| 1 |
>>>>> "UM" == Ulrich Mueller <ulm@g.o> writes: |
| 2 |
|
| 3 |
UM> If we take the second route, then maybe it should be a more general |
| 4 |
UM> solution, i.e. exclude all tiny files (man page or not) from |
| 5 |
UM> compression? |
| 6 |
|
| 7 |
First, from a user’s perspective, not compressing small files is a good |
| 8 |
thing. Man pages perhaps most of all, given makewhatis, et al. |
| 9 |
(Think of all the C₁₂ which won’t be un-sequestered quite so soon. ☺ ;^) |
| 10 |
|
| 11 |
Ideally, there would be some way to configure, per filesystem and/or per |
| 12 |
directory, what constitutes a small file. If the fs uses fixed-size |
| 13 |
blocks then anything already smaller than one block needn’t be compressed. |
| 14 |
OTOH, if the fs supports partial block file packing, then a smaller |
| 15 |
threshold may be better. |
| 16 |
|
| 17 |
Even for large files, if the compression fails to save any blocks then |
| 18 |
it may be better to leave it uncompressed. |
| 19 |
|
| 20 |
That said, some backup strategies may be better served by compressing |
| 21 |
all but the smallest files. |
| 22 |
|
| 23 |
Good heuristics for the default compress-or-don’t threshold should cover |
| 24 |
most systems, but the ability to easily override the default is desirable. |
| 25 |
|
| 26 |
-JimC |
| 27 |
-- |
| 28 |
James Cloos <cloos@×××××××.com> OpenPGP: 1024D/ED7DAEA6 |
Archives