1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
|
Is LSB going to be used in the ISO & the updates?<br><br><div class="gmail_quote">2010/11/8 Luca Berra <span dir="ltr"><<a href="mailto:bluca@vodka.it">bluca@vodka.it</a>></span><br><blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;">
<div class="im">On Sun, Nov 07, 2010 at 11:43:46PM +0100, Renaud MICHEL wrote:<br>
<blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;">
On dimanche 07 novembre 2010 at 22:33, Luca Berra wrote :<br>
<blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;">
> you must include all intermediate updates packages to create the delta.<br>
<br>
you just have to keep track of package 1 and n-1, not all of them<br>
</blockquote>
<br>
No, because a roll back of a file can happen on any update. If a file is changed in release n-2 and rolled back in n-1, if you only consider packages 1 n-1 and n you miss the roll back for the people who updated to n-2 but missed the update to n-1.<br>
</blockquote>
<br></div>
No, sir,<br>
<br>
If n contains n-1 this remains true at every increment of n, so you<br>
never need to look at n-2 because n-1 already contains it.<br>
<br>
But besides the teoric discussions there is no code project to do this,<br>
so all of this discussion is leading nowere.<br>
current deltarpm code works by creating binary deltas between two given<br>
rpms and, besides the fact we are not even sure it works with current<br>
lzma compressed rpms, the points raised by nanar all hold. so i dont'<br>
believe we will use deltarpm as it is.<br><font color="#888888">
<br>
L.</font><div><div></div><div class="h5"><br>
<br>
-- <br>
Luca Berra -- <a href="mailto:bluca@vodka.it" target="_blank">bluca@vodka.it</a><br>
</div></div></blockquote></div><br>
|
