[tex-live] texdoc in luatex

Maarten Sneep maarten.sneep at xs4all.nl
Wed Jun 27 17:22:50 CEST 2007

On Jun 27, 2007, at 17:04, Frank Küster wrote:

> Hi Maarten,
> Did you reply to me in private on purpose?  I wouldn't mind continuing
> this on the list, feel free to quote me there

No, just the setting in my mail client (reply to instead of reply to  
all), and the setting of the mailing list setting the reply-to header  
to the sender only rather than the list.

> Maarten Sneep <maarten.sneep at xs4all.nl> wrote:
>> A similar trick could be used, I think. The trick bit is not to call
>> a plain Darwin 'Mac OS X', unless you don't care about that (i.e.
>> what do you need this information for?)
> I need it for assigning proper viewer commands: Windows has "start % 
> s",
> Debian Linux "see %s", other Unices individual calls like "gv %s".
> That's what the old texdoc script uses, and for "`uname -s` =  
> Darwin" it
> uses "open %s".

That is what I guessed. If you can test for the existence of a  
directory, then it seems that testing for the existence of a file  
should be possible as well. Why not check for the existence of /usr/ 

if [ -x /usr/bin/open -a -x ]

>> Potential directories to test for:
>> /.Spotlight-V100
>> (Mac OS X 10.4.x or later only)
>> /.Trashes
>> /Applications
>> Potential files:
>> /Desktop DB
>> /Desktop DF
>> .DS_Store
> Hm.  I don't want to restrict myself to >= 10.4.x; but the others  
> don't
> really look too safe.  Some other system might have them, too...

I guess that that is always the case. Do you want a listing of any  
particular drectory to see if I missed something?

>> If you don't care about mixing Darwin and Mac OS X, then the
>> pressence of the kernel is probably a decent choice: /mach_kernel, /
>> mach.sym (although this may catch other mach-based systems).
> I fear other systems might not have "open", have they?

Actually, I don't think that open is present on 'pure' Darwin either.


More information about the tex-live mailing list