home *** CD-ROM | disk | FTP | other *** search
Text File | 1994-05-04 | 90.5 KB | 2,098 lines | [ TEXT/R*ch]
Path: ucivax!gateway From: nagel@ICS.UCI.EDU (Mark Nagel) Subject: ARCHIVE: Exception Handler Message-ID: <25660.660009606@ics.uci.edu> Newsgroups: fa.think-c Reply-To: nagel@ICS.UCI.EDU Organization: University of California, Irvine - Dept of ICS Lines: 47 Date: 1 Dec 90 00:02:16 GMT Phone: (714) 856-5039 Date: Fri, 30 Nov 1990 18:24 EST From: "R. Mark Fleming" <FLEMINGM@qucdn.queensu.ca> Subject: Exception Handler ExceptionHandler.sit is a set of routine to enable Think C users to compile and e shading; hidden line, surface, and object removal; matrix data plotting functions; PICT input and output; offscreen and animation facilities; grid backdrops and text labels in 3d space; and lots more. The documentation is good, and you *DO* get souce code. There is a HyperCard XCMD/XFNC version as well. HOWEVER, I have not used it extensively, and you may find the data types not to your liking. There are no liscencing/royalty fees, etc. If I recall correctly, there is a demo file on CompuServe, probably on Symantec forum, or Mac Programming forum. Anyway, if interested, give the guy a call, as he's really nice, and he can tell you better than me whether this is a worthwhile thing. Just thought you might like to know that there is an option. Cheers, David McCormick MIT-EAPS Geology Cambridge, MA 02139 dmac@athena.mit.edu Path: ucivax!gateway From: nagel@ICS.UCI.EDU (Mark Nagel) Subject: ARCHIVE: CAfterDark Message-ID: <23187.660259806@ics.uci.edu> Newsgroups: fa.think-c Reply-To: nagel@ICS.UCI.EDU Organization: University of California, Irvine - Dept of ICS Lines: 74 Date: 3 Dec 90 21:34:11 GMT Phone: (714) 856-5039 Date: Mon, 3 Dec 90 01:50:12 -0600 From: Jeff Francis <jfrancis@umaxc.weeg.uiowa.EDU> Subject: CAfterDark 2.0 - It's all here! CAfterDark 2.0 (c) Copyright Jeff Francis 1990 Introduction ----------- CAfterDark is a base class that allows you to write graphics modules for Berkeley Systems' After Dark in terms of an object based abstraction. The base class, CAfterDark, provides the default behavior for a graphics module by simply blanking the screen. Subclasses can then add or change this basic behaviour. CAfterDark 2.0 now supports After Dark 2.0. This version has been tested on a SE and a IIfx with a 8.24GC card. Disclaimer and Copyright Notice (i.e., # of Lawyers >> # of Programmers!) ------------------------------------------------------------------------ The software and accompanying instructions for use are provided "as is" without warranty of any kind. Further, Jeff Francis, does not warrant, guarantee, or make any representation regarding the use of the software or accompanying instructions in terms of correctness, accuracy, currentness, or otherwise. (c) Copyright Jeff Francis 1990. Permission is granted to make and distribute copies of this software, provided this disclaimer and copyright notice are preserved on all copies. The software may not, however, be sold or distributed for profit, or included with other software which is sold or distributed for profit, without the permission of the author. CAfterDark is free. There is no shareware fee. Files ---- ADGMTypes.h - Enumerations, typedefs and defines for an 'ADgm' code resource. GlueCode.[ch] - Glues the regular "main" interface to our class based interface. You will need to change GlueCode.h so that it creates an instance of your subclass. See GlueCode.h for more information. CAfterDark.[ch] - The base class for all graphics modules CProjectile.* - An example subclass Other Work in Progress --------------------- I'm currently developing what I call "Screen Saver Physics" graphics modules. These savers provide simple animated physical systems. Two of the classes, CAnimator and CAnimatedObject, provide a simple animation facility. Also, some of the CAnimatedObject subclasses may be reusable. I'll make all of this stuff available when I'm finished. FYI - The base class, CAfterDark, was never meant to be very useful on its own. My hope is to develop a general set of CAfterDark subclasses that can be subclassed and composed to build bigger and better After Dark graphics modules with minimal effort. Contact Me At ------------ Jeff Francis CS Student, University of Iowa jfrancis@umaxc.weeg.uiowa.edu (319) 354-0765 [saved as: /mac/think-c/classes/cafterdark-20.hqx; 19K] Path: ucivax!gateway From: nick@lfcs.edinburgh.ac.UK (Nick Rothwell) Subject: CPreferences (again) Message-ID: <1583.9012031644@subnode.lfcs.ed.ac.uk> Newsgroups: fa.think-c L ass. Also, note that the class creates a floating window so it is properly displayed. Thanks to someone on this list (Greg Dow?) for the recommendation. Finally, the initializing method shows how to cobble up a window without an attached document. The programs are copyrighted in case I work up a MacTutor article around them. Otherwise, you have my permission to use them in your programs without further notice. Documentation is still a little shoddy; my apologies. Comments and criticisms to me, please. Martin Minow minow@bolt.enet.dec.com (This file must be converted with BinHex 4.0) :&&4SCA*YEfePG'9b3faKFh-ZFfPd!&0*9#&6593K!!!!!"Gm!!!!!2bM8dP8)3! %!!!AI(*-BA8"!!!!!!!!!!!#%&0`E'PdFd&LEh9d3Qpi,Q0j)%0[EfYLEfpV!!! !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!&4&@&4,38K-!3#MEeF 8ShkTkJ!!!!!!!"8"!!!!!!!!#f!!!"k'!!!!!!!!e(3[9$3!S3,%&$KXdY#C%d6 -QcTdK,c"-e"&JLPei*54-iC0Q$PcG)!B3L50R$*Mk,b4ik,Lb$G`mXK*F`B0(4# T3-6)N3-'L#CKj0"*iqBRd6GhA)!!%B30'a"5D0UF!lA-R)efbT!!D8R3*48dC8$ -3IJ4MGJa*mZi83V#b4XkC865!3ZLidHU#FA)#CS(a*d`EKDk9!Q#c"X3#UQ5fAZ (k"R%4HH',C-e-!JfEf#HS&S&$TN`F%&39NZ(Dm%lBGf-PJ2L6*QEK!'$b2Jjp'M ,Gm!@43hLC"Jb)#5l0*1'64N@BJrc0K0Q69M*K4Nl&NZRMKNc)-#H02fL3B-4DFb 3!#Q$RFL6)9@D&(&#*3L9*%qFH"p+ahL#L`J60h`BF@+$14PYC"G)#C!!C"*++XR aAaeL$(M9(!QimBCUhY8a4aLZHCH!(@qN!Ga5)#k9a%%*,E3I4",KJ8)$p`d4""c jM3&D'K-QS%)B''NNKaeTc,%5#b`Q-!FD+p&aA`*KA%'8BAFN33D,+H`3C"*Z+*4 ''!RT-H1%E#@@)eBpVJ6#3cA"GTJBci%9C"J`*L6M8&b+H"-C2C+9"eA30HB'Ni@ 44j4@))L4ajT+lSQ8Nfb4U*p$+%i8h%a@"AFB@Q@!&PBB3Bla"QDXkFNR#QC!9-G *fB8aaKU1TG$5I5abk1&p*BheaP8SZ%T'P%'ffL%Cpdha'K4b[(('55!e3443+JD *Q4YRh"H'X9@f88FE30khE,0))TZ'Y'e!+@@[V`9(eel-K[8'GY"**NFEEl6afNC X'5&Q'AL%d3C#C35C`,Ah5@C%F@8`SGBCFhh,kT&j2#3(PA"!p+pa,IK`a'X#-cX A#[i#A$("D1#+X-)RpSG(a,l5!Db`a-i",E)S`)"FaXCY2"HZZVjkd@X2Pq'%[@@ XH"m88p!a%l21jVbcZpi'@E+iB4&R(!KZm!`#()$&T@r#SiDF)XNi!hadcffB1S8 48T5"8"i4Qr'ea`IE#LiG90",Kmm*!#fdBrIjk'lFH0#4p(e8FZLFZ$f#i1jFE`# AQeUVJB"eEj8#&mCPIijT4T!!F280JKR"YX%dC!hIj(3CUlDpkpXR$f[9(&CJ@8I 2bNk)E4LYXr(kh`N%rXEJFa9q1*'+kpEiilj*6VPUPQ-ZpqDGIdj8k*X$V)1qr#B J'45c@MQK`3PFV6$$$J-FmF4d!#A('LLBF(hfF,V"GJ*Bbk(e4&bE(+cU)0Aq1XC JB6q(pZk68J2%8)Fcc)&&ApR)T8LPYl#J)5KND%&M6J)F0d`,6A+JLRC)Yj3TV%3 1H@K*!m)!%5)T+!(Q'dT4S)88lfJU*M-T%iZ'!"1C518Q1GP*6hi5&"8D48**@8T 6RK+9-P'PE&ICN9C@p4fed1Pbhb&+4qS`(K#)3&%,F3%D4""&0dbaLPI-MiNB*6) YFK%mAV4G&AP`"-b)!8YDp%%A[aJ@(MaK#N-!bkRQ%-FjUV'1*,(+'P3#KckLNBi Ji%%J0E@A""P5LRp-j""Sb+P#-DQ2haN2FC!!&`3K2+%+92J#&BU!K9!D)6j8F&D Ej"@B62U*Nji%T5K*'8SN&#%*4d"#+R2JRB#iK%T@`P)Dp*!!TV#%$!3TiNi$E08 !,$+%M#R5J3j'*-CRmLG&+`T4J9i8Sbd9*83JZY'AGK3Q13"*Qd-UdT'3!'6*3cd T4#PS`"k#9,D8(-P65%Q4`HTTT(jTKehk$&)kKA+Nc)N-#@@iSF&B53FUQH%0phP SB"ck"S-&b8!R5FP+T*Q%NQ3d35M!8B$)k5-jX#dm)%$"'C!!m"&+VL3+G8M$U4B 6KMZN!!6c2&*LfRN(%23!DQ@i!`SQZDQ9m24p+"LU5qA!8i,FG+G,3NS+)TD%TI+ 8EXlLUC1UG53U5+%+4@$"1Zm6%"!B!6-cBTCISRU()kQJ1qXm!a%%58LZqV0(qQ* E(i*8"MCF"DIkJUUKH[T6e3Ke#%IPRQ#C40@VkSZGE0hUBld+9V'1YDaRI80D(i2 2YYlRVBq9+eeKBPIVS3'[4p*VN!"%bdM3V#4L3b$04Ukk@+NDV,Ch'0pVAV54kT` NICe0%GYmH4q#J-"r!2cFk+!Q0B01V@UP)`KF%j!!'+0*VE""43(I6&De+*'e)%U `d%h83LDcR!X%NaX#&+TQSqP@efXmSqTfekXD&+bcV*++6Nh45jfKB3Zd1Pd)6qh DZcRBY6&NZ&K`*j)#ZmC!"JGQDc+0B`Bk4(L`bDbV[[kRKb)`)3K6H-m3l-TK)b3 "#d8J!K@H!)8JRA3KeR9Aa03JATm'T`cKR3-GBKZBM4L-Z!N`,R,EpcPeXHY`m!U 5G$(hccHNb-B,aS2"U(391[!cI8CfmN53!)2JLf9CZ#"iJ8jXJ"`BX1d*ekRbPGA Aj"59'6NY3!&BEXLrM3#8`@%HF`VBpZ@*Z)#3!$EZ-ajFi"!kU-4c,G#*$'klN!$ 8TBbqBF'Z8"fp1NLrH!k8"JQNUCTT6&F0UlJPm'N0V+mZSk(1kp)b(KUXVcQA#G9 hAV@SlDaU&e$B`[S5p*p*Zq&KHKM%)LEa-%f-BL+%Z+T,b)+,Ibb3!1)@4!JaC30 `S%-ej-d&0'YeLPL)e&2S,2FML,P*iCi(NHJ#Q,S#CQ[%J'@fS26-Zpm&!IRdUkI (U)9(`A+$Z`,6hL"YPk%Sk13R3cR+8RlKP1aKfhDR--bH#9b@"DrP,A0*"ED4Mk% 86Cp%'qU'KekD#Ap#k!eY[('+ZZ!MBb$0BqqcJXe0L1-2GF&ijT!!mPD103%Y,hR (hf$Vb$R'B!C0NFM,C'0Af`3&8BlR@*H5D+226G"+AbI6`cd(N!#VCZJffF'4bRU %1N!3[Eljl(5$2K'Xhq3&2df-eFYJGUh('bL$8`KrFl1T-[5lAh)6HN+*VS+d,f6 YEGpk3D"f`BeBlRK@Z6Zkjl"G'aYfU*CQ012PaZP2"cEGJa9eMb+Fi&-RRF"j,r[ HEA*K*NfB2,LqTi6p$1L@ai$-[Hl`KimpiYJ61m8VE[&pf&Db6ZZ[CfCfqhd,%LX lLEd"HehQVTTCcIRK3CV&Rp@lj4NNSQK()9U"ITfNM`*F*GqAaJfDGG"&Pq9'c9f 3!#&-BTkV'Q8bdjR1PfE*BJblR`A0[d@$,p)5%-qF*S#I"38@4V!AlQ*4pr&Ha[% epE0`Gm-X)89rPmC6khB5920EI#D!""KT9%GrDC1"NXH"2N!%M)& S!34"X5F0'h JX#N$3N`B15d8KS'$"S5C0h*!`%'SN!"K5M)SCBE*if)LL#"Qk'48k5B0R64Kf+6 4%qEU'cFX3)c*bJB%(63dIl*KmqC1'MGR3,3TFrB0Q6Nk5%kKfq4YNc"i8)6T@l9 0R6CK"`1'H6M&$Ved$bCFf&#`PDaebMLHq+,"#$*Pc,bPq3A*&mP&'iD811,Y'$C e3)-3!4*L'4GS4(Kf$9XfE5&e&SDT-dC19pbk'mbK)iFi(B0%+6Zd$8)(L0S43Ha TN!#!c*XkBQBQ''pe$Q('E4jhrakqc2J%jIrL3DrH1hMaj1R-Z3blM(VZGVb4"KR [*B'DG+ZK`0e(9U44aKh[U3"4%@kmpXBF`CA"`S*$!#FFFFE4-Ck%&Q%NKadS!EA KH(1J!C5)ii9aaB"RVCK!Lbqq&`B5CD4a"KTdf)LM($!QX#1&V`RT)T%k@S%N'cC 1XG9ECc`"KeGZ')Q%P(T3'H@8F&Q*TC&@F%NPGjXP%1#!ia'4e)9PS,!Q'@R1f@C 5(18KTi"dlT!!3%!J%'(F(5#id9B))hDQ*TpY$UVJH&+8-8D4%JU(TRTf9X5AAi! pZYjplX9iRQ'),@KIHcV+KpkP!