1 |
On Monday, September 13th, 2021 at 18:04, Rich Freeman <rich0@g.o> wrote: |
2 |
> |
3 |
> On Mon, Sep 13, 2021 at 5:02 PM Michał Górny <mgorny@g.o> wrote: |
4 |
> > |
5 |
> > On Mon, 2021-09-13 at 12:08 +0200, Ulrich Mueller wrote: |
6 |
> > > |
7 |
> > > Also, IIRC one of the goals of the format was to allow partial |
8 |
> > > download |
9 |
> > > of metadata. That will only work if the Manifest file will be the |
10 |
> > > first |
11 |
> > > file in the archive (or at least appear before the image archive). |
12 |
> > |
13 |
> > I disagree. This is solved by having detached metadata signature -- you |
14 |
> > can do a partial fetch and verify the metadata directly. |
15 |
> > |
16 |
> |
17 |
> Another option I've tossed out there in the past is having a content |
18 |
> hash of the metadata and putting that in the filename. That obviously |
19 |
> won't tell you anything about the contents of the file without reading |
20 |
> it, but if you're looking for a file with specific metadata you could |
21 |
> predict its filename. This was intended to work with having multiple |
22 |
> hashes for the same file using subsets of the metadata, using symbolic |
23 |
> links. |
24 |
> |
25 |
> The thinking here is that you'd just hash a subset of metadata useful |
26 |
> for identifying what file you'd want to download, such as CHOST, |
27 |
> linked dependency versions, use flags, etc. You'd probably hash it |
28 |
> with/without stuff like use flags so that you could either take a shot |
29 |
> at getting the file exactly configured how you want, or accepting a |
30 |
> version with any set of flags. |
31 |
> |
32 |
> Of course, this idea goes in direct opposition to your statement about |
33 |
> not wanting to specify the filename. I get that argument. The intent |
34 |
> here was to allow portage to go hunting through trusted repositories |
35 |
> to find packages it can use without having to sync a lot of data - if |
36 |
> you know the exact filename then a simple GET tells you if it is there |
37 |
> or not. |
38 |
|
39 |
Interesting concept, although this should be counted in the |
40 |
binpkg-multi-instance. A predictable configuration hash, rather than |
41 |
relying on index to get the difference between variants. |
42 |
|
43 |
Something like: |
44 |
bar/foo-1.0-r2-e3b0c44298fc1c149afbf4c8996fb9.gpkg.tar |
45 |
|
46 |
Thanks, |
47 |
Sheng Yu |