[SGVLUG] "Survival At C" Discussion Thread

[Dustin Laurence]

The lively discussion after my talk may indicate that there might be
interest in continuing the conversation/debate/outrage, so this is an
open thread for discussion of anything at least vaguely related to my
talk.  Ask/comment/criticize/bellow away. :-)

I'm particularly interested in critiques and suggestions of the
particular topics chosen--I could only fit in three topics, and one
could certainly make a case that other topics should replace one or more
of the ones I chose the next time I give it.  What was or was not useful
to you, and what do you wish I'd covered?

To kick it off, here are some responses to a couple of issues that came
up last night, put together while drinking tea from my speaker's mug. :-)

1. Last night I was asked what to do when the C Exception Handling
Pattern is forbidden by your coding standards.  That's an *excellent*
question since a great many coding standards will flatly forbid it (e.g.
MISRA 2004, though in the current version the prohibition is only
advisory for *forward* gotos), and very often they are enforced without
understanding of the principles standards are supposed to serve.  I need
to address that in any talk that presents the pattern, and I'm grateful
for having the flaw pointed out.

I've run through a few different goto-free implementations and will
offer the final ones for discussion ("Goto Reconsidered"?) on this thread.

2. Someone brought up the "Goto Considered Harmful Considered Harmful"
response to Dijstra.  I looked it up and '"GOTO Considered Harmful"
Considered Harmful' was not Knuth's article, but rather a long letter to
the editor in Communications of the ACM":

http://web.archive.org/web/20090320002214/http://www.ecn.purdue.edu/ParaMount/papers/rubin87goto.pdf

Knuth's response was titled "Structured Programming with goto Statements":

https://www.cs.sjsu.edu/~mak/CS185C/KnuthStructuredProgrammingGoTo.pdf

Does interest in a thirty-year-old food fight constitute a sign of some
kind of borderline personality disorder?  I hope not.

Dustin