Dear Alexel,
even with your suggested correction, it gives exactly the same error.
I think the problem lies in some way in my assignment operator, when I use the overloaded operator= of numeric.


rinterval& rinterval::operator=(const rinterval &rhs)
{
       if (this == &rhs)  return *this;    // Same object?
       else
       {
       try
       {
       this->min=rhs.min; 
// I expect here the assignment operator of numeric to destroy the old instance of min and create a new one
// but it seems like he simply slices rhs.min to fit into min, therefore losing some information about the precision
// of rhs.min; I tried also this using the CLN library and, if I define min as a cl_R and rhs.min is a cl_LF, when I come to this point
// it seems like the precision of min doesn't change, even if the precision of rhs.min is bigger;
// I would like the assignment operator to destroy the former istance of min and build a new one, but it doesn't seem to work
// I tried calling the destructors explictly, but it doesn't work either.
       this->max=rhs.max;
       error=rhs.error;
       }
       catch(std::exception & exc)
       {
           std:cerr << "Exception" << exc.what() << endl;
       }
       return *this;
       }
}

Best regards
Isaia

On Tue, Mar 29, 2011 at 12:00 PM, <ginac-list-request@ginac.de> wrote:
Send GiNaC-list mailing list submissions to
       ginac-list@ginac.de

To subscribe or unsubscribe via the World Wide Web, visit
       https://www.cebix.net/mailman/listinfo/ginac-list
or, via email, send a message with subject or body 'help' to
       ginac-list-request@ginac.de

You can reach the person managing the list at
       ginac-list-owner@ginac.de

When replying, please edit your Subject line so it is more specific
than "Re: Contents of GiNaC-list digest..."


Today's Topics:

  1. assignment (Nisoli Isaia)
  2. Re: assignment (Alexei Sheplyakov)


----------------------------------------------------------------------

Message: 1
Date: Mon, 28 Mar 2011 11:33:53 +0200
From: Nisoli Isaia <nisoli@mail.dm.unipi.it>
To: ginac-list@ginac.de
Subject: [GiNaC-list] assignment
Message-ID: <BANLkTi=ebBvqzhPPdPiZGXfKatwgt64+fw@mail.gmail.com>
Content-Type: text/plain; charset="iso-8859-1"

Dear all,
I have a problem with Ginac.
I'm sorry if my question seems stupid, but I cannot solve the issue.

So, I define a new class:

class rinterval
{
       public:
       numeric min;
       numeric max;
       numeric error;

       //Constructor
       rinterval();
       rinterval(const numeric &, const numeric &, const numeric &);
       //Destructor
       ~rinterval();
       //Friend
       friend const rinterval dynamic(const rinterval &);
       friend std::ostream &operator<< (std::ostream &os, const rinterval
&inter)
       {
       return os << "[" << inter.min << "," << inter.max <<"], error = " <<
inter.error << endl;
       };
       rinterval& operator=(const rinterval &rhs);
};

The involved functions are these

rinterval::rinterval(const numeric &input1, const numeric &input2, const
numeric &miserr)
{
   if (input1>input2)
       {
           min=input2;
           max=input1;
       }
   else
       {
           min=input1;
           max=input2;
       }
   error=miserr;
}

const rinterval dynamic(const rinterval &domain)
{
   numeric a="1.47777777777777777777777777779999";
   rinterval temp(a*domain.min,a*domain.max,domain.error);
   cout << temp;
   return temp;
}

rinterval& rinterval::operator=(const rinterval &rhs)
{
       if (this == &rhs)  return *this;    // Same object?
       else
       {
       try
       {
       this->min=rhs.min;
       this->max=rhs.max;
       error=rhs.error;
       }
       catch(std::exception & exc)
       {
           std:cerr << "Exception" << exc.what() << endl;
       }
       return *this;
       }
}

Now, if inside the main I do

int main (void)
{
numeric a(1,10);
numeric b(2,10);
rinterval inter(a,b,(numeric)0.00001);
inter=dynamic(inter);
cout << inter << endl;
}

