Re: [icu-support] Memory leaks in icu3.6

[Kumar P. <kpr...@ad...>]

Hi George,

Thanks for the reply! Somehow, I overlooked this issue in the test code =
itself :-(. Will take extra care before sending issues to the mailing =
list.


Thanks,
Pratyush

-----Original Message-----
From: icu...@li... =
[mailto:icu...@li...] On Behalf Of George =
Rhoten
Sent: Tuesday, January 29, 2008 10:55 PM
To: ICU support mailing list
Subject: Re: [icu-support] Memory leaks in icu3.6

This isn't Java, which has a garbage collector ;-)  I believe adding =
these=20
lines to your code should fix the problem.

void Test_NumberFormat()
....
    delete numberFormat;
....
}

void Test_tranlist()
....
    delete translit;
....
}

George Rhoten
IBM Globalization Center of Competency/ICU  San Jos=E9, CA, USA
http://www.icu-project.org/



"Kumar Pratyush" <kpr...@ad...>=20
Sent by: icu...@li...
01/29/2008 12:32 AM
Please respond to
ICU support mailing list <icu...@li...>


To
"ICU support mailing list" <icu...@li...>
cc

Subject
[icu-support] Memory leaks in icu3.6






Hi,
=20
I tried following sample code with DevPartner (for VC8) and found the=20
memory leaks.=20
=20
=20
void Test_NumberFormat()
{
      UErrorCode err =3D U_ZERO_ERROR;
      UnicodeString s;
      double d =3D 1234.56;
=20
      NumberFormat* numberFormat =3D NumberFormat::createInstance(err);
      numberFormat->format(d, s);
      UChar const* buf =3D s.getTerminatedBuffer();
=20
      char const* localeID =3D =
numberFormat->getLocaleID(ULOC_VALID_LOCALE,=20
err);
      u_cleanup();
}
=20
Call stack shown is -
=20
Function - File - Line no
=20
Uprv_malloc_3_6 - cmemory.c - 45
UMemory:operator new - uobject.cpp - 60
NumberFormat:makeInstance - numfmt.cpp - 877
NumberFormat:createInstance - numfmt.cpp - 663
NumberFormat:createInstance - numfmt.cpp - 387
=20
Another stack trace is -=20
=20
Uprv_malloc_3_6 - cmemory.c - 45
UMemory:operator new - uobject.cpp - 60
NumberFormat:makeInstance - numfmt.cpp - 893
NumberFormat:createInstance - numfmt.cpp - 663
NumberFormat:createInstance - numfmt.cpp - 387
=20
=20
Another leak that I found is using the code -=20
=20
void Test_tranlist()
{
      UErrorCode err =3D U_ZERO_ERROR;
=20
      int32_t const maxIdNum =3D Transliterator::countAvailableIDs();
=20
      UnicodeString const* translitId =3D=20
&Transliterator::getAvailableID(maxIdNum - 1);
=20
      Transliterator* const translit =3D=20
Transliterator::createInstance(*translitId, UTRANS_FORWARD, err);
      u_cleanup();
}
=20
Call stack shown in this case is -
=20
uprv_mallo_3_6 - cmemory.c - 45
uhash_allocate - uhash.c - 222
uhash_init - uhash.c - 265
uhash_create - uhash.c - 290
uhash_open_3_6 - uhash.c - 537
=20
=20
I found this leak while running with ICU3.4 as well. Are these known=20
problems?=20
=20
=20
Thanks,
Pratyush
 =
-------------------------------------------------------------------------=

This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
icu-support mailing list - icu...@li...
To Un/Subscribe: =
https://lists.sourceforge.net/lists/listinfo/icu-support


-------------------------------------------------------------------------=

This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
icu-support mailing list - icu...@li...
To Un/Subscribe: =
https://lists.sourceforge.net/lists/listinfo/icu-support