Re: slow var to fld trick?
On Tue, 14 Sep 1999, Michael Walas wrote: I have a long handler that puts the contents of 12 different variables (each containing about 15,000 chars) into 12 different flds. It takes 25 seconds for the handler to run on my 6100 Mac running MetaCard 2.2.5 If I take out the 12 lines of the above, the handler runs in less than 1 second. I've tried different fld properties but I can't speedup the "put var into fld myfield" Anybody know any tricks to speed this up? (I know, I need a new computer) The time is probably being spend on the word wrap which is pretty slow on the Mac in 2.2.X. One easy way to speed this up would be set the dontWrap property of the fields to true. If you can't do that, or reduce the amount of text being put into the fields (no way to *see* all that text, is there?), or get a faster machine, or run Linux (which does this *much* faster) on your PPC, your only option is to wait for 2.3 which does text flow anywhere from 3 to 10 times faster. Regards, Scott Scott Raney [EMAIL PROTECTED] http://www.metacard.com MetaCard: You know, there's an easier way to do that...
Re: Metacard on Window NT
Hugh Senior wrote:
[re app hanging on nt]
I'm no NT expert either, but I have been using the NT/MC combination daily
for over a year with no problems. Applications developed under NT have also
worked flawlessly on all client machines (NT, 95, 98) including server
delivery over their networks.
Yes... What I have found with another tool is that apps developed under
NT will work on other Win32, but apps developed on other Win32 will not
necessarily work with NT.
For instance if one is using 16-bit external C code objects, they'll
probably function adequately under 9x, but fail under NT.
And if any part of your code attempts to access hardware directly, NT
will kill it, since that is strictly forbidden in the microkernel
architecture.
Also, as Scott mentioned, securities and permissions are important.
Also, NTFS-formatted drive slices 4 Gb in size might be incorrectly
reporting themselves to your app, causing it to hang up while it tries
to figure out how to run on a drive with something like -2584629593 Mb
available.
A rule of thumb to *always* follow with app development is "port early,
port often". Assuming that just because an app works on one system it
will work on all is a costly mistake. For instance I'd bet you're
looking at a completed app here. Deconstructing it to find the source of
the trouble could take many times longer than it would have done to just
go back and forth as you built, confirming functionality. (Ouch.)
I know that MS claims, more or less, that the Win32 API is consistent
across their 3.75 flavors of it, but Microsoft, inarguably, is staffed
by liars. You can't trust anything they tell you.
I don't know that anyone's going to be able to help you find the
problem... at least not without some more details, such as where the app
is hanging, how "hard" it's hung, system environment ("Win98" is not a
description of a system environment) and so on...
--WthmO
Re: Cascading menus
I think the problem is that you've set the button style to cascading. Try setting it to pulldown or popup instead, but don't make any other changes to the button or script. You will still get the cascading submenus, and the menupick script should now work. I'm not sure why it works this way and not with the "cascading" style; perhaps Scott can explain. Marni [EMAIL PROTECTED] wrote: I have tried making a cascading menu. I created a button named "Test Cascading" and set its style to cascading. I then typed the following into the button contents field. The "Subtest" items have a tab before them. Test 1 Test 2 Test 3 Subtest 1 Subtest 2 Test4 I unchecked all the items in the button behavior (Auto highlight, Auto arm, etc.). I then put the following into the button script: on menuPick thePick beep put thePick end menuPick When I click on the button, a cascading menu appears to the right as desired. When the mouse passes over "Test 3," the cascading menu showing the Subtest 1 2 items appear as desired. So far so good. When I release the mouse on "Test 1," "Test 2," "Test 3," or "Test 4," the computer beeps and the menuPick is displayed in the Message Box. However, when I release the mouse on one of the submenus "Subtest 1," or "Subtest 2," nothing happens. Am I missing something? Philip Chumbley P.S. I'm running MC 2.2.5. Thanks Scott! -- Marni Centor ([EMAIL PROTECTED]) Summit Systems, Inc. 22 Cortlandt Street New York, NY 10007 (212) 896-3466 The views and opinions expressed in this email message are the senders own, and do not necessarily represent the views and opinions of Summit Systems Inc.
