Re: Extra information in the message string of exceptions.

[Jerry <list_email@icloud.com>]

On Friday, October 28, 2022 at 5:35:20 PM UTC-7, Jerry wrote:
> On Thursday, October 27, 2022 at 4:57:08 AM UTC-7, roda...@gmail.com wrote: 
> > On 8/6/22 19:08, Dmitry A. Kazakov wrote: 
> > > On 2022-06-08 09:31, Jerry wrote: 
> > > 
> > >> But has no effect on this program; 
> > >> 
> > >> procedure CE_2 is 
> > >> i : Positive; 
> > >> j : Integer := 1; 
> > >> begin 
> > >> i := -j; 
> > >> end CE_2; 
> > > 
> > > I noticed that it requires a user-defined [sub]type AND nested subprogram: 
> > > 
> > > procedure Test (A : Integer) is 
> > > subtype T is Integer range 0 .. 42; 
> > > i : T; 
> > > begin 
> > > i := A; -- This will work 
> > > end; 
> > > begin 
> > > Test (-1); 
> > > 
> > Are you sure about this ? 
> > 
> > I tried ... 
> > 
> > 
> > procedure Example 
> > is 
> > X : Positive := 5; 
> > begin 
> > loop 
> > X := X - 1; 
> > end loop; 
> > end Example; 
> > 
> > 
> > ... with the -gnateE and saw this result ... 
> > 
> > 
> > [rod@orth example]$ ./example 
> > 
> > raised CONSTRAINT_ERROR : example.adb:6:14 range check failed 
> > value 0 not in 1..2147483647 
> > [./example] 
> > 0x55916975377f Example at example.adb:6 
> > 0x559169753b74 Main at b__example.adb:193 
> > [/usr/lib/libc.so.6] 
> > 0x7f9230be728e 
> > 0x7f9230be7348 
> > [./example] 
> > 0x559169753623 _start at start.S:115 
> > 0xfffffffffffffffe
> So -gnateE works part of the time? 
> Jerry
I simply do not get the robust results others are reporting, and as I noted earlier in this thread, some of the suggested switches are flagged as illegal on macOS

From the docs at https://gcc.gnu.org/onlinedocs/gcc-4.9.4/gnat_ugn_unw/Symbolic-Traceback.html#Symbolic-Traceback:
"Note that this feature is not supported on all platforms. See GNAT.Traceback.Symbolic spec in g-trasym.ads for a complete list of currently supported platforms."

I'm using an older version of GNAT (2017) on Mac and the list of supported platforms is actually in s-trasym.ads and reads as such:
--  The full capability is currently supported on the following targets:

--     HP-UX ia64
--     GNU/Linux x86, x86_64, ia64
--     FreeBSD x86, x86_64
--     Solaris sparc and x86
--     Windows

Maybe newer versions of GNAT are more Mac-friendly. So in order to get both extra information and symbolic tracebacks I've had to write my own garbage-y handler that redirects text output to a file, calls system stuff such as tail and atos, bla bla bla. I hope atos is still supported when I eventually move to ARM. atos I think is Apple's thing that works like addr2line.

BTW, for others attempting this on macOS, turn of address randomization with -no_pie. That is, in my gpr, I have this:
	package Linker is
		for Default_Switches ("Ada") use ("-Wl,-no_pie"); 
	end Linker;