<div dir="ltr"><div dir="ltr"><br></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Wed, 6 Mar 2024 at 04:12, Reinhard Kotucha <<a href="mailto:reinhard.kotucha@gmx.de">reinhard.kotucha@gmx.de</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">On 2024-03-06 at 00:21:07 +0000, David Carlisle wrote:<br>
<br>
> If openin_any = p is in force the calling tex system wouldn't be able to<br>
> > read .foo but (currently at least) it shows in the directory listing.<br>
><br>
> I think that's compatible with openin_any=p (which actually is in force<br>
> above) otherwise we would (I think) have to add a call to<br>
><br>
> kpsewhich -var-value 'openin_any'<br>
><br>
> to disable --all in that case.<br>
><br>
> or using the lua interface rather than calling the command<br>
><br>
> kpse.set_program_name("kpsewhich")<br>
> if (kpse.var_value('open_in') ~= 'a')<br>
> then<br>
> is_all=false<br>
> end<br>
<br>
The latter variant is preferred. Using the kpse library is much more<br>
efficient than forking sub-processes.<br>
<br></blockquote><div><br></div><div>Yes but the question is mostly do we need to do that at all?</div><div><br></div><div>If openin_any is not "a" then clearly you can't read dot files but it isn't clear if they </div><div>have to be excluded from a directory listing.</div><div><br></div><div>It's no problem to implement that (as discussed the cost of an extra kpse initialisation)</div><div>but that's not the interpretation I would naturally have from</div><div><br></div><div>% r (restricted) : disallow opening dot files</div><div><br></div><div><br></div><div>David</div><div><br></div><div><br></div><div> <br></div></div></div>