
>
> Raymond
>
> > From: Jacqueline Landman Gay <[EMAIL PROTECTED]>
> > Subject: Backdrops
> > Date: Tue, 23 Jan 2001 14:38:44 -0600
> > MIME-Version: 1.0
> > Content-Type: text/plain; charset=us-ascii
> > Content-Transfer-Encoding: 7bit
> >
> > I have a stack that displays a backdrop and hides the
> menubar. It works
> > fine except when the monitor is set to 640x480 (which is
> the same size
> > as the stack). In that case, the stack is clipped off on
> the right and
> > bottom, eliminating much of the card. The idea is to fill
> the screen
> > with the stack.
> >
> > I have tried using a script that checks the screenrect and
> sets the
> > backdrop to "none" if the screenrect is the same size as
> the stack, but
> > it doesn't solve the problem. At 640x480 resolution, the
> stack gets cut
> > off no matter what. There is room at the edges of the
> monitor for the
> > rest of the stack, it just isn't being drawn to the
> screen. Instead I
> > can see through to the Finder at these edges (or I see the
> backdrop if I
> > don't use the script that checks the size of the
> screenrect.)
> >
> > Solutions?
> > --
> > Jacqueline Landman Gay |
> [EMAIL PROTECTED]
> > HyperActive Software   | 612-724-1596
> > Custom hypermedia solutions|
> http://www.hyperactivesw.com
> >
>
>
> Archives: http://www.mail-archive.com/metacard@lists.runrev.com/
> Info: http://www.xworlds.com/metacard/mailinglist.htm
> Please send bug reports to <[EMAIL PROTECTED]>, not this list.
>


Archives: http://www.mail-archive.com/metacard@lists.runrev.com/
Info: http://www.xworlds.com/metacard/mailinglist.htm
Please send bug reports to <[EMAIL PROTECTED]>, not this list.