[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[UDUNITS #MGE-245661]: perl-package for udunits2? debian packages?

This archive contains answers to questions sent to Unidata support through mid-2025. Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated. For the most up-to-date information on the use of NSF Unidata software and data services, please consult the Software Documentation first.

  • Subject: [UDUNITS #MGE-245661]: perl-package for udunits2? debian packages?
  • Date: Fri, 31 Jul 2009 10:20:07 -0600
Heiko,

> while working with the perl-udunits2 package, I recognized the following;
> 
> • ut_free_system: segmentation-fault when freeing same system* twice

That's not a bug.  It's like passing a pointer to the function free() twice, 
resulting in undefined behavior.  The description of the ut_free_function() 
says this:

    Use of [the pointer argument] after this function returns results in 
undefined behavior.

The documentation of the free() function says something similar.

You should modify your code so that it calls ut_free_system() only once for 
each pointer.  Alternatively, set the value of the pointer to NULL upon return 
from the ut_free_system() function.

> • several errors in definition files, e.g. Definition of "microns" in
> "/home/heikok/Programme/Perl/Udunits/share/udunits2-common.xml", line
> 296, overrides prefixed-unit "1e-15 second"

Those messages are warnings: they don't indicate an error.  Before the parser 
encountered the "micron" unit, it would have parsed the string "microns" as 
"micro nanoseconds" or "1e-15 seconds".  The parser is letting you know that 
this will no longer happen.  You can disable these warning messages via the 
function ut_handle_error_message() if you wish.

> • chapter 5 in c-api doc (ut_parse) should be merged with 3.2 (ut_get_unit*)

I wondered about that.  The ut_parse() and ut_get_unit() functions are 
different, however.  The ut_parse() function accepts arbitrary string unit 
specifications, which includes prefixes and mathematical operations; whereas, 
the ut_get_unit() function requires a simple, single unit specification -- no 
prefixes or operations allowed.

Maybe I should switch chapters 4 and 5 so that the chapter on parsing an 
arbitrary unit specification would follow the chapter on getting a particular 
unit from a unit system.  What do you think?

> Heiko

Regards,
Steve Emmerson

Ticket Details
===================
Ticket ID: MGE-245661
Department: Support UDUNITS
Priority: Normal
Status: Closed