AVR-Chat

--- In AVR-Chat@yahoogroups.com, David VanHorn <microbrix@...> wrote:
>
> > My opinion is that most experienced, or even semi-experienced programmers
> > follow, if unconsciously, a set of "standards" like this. Some of these
> > things are for your own sanity. Some are for consistency if the project
> > ever needs to be handed off to someone else. Many seek to reduce bugs and
> > a few deal with core safety issues.
> >
> > I would certainly like to read what they have but its not worth $100 to me.
> 
> 
> There's at least a partial list here:
> http://www.abxsoft.com/misra/misra-test.html
> 
> I like the one about not using any octal constants other than Zero.
> 

http://www.misra.org.uk/

The MISRA standards are from the auto industry (and cost money also)
but are hard to use 100% for most people. Even the Netrino people
don't follow them 100% in their standard

 [Our few known differences of opinion with [MISRA04] are 
  identified in the footnotes to this standard.]

I have worked places that had a group stick to a single
coding standard for more than a few projects but it was 
usually a smaller subset of the stuff mentioned on the 
Netrino web page (or the MISRA standards).

If it is a sane set of rules covering brace placement 
and naming standards and sticking to use of typedefs in
stdint.h most programers would I think see some gain
in ease of interaction and reuse.

The hardest parts to get agreement on are brace placement
and Camel vs underscore word seperation. After those two
the naming conventions seem to be much easier but maybe that
is because everyone except the zelots is sick of the whole
subject and just wants to get it over with <g>.