On Nov 22, 2007, at 6:26 PM, Anders F Björklund wrote:

> Jeff Johnson wrote:
>
>> I'm still debating xar <-> rpmio Payload compression, as well as  
>> whether
>> rpmio should swallow libxar, or vice versa. There's some twists  
>> and turns
>> in the code paths I'd like to see, but that's likely just my  
>> fetishes.
>>
>> I'll hack out something this weekend, the problem is more theoretical
>> than what is needed soonest.
>
> There's a couple of WITH_XAR define shields missing for the rpmxar
> functions, in order to make %standalone compile (i.e. without xar)
>

Yah, no worries. I will make sure that those who insist on ripping
out one of the two primary ROADMAP goals (i.e. support for XAR format)
in rpm-5.0 are happy! happy! happy!

> Not sure if anything in rpmwf is actually needed without xar ?
> At least it didn't complain when I conditionalized all of it...
>

Nod. rpmwf is likely history, now that I have rpmxar.c drilled into  
rpmio,
rpmwf will be used by txar.c and nothing else.

What remains in rpmxar hacking is to control for (and maintain  
streaming)
the iteration over the "magic" XAR archive file names
     Lead
     Signature
     Header
     Payload

At that point, the generalization to any archive format, not just  
XAR, becomes
quite doable. In fact, I may do tar/cpio archives as well as XAR in  
the next week,
actually easier than XAR because there's no predefined compression  
buried
behind the archive format.

Hint: *.deb uses AR format with "magic" file names, fundamentally no  
different
(and easier) than what rpm-5.0 has (almost) implemented with the  
"Leopard-like" XAR
wrapper format.

Debian Borg beware ;-) And I'm going to be ecstatic sending Joey Hess  
new
bulltet items for his rpm vs. apt comparison page. When the time  
comes of course ...

73 de Jeff