home *** CD-ROM | disk | FTP | other *** search
- Xref: sparky comp.sys.mac.system:15484 comp.sys.mac.programmer:20238
- Path: sparky!uunet!mcsun!Germany.EU.net!Informatik.Uni-Dortmund.DE!siggi!maennig
- From: maennig@siggi.informatik.uni-dortmund.de (Olaf Maennig)
- Newsgroups: comp.sys.mac.system,comp.sys.mac.programmer
- Subject: SADE --- the worst debugger I ever used!!!!!
- Date: 23 Dec 1992 10:57:25 GMT
- Organization: CS Department, Dortmund University, Germany
- Lines: 77
- Sender: maennig@siggi (Olaf Maennig)
- Distribution: world
- Message-ID: <1h9gmlINNk3f@fbi-news.Informatik.Uni-Dortmund.DE>
- Reply-To: maennig@siggi.informatik.uni-dortmund.de
- NNTP-Posting-Host: siggi.informatik.uni-dortmund.de
-
- Hi! (frustrated greeting)
-
- I have developed programs on several Sun SPARCs and DEC so far. A month ago
- I started in a new job, in which I have to program the Mac (System 7) with
- MPW C/C++. I have never used a Macinotsh before, and I started with a high
- motivation due to the knowledge that the Mac is supposed to be one of the
- user-friendliest computers.
-
- This motivation turned into frustration soon. I find the GUI extremely slow
- and in comparison to X11 it is in my opinion really poor. But that's another
- thing and not the subject here.
-
- What's disgusting to me is the SADE debugger. I work with version 1.3 and I
- think I've never sweared so much on a single morning. Maybe I'm using the
- program the wrong way, but first of all is the manual the most unuseful
- thing in that package. It's full of mistakes, contradictions, and most
- important: it's incomplete.
-
- An example? See page 59: "Referencing structured types". The selection,
- pointer dereference, and array access operators are listed. "The rest of
- this section illustrates the use of all three of these operators." Ha! I
- want to know where the array access operator is illustrated?! That's what
- I'm interested in at the moment. I don't achieve to get the content of an
- character array (not the dereference of an char* pointer!).
-
- What I get after selecting the variable "array" and choosing the menu "show
- value" is:
- `array = 84, ... # content: "Text"
-
- How nice would it be to get
- `array = ['T' 'e' 'x' 't']
- or the like.
-
- I REALLY don't like to step through the array manually with
- `array[1] <ENTER>
- and get a number which I have to look up in an ASCII table!!!!!!
-
- By the way, here we see another mistake in the manual. Somewhere it says
- that arrays are 1-based in SADE. That's wrong. Indeed, arrays start with
- index 0, as usual.
-
- So, how can I tell the debugger to display my array?
-
- Another thing is that arrays aren't even accessible when their sizes are
- "too big". When I initialize a 512-byte char array in the beginning, SADE
- simply states that the variable is too big to fit into the environment.
- HORRIBLE!!
-
- Then, what will I have to do when an error occurs which makes the examined
- program crash? I will have to restart it. That's fine. But when I want to
- debug a program which runs as an MPW tool, I will kill the whole MPW with
- "kill" or "untarget". Re-start MPW, correct the program, compile and debug
- once more. When there is still the error --- crash! Kill whole MPW,
- re-start, ... BULLSH...!!!
-
- Finally, I never saw a program which uses windows in such a heavy way. For
- nearly every little output a single window is opened. That's alone is not
- that bad, but when I trace my program it's sooooo slow. Additionally, before
- (or after?) each trace step the source file window is moved into the
- background, all window frames are drawn new, and right after that the source
- file window is moved to the top, and its contents are re-drawn. I never saw
- such a stupid thing! Not only it is stupid, it is additionally slow.
-
- Okay, that's what I went through the last few weeks. Now, the most important
- question to me: IS THERE ANOTHER AND BETTER DEBUGGER OUT THERE?
- If no, I would like to get in contact with other SADE users. Maybe they can
- give me solutions to the described problems.
-
- If you want to prevent a suicide, help a little German Macintosh programmer!
- --
- ---------------------------------------------------------------------------
- _ Department of Computer Science IV
- / \ |\/| University of Dortmund, Germany
- \_/laf | |aennig e-mail: maennig@siggi.informatik.uni-dortmund.de
- -----------------------------------------------------------
- "To understand recursion you need to understand recursion."
- ---------------------------------------------------------------------------
-