$+U+4hmCHETUIL4LCP%#@c@af-0!-'G!!!: Path: ucivax!gateway From: cohen@excalibur.itd.nrl.navy.MIL (Neil Cohen) Subject: Beginner's programming question... Message-ID: <9012041829.AA01264@constellation.arpa> Newsgroups: fa.think-c Lines: 45 Date: 4 Dec 90 18:36:50 GMT Hi, I'm an experienced (Unix) C/C++ programmer, with limited Mac programming, and almost no experience in the graphics area. I could use a 'jump-start' on the problem I'm currently trying to solve. The problem is basically simple. I need to draw a directed graph in a window. Circles or squares for the nodes, straight lines for the arcs connecting the nodes. Eventually, I'll need to be able to scroll larger graphs that won't fit in a window, and click on nodes to bring up information about them, but for now, just getting the graph displayed would be fine. My hope is to try to learn to use the Think Class Library in the process of building this program. Questions: 1) Is the TCL likely to help or get in my way for something like this? 2) a) Should I use QuickDraw to create the circles? b) Do I want the nodes (circles) to be a resource? If so, how do I access/display it? 3) I'd like to be able to put an identifier (number) inside the circles too - what is the best way to do that? I'm swamped with documentation - more than I can assimilate quickly. This includes Inside Mac, Mac Revealed, the Think manuals etc. Pointers to the right parts of the documents to read, or any other help (including sample code fragments) would be appreciated. Please email responses. I'll pass along anything I learn to anyone who requests it. Thanks in advance, nbc NAME: Neil B. Cohen (Tracor Applied Sciences) PHONE: 703-444-4610 DOMAIN: nbc@excalibur.itd.nrl.navy.mil ************************************************************* * Murphy's Philosophy: Smile - tomorrow will be worse... * * * * O'Tooles Commentary: Murphy was an optimist! * ************************************************************* Path: ucivax!gateway From: dmitri@bolvan.ph.utexas.edu (Dmitri Linde) Subject: Writing resources to a file Message-ID: <9012050446.AA15969@bolvan.ph.utexas.edu.ph.utexas.edu.> Posted-Date: Tue, 4 Dec 90 22:46:54 CST Newsgroups: fa.think-c Lines: 29 Date: 5 Dec 90 18:02:53 GMT Hello! Here is another problem, solution for which may be interesting to many people in this list, just like all problems posted here... My program needs to write a resource to a file, when saving. I made a class identical to CDocument, but I declared itsFile as CResFile (is in More Classes). Then I used that class with Starter project to make CStarterDoc subclass of the class, instead of CDocument. I then changed the following code in DoSaveAs method: itsFile = new(CResFile); itsFile->IResFile(); itsFile->SFSpecify(macSFReply); itsFile->CreateNew(gSignature, 'INIT'); gError->CheckOSError(itsFile->Open(fsRdWrPerm)); I always get OSError ID -39 I can't find the definition of -39 in InsideMac DA and in Think C headers, maybe it is documeneted in Inside Mac? So: 1)What is -39 2)What is worng with that minor change in CStarterDoc? Any Ideas??? Thank you very much, - Dmitri Linde dmitri@bolvan.ph.utexas.edu P.S. Please mail to me, and I will sumurize to list if there is interest. When I finish the this fun project I will post it to the list :-) Path: ucivax!gateway From: nick@lfcs.edinburgh.ac.UK (Nick Rothwell) Subject: CPreferences and other goodies Message-ID: <3818.9012051500@subnode.lfcs.ed.ac.uk> Newsgroups: fa.think-c Lines: 75 Date: 5 Dec 90 18:10:20 GMT Thanks to those who gave me feedback about the CPreferences code. It now seems to work OK, so I'll send it to Mark Nagel who can secrete it somewhere sensible. In fact, I've thrown in some other bits and pieces as well. Here's the README file from the bundle: [Cue: pause to change it from a TeachText "ttro" file back into "TEXT"...] --cut-- Here are some odds and ends from my MIDI Generic Editor/Librarian. The most useful thing (I guess) is CPreferences. It provides a standard way to create a preferences file and read and write distinct preference information (in the form of resources, although the application doesn't see the resource information directly). The file is put into a folder called Preferences in the System Folder (Preferences is created if necessary); the file name is the name of the application, so watch out for the .<pi> for THINK C projects (I may strip this off automatically sometime soon). CPreferences seems pretty stable, although the HFS stuff was lifted blind from CInstance by Sven Axelsson; but I found that rather confusing and way too clever for what I wanted. CInstance was also buggy - thanks to Martin Minow for setting me right. Other goodies - well, a copy of CUtility (which CPreferences needs). Nothing really world-shattering here, just some random odds and ends, although it does have an interface to the Notification Manager (needed for the MIDI stuff). Oh, I'll also include CDiagnostic. This implements a diagnostic text window as a CDirector. However, it's not quite that straightforward, since you might want to print diagnostics in the middle of doing something graphical, and the act of printing a diagnostic would screw the graphical environment. So, CDiagnostic works in conjunction with CDiagChore, which is an idle chore which looks periodically at the diagnostic object to see if any text needs printing while awaiting events. CDiagnostic also throws away text as it fills up - not very elegantly (the last line of text moves around), but it does the job. Note also that, because CDiagnostic is a CDirector, quitting the application might nuke CDiagnostic first (if it's the top window), and screw up diagnostic messages from other windows ("I'm deactivating now", "exiting application" and so on). Complain to Greg Dow about this, not me... :-) Anyway, that's why CDiagnostic is accessed via global procedures (g_Diagnostic()) instead of method calls to a global object. Oh yes, g_Diagnostic() also takes varargs, something you don't get in methods. While I'm here, I suppose I could throw in the MIDI Manager interface as well. Channelised echo routines, plus asynchronous System Exclusive message capture and a degree of asynchronous transmission of messages (subject to size of output buffer) - the latter is a new feature over my previous posting of the MIDI Manager code (last year sometime). There are extensive comments in CMIDI.c, since there are a lot of gotchas involved in interrupt-level driving (as you might expect). One drawback of CMIDI - transmission of large amounts of data (i.e. larger than the output buffer) causes a spin-wait. This will annoy intensely any other MIDI Manager applications (such as sequencers) which wish to look at the data and need to allocate memory at event-poll time to do so. I shall have to think of a way of decoupling the transmission routines to get around this sometime (or, Apple will have to give us pre-emptive multitasking...). Don't be put off by the "NEW", "OVERRIDE" and "PRIVATE" keywords. They're just noise. They *should* be part of any self-respecting object language, though. "LOCAL" is just aliased to "static". I haven't bothered including the various constants or the resource files. It's obvious stuff, folks. Besides, you don't want *everything* on a plate, do you? That would make it too easy. ("Playing strip poker with an exhibitionist somehow removes the challenge..." - Metropolitan.) Feedback, bug reports, etc. welcome. Merry Christmas. Nick Rothwell LFCS, Edinburgh University. nick@lfcs.ed.ac.uk. 5th December 1990. Path: ucivax!gateway From: nagel@ICS.UCI.EDU (Mark Nagel) Subject: ARCHIVE: HyperCard 2.0 XCMD interface Message-ID: <17171.660419730@ics.uci.edu> Newsgroups: fa.think-c Reply-To: nagel@ICS.UCI.EDU Organization: University of California, Irvine - Dept of ICS Lines: 18 Date: 5 Dec 90 18:12:23 GMT Phone: (714) 856-5039 From: Nigel Perry <np@doc.imperial.ac.uk> Date: Mon, 29 Oct 90 09:28:04 GMT Subject: (for HC 2.0) HyperXCmd.h etc. for Think C Hi, Below is a binhex'ed compactor archive containing a version of HyperXCmd.h for HyperCard 2.0 which uses #define macros and runs under Think C. If you prefer to have interface functions rather than macros a small time in Emacs should solve your needs - if you store your paramPtr in a register variable I think you'll find the macros produce more compact code - but you do lose argument type checking... Also included is a small Think C project used to test out the HyperXCmd.h and the original Apple versions (for MPW) and documentation. Enjoy. Nigel [saved as: /mac/think-c/code/hyper-xcmd.hqx; 71K] Path: ucivax!gateway From: dmitri@bolvan.ph.utexas.edu (Dmitri Linde) Subject: Bug in TCL Message-ID: <9012061511.AA16707@bolvan.ph.utexas.edu.ph.utexas.edu.