[comp.mail.mush] Re^2: Mush 7.0, PatchLevel 4, and folders.

sbanner1@uvicctr.UVic.CA.UUCP (S. John Banner) (03/03/90)

schaefer@ogicse.ogi.edu (Barton E. Schaefer) writes:

>In article <929@uvicctr.UVic.CA.UUCP> sbanner1@uvicctr.UVic.CA.UUCP (S. John Banner) writes:
>} mention.  On line 35 of folders.c, is a the line:
>} 
>}     } else if (!tempfile || !*tempfile) {

>If it actually generated an *error* rather than a *warning*, you have a
>broken compiler.
Yep, an error, compiler exits with a non-zero code, and the make stops,
with no .o file produced.  Most annoying of it.
The message from the make is:
"folders.c", line 35: operand of ! must be arithmetic or pointer
*** Error code 1


>I can't decide whether to remove this or not.  Clearly, as it stands, it
>isn't *necessary* -- but it also doesn't affect the correctness of the
>code.  If the test were removed, and later the declaration of "tempfile"
>should change to become a pointer (to a malloc'd array), the remaining
>test of !*tempfile would break.

>All in all, I think it's doing more good than harm, but the suggestion
>is taken under advisement.
That's all I can ask.  So long as it is not something that was put there
in the past, and is left over because it was overlooked (which it may
still be, but at least is is no longer forgotten/lost if it was).

sbanner1@uvicctr.UVic.CA.UUCP (S. John Banner) (03/03/90)

schaefer@ogicse.ogi.edu (Barton E. Schaefer) writes:

>In article <929@uvicctr.UVic.CA.UUCP> sbanner1@uvicctr.UVic.CA.UUCP (S. John Banner) writes:
>} mention.  On line 35 of folders.c, is a the line:
>} 
>}     } else if (!tempfile || !*tempfile) {


>broken compiler.
Yep, an error, compiler exits with a non-zero code, and the make stops,
with no .o file produced.  Most annoying of it.
The message from the make is:
"folders.c", line 35: operand of ! must be arithmetic or pointer
*** Error code 1


>I can't decide whether to remove this or not.  Clearly, as it stands, it
>isn't *necessary* -- but it also doesn't affect the correctness of the
>code.  If the test were removed, and later the declaration of "tempfile"
>should change to become a pointer (to a malloc'd array), the remaining
>test of !*tempfile would break.

>All in all, I think it's doing more good than harm, but the suggestion
>is taken under advisement.
That's all I can ask.  So long as it is not something that was put there
in the past, and is left over because it was overlooked (which it may
still be, but at least is is no longer forgotten/lost if it was).

				 sjb.