I get the following error:
*** glibc detected *** ./invmeasure.exe: invalid fastbin entry (free):
0x00000000008ec340 ***
======= Backtrace: =========
/lib/libc.so.6(+0x774b6)[0x7f2c9b11a4b6]
/lib/libc.so.6(cfree+0x73)[0x7f2c9b120c83]
/usr/lib/libcln.so.6(+0x6fa0d)[0x7f2c9ad91a0d]
/usr/lib/libcln.so.6(_ZN3cln11print_floatERSoRKNS_20cl_print_float_flagsERKNS_4cl_FE+0x29)[0x7f2c9ad92d39]
/usr/lib/libginac-1.5.so.0(+0x1ddb5b)[0x7f2c9bfbfb5b]
/usr/lib/libginac-1.5.so.0(_ZNK5GiNaC7numeric13print_numericERKNS_13print_contextEPKcS5_S5_S5_j+0x97)[0x7f2c9bfc0857]
/usr/lib/libginac-1.5.so.0(_ZNK5GiNaC7numeric8do_printERKNS_13print_contextEj+0x28)[0x7f2c9bfc0e48]
/usr/lib/libginac-1.5.so.0(_ZNK5GiNaC5basic14print_dispatchERKNS_10class_infoINS_24registered_class_optionsEEERKNS_13print_contextEj+0x18b)[0x7f2c9be6c62b]
/usr/lib/libginac-1.5.so.0(_ZN5GiNaClsERSoRKNS_2exE+0x9e)[0x7f2c9bfe5c2e]
./invmeasure.exe[0x404261]
./invmeasure.exe[0x403808]
/lib/libc.so.6(__libc_start_main+0xfe)[0x7f2c9b0c1d8e]
./invmeasure.exe[0x403439]
======= Memory map: ========
00400000-00407000 r-xp 00000000 08:07 66333
/home/isaia/Dropbox/HTGFL/GN-Comput_inv_1D/codice/invmeasure.exe
00607000-00608000 r--p 00007000 08:07 66333
/home/isaia/Dropbox/HTGFL/GN-Comput_inv_1D/codice/invmeasure.exe
00608000-00609000 rw-p 00008000 08:07 66333
/home/isaia/Dropbox/HTGFL/GN-Comput_inv_1D/codice/invmeasure.exe
008c7000-00909000 rw-p 00000000 00:00 0
[heap]
7f2c94000000-7f2c94021000 rw-p 00000000 00:00 0
7f2c94021000-7f2c98000000 ---p 00000000 00:00 0
7f2c9a8be000-7f2c9a91d000 r-xp 00000000 08:07 1523786
/usr/lib/libgmp.so.3.5.2
7f2c9a91d000-7f2c9ab1c000 ---p 0005f000 08:07 1523786
/usr/lib/libgmp.so.3.5.2
7f2c9ab1c000-7f2c9ab1d000 r--p 0005e000 08:07 1523786
/usr/lib/libgmp.so.3.5.2
7f2c9ab1d000-7f2c9ab1e000 rw-p 0005f000 08:07 1523786
/usr/lib/libgmp.so.3.5.2
7f2c9ab1e000-7f2c9ab20000 r-xp 00000000 08:07 262984
/lib/libdl-2.12.1.so
7f2c9ab20000-7f2c9ad20000 ---p 00002000 08:07 262984
/lib/libdl-2.12.1.so
7f2c9ad20000-7f2c9ad21000 r--p 00002000 08:07 262984
/lib/libdl-2.12.1.so
7f2c9ad21000-7f2c9ad22000 rw-p 00003000 08:07 262984
/lib/libdl-2.12.1.so
7f2c9ad22000-7f2c9ae98000 r-xp 00000000 08:07 1549774
/usr/lib/libcln.so.6.0.1
7f2c9ae98000-7f2c9b097000 ---p 00176000 08:07 1549774
/usr/lib/libcln.so.6.0.1
7f2c9b097000-7f2c9b09b000 r--p 00175000 08:07 1549774
/usr/lib/libcln.so.6.0.1
7f2c9b09b000-7f2c9b0a1000 rw-p 00179000 08:07 1549774
/usr/lib/libcln.so.6.0.1
7f2c9b0a1000-7f2c9b0a3000 rw-p 00000000 00:00 0
7f2c9b0a3000-7f2c9b21d000 r-xp 00000000 08:07 262821
/lib/libc-2.12.1.so
7f2c9b21d000-7f2c9b41c000 ---p 0017a000 08:07 262821
/lib/libc-2.12.1.so
7f2c9b41c000-7f2c9b420000 r--p 00179000 08:07 262821
/lib/libc-2.12.1.so
7f2c9b420000-7f2c9b421000 rw-p 0017d000 08:07 262821
/lib/libc-2.12.1.so
7f2c9b421000-7f2c9b426000 rw-p 00000000 00:00 0
7f2c9b426000-7f2c9b43b000 r-xp 00000000 08:07 262222
/lib/libgcc_s.so.1
7f2c9b43b000-7f2c9b63a000 ---p 00015000 08:07 262222
/lib/libgcc_s.so.1
7f2c9b63a000-7f2c9b63b000 r--p 00014000 08:07 262222
/lib/libgcc_s.so.1
7f2c9b63b000-7f2c9b63c000 rw-p 00015000 08:07 262222
/lib/libgcc_s.so.1
7f2c9b63c000-7f2c9b6be000 r-xp 00000000 08:07 262828
/lib/libm-2.12.1.so
7f2c9b6be000-7f2c9b8bd000 ---p 00082000 08:07 262828
/lib/libm-2.12.1.so
7f2c9b8bd000-7f2c9b8be000 r--p 00081000 08:07 262828
/lib/libm-2.12.1.so
7f2c9b8be000-7f2c9b8bf000 rw-p 00082000 08:07 262828
/lib/libm-2.12.1.so
7f2c9b8bf000-7f2c9b9a7000 r-xp 00000000 08:07 1523946
/usr/lib/libstdc++.so.6.0.14
7f2c9b9a7000-7f2c9bba6000 ---p 000e8000 08:07 1523946
/usr/lib/libstdc++.so.6.0.14
7f2c9bba6000-7f2c9bbae000 r--p 000e7000 08:07 1523946
/usr/lib/libstdc++.so.6.0.14
7f2c9bbae000-7f2c9bbb0000 rw-p 000ef000 08:07 1523946
/usr/lib/libstdc++.so.6.0.14
7f2c9bbb0000-7f2c9bbc5000 rw-p 00000000 00:00 0
7f2c9bbc5000-7f2c9bbdd000 r-xp 00000000 08:07 262823
/lib/libpthread-2.12.1.so
7f2c9bbdd000-7f2c9bddc000 ---p 00018000 08:07 262823
/lib/libpthread-2.12.1.so
7f2c9bddc000-7f2c9bddd000 r--p 00017000 08:07 262823
/lib/libpthread-2.12.1.so
7f2c9bddd000-7f2c9bdde000 rw-p 00018000 08:07 262823
/lib/libpthread-2.12.1.so
7f2c9bdde000-7f2c9bde2000 rw-p 00000000 00:00 0
7f2c9bde2000-7f2c9c0c1000 r-xp 00000000 08:07 1524595
/usr/lib/libginac-1.5.so.0.2.0
7f2c9c0c1000-7f2c9c2c0000 ---p 002df000 08:07 1524595
/usr/lib/libginac-1.5.so.0.2.0
7f2c9c2c0000-7f2c9c2cc000 r--p 002de000 08:07 1524595
/usr/lib/libginac-1.5.so.0.2.0
7f2c9c2cc000-7f2c9c2cd000 rw-p 002ea000 08:07 1524595
/usr/lib/libginac-1.5.so.0.2.0
7f2c9c2cd000-7f2c9c2cf000 rw-p 00000000 00:00 0
7f2c9c2cf000-7f2c9c2ef000 r-xp 00000000 08:07 262975
/lib/ld-2.12.1.so
7f2c9c4c3000-7f2c9c4ca000 rw-p 00000000 00:00 0
7f2c9c4ec000-7f2c9c4ef000 rw-p 00000000 00:00 0
7f2c9c4ef000-7f2c9c4f0000 r--p 00020000 08:07 262975
/lib/ld-2.12.1.so
7f2c9c4f0000-7f2c9c4f1000 rw-p 00021000 08:07 262975
/lib/ld-2.12.1.so
7f2c9c4f1000-7f2c9c4f2000 rw-p 00000000 00:00 0
7fff794fd000-7fff7951e000 rw-p 00000000 00:00 0
[stack]
7fff795c7000-7fff795c8000 r-xp 00000000 00:00 0
[vdso]
ffffffffff600000-ffffffffff601000 r-xp 00000000 00:00 0
[vsyscall]