> Posted-Date: Thu, 6 Dec 90 09:11:35 CST Newsgroups: fa.think-c Lines: 22 Date: 6 Dec 90 15:20:28 GMT Hi! Greg Dow said there is a bug in CResFile. If you ever use it then replace open method of CResFile, which is: if (refNum == -1) { refNum = 0; } return( ResError() ); should be: if (refNum == -1) { refNum = 0; return( ResError() ); } else { return(noErr); } since ResError() should only be called when refNum is -1 - Dmitri dmitri@bolvan.ph.utexas.edu Path: ucivax!gateway From: kg1a+@andrew.cmu.edu (Kevin Michael Goldsmith) Subject: question Message-ID: <QbLcxpS00WAt8AX0ZM@andrew.cmu.edu> Newsgroups: fa.think-c Lines: 13 Date: 6 Dec 90 18:40:24 GMT I have think-c installed on my hard disk, sometimes I want to work on a program on campus, so I have copied a "small think c" disk which has all the necessary files to compile my program and also the source for my program. The problem is this, whenever I try to compile my program, I get the error "Cannot open #include'd file", even though the file is sitting in the same directory, and isn't damaged. I have been doing this for months, and this problem just came up this week. Anybody know what is going on? Kevin Goldsmith kg1a+@andrew.cmu.edu kmg@isl1.ri.cmu.edu Disclaimer: Disclaimer, I don't need no stinkin disclaimer. Path: ucivax!gateway From: nagel@wintermute.ICS.UCI.EDU (Mark Nagel) Subject: ARCHIVE: CPreferences, 4+, InsideTCL, CTrace Message-ID: <18350.660509592@wintermute.ics.uci.edu> Newsgroups: fa.think-c Reply-To: nagel@ICS.UCI.EDU Organization: University of California, Irvine - Dept of ICS Lines: 142 Date: 6 Dec 90 18:55:52 GMT Phone: (714) 856-5039 From: nick@lfcs.edinburgh.ac.UK (Nick Rothwell) Subject: CPreferences and other goodies Date: 5 Dec 90 18:10:20 GMT Thanks to those who gave me feedback about the CPreferences code. It now seems to work OK, so I'll send it to Mark Nagel who can secrete it somewhere sensible. In fact, I've thrown in some other bits and pieces as well. Here's the README file from the bundle: [Cue: pause to change it from a TeachText "ttro" file back into "TEXT"...] --cut-- Here are some odds and ends from my MIDI Generic Editor/Librarian. The most useful thing (I guess) is CPreferences. It provides a standard way to create a preferences file and read and write distinct preference information (in the form of resources, although the application doesn't see the resource information directly). The file is put into a folder called Preferences in the System Folder (Preferences is created if necessary); the file name is the name of the application, so watch out for the .<pi> for THINK C projects (I may strip this off automatically sometime soon). CPreferences seems pretty stable, although the HFS stuff was lifted blind from CInstance by Sven Axelsson; but I found that rather confusing and way too clever for what I wanted. CInstance was also buggy - thanks to Martin Minow for setting me right. Other goodies - well, a copy of CUtility (which CPreferences needs). Nothing really world-shattering here, just some random odds and ends, although it does have an interface to the Notification Manager (needed for the MIDI stuff). Oh, I'll also include CDiagnostic. This implements a diagnostic text window as a CDirector. However, it's not quite that straightforward, since you might want to print diagnostics in the middle of doing something graphical, and the act of printing a diagnostic would screw the graphical environment. So, CDiagnostic works in conjunction with CDiagChore, which is an idle chore which looks periodically at the diagnostic object to see if any text needs printing while awaiting events. CDiagnostic also throws away text as it fills up - not very elegantly (the last line of text moves around), but it does the job. Note also that, because CDiagnostic is a CDirector, quitting the application might nuke CDiagnostic first (if it's the top window), and screw up diagnostic messages from other windows ("I'm deactivating now", "exiting application" and so on). Complain to Greg Dow about this, not me... :-) Anyway, that's why CDiagnostic is accessed via global procedures (g_Diagnostic()) instead of method calls to a global object. Oh yes, g_Diagnostic() also takes varargs, something you don't get in methods. While I'm here, I suppose I could throw in the MIDI Manager interface as well. Channelised echo routines, plus asynchronous System Exclusive message capture and a degree of asynchronous transmission of messages (subject to size of output buffer) - the latter is a new feature over my previous posting of the MIDI Manager code (last year sometime). There are extensive comments in CMIDI.c, since there are a lot of gotchas involved in interrupt-level driving (as you might expect). One drawback of CMIDI - transmission of large amounts of data (i.e. larger than the output buffer) causes a spin-wait. This will annoy intensely any other MIDI Manager applications (such as sequencers) which wish to look at the data and need to allocate memory at event-poll time to do so. I shall have to think of a way of decoupling the transmission routines to get around this sometime (or, Apple will have to give us pre-emptive multitasking...). Don't be put off by the "NEW", "OVERRIDE" and "PRIVATE" keywords. They're just noise. They *should* be part of any self-respecting object language, though. "LOCAL" is just aliased to "static". I haven't bothered including the various constants or the resource files. It's obvious stuff, folks. Besides, you don't want *everything* on a plate, do you? That would make it too easy. ("Playing strip poker with an exhibitionist somehow removes the challenge..." - Metropolitan.) Feedback, bug reports, etc. welcome. Merry Christmas. Nick Rothwell LFCS, Edinburgh University. nick@lfcs.ed.ac.uk. 5th December 1990. [saved as: /mac/think-c/classes/cpreferences.hqx; 41K] /*****************************************************************************/ Date: Thu, 06 Dec 90 11:36:53 From: "David S. McCormick" <dmac@eagle.mit.EDU> Subject: 3 New ThinkC related uploads to /incoming (1) 4Plus.sit.Hqx This is a binhex 4.0 stuffit file containing an INIT which adds some great functions to the Think C editor, the most relevant of which is the TCL Browser facility which allows one to get a popup menu of the ancestors, siblings, or progeny of any class once the file has been compiled and parsed by the 4+ program. It's not a browser which allows you to see all the methods of a class concurrently and visually (the way in which I believe Think Pascal allows), but you do get a popup menu function (similar to ThinkC headers popup menu) for whatever file is currently active. (a note of caution: the function/method parser does not get all the functions in my files or some TCL function on my machines SE/30 and IIfx but I may have some INIT conflict which I have been too lazy to track down; However the CMarker (another program) popup menu function has worked flawlessly and can work at the same time to offset this problem). The other nice features are a Quick-Keys-like user-configurable keystrokes for any 4+ function, macro recording, and many editor extensions like page-up/down, marking places in code, etc. The browser has worked flawlessly, though. It's a $40 shareware product. (2) InsideTCL.sit.Hqx This is a revamp of Bernard Gallet's phenomenally useful Inside Mac DA for the TCL. This is an extensible product, so you can add your own classes as you write them. There are some bugs which the new author documents, but it is really nice for seeing what the classes are and what are there methods and prototypes all from the Apple menu. It's a $20 shareware product for the original Inside Mac DA author (I highly recommend getting the original). (3) ctrace.sit.Hqx This is a direct upload of an article by William D. Cramer in November 1990 Dr. Dobbs Journal (#170) p. 44-55, code listing p.116-120. Cramer implements CTrace, a message logging class for debugging TCL applications, allowing one to throw up an extra window in your application allowing (essentially) the functionality of printf() in command line C for debugging. Built-in logging include provisions for function entry/exit, error messages, warnings, etc., all of which are extensible. I have not tested this yet, but it sounded of general interest. This is free code. Hope you find these useful. Cheers, David S. McCormick MIT-EAPS Geology Bldg. 54-1016 Cambridge, MA 02139 617-253-9852 dmac@eagle.mit.edu dmac@athena.mit.edu [saved as: /mac/think-c/compiler/4plus.hqx; 120K] [saved as: /mac/think-c/compiler/inside-tcl.hqx; 136K] [saved as: /mac/think-c/classes/ctrace.hqx; 28K] Path: ucivax!gateway From: cohen@excalibur.itd.nrl.navy.MIL (Neil Cohen) Subject: Drawing Elipses Message-ID: <9012062146.AA03092@constellation.arpa> Newsgroups: fa.think-c Lines: 16 Date: 6 Dec 90 21:50:35 GMT Hi, I'm asking this question for someone who will be joining this mailing list shortly. Please respond to him directly. Thanks. His question is: In Think C 4.0, how does one draw an elipse that is 'canted' - that is, the major and minor axes are not parallel to the X and Y axes of the window. Send email to: Tom Kratzke (Metron Inc.) at kratzke@gwuvm.bitnet nbc Path: ucivax!gateway From: autodesk!ceili!bobert@uunet.uu.NET (Robert Murphy) Subject: Re: question Message-ID: <9012070039.AA00952@ceili.YP.acad> Newsgroups: fa.think-c Lines: 21 Date: 7 Dec 90 01:00:45 GMT >I have think-c installed on my hard disk, sometimes I want to work on >a program on campus, so I have copied a "small think c" disk which has >all the necessary files to compile my program and also the source for my >program. The problem is this, whenever I try to compile