[Nmh-workers] too many uninitialized variables in nmh-1.1 RC4

[Igor Sobrado]

Hi!

I tried sending this email to Jon Steinhart, but his spam filter is doing
a fine work to stop all the messages I sent him today.  :-(

Perhaps it is useful for other nmh developers on this mailing list too:


I am building the latest release candidate of nmh-1.1.  From the output of
the compiler I see that there are *a lot* of uninitialized variables.  Really
a lot of them!  I will use this nmh release candidate on a machine were I am
not receiving important email (I prefer nmh-1.0.4 on production machines
until nmh-1.1 is released).  I will provide some feedback if I see something
wrong.

Perhaps someone should fix these uninitialized variables before nmh-1.1 is
released.  C is a bit tricky in this matter and the standard does not assign
a default value to these variables (even if some compilers do).  A wrong but
valid value could provoke an unexpected behaviour in nmh.

Nice to see that nmh is alive!

Best regards,
Igor.