home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.arch
- Path: sparky!uunet!uunet.ca!geac!itcyyz!yrloc!rbe
- From: rbe@yrloc.ipsa.reuter.COM (Robert Bernecky)
- Subject: Re: What would you like in a debugger?
- Message-ID: <1992Aug13.144912.8504@yrloc.ipsa.reuter.COM>
- Reply-To: rbe@yrloc.ipsa.reuter.COM (Robert Bernecky)
- Organization: Snake Island Research Inc, Toronto
- References: <19920809.175600.935@almaden.ibm.com> <343@moene.indiv.nluug.nl> <chris.0qws@genly.uucp> <1992Aug12.011403.16934@technix.mn.org>
- Date: Thu, 13 Aug 92 14:49:12 GMT
- Lines: 47
-
- In article <1992Aug12.011403.16934@technix.mn.org> bret@technix.mn.org (Bret Indrelee) writes:
- >
- >I don't know how many debuggers I have given up on because they decided
- >to display the contents of a pointer, rather than the value of the pointer.
- >
- >If you keep with the syntax of the native language, the programmer should
- >be able to see that they asked for the pointer rather than the value at
- >the pointer. If you try to 'fix their error' it makes it really nasty
- >to really see the value of a pointer.
- >
- >Go chase a linked list bug if you don't think you need to see the value
- >of a pointer. Or use mmap() with your favorite memory-mapped device. Or
- >any number of other similair things...
- >
- >If you can manage it, another feature that would be nice is disassembing
- >code when someone asks you to display the contents of a pointer to function.
- >If your feeling really hot, try to relate it to the source line in the
- >program...just make sure that this 'feature' can be disabled for when there
- >is a misinterpretation.
-
- I think the important point here is that no debugger designer can
- determine which debugging facilities are adequate for a specific
- debugging job. Because of this (and for convenience), we at I.P. Sharp
- Associates (May it rest in peace, stomped as it was by Big Feet)
- developed a system which combined interactive debugger and dump
- analysis facilities, written entirely in SHARP APL.
-
- Since everything was just APL functions, it was trivial to
- build a special purpose function on the fly for your particular
- problem. So, special purpose formatters (funny translations,
- linked list chasers, table transposers, etc) are trivial to
- build.
- This work was incredibly successful, and I strongly recommend
- that those of you who are in possession of USABLE interactive
- systems consider the development and/or integration of interactive
- tools to allow on-the-fly customization of dump/debug tools to
- make the programmer's job easier.
-
- Bob
-
-
-
- Robert Bernecky rbe@yrloc.ipsa.reuter.com bernecky@itrchq.itrc.on.ca
- Snake Island Research Inc (416) 368-6944 FAX: (416) 360-4694
- 18 Fifth Street, Ward's Island
- Toronto, Ontario M5J 2B9
- Canada
-