my program, I get >the error "Cannot open #include'd file", even though the file is sitting in the >same directory, and isn't damaged. I have been doing this for months, and >this problem just came up this week. Anybody know what is going on? > >Kevin Goldsmith >kg1a+@andrew.cmu.edu >kmg@isl1.ri.cmu.edu I've had this sort of problem, too. It appears that Think C projects store a path to each of their component files, including headers, so that if you change the path at all (move the file, rename a folder or disk), Think C gets lost. The fix is to do a "Remove Objects" on the project and rebuild it; this will force Think C to go looking for the source files and libraries. - Bob Murphy bobert@autodesk.com bobert@well.sf.ca.us Path: ucivax!gateway From: eaiu088@orion.oac.UCI.EDU (Alisa Tannira) Subject: Calling code resources from Excel. Message-ID: <9012061906.aa23808@orion.oac.uci.edu> Newsgroups: fa.think-c Lines: 12 Date: 7 Dec 90 03:08:47 GMT I have a simple question. Has anybody had any success calling code resources from Excel? Excel supposedly has a routine called REGISTER(lib,proc,arg). The manual is not specific, but I think "lib" is supposed to be a code resource containing the procedure you wish to call. Does ThinkC produce a code resource that can be read by Excel? Or conversely, is Excel smart enough to read ThinkC's code resources. :) Any help is GREATLY appreciated. Alisa Path: ucivax!gateway From: bootsie!olson@ICS.UCI.EDU (Eric Olson) Subject: Re: Cannot open included file Message-ID: <9012070455.AA03415@bootsie.UUCP> In-Reply-To: Root Maytall's message of Dec 6, 11:42pm. X-Mailer: Mail User's Shell (6.4 2/14/89) Newsgroups: fa.think-c Reply-To: olson@endor.harvard.edu Lines: 19 Date: 7 Dec 90 05:59:32 GMT >I've had this sort of problem, too. It appears that Think C projects store >a path to each of their component files, including headers, so that if you >change the path at all (move the file, rename a folder or disk), Think C >gets lost. The fix is to do a "Remove Objects" on the project and rebuild >it; this will force Think C to go looking for the source files and libraries. > You can also Uncheck "Quick Scan" in the "Make..." dialog and then click "Use Disk". This will recache all the stored file locations, and will only recompile the ones that have changed or moved. -Eric -- Eric K. Olson, Editor, Prepare() NOTE: olson@bootsie.uucp will not work! Lexington Software Design Internet: olson@endor.harvard.edu 72A Lowell St., Lexington, MA 02173 Usenet: harvard!endor!olson (617) 863-9624 Bitnet: OLSON@HARVARD Path: ucivax!gateway From: minow@bolt.enet.dec.COM ("Martin Minow, ML3-5/U26 07-Dec-1990 0657") Subject: Multi-window display under TCL -- help! Message-ID: <9012071215.AA26840@decpa.pa.dec.com> Newsgroups: fa.think-c Lines: 32 Date: 7 Dec 90 12:18:37 GMT I am obviously doing something stupid but haven't the foggiest idea what. My application has a main window (document->window) that works ok. If I have several documents, all of the windows seem to work normally. I also have a status window that is independent of the documents. If I click on a button in the window, the document sends a message to the status window which shows and draws itself. This also seems to work ok. However, if I then click on the document's window again, only the controls (buttons and scroll-bars) are redrawn; my text items are not updated. Furthermore; this only happens if I show the status window. If I put a desk accessory window up (without showing the status window), or switch applications, everything is redraw ok. Also, if I have two document windows up, and show the status window, *only* the document that triggered the status window is incorrect: the other window continues to be redrawn correctly. The window has CPane's (mostly text), a ListManager class, and buttons. Clearly, I'm doing something dumb, but don't have any clues. The status window is created by the application, and the document causes it to be displayed by calling an application-specific message. I've also tried handling the status window in a CChore, but that didn't help either. The Debugger shows my CPane's drawing methods are being called. Any ideas? Martin. minow@bolt.enet.dec.com Path: ucivax!gateway From: nick@lfcs.edinburgh.ac.UK (Nick Rothwell) Subject: "Not a text file." Huh... Message-ID: <5461.9012071250@subnode.lfcs.ed.ac.uk> Newsgroups: fa.think-c Lines: 12 Date: 7 Dec 90 14:39:13 GMT A couple of suggestions for THINK C error messages. If you include a precompiled header, but after other #includes, or even after some #defines, how about saying "Precompiled header must be first statement in file," rather than "not a text file." If you try to use a precompiled header in a project where you've forgotten to uncheck the MacHeaders option, how about "Precompiled header cannot be used with MacHeaders enabled," rather than (wait for it:) "not a text file." Nick. Path: ucivax!gateway From: nick@lfcs.edinburgh.ac.UK (Nick Rothwell) Subject: Re: Multi-window display under TCL -- help! Message-ID: <5586.9012071442@subnode.lfcs.ed.ac.uk> Newsgroups: fa.think-c Lines: 23 Date: 7 Dec 90 17:01:36 GMT I had the same problems. Trouble is if you draw into the status window, you need another Prepare() to carry on drawing into the original window. I had all sorts of problems like this (scroll-bars not appearing, and so on), and finally fixed it using the trick of deferring diagnostics - see CDiagnostic for details. The downside is that you only see diagnostics when the program goes though the event loop again (or some time thereafter), and not at the point of call. So, some care is required. Actually (come to think of it), I had this problem with a status window as well. The status window was supposed to be permanently active, and needed to accept clicks to check boxes and radio buttons, without screwing the environment of the document window. My solution was to implement CDeferredRadioGroup (and so on), such that clicks wouldn't call CRadioGroup::SetStation(), but would wait until an urgent chore made them update themselves. This Prepare()/Draw() business is probably the second biggest gotcha in the TCL (well, to be fair, it's an issue with the ToolBox). The biggest gotcha is when windows start getting closed and deactivated... Nick. Path: ucivax!gateway From: eacj@theory.tn.cornell.edu (Julian Vrieslander) Subject: Conflict Between 4Plus and QM Menu Message-ID: <9012072005.AA17128@theory.TN.CORNELL.EDU> Newsgroups: fa.think-c Lines: 9 Date: 7 Dec 90 20:04:50 GMT The 4Plus INIT looks like a very useful hack, but I'm having a problem with it. When I try to launch THINK C I get an immediate crash (Bus Error). After trial and error, I found that there was a conflict between 4Plus and QM Menu, an INIT that is part of CE Software's QuickMail system. QM Menu installs a QuickMail menu that appears in all applications. I tried renaming QM Menu so it would load before 4Plus, but that does not help. If I remove the QM Menu INIT, the crashes do not occur. Looks like both INITs are fighting over the Menu Bar. Any ideas? Path: ucivax!gateway From: KRATZKE%GWUVM.BITNET@cunyvm.cuny.edu Subject: list Message-ID: <9012071808.aa08883@ICS.UCI.EDU> Newsgroups: fa.think-c Lines: 2 Date: 8 Dec 90 02:09:31 GMT My friend gave me this e-mail address and told me that I could get e-mail from you by sending a message to you. If so, thanks!! Path: ucivax!gateway From: fri0@midway.uchicago.edu ("Christian E. Fritze") Subject: Re: Conflict Between 4Plus and QM Menu Message-ID: <CMM.0.88.660632494.fri0@quads.uchicago.edu> Newsgroups: fa.think-c Lines: 13 Date: 8 Dec 90 05:06:33 GMT (I think I sent Julian two replies before realizing things weren't bouncing back to the Think-C list. Anyway...) I agree 4Plus looks very useful. I get an address error on my SE on startup even if 4Plus is the only Ioad (other than MacsBugs 6.1). The crash drops me into MacsBugs even before the 4Plus Icon is displayed at the bottom of my screen. I'm running Sys 6.0.7 and MultiFinder 6.1b9. Anyone else having similar troubles? I've even tried retrieving 4Plus twice from archive... Christian E. Fritze | AOL:geneman University of Chicago | fri0@midway.uchicago.edu Molecular Genetics and Cell Biology | "No one ever died of laughing" -M.B. -- Path: ucivax!gateway From: sarachan@wsqtbf.crd.ge.COM ("B. D. Sarachan") Subject: (none) Message-ID: <9012101443.AA01656@wsqtbf.steinmetz> Posted-Date: Mon, 10 Dec 90 09:43:38 EST Newsgroups: fa.think-c Lines: 18 Date: 10 Dec 90 14:48:43 GMT <fri0@midway.uchicago.edu> writes: >I agree 4Plus looks very useful. I get an address error on my SE on startup >even if 4Plus is the only Ioad (other than MacsBugs 6.1). The crash drops me >into MacsBugs even before the 4Plus Icon is displayed at the bottom of my >screen. I'm running Sys 6.0.7 and MultiFinder 6.1b9. Anyone else having >similar troubles? I've even tried retrieving 4Plus twice from archive... Whenever I try to bring up 4Plus on an SE, the system freezes on startup. This happens even when I boot from a floppy containing only a System Folder with System, Finder, and 4Plus. Brion Sarachan General Electric CR&D sarachanbd@crd.ge.com Path: ucivax!gateway From: minow@bolt.enet.dec.COM ("Martin Minow, ML3-5/U26 