1 |
On Sun, Jul 30, 2017 at 10:53 AM, John Covici <covici@××××××××××.com> wrote: |
2 |
> |
3 |
> Thanks. clone-depth seems not to be available, so I amnot sure whats |
4 |
> best here. I thinkI like the history, so I will see how to do a git |
5 |
> clone. I do havethe type as git in the gentoo.conf, but I don't know |
6 |
> what happened to clone-depth -- its not in the portage man page. |
7 |
> |
8 |
|
9 |
Feel free to attach the file and we can take a quick look at it. I'm |
10 |
also not sure if it only applies on the initial sync. I think the |
11 |
option has been around for a while so I'm not sure why it isn't |
12 |
working if you really do have the type set to git. |
13 |
|
14 |
And you can always browse the history from the web viewers. The |
15 |
reason that we ended up removing Changelogs is that they end up adding |
16 |
a lot of bandwidth to the sync process (and of course they take up |
17 |
space), when you only rarely look at them. Even if you occassionally |
18 |
do look at one, in order for it to be there you have thousands of them |
19 |
constantly being synced. |
20 |
|
21 |
The other challenge was that with the way rsync worked it greatly |
22 |
increases the bandwidth transferred if new entries are added to the |
23 |
top (if they were added to the bottom rsync would only transmit the |
24 |
last block of the file). However, from a convenience standpoint it is |
25 |
usually nicer to have new entries at the top. |
26 |
|
27 |
The general sense is that Changelogs represent the old way of doing |
28 |
things. Most projects have gone away from having them, or they just |
29 |
auto-generate them from git. I don't think most projects routinely |
30 |
distribute them either - they just stick them on a webpage and only |
31 |
people who care about them look at them. The linux kernel only |
32 |
includes the changes in the last release in their change logs as well |
33 |
(which is nothing more than a dump of git log). |
34 |
|
35 |
Encouraging users to use git is also a relatively positive thing, |
36 |
because this is a tool with extremely widespread use. The exact same |
37 |
commands you use to find out what is going on in the Gentoo repo will |
38 |
serve you just as well when trying to figure out what is going on in |
39 |
some other project. It also makes it trivial to do historical |
40 |
checkouts - users sometimes are looking for old portage snapshots to |
41 |
try to deal with updating outdated systems, and with git this is |
42 |
pretty trivial to accomplish. |
43 |
|
44 |
-- |
45 |
Rich |