Wireshark-dev: Re: [Wireshark-dev] [Wireshark-commits] rev 39559: /trunk/epan/ /trunk/epan/: pr
Date Index · Thread Index · Other Months · All Mailing Lists
Date Prev · Date Next · Thread Prev · Thread Next
From: Jeff Morriss <jeff.morriss.ws@xxxxxxxxx>
Date: Tue, 25 Oct 2011 14:43:25 -0400
Guy Harris wrote:

On Oct 25, 2011, at 10:49 AM, Stig Bjørlykke wrote:


But now I don't understand the question...


Sorry, I misread the checkin comment

	Allow signed integers displayed as DEC_HEX.

as

	Allow signed integers displayed as BASE_HEX.

It might have been a bit harder to misread if it were

	Allow signed integers displayed as BASE_DEC_HEX.

We should probably allow BASE_HEX_DEC as well.


That would, as per this comment:


                /*      Hexadecimal and octal are, in printf() and everywhere
                 *      else, unsigned so don't allow dissectors to register a
                 *      signed field to be displayed unsigned.  (Else how would
                 *      we display negative values?)
                 *
                 *      If you want to take out this check, be sure to fix
                 *      hfinfo_numeric_format() so that it does not assert out
                 *      when trying to construct a hexadecimal representation of
                 *      FT_INT*.
                 */
also require modifying hfinfo_numeric_format() to be able to generate representations of negative hexadecimal numbers (so as to prevent "Apply as filter" on a signed BASE_HEX_DEC field whose decimal value is negative from causing an abort). 

If that's done we may as well let in BASE_HEX and BASE_OCT too.
This stuff came in rev 24643/bug 1539. (See the later comments in the bug for at least one developer who wasn't happy about these checks being put in.)