09-Dec-1990 2301") Subject: Multi-window display problem solved Message-ID: <9012101805.AA07695@decpa.pa.dec.com> Newsgroups: fa.think-c Lines: 24 Date: 10 Dec 90 18:10:11 GMT Last week, in desperation, I wrote: >I am obviously doing something stupid but haven't the foggiest idea what. I was trying to display a status window together with my document window, and discovered that if my status window was shown, the document window wouldn't update. As it turned out, a piece of half-written code (I can type much faster than I can think) set RGBForeColor and RGBBackColor on the status window -- at least that was the intention. Unfortunately, I forgot to invoke Prepare() so the port was still pointing to the document window. To make matters worse, I hadn't written the color initialization code, so both colors (for the document window ) were set to white. It only took a week to find this silly bug. I can hardly wait for the real bugs to creep out of my program. Let this be a warning to you. Thanks for the good suggestions from several folk. Martin. minow@bolt.enet.dec.com Path: ucivax!gateway From: paco@neuromancer.sps.mot.COM (Paco Xander Nathan) Subject: 4Plus Message-ID: <9012101841.AA12835@neuromancer> Newsgroups: fa.think-c Lines: 25 Date: 10 Dec 90 19:37:37 GMT This 4Plus business looks great. I've got copies of it from three different sources now: here (think-c%ics.uci), CompuServe & internal Motorola, but it won't do anything... I can opt-cmd-click the INIT to launch the configuration mode, no problem. And of course the INIT icon shows up okay on the screen during startup. But there's no effect to ThC... No menus, no special key mapped actions, nada. This is running on a IIx with 6.0.5 and ThC 4.0.2 Are there any other INITs known to make 4Plus null and devoid? Thanks - paco ---- Internet: paco@neuromancer.sps.mot.com Voice Mail: 1.512.891.2973 Neural Networks & Virtual Reality Development Center for Emerging Computer Technologies Motorola, Inc. Austin, Texas, USA Path: ucivax!gateway From: FLEMINGM@qucdn.queensu.ca ("R. Mark Fleming") Subject: re: 4plus Message-ID: <9012101354.aa28959@ICS.UCI.EDU> Newsgroups: fa.think-c Lines: 25 Date: 10 Dec 90 22:01:59 GMT I have been able to install and uses 4Plus with Think C version's V4.0.2 and V4.0.4 with no problem assuming I try to operate on oop's code. I also have CMaker Window Def, installed in Think C too... If you ask it to update the cross ref, on a non 'oop TCL' project it crashes. I have not tested it extensively, but I do have a few other init loading with it with no apparent conflicts. System Configuration: Macintosh Portable HD40, 2Mb RAM, OS V6.0.5, MultiFinder. Init's I can think of: GateKeeper, GateKeeper Aid v1.1, Public Folder, Pyro v3.xx, MacBug V6.2B3, Responder v1.1.1, Flash iT I feel that it has LOT's of potential, but needs some more error trapping. I look forward in see the next version of 4Plus, and at that time I will do some extensive testing and then start using it Cheers, Mark <<<<<<<<<<<<<<<=========================================>>>>>>>>>>>>>>>>> Mark Fleming, Macintosh Support, Office Systems, Computing & Communicat. Services, Queen's University at Kingston, Phone: (613) 545-2039 Bitnet: FLEMINGM@QUCDN or Mark.Fleming@QueensU.CA AppleLink: CDA0448 <<<<<<<<<<<<<<<=========================================>>>>>>>>>>>>>>>>> Path: ucivax!gateway From: ech@pegasus.att.COM Subject: Re: Conflict Between 4Plus and QM Menu Original-From: ech (Ned Horvath) Lines: 17 Date: 10 Dec 90 23:00:27 GMT Message-ID: <9012101459.aa06425@ICS.UCI.EDU> In-Reply-To: your message <internet3420640530> of Sat Dec 8 05:06:33 GMT 1990 >To: att!ics.uci.edu!think u to wait, go have a coke on the rocks. Check out what was the last trap that was run before the system hung. If it kills MacsBug, you have to be there to look at the trap names while they are displayed. When you know what the last trap was, you are pretty close to what is happening in the crash (it might not be the last one). If this still doesn't work, you might want to trap calls to SetTrapAddress (and the N-counterpart of that). Of course, you only have to do this if you want to know why the init is crashing. When you know why it is crashing, it's usually quite easy to make a quick binary patch to fix it. The INIT is 4Plus is only 886 bytes long, so it's quite probable that the problem with Macintosh SEs is quite easy to fix. Use MPW DumpObj to disassemble, if all else fails... (We all have a "Highly confidential, please don't copy" version of MPW 1.0B2 available from the early days, don't we?) P.S. Would you be interested in a Digital Simulator written in Think C? It doesn't use any object stuff, since I wrote it for X Windows. I used Think C and the Mac, but the user interface is quite minimal. I'll have the program running under unix in a week or so, but at the moment it runs quite well on the Macintosh. (It doesn't really like the Mac File System, but since you are programmers, one of you could easily fix it. Currently it only knows about a single folder for subcomponents.) It's slightly more advanced than DigSim 2.0. ____________________________________________________________________________ / Juri Munkki / Helsinki University of Technology / Wind / Project / / jmunkki@hut.fi / Computing Center Macintosh Support / Surf / STORM / ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Path: ucivax!gateway From: cohen@excalibur.itd.nrl.navy.MIL (Neil Cohen) Subject: Beginner's questions - thanks! Message-ID: <9012111346.AA06432@constellation.arpa> Newsgroups: fa.think-c Lines: 35 Date: 11 Dec 90 13:49:33 GMT Hi, I'd like to thank all those who responded to my questions about getting started with Graphics and the TCL. I was referred to the following books for sample code and information on drawing objects in a window: Programming Primer (I & II) - Author David Owen?? QuickDraw (in Inside Mac) TCL Manual pages The consensus was that I just need to get over the (steep) TCL learning curve, but the results will be worth the effort. Sample code from apple.com Most of what I found there seemed to be for MPW, but there is quite a bit of code lying around, and I may have looked in the wrong places. I have managed to get a simple program running using the TCL. I expect to be back soon with more questions... I'll send a bit more detail to those who asked me for copies of my replies. Thanks again to all who responded, nbc NAME: Neil B. Cohen (Tracor Applied Sciences) PHONE: 703-444-4610 DOMAIN: nbc@excalibur.itd.nrl.navy.mil ************************************************************* * Murphy's Philosophy: Smile - tomorrow will be worse... * * * * O'Tooles Commentary: Murphy was an optimist! * ************************************************************* Path: ucivax!gateway From: gd1a+@andrew.cmu.edu ("Glen M. Devries") Subject: THINK C 4.0.4 Message-ID: <8bNEmf_00WBMQ1yUgD@andrew.cmu.edu> Newsgroups: fa.think-c Lines: 5 Date: 11 Dec 90 16:48:29 GMT Where (besides the System 7 beta CD-ROM) can one obtain the 4.0.4 updater? Could someone please point me in the right direction or post it on comp.sys.binaries.mac. -Glen Path: ucivax!gateway From: fri0@midway.uchicago.edu ("Christian E. Fritze") Subject: Re: 4Plus & 68000 Macs = CRASH? Message-ID: <CMM.0.88.660933910.fri0@quads.uchicago.edu> Newsgroups: fa.think-c Lines: 10 Date: 11 Dec 90 16:48:42 GMT Before I saw the posting about using MacsBug to track down the specific bug in 4Plus, I sent a message to the author on CompuServe. (Well, I hope it's the author, sent it to the uploader of 4Plus on CompuServe). If I get anything help back, I'll sent it along. Meanwhile, should anyone else figure out a patch, let us know. Christian E. Fritze | AOL:geneman University of Chicago | fri0@midway.uchicago.edu Molecular Genetics and Cell Biology | "No one ever died of laughing" -M.B. -- Path: ucivax!gateway From: ech@pegasus.att.COM Subject: Re: THINK C 4.0.4 Original-From: ech (Ned Horvath) Lines: 22 Date: 11 Dec 90 20:05:15 GMT Message-ID: <9012111203.aa06924@ICS.UCI.EDU> In-Reply-To: your message <internet3451919020> of Tue Dec 11 16:48:29 GMT 1990 >To: att!ics.uci.edu!think-c Content-Type: Text Content-Length: 761 Newsgroups: fa.think-c Message-Version: 2 Email-Version: 2 UA-Message-ID: <MAC-1.3.4A1-618034-ech-697> UA-Content-ID: <PMX-PC-2.01A-000144-pegasus1-ech-3066> ------------ Original Message ------------- Where (besides the System 7 beta CD-ROM) can one obtain the 4.0.4 updater? Could someone please point me in the right direction or post it on comp.sys.binaries.mac. -Glen --------- End of Original Message --------- Weasel answer follows: The distribution of 7.0b1 is limited by Apple to certified developers; the ThC updates are only needed by people with 7.0b1. The inexorable logic which follows is that if you don't have it, you don't need it. This is Symantec's line as well as Apple's. You could hassle Symantec tech support, and they may or may not respond. But everybody who has the 7.0b1 is under nondisclosure... Sorry. I don't like it either. =Ned Horvath= Path: ucivax!gateway From: ar4@sage.cc.purdue.edu (Piper Keairnes) Subject: ThinkC 4.0.4 problem Message-ID: <9012112151.AA12305@sage.cc.purdue.edu> Newsgroups: fa.think-c Lines: 19 Date: 11 Dec 90 21:52:15 GMT I recently updated ThinkC to 4.0.4 on my hard disk. I ran into a recurring problem: Compiling gives several "does not match prototype errors" I finally discovered the root of my problem (as I perceive it). I replaced the updated MacHeaders file with an original from the 4.0 version. The next compilation worked fine. I tried precompiling the 4.0.4 'Mac #includes.c' file and using that. Compile errors reappeared. Replaced with the old MacHeaders and compile worked again. Has anyone else experienced these difficulties? If so... can you offer a suggestion as to how to solve them? Or just tell me I'm stupid... well, on second thought don't. ;-) _____ Piper Keairnes ar4@sage.cc.purdue.edu (Purdue University) Path: ucivax!gateway From: paco@neuromancer.sps.mot.COM (Paco Xander Nathan) Subject: 4Plus, ThC4.0.4, CCursor Message-ID: <9012112123.AA13253@neuromancer> Newsgroups: fa.think-c Lines: 14 Date: 11 Dec 90 22:22:53 GMT Thanks for the direct replies about 4Plus. Finally got it working. It seems theres a few versions running about, so check closely: I've gotten v1.3 to work on my sys: Mac IIx, System 6.0.5, ThC 4.0.2 w/ Gatekeeper & Gatekeeper Init v 1.1.1 no QM, etc. Gatekeeper doesn't mention any vetoes though, but I still don't get the menus from 4+. Just press the "help" key to check if 4+ installs. With mention of ThC 4.0.4, was there any distribution of a 4.0.3 version? I admit to blowing off the 7.0 beta CD... I absolutely love using Resourceror!!!! Well worth $256 of hard earned cash to save so much time and frustration with an excellent Mac product. Has anybody got a good means of capturing color cursors for it? Path: ucivax!gateway From: FLEMINGM@qucdn.queensu.ca ("R. Mark Fleming") Subject: RE:4Plus & 68000 CPU Message-ID: <9012111420.aa26042@ICS.UCI.EDU> Newsgroups: fa.think-c Lines: 9 Date: 11 Dec 90 22:43:54 GMT I have used 4Plus on the Macintosh Portable which is a 68000 based system. running v6.0.5 & Think C v4.0.2. Mark <<<<<<<<<<<<<<<=========================================>>>>>>>>>>>>>>>>> Mark Fleming, Macintosh Support, Office Systems, Computing & Communicat. Services, Queen's University at Kingston, Phone: (613) 545-2039 Bitnet: FLEMINGM@QUCDN or Mark.Fleming@QueensU.CA AppleLink: CDA0448 <<<<<<<<<<<<<<<=========================================>>>>>>>>>>>>>>>>> Path: ucivax!gateway From: siegel@das.harvard.edu (Rich Siegel) Subject: Re: THINK C 4.0.4 Message-ID: <9012120243.AA19746@das.harvard.edu> Newsgroups: fa.think-c Lines: 16 Date: 12 Dec 90 02:47:49 GMT In fact, you've got it precisely right. As of now, all version of THINK C *after* 4.0.2 are intended only to address compatibility issues with pre- release versions of System 7.0, and offer no benefits for people not running that version of the system. (And the logic still applies: System 7 prereleases are provided only to Apple Partners and Apple Associates under the injunction that they are not to distribute it, so anyone who doesn't have a System 7.0 CD theoretically got it through someone's violation of this agreement. Likewise, all Partners and associates have Applelink accounts, so Applelink is the preferred medium to distribute System 7.0-related materials.) If you hassle the Tech Support dudes, you'll be wasting your time and theirs. R. Path: ucivax!gateway From: siegel@das.harvard.edu (Rich Siegel) Subject: Re: ThinkC 4.0.4 problem Message-ID: <9012120244.AA19749@das.harvard.edu> Newsgroups: fa.think-c Lines: 6 Date: 12 Dec 90 02:48:05 GMT It's my understanding that 4.0.4 was supplied with updated Toolbox interfaces and a new MacHeaders file. You should check your existing code agains the new interfaces. R. Path: ucivax!gateway From: tarr-michael@cs.yale.edu (michael tarr) Subject: 4Plus and Precompiled Headers Full-Name: michael tarr Message-ID: <9012130219.AA08238@thailand.CS.YALE.EDU> Newsgroups: fa.think-c Lines: 11 Date: 13 Dec 90 02:19:59 GMT 4Plus works fine if MacroMaker is not loaded. But I use my own precompiled heeader (rather than <MacHeaders> and 4Plus only seems to show .h files from <MacHeaders> or explicitly included files. So I am left with no way of quickly getting a list of .h files in a menu... Any suggestions? Mike Tarr Department of Psychology tarr@cs.yale.edu Yale University Path: ucivax!gateway From: paco@neuromancer.sps.mot.COM (Paco Xander Nathan) Subject: Resorceror Message-ID: <9012130524.AA13844@neuromancer> Newsgroups: fa.think-c Lines: 78 Date: 13 Dec 90 14:07:28 GMT I was recently asked to post a more detailed review following my recent wildly blatant endorsement for a new product known as Resorceror. First, I sspeelled it wrong. It's spelled as "Resorceror" and the authors are quite emphatic on this point. Secondly, here is the contact info: Mathemaesthetics, Inc. P.O. Box 67-156 Chestnut Hill, MA 02167 USA Phone: 1.617.738.8803 AppleLink: RESORCERER Internet: RESORCERER@AppleLink.apple.com Okay. It's still in beta, but quite stable. We use it here for development; using almost anything else would be a risk. Here's my random thoughts on the subject: Buy it for $256, then throw away ResEdit vX.X quickly. The product shows all resources for any file you choose. No more "Shall I create a resource fork?" nonsense. It also shows and edit all file info. It displays rsrc in sorted order; you specify which order, ie. number, title, etc. The rsrc are arrange in a tree structure, very easy to see. I tend to only need 4 windows open at a time. Lists are displayed visually whenever appropriate. The editors are all quite smart, ie. BNDL editor manages the signature rsrc + FREF + ICN# + icn8 + icn4 in one display. Dialogs. It manages DLOG, DITL & DLGX all in one effort. It interactively simulates a dialog, showing ModalDialog return values. It creates TCL, MPW, etc. code to run the dialog, either modeless or modal. It automatically creates popup menu items & user defined list items. Throw out ResMenu (it was nice though) because the menu editor is brilliant. All icon, icn8, cicn, CURS, etc. editors will capture from the screen, even in color. Except for color cursors (I'll have to link 'em). Windows. Takes care of WDEF's, manages color tables, generates code, etc. ThC/TCL templates. Knows them well. Very nice editors for panes, borders, etc. Any templates. Edits them well, using an extended version of ResEdit TMPL language. I am only a user of the product and probably have gotten a few issues jumbled, but you'll like this product if you spend much time using ThC. paco. ---- Internet: paco@neuromancer.sps.mot.com Voice Mail: 1.512.891.2973 Neural Networks & Virtual Reality Development Center for Emerging Computer Technologies Motorola, Inc. Austin, Texas, USA (c)1990, PXN. B*B*! Subject to Public Law 99-508 Path: ucivax!gateway From: bin@primate.wisc.edu (Brain in Neutral) Subject: TransSkel users please read Message-ID: <9012132255.AA01211@uakari.primate.wisc.edu> Newsgroups: fa.think-c Lines: 18 Date: 13 Dec 90 22:59:33 GMT It's time (now that I'm Macintosh programming again) to revise TransSkel. The current version is 2.01, available via anonymous ftp at indri.primate.wisc.edu [128.104.230.11] in THINK C and Pascal versions. The major glaring omissions are support for MultiFinder and hierarchical menus. There are also several minor omissions I can think of (e.g., providing a hook to be called before menu selections are executed). But of course I don't think of everything. My request: if you are a TransSkel user and have suggestions or comment about what should be added or changed, please let me know. If you don't know what TransSkel is, don't ask. You don't want to know. -- Paul DuBois dubois@primate.wisc.edu Path: ucivax!gateway From: dmac@eagle.mit.edu ("David S. McCormick") Subject: At what level should methods be over-ridden in TCL? Message-ID: <9012151833.AA28667@EAGLE.MIT.EDU> Newsgroups: fa.think-c Lines: 16 Date: 15 Dec 90 18:36:35 GMT I am taking my first furtive steps into the TCL and I am having difficulty trying to figure out at what part the hierarchy methods should be over-ridden. For example, if a menu item is invoked to show a window that may be hidden, should I override the CDocument::DoCommand() or the CApplication::DoCommand()? (AppMaker is helpful in this regard). In other words, are there guidelines to figure out what each item should handle? If this is a FAQ just refer me somewhere else. Any help greatly appreciated. Cheers and Happy Holidays, David S. McCormick MIT-EAPS Geology dmac@eagle.mit.edu dmac@athena.mit.edu Path: ucivax!gateway From: mwg@med.pitt.edu ("Michael W. Groff") Subject: TCL class/instance browser Message-ID: <9012170324.AA13997@earth.med.pitt.edu> Newsgroups: fa.think-c Lines: 10 Date: 17 Dec 90 03:28:16 GMT Does anyone have a recommendation for a good class/instance browser for Think C and the TCL? I would like to be able to stop my program and find out the name of the last n methods that were called. Also it would be helpful to know which instances made the calls- as opposed to which objects. I have trouble with Think's debugger because it seems that the scope of the instance is always out of bounds by the time the method kicks in. Any tips or pointers? Thanks in advance. mike groff mwg@med.pitt.edu Path: ucivax!gateway From: nagel@ICS.UCI.EDU (Mark Nagel) Subject: ARCHIVE: list archives available Message-ID: <8922.661468419@ics.uci.edu> Newsgroups: fa.think-c Reply-To: nagel@ICS.UCI.EDU Organization: University of California, Irvine - Dept of ICS Lines: 24 Date: 17 Dec 90 21:18:31 GMT Phone: (714) 856-5039 I've placed the mailing list archives in the think-c archive area under the directory mac/think-c/archives. Each month is stored compressed as mmmyy.Z (e.g. sep90.Z) in MMDF mailbox format. That format is easier for me to generate, but if it is a large problem, I can change them to standard mailbox format. For those who don't know, MMDF format is essentially: ^A^A^A^A message 1 ^A^A^A^A ^A^A^A^A message 2 ^A^A^A^A ^A^A^A^A ... ^A^A^A^A ^A^A^A^A message N ^A^A^A^A The ^A characters above are actual control characters, not the sequence "^A". Mark Path: ucivax!gateway From: hirose@sws.cpd.mei.co.jp (Masato Hirose) Subject: Pro.Prog.Extender? Message-ID: <9012180910.AA00443@coron.sws.cpd.mei.co.jp> Newsgroups: fa.think-c Lines: 5 Date: 18 Dec 90 09:12:05 GMT Does anyone know about Professional Programmer's Extender? I hear it includes think-C source code and it is very useful. Really? How much? Where can I buy? --Masato Hirose (hirose@sws.cpd.mei.co.jp) Path: ucivax!gateway From: mrose@cheetah.ca.psi.COM (Marshall Rose) Subject: MacTCP and THINK-C Message-ID: <9075.661544625@cheetah.ca.psi.com> Newsgroups: fa.think-c Lines: 15 Date: 18 Dec 90 18:28:06 GMT [ sorry to bother the list with a question I'm sure has been asked a dozen times, but I can't find the answer in the archives here... ] OK, so I am having a great time with LSC4.0 and now what to write a program using MacTCP. How does one go about doing this? The manual doesn't seem to talk about how to "import" things like the MacTCP library. My guess is that I need some kind of include file that defines both some C structs and some pascal routine prototypes. Can anyone help? Thanks, /mtr Path: ucivax!gateway From: zaccone@bucknell.edu (zaccone - 1393) Subject: Problem with code resource Message-ID: <9012182030.AA15691@sol.bucknell.edu> Newsgroups: fa.think-c Lines: 17 Date: 18 Dec 90 20:34:16 GMT I am having a problem creating a code resource with THINK C. This is my first attempt at this, so I think that my problem is probably elementary. First, I wrote my code, set the project type to application, included the ANSI library and tested it. Then, I removed the objects, changed the type to code resource, removed the ANSI library and included ANSI-A4, set the creator, type, id, and file type. Finally, I selected "Build Code Resource...". THINK C complains because I "Can't define classes in this project". I'm not using classes! The manual hints that setting the multi-segment flag might help (even though it is one segment). If I set this flag, the OK button doesn't highlight. Can anyone tell me what I'm doing wrong? I'm out of ideas! Rick Zaccone zaccone@bucknell.edu Path: ucivax!gateway From: GFT_ROBERT@gsbvxb.uchicago.edu (opcode ranger) Subject: How does one get Profile to print to a file? Message-ID: <901218180354.2ce016d1@GSBACD.UCHICAGO.EDU> Newsgroups: fa.think-c Lines: 12 Date: 19 Dec 90 00:08:17 GMT X-Vmsmail-To: @THINKC I've written a fairly substantial Macintosh app. I'd like to get a performance profile on it, and I was trying to use THINK C's optional Profiler() library. When I use it a 'console' window pops up in the back of my program while it's running and spits out the function calls (I set _trace to true), and when it's done it spits out the stats to the 'console' window which then promptly vanishes. All in all fairly useless as it stands now: I don't know how to save the stats. Has anyone figured out a way to do this to a Mac (i.e. non-console-oriented) app? All I'd like it to do is print it to a file "Profile.out" or something. Robert Path: ucivax!gateway From: GFT_ROBERT@gsbvxb.uchicago.edu (opcode ranger) Subject: Emily Litella speaks (re: How does one get Profile to print to a file?) Message-ID: <901218193046.2ce018f7@GSBACD.UCHICAGO.EDU> Newsgroups: fa.think-c Lines: 7 Date: 19 Dec 90 01:34:51 GMT X-Vmsmail-To: @THINKC Never mind! :-> With the help of a friend more standard-C-knowledgeable than I I figured out how to get profiler to write to a file. I just went into the profile source code and opened a file and changed all printf's to fprintf's. Simple. Robert Path: ucivax!gateway From: puklich@plains.nodak.edu (Blayne Puklich) Subject: Re: How does one get Profile to print to a file? Message-ID: <9012190214.AA14112@plains.NoDak.edu> In-Reply-To: <901218180354.2ce016d1@GSBACD.UCHICAGO.EDU>; from "opcode ranger" at Dec 19, 90 12:08 am X-Mailer: ELM [version 2.3 PL0] Newsgroups: fa.think-c Lines: 35 Date: 19 Dec 90 02:18:37 GMT > I've written a fairly substantial Macintosh app. I'd like to get a performance > profile on it, and I was trying to use THINK C's optional Profiler() library. > When I use it a 'console' window pops up in the back of my program while it's > running and spits out the function calls (I set _trace to true), and when it's > done it spits out the stats to the 'console' window which then promptly > vanishes. All in all fairly useless as it stands now: I don't know how to save > the stats. > > Has anyone figured out a way to do this to a Mac (i.e. non-console-oriented) > app? All I'd like it to do is print it to a file "Profile.out" or something. Actually it's pretty easy. Using one of the old UNIX tricks, do the following two calls at the very start of your application: fclose(stdout); freopen("Profile.out", "w", stdout); That's it. Has anyone noticed that fopen() won't return file descriptors from 0 through 3? By my count, only 0 through 2 are used, and 3 should be the first one available to the first fopen(). Look at __getfile() (in iomisc.c) to see what I mean. What one should be able to do is fclose(stdout); fopen("Profile.out", "w"); and it should reassign stdout to Profile.out. The lowest numbered file descriptor should get reassigned -- after all, that's how parent and child processes frequently talk to each other under UNIX (not that I want my Mac to be UNIX, unless it's A/UX). -- ||+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++|| || Blayne Puklich puklich@Plains.NoDak.EDU "I think I'm going bald..." || || NDSU, Fargo, ND (701) 237-4408 -- Rush, Caress of Steel, || || 1975 || ||-------------------------------------------------------------------------|| Path: ucivax!gateway From: phils@chaos.cs.brandeis.edu (Phil Shapiro) Subject: How does one get Profile to print to a file? Message-ID: <9012191835.AA02103@chaos.cs.brandeis.edu> In-Reply-To: Blayne Puklich's message of 19 Dec 90 02:18:37 GMT <9012190214.AA14112@plains.NoDak.edu> Newsgroups: fa.think-c Lines: 16 Date: 19 Dec 90 18:40:58 GMT > Has anyone figured out a way to do this to a Mac (i.e. non-console-oriented) > app? All I'd like it to do is print it to a file "Profile.out" or something. Actually it's pretty easy. Using one of the old UNIX tricks, do the following two calls at the very start of your application: fclose(stdout); freopen("Profile.out", "w", stdout); You don't have to bother closing the stream you're re-opening, since freopen() does it for you. If you don't mind using the console library, you could always do: cecho2file("Profile.out", 0, stdout); -phil Path: ucivax!gateway From: phils@chaos.cs.brandeis.edu (Phil Shapiro) Subject: Problem with code resource Message-ID: <9012191843.AA02167@chaos.cs.brandeis.edu> In-Reply-To: zaccone - 1393's message of 18 Dec 90 20:34:16 GMT <9012182030.AA15691@sol.bucknell.edu> Newsgroups: fa.think-c Lines: 28 Date: 19 Dec 90 18:46:28 GMT From: zaccone - 1393 <zaccone@bucknell.edu> Newsgroups: fa.think-c Date: 18 Dec 90 20:34:16 GMT I am having a problem creating a code resource with THINK C. This is my first attempt at this, so I think that my problem is probably elementary. [ ... stuff deleted ... ] Finally, I selected "Build Code Resource...". THINK C complains because I "Can't define classes in this project". I'm not using classes! The manual hints that setting the multi-segment flag might help (even though it is one segment). If I set this flag, the OK button doesn't highlight. Can anyone tell me what I'm doing wrong? I'm out of ideas! I don't know why you're getting that error message. I would try creating a new project and starting with that. The reason the OK button is dimmed when you check the Multi Segment flag is that Think C uses an owned resource scheme to keep track of multi segment code resources. This means that you're limited to a 6 bit integer to specify the "owning" resource ID, so your code resource's ID must be in the range 0..63. For more info about owned resources, look at IM I-109. -phil Path: ucivax!gateway From: dmitri@bolvan.ph.utexas.edu (Dmitri Linde) Subject: RE: code resources Message-ID: <9012192345.AA21931@bolvan.ph.utexas.edu.ph.utexas.edu.