A SpreadSheet Example
I would like to add a hearty "AMEN!" to the accolades being given Scott Raney and his product. In a message dated 13/9/99 1:04:54 am Richard Gaskin wrote: The MetaCard engine is cool enough for the money, but the support has been a mind-bending good time (how on earth does Raney stay on top of all his email?) I just want to chip in and support this. I started to play with the starter kit towards the end of last year. I still can't afford the full version yet, but as soon as I can The fact is, I haven't had so much fun with any product since Hypercard arrived with my first Mac. I'd also like to thank Kevin Miller at Cross Worlds Computing for his kind support and encouragement. He is a great guy to have put up with me as he has. The more I play with MetaCard, the better I like it. In fact, I work with MetaCard more than with HyperCard now. I haven't has so much fun doing scripting in ages. I have created a stack illustrating just how powerful even the Starter Kit can be. In the general theme of Jacqueline Gay's PseudoTable and Marni Centor's PseudoTable Plus, my SpreadSheets stack illustrates how to take a single MetaCard field and create a spreadsheet object. I've tried to give clear instructions on scripting, and hints as to how to get the most use out of a limited number of lines of programming. You can also create your own spreadsheet object, copy it, and paste it into your own stack! Kevin has graciously posted it on his site. You can download it from http://www.xworlds.com/metacard/contributorfiles/spreadsheet.zip By the way, if you can think of ways to make it better, please let me know. I am still learning. Like Richard Gaskin, I still can't afford the full version yet, but as soon as I can, I will! But in the mean time, I have learned that you can do some serious and amazing things with even the Starter Kit. Awesome product, people! Raymond Griffith
Re: on topic? httpProxy or worse
On Tue, 14 Sep 1999, Kevin Miller wrote: On Tuesday, Sept 14 1999, Scott Raney wrote: Are you sure you got the address of the proxy server correct, complete with port? Is it the same as the settings used in Netscape, for example? Also note that only one type of HTTP proxy is currently supported, the one where you have a specific host name or IP address. The other kind has you access a file on your side of the firewall that contains the address of the proxy. If that's the kind your browser is set up to use, you can usually get the address of the other kind by asking your system administrator. System administrators often supply these nowadays to make it easier to configure a client. (Having recently upgraded our LAN I've spent a while fighting this stuff.) However, that address only refers to a configuration file that contains all the other addresses, so you should be able to open/download that and look, or guess the proxy from the address for that file (it will typically be the same IP with whatever port you use for HTTP, usually 80 or 8080). Any plans to support this kind of auto-configuration in MetaCard? Even if we solve this problem, you're still left with the one of figuring out what to set the httpProxy property to, which in some cases is a rather difficult problem to solve. Current thinking is to move HTTP support out of the engine entirely and make it a standard scripted solution using the new socket features. This will make it much easier to add new features like these, and allow people with special needs to build custom solutions. Regards, Scott Scott Raney [EMAIL PROTECTED] http://www.metacard.com MetaCard: You know, there's an easier way to do that...
RE: Metacard on Window NT
From: [EMAIL PROTECTED] (Christoph Wollek) Subject: Metacard on Window NT Date: Tue, 14 Sep 1999 15:00:35 + ..does Metacard run under win NT? I developed a standalone with win 98, but a friend told me, under win NT the program hung up totally. (The standalone gets text from other stacks in the same folder.) Sorry, I don't now the problem in your case but my experience with MetaCard under NT are: * runs perfectly * it takes longer to open a stack (vs. win95) * if doesn't display UniCode characters (vs. japanese win95) And I don't now if this is special only to NT. Sometimes on a certain PC there is problem refreshing stacks (colors). Maybe that's just a problem of an old graphic driver. Regards, Ulf
Re: Odd button behavior
On Monday, Sept 13 1999, David C. Tremmel wrote: I'm new to the list, so forgive me if this has already been covered. I am writing an application that, in part, allows users to choose what information they would like to see displayed from a text file containing many possible choices. I first wrote this using a scrolling field set to list behavior, so that by selecting a line in the field the user could display the information about what was on that line. This was done with the field's script, and it worked fine. However, I wanted to use a button - set to a pop-up or other type of menu - as the method for making this selection. I did this, and copied the script with some minor modifications (since I was dealing with a button, not a field, and which I know work) to the button. However, now the script will not work. The problem appears to be that the result gets set to "eof" the very first time it reads from the file, even though it should take over thirty reads before it reaches the end of the file. This of course kicks me out of my read loop informs me that it couldn't find what I was looking for. However, the field script in the same stack (!) works great. I've tried the button as a pulldown, popup, option, and even combobox, all with the same result. Is something going on with the "result" constant that I don't understand? Could you possibly post the script? Also, I have a problem with pulldown, popup, and option buttons. If they have more choices than can be accomodated on the screen, those choices beyond the bottom edge of the screen cannot be selected. Normally, moving the mouse to the bottom of the popup would cause the menu to scroll so you could get to the hidden choices, but this does not occur in MetaCard. I can only do this with a ComboBox, but I really don't want to use this because there's no way to keep the user from editing in the box, which I don't want them to do. I'm using version 2.2.5 on a Mac, System 8.6. 2.2.5 doesn't support scrolling menus. You can workaround this by using the free scripted scrolling menu utility on our web site, or waiting until 2.3 is released, which will have native scrolling menu support on all platforms. Regards, Kevin Any insights would be greatly appreciated. David Tremmel Kevin Miller [EMAIL PROTECTED] http://www.xworlds.com/ Cross Worlds Computing, MetaCard Distributors, Custom Development. Tel: +44 (0)131 672 2909. Fax: +44 (0)1639 830 707.
