Re: Code problems with Perform Thru Exit causes fall through

On Jul 28, 1:45 pm, "tlmfru" <la...@xxxxxxx> wrote:

SECTIONs do have their uses, for example segmentation. As I pointed
out above I dislike things that can easily lead to errors and that the
compiler won't tell you about. In fact I would use that as a
definition of what is _not_ 'easy to maintain' and _not_ 'perfectly
effective'.

The compiler can't tell if you should have used a minus instead of a plus
sign. Better ban it!

While the comiler won't tell you that a GO TO has been used, or a
SECTION or a THRU, the presence of these is easy to determine with
grep or similar.

Perhaps you find that these errors are simply a part of 'normal
programming' and find them easy, after all what would you do for the
rest of the week if they did not occur.

I don't find "these errors" happening at all. I have heard of the concept
called "testing" and use it all the time. I find that "testing" locates all
these errors.

While testing may well find many programming errors it is unlikely to
find _all_ of them. For example if there is a GOTO X-exit in the scope
of a PERFORM A THRU A-exit (due to copy-paste from another) then it
will only be 'found' if that particular branch is actually tested. As
many of these may be rare errors or exceptions it may not be found
even in live running for months.

Of course automated tools can be used to find these if they have been
done in a standard way. Was it Judson who had this.

I'm a big advocate of "testing". I suppose you could be
right, though: if I didn't use "testing" then I would have a lot more to do
during the week. (FWIW, I don't use PERFORMED sections).

The only purpose of doing PERFORM A THRU A-exit is allow a

All sarcasm aside, you are making the mistake of assuming that since you do
things one way, that it is the only way to do it.

Actually, if you read the messages instead of reacting you would have
noticed that I was replying to a person who probably had one or more
of those exact programming errors. It was a specific message to them
to check for those errors. I also stated how _I_ avoided them
_without_ claiming that is was "the only way".

Perhaps you could specify
what is acceptable program lifetime to qualify as the right way to do
things: I have several systems which are now over 15 years old and running
successfully - which accepted Y2K without changes - which have not been
changed at all except for spec. changes - different data formats, changed
calculations, a new report, etc. These systems use all sorts of syntax
which you would undoubtedly condemn. Your privelege. They work. Enough
said.

The discussion was not only about whether it worked, it was also about
"Is it easy to maintain?". In at least one case, the origin of this
thread, the answer seems to be NO, but we haven't heard the resolution
yet.

Does it follow the installation standards?
Discussion over!

Yes, that is often used to protect poor programming styles.

Also used to provoke thoughtless responses. D'you hold that installation
standards are automatically wrong? Or only when you disagree with them?

I have no views on installation standards themselves, only on the
closed minds that call upon them as a reason to shut off thinking.


.

Relevant Pages

  • Re: Why not add namespace feature into standard C?
    ... features, I can't understand why not add this feature into standard C. ... Because the standards people were more concerned with adding arbitrary ... All the compiler vendors have balked. ...
    (comp.lang.c)
  • Re: SOURCEFORMAT AND COPY LIBRARY
    ... Jimmy? ... Oh that human communication problem! ... version of the compiler, ... I still LIKE YOU :-) - but you are very passionate about the Standards. ...
    (comp.lang.cobol)
  • Re: Testing ham radio software on Sun hardware.
    ... Is that like the way Micro$oft adheres to the W3C standards? ... It sometimes means code built with gcc will not compile on better compilers. ... The other issue is that people often write code that works on Linux, but makes use of Linux specific header files. ... If anyone has some open-source UNIX or Linux code they would like to test on a Sun, using a commercially produced C compiler, let me know. ...
    (uk.radio.amateur)
  • Re: COBOL "non-myth" confirmed - Index and subscripts (MF on Windows)
    ... and will not, by itself, have subscripts catch up to indexes. ... by compiling with optimization off. ... I'll wager 100US it shows the offset from the beginning of the table on my compiler, ... IBM _are_ in the business of writing standards. ...
    (comp.lang.cobol)
  • Re: Moving from 8051 to AVR
    ... particular task -- all of which adhear to the standards. ... incrementing a pointer through an array. ... If the compiler can figure out what is happening, ... An architecture which is C-Friendly will simplify writing the compiler ...
    (comp.arch.embedded)