> Posted-Date: Wed, 19 Dec 90 17:45:54 CST Newsgroups: fa.think-c Lines: 8 Date: 20 Dec 90 00:38:32 GMT I had a similar problem. I added TCL headers to Macheaders. I did this to speed up compilation of TCL project by about 2 times. But when I built code resources, it complained "Caan't build classes..." I removed TCL headers and it worked fine... - Dmitri dmitri@bolvan.ph.utexas.edu Path: ucivax!gateway From: mkelly@cs.uoregon.edu Subject: CDesktop Message-ID: <9012200702.AA24751@pragmatix.cs.uoregon.edu> Newsgroups: fa.think-c Lines: 49 Date: 20 Dec 90 07:02:58 GMT I've found a possible bug in CDesktop. The procedure RemoveWind is: void CDesktop::RemoveWind( CWindow *theWindow) /* Window object to remove */ { if (theWindow->IsVisible()) { HideWind(theWindow); /* Hide window before removing it */ } itsWindows->Remove(theWindow); } If this procedure removes the last window from the desktop, the CDesktop instance variable 'topWindow' should go to NIL, shouldn't it? It doesn't; it remains whatever it was before the call. I encountered this problem by making this call from a document: gApplication->itsDirectors->DisposeItems(); gDesktop->topWindow remained the same after the call. This eventually caused a crash, since the window had been disposed of. I changed RemoveWind to void CDesktop::RemoveWind( CWindow *theWindow) /* Window object to remove */ { if (theWindow->IsVisible()) { HideWind(theWindow); /* Hide window before removing it */ } itsWindows->Remove(theWindow); if ( itsWindows->numItems == 0 ) topWindow = NULL; } This has apparently fixed the problem, since the program no longer crashes. So, was the problem with CDesktop, or am I not doing something correctly? Also, is there a way to make a window completely ignore clicks other than overriding CDesktop::DispatchClick ? Thanks, Mike. Michael A. Kelly | "Fish heads, fish heads, Internet: mkelly@cs.uoregon.edu | Roly-poly fish heads, America Online: Michael792 | Fish heads, fish heads, Compu$erve: 73567,1651 | Eat them up, yum!" - Barnes & Barnes Path: ucivax!gateway From: phils@chaos.cs.brandeis.edu (Phil Shapiro) Subject: CDesktop Message-ID: <9012202049.AA18758@chaos.cs.brandeis.edu> In-Reply-To: mkelly@cs.uoregon.edu's message of 20 Dec 90 07:02:58 GMT <9012200702.AA24751@pragmatix.cs.uoregon.edu> Newsgroups: fa.think-c Lines: 41 Date: 20 Dec 90 20:54:45 GMT From: mkelly@cs.uoregon.edu Newsgroups: fa.think-c Date: 20 Dec 90 07:02:58 GMT I've found a possible bug in CDesktop. The procedure RemoveWind is: void CDesktop::RemoveWind( CWindow *theWindow) /* Window object to remove */ { if (theWindow->IsVisible()) { HideWind(theWindow); /* Hide window before removing it */ } itsWindows->Remove(theWindow); } If this procedure removes the last window from the desktop, the CDesktop instance variable 'topWindow' should go to NIL, shouldn't it? It doesn't; it remains whatever it was before the call. The method CDesktop:HideWind() will correctly set the topWindow to the first visible window in the itsWindows list. The code is: if (theWindow == topWindow) { /* This was the top window */ /* Find the new top top window */ topWindow = (CWindow*) itsWindows->FirstSuccess(Window_IsVisible); If FirstSuccess() fails, it should return NIL. Maybe you're removing a window that isn't visible -- but in that case, it shouldn't be the top Window. Also, is there a way to make a window completely ignore clicks other than overriding CDesktop::DispatchClick ? The drag, close, grow, and zoom clicks are all handled by the Desktop. If you want to disable these functions, or change their behavior, you must override the CDesktop methods. I guess the author of the TCL considered the behavior of these functions to be an attribute of the type of Desktop you're using -- or maybe it was just the easiest way :-). -phil Path: ucivax!gateway From: paco@neuromancer.sps.mot.COM (Paco Xander Nathan) Subject: re: How does one get Profile to print to a file? Message-ID: <9012202010.AA16248@neuromancer> Newsgroups: fa.think-c Lines: 69 Date: 20 Dec 90 23:08:38 GMT You can also punch directly down to using the VIA timer to gather performance stats. We use this method here since it's more flexible for analyzing special tasks, critical loops, etc. rather than trying to work around the ThC profiler. NB: you need to run/stop the timer once before measure the first time and you can't leave the timer running since it tends to crash other program functions like SoundMgr. paco ---- Internet: paco@neuromancer.sps.mot.com Voice Mail: 1.512.891.2973 Neural Networks & Virtual Reality Development Center for Emerging Computer Technologies Motorola, Inc. Austin, Texas, USA (c)1990, PXN. B*B*! Subject to Public Law 99-508 It's nearly Yule, so happy holidays to all! /****** Snip, snip ****** Snip, snip ****** Snip, snip ****** Snip, snip ******/ #include <profile.h> /**************** INITIALIZE ********************/ matchCount = 0L; matchTicks = 0L; start_VIA_timer(); (void) VIA_ticks(); stop_VIA_timer(); /**************** MEASURE ********************/ register long lastTick; register long startTick; start_VIA_timer(); startTick = VIA_ticks(); /** run something **/ lastTick = VIA_ticks(); stop_VIA_timer(); if (matchTicks > (MAXLONG - 500)) { matchTicks = 0; matchCount = 0; } matchTicks += (lastTick - startTick); matchCount++; /**************** REPORT ********************/ float matchTime = 0.0; if (matchCount > 0L) matchTime = ((float) matchTicks / (float) matchCount * 0.00128) + 0.5; MoveTo(5, 50); sprintf(theText, "%7.3f msec Match Mean", matchTime); DrawString((StringPtr) CtoPstr(theText)); Path: ucivax!gateway From: Tom_Johnson@chip.cs.ucla.edu (Tom Johnson) Subject: Think C and Graf3d Message-ID: <9012271010.aa28086@ICS.UCI.EDU> Newsgroups: fa.think-c Lines: 9 Date: 27 Dec 90 18:11:31 GMT Think C and Graf3d Has anyone out there tried to do Graf3d stuff in Think C? I noticed that Symantec provides the Graf3d library with Think Pascal, but not with Think C. I'd be most appreciative if anyone could point me in the direction of some C sample code for this stuff. Tom Path: ucivax!gateway From: mwg@med.pitt.edu ("Michael W. Groff") Subject: background processing w/ TCL Message-ID: <9012281631.AA09600@earth.med.pitt.edu> Newsgroups: fa.think-c Lines: 7 Date: 28 Dec 90 16:37:44 GMT Please help me!! I am converting a unix program to run on the mac. How do I support multifinder and background processing with the Think Class Lib? Give me all your suggestions and I will post a summary if there is demand. Thanks in advance, mike groff mwg@med.pitt.edu Path: ucivax!gateway From: resnick@cogsci.uiuc.edu (Pete Resnick) Subject: THINK C and MacTCP Message-ID: <9012290104.AA17012@tarski.cogsci.uiuc.edu> X-Mailer: ELM [version 2.3 PL2] Newsgroups: fa.think-c Lines: 49 Date: 29 Dec 90 01:06:59 GMT Here is what to do to the MacTCP header files that are provided: In MacTCPCommonTypes.h, change the "ifndef/include" definition to look like this: #ifndef _MacTypes_ #include <MacTypes.h> #endif /* _MacTypes_ */ In UDPPB.h, comment out the word "pascal" in the declaration of UDPNotifyProc so that it looks like this (THINK C doesn't like a typedefed pascal funtion): typedef /*pascal*/ void (*UDPNotifyProc) ( StreamPtr udpStream, unsigned short eventCode, Ptr userDataPtr, struct ICMPReport *icmpMsg); Do the same thing in TCPPB.h for TCPNotifyProc: typedef /*pascal*/ void (*TCPNotifyProc) ( StreamPtr tcpStream, unsigned short eventCode, Ptr userDataPtr, unsigned short terminReason, struct ICMPReport *icmpMsg); You will not be able to use these typedefs in your own programs; they are only to deal with other declarations in your header files. In your code, declare your notify procs as 'pascal void whatever' with all of the parameters as they are declare in the typedef. That's all I have found so far, except that in my copy of TCPPB.h, instead of the csCode for receive buffer return being TCPBfrReturn as documented, it was TCPRcvBfrReturn. I changed it to TCPBfrReturn. There were also a couple of changes in the DNR stuff, but these have been hacks so far. I will send them if they become more reasonable. One which you should watch out for is that instead of OpenRFPerm, you must use OpenResFile. Currently I don't know why. pr -- Pete Resnick (...so what is a mojo, and why would one be rising?) Graduate assistant - Philosophy Department, Gregory Hall, UIUC System manager - Cognitive Science Group, Beckman Institute, UIUC Internet/ARPAnet/EDUnet : resnick@kant.cogsci.uiuc.edu BITNET (if no other way) : FREE0285@UIUCVMD Path: ucivax!gateway From: mrose@cheetah.ca.psi.COM (Marshall Rose) Subject: Re: THINK C and MacTCP Message-ID: <26779.662434836@cheetah.ca.psi.com> In-Reply-To: Your message of Fri, 28 Dec 90 19:04:54 -0600. <9012290104.AA17012@tarski.cogsci.uiuc.edu> Newsgroups: fa.think-c Lines: 6 Date: 29 Dec 90 01:49:06 GMT Thanks. Can you summarize what the DNR problems are? I'm not using it yet, but would like to know what to expect. Thanks, /mtr Path: ucivax!orion.oac.uci.edu!teri.bio.uci.edu!bdugan From: bdugan@teri.bio.uci.edu (Bill Dugan) Subject: GWorld Nntp-Posting-Host: teri.bio.uci.edu Message-ID: <277D5262.6723@orion.oac.uci.edu> Newsgroups: fa.think-c Keywords: GWorld 32-bitquickdraw Reply-To: bdugan@teri.bio.uci.edu (Bill Dugan) Organization: University of California, Irvine Lines: 6 Distribution: na Date: 30 Dec 90 02:35:15 GMT Question: I am trying to rewrite some graphics routines using the new GWorld and I get link errors with my "include the world" project. Is there some .h file I'm missing, or an update to THINK C? bill 
