diff options
Diffstat (limited to 'zarb-ml/mageia-discuss/attachments/20101109/35e2797c/attachment.html')
| -rw-r--r-- | zarb-ml/mageia-discuss/attachments/20101109/35e2797c/attachment.html | 31 |
1 files changed, 31 insertions, 0 deletions
diff --git a/zarb-ml/mageia-discuss/attachments/20101109/35e2797c/attachment.html b/zarb-ml/mageia-discuss/attachments/20101109/35e2797c/attachment.html new file mode 100644 index 000000000..a1221f9bf --- /dev/null +++ b/zarb-ml/mageia-discuss/attachments/20101109/35e2797c/attachment.html @@ -0,0 +1,31 @@ +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> |