On Fri, 29 Jun 2001 07:28:06 -0500, Shahan, Ray
<[log in to unmask]> wrote:

>Since I'm responsible, in part, for this thread...I'll try to end it with
>one last explanation:
[snip example]
>
>Now, have I stated the "READABILITY" factor enough times, and in enough
ways
>to point out that the END-INSPECT scope terminator, while not necessary for
>correct program execution, would be nice to have, so that all the code
could
>be written using scope terminators to help with the readability of the code
>overall?  :-)  :-)  :-)

So how about a bit of compromise?  Put a "*" in column 7 (the comment
indicator) on any line that has END-INSPECT.  A caveat however: if this is
the last "statement" within a paragraph, either the preceding line should
have a terminating ".", or you would(should) have a free-floating "." on
the line following  [aside: I've seen code here where the programmer went
to great pains to ensure there were free-floating "."'s in paragraphs AND
that this was the only "." in the paragraph...]  A lesser caveat (although
I don't know that anyone would actually do this) is that the "end-inspect"
obviously couldn't be on the same line as any inspect clauses, as in:

   INSPECT some-string REPLACING ALL whats BY evers END-INSPECT

If you think you might have difficulty remembering to insert the comment
indicator, perhaps you could write some form of pre-compiler that looks for
uncommented END-INSPECT phrases and re-writes the line with the appropriate
comment indicator.  Just thinking off the top of my head, perhaps even
qedit can do this via a semi-complex "change" command, something like:

   /c 7/7,"*","end-inspect"(u)

[which could be embedded into a compilation jobstream so you wouldn't have
to "remember" to do this pre-compilation step...]

Tom

* To join/leave the list, search archives, change list settings, *
* etc., please visit http://raven.utc.edu/archives/hp3000-l.html *