Re: Why info files need to be named info-*.gz?
Jörg Sommer <joerg@alea.gnuu.de> writes:
> Russ Allbery <rra@debian.org> wrote:
>> Jörg Sommer <joerg@alea.gnuu.de> writes:
>>> They are compressed and end in .gz.
>>> % ls /usr/share/info/jed.*
>>> /usr/share/info/jed.1in.gz /usr/share/info/jed.3in.gz
>>> /usr/share/info/jed.2in.gz /usr/share/info/jed.info.gz
>>> The lintian check looks so:
>>> unless ($infoext && $infoext =~ /info(-\d)?/) { # it's not foo.info
>>> unless (!@fname_pieces) { # it's not foo{,-{1,2,3,...}}
>>> tag "info-document-has-wrong-extension", "$file";
>>> }
>>> }
>> Oh, that. I don't know, actually; it's been in lintian from long
>> before I started working on it. Have you checked that all info readers
>> (info, pinfo, and C-h i from inside Emacs) can find those info files
>> correctly and understand their sequencing when doing searches from
>> inside the info reader?
> I don't why you question this. The main document in named properly
> (jed.info.gz) and the other files are referenced there by there name.
I question this because someone at some point went to the work of writing
the test into lintian and usually people don't do things for no reason at
all. Since I have no idea why that test is there, I worry about just
removing it without some verification that it isn't going to break
anything.
However, it is certainly a good point that the info file format contains
the references to the other files. So off-hand, I can't think of any
reason why this should be required, and therefore can't think of any
reason why this test should be here. That makes me nervous, but I suppose
I can remove it and see if we rediscover why it was there later.
--
Russ Allbery (rra@debian.org) <http://www.eyrie.org/~eagle/>
Reply to: