Celestin Apprentice 5/Apprentice-Release5.iso/Information/THINK C Digest/1990/90-09

Text File | 1995-12-31 | 26KB | 662 lines

Path: ucivax!gateway
From: [e-mail] (Mark Nagel)
Subject: ARCHIVE: Prepare Demo 1.3
Message-ID: <[e-mail]>
Newsgroups: fa.think-c
Reply-To: [e-mail]
Organization: University of California, Irvine - Dept of ICS
Lines: 56
Date: 13 Sep 90 22:52:15 GMT
Phone: [phone]

A new demo issue of Prepare has been released:

/*****************************************************************************/

This is the latest demo issue of Prepare(): The Journal of Think Class Library
Programming.  This demo issue includes, among other things:

- A demo of The Prepare() View Editor, a full-fledged editor for hierarchies
of Think C Panes.  By compiling your objects into the View Editor (or
compiling the View Editor into your project!), you can edit any subclass of
CPane in a MacDraw-like way.  The Prepare() View Editor includes a
ResEdit-like "TMPL" processor that edits arbitrary resources.

- Patches for the Think Class Library making IViewTemp() work for all
subclasses of CPane, and adding a new method, WViewTemp(), which writes
the View's template data back into a buffer.  These patches are included
in their entirety in the demo issue.

- Bug Alerts, including patches and work-arounds for several Macintosh
Toolbox and Think Class Library bugs and misfeatures.

- Samples from the Manual Pages for Prepare().

Prepare() is a bimonthly journal containing source code for use with
Symantec's THINK C version 4 and the Think Class Library.  Each issue
contains many new classes for use in your own programs (the Prepare()
Class Library), detailed manual pages for each class, and one or more
demonstration applications that show you how to use them.  The entire
issue is distributed solely on diskette.  The Manual pages can be
printed for easy reference, or read on-line.

Subscribers license the code in Prepare() via an agreement very similar
to Symantec's THINK C license:  no licensing fees are required for
compiled applications written using the code in Prepare().  The source
code and documentation in Prepare() is copyrighted by Lexington Software
Design and licensed to the subscriber and may not be redistributed.

Prepare() is an independent journal not affiliated with Symantec
Corporation.

Demo Issues of Prepare() are available on Compuserve (in the SYMANTEC
forum) and on the Internet via anonymous FTP from ics.uci.edu in
the files /mac/think-c/Prepare-Demo-1-*.hqx (where * is any digit).

A one year subscription to Prepare() is $495 in the US & Canada,
$545 international.  Educational discounts are available.  Subscription
inquiries should be directed to the phone number and address below.

For further information, please call:  Prepare(), Lexington Software Design,
72A Lowell Street, Lexington, MA  02173, [phone].

/*****************************************************************************/

[saved as: mac/think-c/Prepare-Demo-1-3.hqx; 576K]

Mark










Path: ucivax!gateway
From: [e-mail] (Charles Bazaar X5416)
Subject: Addition to list
Message-ID: <[e-mail]>
Newsgroups: fa.think-c
Lines: 1
Date: 14 Sep 90 21:12:25 GMT

    [e-mail]










Path: ucivax!gateway
From: [e-mail] (Mark Nagel)
Subject: gatewaying from comp.sys.mac.programmer...
Message-ID: <[e-mail]>
Newsgroups: fa.think-c
Reply-To: [e-mail]
Organization: University of California, Irvine - Dept of ICS
Lines: 12
Date: 21 Sep 90 17:24:31 GMT
Phone: [phone]

Considering that this list is, um, underutilized, I thought that
people might want me to select articles posted to
comp.sys.mac.programmer related to Think C and gateway them here, at
least for a while.  It's kinda eerie having a mailing list of over
230 people and 5 exploders with absolutely no discussion... :)  Of
course, we don't want dozens of mostly meaningless messages per day,
either, like with the old list.

Let me know what you think...

Mark
Think C List pseudo-moderator










Path: ucivax!gateway
From: convex![e-mail] (Russell Donnan)
Subject: EPROM Programmer
Message-ID: <9009211845.AA09176@magnum>
Newsgroups: fa.think-c
Lines: 11
Date: 21 Sep 90 19:39:30 GMT


Anyone out there written any Think C code to talk to an EPROM programmer.
I don't really care what kind of programmer, either.  I'll buy the appropriate
programmer, if I have some code to start with or at least some libraries
to help out.

If you aren't willing to part with any code, but have done the project, I
would still like to hear from you!

Thanks,
-Russ  ([e-mail])










Path: ucivax!gateway
From: [e-mail] ("Robert E. Webber")
Subject: Editing ThinkC
Message-ID: <[e-mail]>
Newsgroups: fa.think-c
Lines: 19
Date: 21 Sep 90 20:05:46 GMT


Speaking of comp.sys.mac.programmer messages, perhaps one of the most
interesting discussion of ThinkC that I have seen there in recent weeks
raised the question of altering the default behavior on various menus
(specifically the question of replacing the old program with the new
program when recompiling).  This can apparently be done by editing
the appropriate resource file.

This, in turn, raises the issue of just what hooks actually exist in
the ThinkC interface.  For example, how much can one change the behavior
of the ThinkC editor?  Can one set up an alternative editor so that it
works as closely with the project manager as ThinkC's editor does?  Can
one have programs launch compilations without going thru the mouse interface,
i.e., can a program answer the questions that ThinkC normally asks the user
via the console so that a series of compilations could be set up to run
unattended?

--- BOB ([e-mail])











Path: ucivax!gateway
From: [e-mail] (kevin p burke)
Subject: mac tcp  for think c
Message-ID: <[e-mail]>
X-Mailer: Mail User's Shell (6.4 2/14/89)
Posted-Date: Fri, 21 Sep 90 17:40:51 -0400
Newsgroups: fa.think-c
Lines: 9
Date: 21 Sep 90 22:14:11 GMT

does semantic produce a library to use with mac tcp? what do I do to utilize it?
-k

--
Kevin P. Burke, Corporate Mystic, Litle & Co. | POB C26, Hanover, NH 03755-6000
-------------------------------------------------------------------------------
  "Just Do It."                               | [phone]/5558
-------------------------------------------------------------------------------
[e-mail]                    {backbone}!dartvax.dartmouth.edu!litle!kevin










Path: ucivax!gateway
From: [e-mail] (Bruce)
Subject: Re: gatewaying from comp.sys.mac.programmer...
Message-ID: <2862940762-947271@PCAI>
In-Reply-To: Msg of 21 Sep 90 17:24:31 GMT from Mark Nagel <[e-mail]>
Newsgroups: fa.think-c
Lines: 19
Date: 22 Sep 90 02:09:36 GMT

Mark (et. al.),

I know almost nothing about the mechanics of operating this list, but it
seems to me that it is so "underutilized" because when someone posts a
question, there is no mechanism for the replies to automatically be sent
to the list.  Unless someone replying to a message explicitly includes a
"Cc: [e-mail]", the rest of the list never sees
anything beyond the original posting.  If there were some way to cause
everyone's mailers to automagically include such a line when replying to
a message, we might actually see some discussion going on here.

Re: gatewaying articles from comp.sys.mac.programmer.  For myself, such
a service would be redundant, since I already read that newsgroup on a
semi-regular basis, but I'm not opposed to the idea if there is a
significant portion of the list members who don't have usenet access.

Cheers,
Bruce Florman
[e-mail]










Path: ucivax!gateway
From: [e-mail] (Ben Haller)
Subject: Re: gatewaying from comp.sys.mac.programmer...
Message-ID: <[e-mail]>
Newsgroups: fa.think-c
Lines: 9
Date: 22 Sep 90 02:24:55 GMT

> Considering that this list is, um, underutilized...
> [...]
> Let me know what you think...
I don't really care.  I read comp.sys.mac.programmer, so I guess it would
be somewhat annoying to have to read all that stuff twice.  Maybe most
everybody on this list does, so in that case it would really be a waste
of time.

-Ben Haller ([e-mail])










Path: ucivax!gateway
From: bushido![e-mail] ("Michael A. Nowak")
Subject: THINK C 4.0 under A/UX
Message-ID: <[e-mail]>
X-Mailer: Mail User's Shell (6.3 6/25/88)
Newsgroups: fa.think-c
Lines: 12
Date: 22 Sep 90 18:03:28 GMT

I've initiated some discussion of the problems of using THINK C 4.0 and
the debugger under A/UX.  From what I can tell, some people have gotten
it to work.  If you've gotten THINK C and the debugger to work under A/UX,
could you drop me a line, telling me under what circumstances you've
gotten it to work?

I appreciate your help!

Michael Nowak
--------------------------------------------------------------------------
                                                           Bushido Systems
                                               [e-mail]










Path: ucivax!gateway
From: [e-mail] (Mark Nagel)
Subject: Re: gatewaying from comp.sys.mac.programmer...
Message-ID: <[e-mail]>
In-reply-to: Your message of Fri, 21 Sep 90 16:19:22 -0500.
     <2862940762-947271@PCAI>
Newsgroups: fa.think-c
Reply-To: [e-mail]
Organization: University of California, Irvine - Dept of ICS
Lines: 23
Date: 22 Sep 90 19:34:23 GMT
Phone: [phone]

    I know almost nothing about the mechanics of operating this list, but it
    seems to me that it is so "underutilized" because when someone posts a
    question, there is no mechanism for the replies to automatically be sent
    to the list.  Unless someone replying to a message explicitly includes a
    "Cc: [e-mail]", the rest of the list never sees
    anything beyond the original posting.  If there were some way to cause
    everyone's mailers to automagically include such a line when replying to
    a message, we might actually see some discussion going on here.

If anyone has problems like this, please talk to your postmaster.
This should not happen.  I replied to this message simply by typing
'repl' (we use MH), but any mailer should be intelligent enough to
do the right thing.  The relevant headers from your message looked
like:

    From:    Bruce <[e-mail]>
    To:      Mark Nagel <[e-mail]>
    Cc:      [e-mail]

When I reply to that, it included think-c as a recipient.  Sorry for
the meta-discussion...

Mark










Path: ucivax!gateway
From: [e-mail] (Kevin Purcell)
Subject: Re: gatewaying from comp.sys.mac.programmer...
Message-ID: <[e-mail]>
In-Reply-To: Your message of 22 Sep 90 02:09:36 GMT
Newsgroups: fa.think-c
Lines: 47
Date: 23 Sep 90 15:00:02 GMT

On 22 Sep 90 02:09:36 GMT Bruce ([e-mail]) said:

>Mark (et. al.),
>
>I know almost nothing about the mechanics of operating this list, but it
>seems to me that it is so "underutilized" because when someone posts a
>question, there is no mechanism for the replies to automatically be sent
>to the list.  Unless someone replying to a message explicitly includes a
>"Cc: [e-mail]", the rest of the list never sees
>anything beyond the original posting.  If there were some way to cause
>everyone's mailers to automagically include such a line when replying to
>a message, we might actually see some discussion going on here.

I agree, this problem could be alleviated if folks who ask a question would
post a summary back to the list of the answers they get. I read lists like
this because they often tell me things that *might* be useful in the future,
or give me ideas to do things which might not have crossed my mind.

So *please* folks, post summaries of the replies to your questions. If you
asked one in the past post a summary *now*.

And put a Cc: or a Reply-to: in the header, it will liven up the list.

>
>Re: gatewaying articles from comp.sys.mac.programmer.  For myself, such
>a service would be redundant, since I already read that newsgroup on a
>semi-regular basis, but I'm not opposed to the idea if there is a
>significant portion of the list members who don't have usenet access.

Though I have access to net news I rarely get time to read all the articles
on think-c (or that are think-c releated). A clipping service to post
relevant bits here (such as a complete question and answer session on a given
topic in one posting) would be very useful, even to those with netnews access.

>
>Cheers,
>Bruce Florman
>[e-mail]

More suggestions, post to the list!

-kgp

Kevin G. Purcell       | [e-mail]
Surface Science,       |
Liverpool University   | Programming the Macintosh is easy if you understand
Liverpool L69 3BX      | how the Mac works and hard if you don't. -- Dan Allen










Path: ucivax!gateway
From: [e-mail] (Monty Solomon - Temp Consultant)
Subject: Re: gatewaying from comp.sys.mac.programmer...
Message-ID: <[e-mail]>
Newsgroups: fa.think-c
Lines: 8
Date: 23 Sep 90 23:59:59 GMT

> If anyone has problems like this, please talk to your postmaster.
> This should not happen.  I replied to this message simply by typing
> 'repl' (we use MH), but any mailer should be intelligent enough to
> do the right thing.  The relevant headers from your message looked
> like:

You should use the replyall command or its equivalent instead of
just reply.










Path: ucivax!gateway
From: metaphor!jobone.metaphor.com![e-mail] (Mike Greenawalt)
Subject: comp.sys.mac.programmer proposal
Message-ID: <[e-mail]>
Newsgroups: fa.think-c
Lines: 6
Date: 24 Sep 90 01:05:35 GMT

I guess I just have to ask the obvious question.  Why is there not a Think C
newsgroup in Usenet?  Seems to me that Think C is as popular and widespread in
interest as a number of other topics that are included.

-- Mike Greenawalt











Path: ucivax!gateway
From: [e-mail] (Nick Rothwell)
Subject: Discussion time
Message-ID: <[e-mail]>
In-Reply-To: Mark Nagel's message of 21 Sep 90 17:24:31 GMT
             <[e-mail]>
Newsgroups: fa.think-c
Lines: 28
Date: 24 Sep 90 10:45:17 GMT

Ok then, to kick off some discussion: what do you think of the THINK Class
Library? I've been Mac programming (with THINK C 3.0) for a couple of
years, and I'm just getting into OOP and the Class Library now.

Initial impressions: The TCL is very impressive and comprehensive. But, it
doesn't really make the Mac any less of a complicated machine to program:
it just goes some way towards managing the complexity.

The TCL seems rather too complicated in some ways, with invariants and
things which have to be done in certain orders (none of which come
to mind right now - just a general impression). Also, I don't understand
why it should have *any* globals (gXXX) at all; maybe this is just because
it's not possible to redefine initialisation methods to pass in all the
objects required by a newly created object. Shame, though.

Documentation: pretty good, although I've found I've had to try a lot of
things where the documentation doesn't go into enough detail. One thing
I'd like to see, alongside the inheritance diagram: a *clear* list of
what methods are defined, and overridden, for all the class objects. The
listings provided seem incomplete and inconsistent (some of the struct
listings seem include purely inherited methods as well), although I haven't
spend a lot of time examining them yet. The other annoying thing is that
the methods listed include ones which are private, and ones which shouldn't
be overwritten - these should be marked clearly and concisely (I have
the noise-words NEW, OVERRIDE and PRIVATE in all my class descriptions
and methods).

        Nick.










Path: ucivax!gateway
From: [e-mail]
Subject: Re: gatewaying from comp.sys.mac.programmer...
Message-ID: <[e-mail]>
Newsgroups: fa.think-c
Lines: 16
Date: 24 Sep 90 14:22:55 GMT

>Re: gatewaying articles from comp.sys.mac.programmer.  For myself, such
>a service would be redundant, since I already read that newsgroup on a
>semi-regular basis, but I'm not opposed to the idea if there is a
>significant portion of the list members who don't have usenet access.

   I read the newsgroup, and  don't want the extra mail traffic. If this is
what the mailing list is to be used for, please remove me from the list.

   Thanks

    James D. Meiss
    Program in Applied Mathematics
    University of Colorado, Boulder

    [e-mail]











Path: ucivax!gateway
From: [e-mail]
Subject: Re: Editing Think C
Message-ID: <[e-mail]>
X-Mailer: ELM [version 2.3 PL6]
Newsgroups: fa.think-c
Lines: 26
Date: 24 Sep 90 14:58:41 GMT


= This, in turn, raises the issue of just what hooks actually exist in
= the ThinkC interface.  For example, how much can one change the behavior
= of the ThinkC editor?

Well, I wouldn't call them hooks, but there are ways to tweak the editor so
it does a little more than normal.  I've written an init called Kiss which
adds several useful functions to the editor -- move by word, page up & down,
set mark, parenthesis kissing, that sort of thing.  If anyone wants it,
send me mail and I'll send you a copy.

= Can one set up an alternative editor so that it
= works as closely with the project manager as ThinkC's editor does?  Can
= one have programs launch compilations without going thru the mouse interface,
= i.e., can a program answer the questions that ThinkC normally asks the user
= via the console so that a series of compilations could be set up to run
= unattended?

I think not.  This has been a recurring topic on comp.sys.mac.programmer,
but I don't think this sort of hook will make it into a new version any time
soon.

Alex Chaffee
chaffee@reed.{UUCP,BITNET}
Reed College, Portland OR 97202
____________________










Path: ucivax!gateway
From: [e-mail] (Mark Nagel)
Subject: results
Message-ID: <[e-mail]>
Newsgroups: fa.think-c
Reply-To: [e-mail]
Organization: University of California, Irvine - Dept of ICS
Lines: 9
Date: 24 Sep 90 18:22:59 GMT
Phone: [phone]

I've received many comments about the proposed extraction of Think C
articles from c.s.m.programmer for distribution here, and the
sentiments seem about equally split, although those opposed are
mostly extremely opposed and those for are pretty much like "sure,
that would be ok."  For now, I will leave things as they are, so if
the list is quiet, then as someone so aptly put it, perhaps there is
simply nothing to say...

Mark










Path: ucivax!gateway
From: metaphor!jobone.metaphor.com![e-mail] (Mike Greenawalt)
Subject: Re: Discussion time
Message-ID: <[e-mail]>
Newsgroups: fa.think-c
Lines: 61
Date: 24 Sep 90 18:28:15 GMT


Nick Rothwell <[e-mail]> writes:

>  Ok then, to kick off some discussion: what do you think of the THINK Class
>  Library? I've been Mac programming (with THINK C 3.0) for a couple of
>  years, and I'm just getting into OOP and the Class Library now.

Yeah, me too.  I have been a Think C user since Version 1.02.  Version 4 with
the Class Library came along at a very opportune time.  The time for OOP has
arrived, and I needed to get my head around the concepts.  The THINK Class
Library has been a very good tool for that.  Of course, OOP is not the
difficulty, rather it is OOD that is the hard part.

>  Initial impressions: The TCL is very impressive and comprehensive. But, it
>  doesn't really make the Mac any less of a complicated machine to program:
>  it just goes some way towards managing the complexity.

The Macintosh *is* a complicated machine to program.  I have used other
simplifying aids (i.e. Programmer's Extender) as well, and even though one can
indeed implement a lot of functionality with fewer lines, no less understanding
of what is going on is needed if one is to achieve success.

Where such packages achieve a reduction in complexity, they simultaneously
sacrifice flexibility.  The TCL has a particular view of the way things should
work.  I found that with the very first application I wanted to do, which I
did not think was particularly exotic, I crashed headlong into conceptual
barriers imposed by the design of the TCL.  What I wanted to do is demonstrably
possible on the Mac, but not using the TCL as it is delivered.  The TCL puts
up a pretty vanilla Mac interface.  But that is as it should be.

But, the TCL allowed me to quickly and easily overcome these limitations using
subclassing and method redefinition.  The task was much cleaner than using an
extender library and trying to work around its limitations.  For me, this is
where OOP wins hands down!

    ...

