Re: Documenation update

Wiebe Cazemier wrote:
> About all the man page stuff. I think we can ignore my statements on
> it... I use(d) vim as man browser, which, I just discovered, messed up
> the entire layout. Using the normal man browser, the man page is clear
> enough.

Yep, OK. And we have now solutions if dar/libdar receive even more 
options to keep the man documentation readable ;-)

Regards,
Denis.


> 
> 
> Denis Corbin wrote:
> 
> 
>>I can now see something interesting to do about man pages :
>>
>>dar.1 -> general presentation of the command line tool
>>dar_filtering.1 -> all that concerns the filtering -am, -I -X, -g, -P,
>>-[, -] -u, -U, -ac, -an
>>dar_ea -> all that concerns EA
>>dar_create.1 -> all that concerns archive creation (differential/full
>>backup/snapshot)
>>dar_isolate.1 -> all that concerns catalogue isolation
>>dar_merge.1 -> all that concerns archive merging
>>dar_test.1 -> all that concerns testing
>>dar_diff.1 -> all that concerns diffing
>>dar_list.1 -> all that concerns listing (normal,tree-like,XML listings)
>>
>>with cross references between theses pages. 
> 
> 
> No "extraction" man page?
> 
> But I'm not sure I like this approach. For example, you'll have issues
> with where to put options like --alter=atime, which are for create and
> compare. With this approach, you run the risk of not putting something
> where it is expected. A user may end up not reading an important option,
> because he's in the wrong manpage, which could make his backup unusable
> without his knowledge, or mess up his system (the --alter=ctime comes to
> mind...)
> 
> My discovery about vim messing up the layout negates any problems I had
> with the man page.
> 
> 
> Denis Corbin wrote:
> 
> 
>>the -am description refers to a section above in the man page about
>>file filtering, where is described two ways of building a filter based
>>on -I -X, -P, -g -[ and -] options. Does this paragraph is not correct
>>or could not you find the reference given in the option description
>>(in that case a separated man page would be more interesting)? 
> 
> 
> With the help of Wesley's elaborate explanation, I'll try to figure the
> option out, and come up with some documentation about it.
> 
> 
> Wesley Leggette wrote
> 
> 
>>I don't know how Denis wants to handle it, but it is sometimes
>>advantageous to provide patches. Do you know how to do this?
>>
> 
> 
> Denis Corbin wrote:
> 
> 
>>Definitively yes. This would avoid me forgetting treating tasks and
>>eases the review of changes. As I cannot treat patches as soon as they
>>are available, using the sourceforge tracker lets the possibility to
>>other users to know about pending patches and let users applying them
>>while they are not yet integrated into CVS.
>>
>>But that's OK to post here (as we have asked you to do so :-) )
> 
> 
> I have no problem with writing patches. I've checked out the source and
> will make patches. Do you want me to submit the sections I've already
> posted here to the patch tracker?
> 
> Using CVS instead of SVN is new to me, so if I run into problems, I'll
> ask :)
> 
> Wesley, I think I covered/used everything you said in your message. I
> may get back to the explanation of --alter=mask when I get to rewriting
> that section of the manpage. I hope you don't feel too much ignored :).

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.6 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFEJvPspC5CI8gYGlIRAo0wAKDKxiJj2Mbmn6irtUEQkOpaE8HCtACfZAJw
KoBLH2UIBQULhBDCAUcM4Y8=
=ZrqX
-----END PGP SIGNATURE-----