home *** CD-ROM | disk | FTP | other *** search
Text File | 1991-03-06 | 2.8 KB | 55 lines | [TEXT/GEOL] |
- Item 2225615 13-Feb-91 07:34PST
-
- From: PHAROS.TECH Pharos Tech, Tech Staff,PRT
-
- To: MACAPP.TECH$ MacApp Technical
-
- ------------------------------------------------------------------------------
-
- Sub: RE: MacApp and C++
-
- From: Schmitz, Scott D. on Wed, Feb 13, 1991 10:40 AM
- Subject: RE: MacApp and C++
- To: MacApp
-
- Steve,
- I don't know how much effort is involved in keeping two versions of the source
- code around. However, if it is significant I would rather Apple spend its time
- on other things. This comes from a Die Hard Object Pascal fan. Here are a few
- alternative things I would rather see done:
-
- • Integrate ViewEdit with ResEdit.
- • Provide Integrated Source Level Debugging. I mean provide something as
- powerful as THINK Pascal's debugger to MPW. This is a difficult task and not
- to be considered lightly. Several companies have attempted this and failed.
- Two examples of a good idea implemented poorly are SADE and the THINK C
- debugger. Glad to hear that this is already in the works.
- • Harden the syntax error checking of the C and C++ compilers. One solution
- would be to add more of lint functionality into the compiler. OK, go ahead and
- make this stuff flippable through a compiler directive. As a Pascal person
- moving to C I am constantly disgusted with the poor error checking that C does.
- It's amazing what the compiler will take. If Apple wants it's existing
- knowledge base of MacApp programmers to think kind thoughts on this C++
- transition, then add some of the things Pascal people like to C++.
- • When the C compiler finds an error it reminds me of an old lady tripping and
- falling down 40 flights of stairs. Like the old lady, the C compiler has a
- hard time recovering from an initial syntax error. The first error triggers a
- false assumption by the compiler and cascades perhaps 10 errors for that line
- and then all the code after than line is incorrectly understood based on the
- initial incorrect assumption. A good compiler is able to recover from an
- initial syntax error enough so that it can continue and produce useful error
- conditions for other possible errors.
- • Either make faster hardware or speed up the compilers. Part of the basis of
- user friendliness is interaction. I feel like I am using punched cards in
- batch mode when it takes so long to do a compile.
- • We all bought Macs not UNIX machines. How about a Mac interface for our
- programmer environment? I LIKE the Mac interface. Why am I not allowed to use
- it when I program? MPW is just not user friendly. Why have I never seen a
- modeless dialog box when using MPW? I can create a make file in Lightspeed
- Pascal but in MPW, I cut and paste an ond one when I need a new one. I'll be
- honest, I don't fully understand how to write a Make file in MPW.
-
- Scott Schmitz
-
-
-