1 |
Il 12/11/18 16:40, Michał Górny ha scritto: |
2 |
> On Mon, 2018-11-12 at 01:21 +0100, Francesco Riosa wrote: |
3 |
>> Il giorno dom 11 nov 2018 alle ore 22:17 Michał Górny <mgorny@g.o> |
4 |
>> ha scritto: |
5 |
>> |
6 |
>>> On Sun, 2018-11-11 at 21:53 +0100, Michał Górny wrote: |
7 |
>>> [...-] |
8 |
>>> Of course, one option would be to use ZIP ;-). |
9 |
> I wasn't serious there. |
10 |
> |
11 |
>> Zip archives have another big advantage; there is an index of files, so |
12 |
>> listing the archive contents and extracting a single file is very fast and |
13 |
>> does not depend from it's position in the archive. |
14 |
>> The big disadvantage is that only "desktop" profile has unzip by default |
15 |
>> |
16 |
> The two main problems with ZIP is that: |
17 |
> |
18 |
> 1. As you noted, it's not present in core system packages. |
19 |
> |
20 |
> 2. It uses trailer format which means that you need to fetch the whole |
21 |
> file before being able to process it. |
22 |
|
23 |
Well, with some protocols (HTTP-1.1) and with a well behaving server |
24 |
(content-length) this is doable, but limited indeed. |
25 |
However the same is true for the tarball if the order of the contained |
26 |
files is uncertain. |
27 |
|
28 |
Since a tar file is sequential even to know contained files you need to |
29 |
seek into the tarball and the number of seeks is dependant from the |
30 |
number of files included in the tarball before the wanted one. |
31 |
|
32 |
|
33 |
> |
34 |
> There was also some patent hassle back in the day but I think it's no |
35 |
> longer applicable today. |
36 |
> |