Re: avoid warnings about slice vs. base name in single slice mode

Denis Corbin <[email protected]>
Newsgroups gmane.comp.sysutils.backup.dar.support
Organization http://dar.linux.free.fr/
Message-ID <[email protected]>
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA256

On 25/08/2019 16:24, Matus UHLAR - fantomas wrote:
>>>> On 31/07/2019 19:59, Matus UHLAR - fantomas wrote:
>>>>> is it possible to avoid warnings about "seems more to be a
>>>>> slice name than a base name." when slice size is not
>>>>> defined?
> 
[...].
>> 
>> OK for you? If so, please fill a feature request at 
>> https://sourceforge.net/p/dar/feature-requests/
> 
> been there, done that, thank you.
> 
> The option would be ignored if '-s' option is used (maybe with
> non-zero value, how does dar handle this case?)

The option would not concern archive creation (merging, backup,
isolation) but only archive that should exist in regard to the option
they are provided to (-A -@ -l -t -d -x, ...). It would work that way:

If the provided basename does not lead dar to find any slice name
(adding .*.dar to the basename) dar would check the following:
- - if the basename ends by a dot, it would retry the process removing
the ending dot, assuming this was the result of a shell completion
when more than one slice is present in the directory.
- - if the basnename ends by ".*.dar" dar will assume a slice name has
been provided and would retry removing ending the ".*.dar" to get the
basename.

in both case, dar will issue a warning telling the modification made
on the provided basename

after that if there is still not slice name corresponding to the
basename, dar would report the missing slices as of today.

> 
> the option would expect full file names when creating or
> reading/updating archives.

it would expect either basename or basename+'.' or basename+'.*.dar'
when providing the name of an existing archive.

But for archive creation, it would only use the provided name as
basename as of today (else this would collide with the user freedom to
define archive basename ending by a dot or '.*.dar').

> 
> autodection of slices when archive to be opened doesn't exist would
> be usefull though.
> 

the problem is: base on what to assume the user wants another thing
than the name he provided? A software can assume there is a mistake
from the user if and only if the order is not possible to execute
(missing slices for a given basename).

But here, as there is nothing yet, why to assume that the user did not
want a basename ending by a dot? Because, me as developer here an now,
I would know better than any user in the world and in any possible
context, what he wants to do?

Sorry I can't do that :-)


Regards,
Denis
-----BEGIN PGP SIGNATURE-----
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=oqa4
-----END PGP SIGNATURE-----
lmpx.com only provides a reader for public news (NNTP) servers. It is not affiliated with the servers or forums shown here and is not responsible for the content of articles, which is written by their respective authors.