I don't understand why, isn't the destruction of the old instances of min
and max done by the assignment operator for the type numeric?

Thank you in advance
Isaia
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.cebix.net/pipermail/ginac-list/attachments/20110328/e099292a/attachment-0001.html>

------------------------------

Message: 2
Date: Mon, 28 Mar 2011 16:46:48 +0300
From: Alexei Sheplyakov <alexei.sheplyakov@gmail.com>
To: GiNaC discussion list <ginac-list@ginac.de>
Subject: Re: [GiNaC-list] assignment
Message-ID:
       <AANLkTimV2YQ0Ut2OEhUDbuq+Xh3Qzj96Z28A61ZsXKO8@mail.gmail.com>
Content-Type: text/plain; charset=UTF-8

Hello,

> rinterval inter(a,b,(numeric)0.00001);
                                ^^^^^^^^^^^^^^^^^^^

Casting double to GiNaC::numeric is absolutely wrong. For starters,
sizeof(numeric) != sizeof(double), numeric is reference counted, and
double is not, and so on. Just don't do this. Use the normal constructor
instead, that is

 rinterval inter(a, b, numeric(0.00001));

Best regards,
        Alexei


------------------------------

_______________________________________________
GiNaC-list mailing list
GiNaC-list@ginac.de
https://www.cebix.net/mailman/listinfo/ginac-list


End of GiNaC-list Digest, Vol 69, Issue 3
*****************************************