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

Re: Important Bug Fix: xorriso-0.2.8.pl01



Hi,

> This is correct behavior from FreeBSD!

I hope the FreeBSD cd9660 folks see it differently. :)
About all other BSDs already seem to like it.
What about Solaris ?


> The "IEEE_1282" draft was withdrawn with RR-V1.10

It is not mentioned there, indeed.

> Mkisofs implements RR-V1.12

RRIP 1.12 does not mention "RRIP_1991A" any more.

>From ftp://ftp.ymi.com/pub/rockridge/rrip112.ps :
"
4.3 Specification of the ER System Use Entry Values for RRIP
[...]
The content of the Extension Identifier field shall be "IEEE_P1282".
[...]
The recommended content of the Extension Descriptor is "THE IEEE P1282
[...]
Note: Upon adoption as an IEEE standard, these lengths will each decrease by 1.
"

So we have "RRIP_1991A", "IEEE_P1282" and "IEEE_1282"
as potential signature for Rock Ridge extensions.

A rugged ISO 9660 / ECMA-119 reader should accept them all.
(It should better not draw much conclusions from them,
 though.)

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

Some observations from mkisofs images:

There is field RR. Not even RRIP-1.10 mentions it.
One has to google for old RRIP-1.09 to learn that it
was a quite useless optional flag array which announced
what other fields are to be expected.

The PX fields have length 44 as prescribed in RRIP-1.12
and not 36 as in RRIP-1.10.
But then the signature should either be "IEEE_P1282" or
"IEEE_1282" as prescribed in that document.
(Thanks to my peers i dropped the idea to flatly change
 xorriso's signature to "RRIP_1991A". I am heavily testing
 with that slightly ill setting, though.)

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

Have a nice day :)

Thomas


Reply to: