[GRASS-user] FYI - Output from compiling most recent release 7.0

[Glynn Clements]

Craig Aumann wrote:

> Did it and attached!

Apart from the TIFF errors discussed elsewhere, I see few warnings
which indicates real errors.

raster/r.statistics/o_sum.c:

o_sum.c: In function ‘o_sum’:
o_sum.c:47:5: warning: ‘stat’ is used uninitialized in this function [-Wuninitialized]

imagery/i.segment/create_isegs.c.

create_isegs.c: In function ‘calculate_shape’:
create_isegs.c:965:14: warning: ‘pl1’ is used uninitialized in this function [-Wuninitialized]
create_isegs.c:965:14: warning: ‘pl2’ is used uninitialized in this function [-Wuninitialized]
create_isegs.c:967:17: warning: ‘n2’ is used uninitialized in this function [-Wuninitialized]
create_isegs.c:967:17: warning: ‘n1’ is used uninitialized in this function [-Wuninitialized]
create_isegs.c:967:31: warning: ‘s2’ is used uninitialized in this function [-Wuninitialized]
create_isegs.c:967:31: warning: ‘s1’ is used uninitialized in this function [-Wuninitialized]
create_isegs.c:968:17: warning: ‘e2’ is used uninitialized in this function [-Wuninitialized]
create_isegs.c:968:17: warning: ‘e1’ is used uninitialized in this function [-Wuninitialized]
create_isegs.c:968:31: warning: ‘w2’ is used uninitialized in this function [-Wuninitialized]
create_isegs.c:968:31: warning: ‘w1’ is used uninitialized in this function [-Wuninitialized]
create_isegs.c:983:11: warning: ‘count1’ is used uninitialized in this function [-Wuninitialized]
create_isegs.c:983:11: warning: ‘count2’ is used uninitialized in this function [-Wuninitialized]

The other -Wuninitialized warnings are all "may be used ...", which
usually just indicates that they are conditionally initialised, and
the compiler cannot be sure that they'll always be initialised in the
cases where they're used (either because the logic is non-trivial or
the usage is only safe when arguments have valid values).

There are quite a few -Wformat warnings, most of which indicate real
errors (mostly arising from assumptions that int, long, size_t and
off_t are all 32-bit).

The -Wformat-security warnings are mostly noise caused by the
localisation macro. r57238 was supposed to have fixed this; I don't
know why it hasn't.

The -Wunused-variable warnings are probably harmless. Likewise for
-Wpointer-sign.

-- 
Glynn Clements <glynn at gclements.plus.com>