>  Documentation: pretty good, although I've found I've had to try a lot of
>  things where the documentation doesn't go into enough detail. One thing
>  I'd like to see, alongside the inheritance diagram: a *clear* list of
>  what methods are defined, and overridden, for all the class objects. The
>  listings provided seem incomplete and inconsistent (some of the struct
>  listings seem include purely inherited methods as well), although I haven't
>  spend a lot of time examining them yet. The other annoying thing is that
>  the methods listed include ones which are private, and ones which shouldn't
>  be overwritten - these should be marked clearly and concisely (I have
>  the noise-words NEW, OVERRIDE and PRIVATE in all my class descriptions
>  and methods).

I have found the documentation to be weak in explaining how the TCL classes
interact, and sometimes it is important to know.  I had to resort to making
listings of all the classes, to which I refer frequently during the course of
a design.

What I wish for is better online support.  A class browser is needed.
Debugger support that interprets a pointer to an object to tell you its class
would be very useful.

-- Mike Greenawalt












Path: ucivax!gateway
From: metaphor!jobone.metaphor.com![e-mail] (Mike Greenawalt)
Subject: Re: Discussion time
Message-ID: <[e-mail]>
Newsgroups: fa.think-c
Lines: 24
Date: 26 Sep 90 00:12:02 GMT

>
>  I know alot of talk has been going around about the Think Class Library,
>  some people like it (me included) and some people think it is not sophisticated
>  enough.  My question is "Is their enough people using it to start some sort of
>  group or something for those of us wanted to learn more but cannot (or afford)
>  other publications devoted to the TCL?"
>          Thanks,
>          Johnathon Suker
>

How many publications are out there which are devoted to TCL?  I have seen demo
versions of the Prepare() electronic publication, and I recently got a mailer
from Symantec which promoted a new one called SPLash (or something like that),
but I have not seen any real books yet.  At $495, Prepare() is a bit too steep
for my blood, but the $30 for SPLash (wish I could really remember the name :)
seemed OK.  I signed up sight unseen.  I suspect that you get what you pay for,
however.

TCL is pretty basic, but it gives a base for lots of expansion.  Good mech-
anisms are needed for sharing new classes that are based on it.  Right now, I
would really like to have a class for multi-windowed documents.  Anyone out
there have one they would be willling to share?

-- Mike










Path: ucivax!gateway
From: [e-mail]
Subject: Re: Discussion time
Message-ID: <[e-mail]>
In-reply-to: Your message of Tue, 25 Sep 90 16:22:08 -0700.
             <[e-mail]>
Newsgroups: fa.think-c
Lines: 4
Date: 26 Sep 90 00:17:41 GMT

Can you send me more information about Splash or let me see the copy you have.
I already get Develop (mainly for the CDs) and 30 is okay with me.
        Thanks
            Johnathon]










Path: ucivax!gateway
From: [e-mail]
Subject: SPLAsh information
Message-ID: <[e-mail]>
Newsgroups: fa.think-c
Lines: 27
Date: 26 Sep 90 06:22:30 GMT

A couple people have posted information asking about SPLAsh. SPLAsh is an
association for programmers who use THINK C and THINK Pascal. SPLAsh publishes
a quarterly journal, which includes a floppy disk containing source code
from the articles--in both C and Pascal--and other goodies. SPLAsh will also
host events at major Mac shows and provide limitied online support (mostly
on America Online). Membership is $30 per year. It is an independent group,
but is endorsed by Symantec.

The journal will cover all aspects of programming, although there will be a
strong emphasis on the THINK Class Library (TCL). I am the editor of the
journal and also wrote the TCL under contract for Symantec. I plan on
introducing new classes for SPLAsh and writing regular articles explaining
my design philosophy.

I'm probably going to regret this offer, but anyone who would like to receive
a copy of the first journal issue (this is a mini-issue which was given out
for free at MacWorld Boston) should send me their name and US Mail address.
I can be reached at this account ([e-mail]) or at

  SPLAsh Resources
  1678 Shattuck Ave #302
  Berkeley, CA  94709
  [phone]

Once again, this is a FREE offer.

-- Gregory Dow