| 1 |
-----BEGIN PGP SIGNED MESSAGE----- |
| 2 |
Hash: SHA512 |
| 3 |
|
| 4 |
Amadeusz Żołnowski: |
| 5 |
> Göktürk Yüksek <gokturk@g.o> writes: |
| 6 |
>>> + cat "${f}" >"${tmpf}" || return 1 |
| 7 |
>> Why shell redirection with cat instead of cp? both are in |
| 8 |
>> coreutils. |
| 9 |
> |
| 10 |
> I thought cp could overwrite file mode of already existing tmp |
| 11 |
> file, but actually it doesn't, so cp can be here as well. Is there |
| 12 |
> actual benefit of using cp here? |
| 13 |
> |
| 14 |
There could be some performance implications. cat will usually do |
| 15 |
slow, buffered I/O. cp tries to be smarter with allocation, i.e. it |
| 16 |
may take advantage of the btrfs specific clone to do a O(1) copy. |
| 17 |
|
| 18 |
>> Also, wouldn't the absence of 'die' cause silent breakages? |
| 19 |
> |
| 20 |
> I want to caller decide whether die or not and what error messge |
| 21 |
> to give. |
| 22 |
> |
| 23 |
Maybe that should be part of the documentation. People are more likely |
| 24 |
to assume that it will die on its own, I think. Is there a case in |
| 25 |
which you would not want to die? |
| 26 |
|
| 27 |
> Cheers, |
| 28 |
> |
| 29 |
|
| 30 |
-----BEGIN PGP SIGNATURE----- |
| 31 |
|
| 32 |
iQEcBAEBCgAGBQJXPPtVAAoJEIT4AuXAiM4zhX4H+gNivSvNvQzf3tsjmbVatGWB |
| 33 |
DLIvIxUY7h79c0YDTXnaU+4LLf7CvED26XT9QNWp297OfNpsXKWGTR5GbdG+w0ab |
| 34 |
SR490l9Uh/qn8HBKvh1hmPTEyeRHXm0ZoQ6/jrgwlz7ehSawQlzGRUPqFOHj88Fc |
| 35 |
qwWhKk1p2fc7TUnRO0SUY/xcRyZeEyUqEG3ueVYfPlTo10rC5+B20IFjnO09vwbT |
| 36 |
INV27KHqjLnpRT4xGWvCyCiEfw6DDypxZ/PKfr4c7omHF8OF+5GGnf+FFZ8UJFkt |
| 37 |
93TVIPDeUC9XeAsmqirgN5LuYWsNNUwTEdiDfjglLrTb6OgXA0eIIwXsGH/G3d4= |
| 38 |
=J3uT |
| 39 |
-----END PGP SIGNATURE----- |
Archives