Hi Eric, Le vendredi 08 novembre 2013 à 08:11 -0700, Eric Blake a écrit : > On 11/07/2013 11:14 AM, Benjamin Cama wrote: > > diff --git a/gnulib-tool b/gnulib-tool > > index 669931c..55fbee3 100755 > > --- a/gnulib-tool > > +++ b/gnulib-tool > > @@ -2954,7 +2954,7 @@ ba > > fi > > done > > # Add the dummy module, to make sure the library will be non-empty. > > - if test -z "$have_lib_SOURCES"; then > > + if test -n "$have_lib_SOURCES"; then > > if func_acceptable "dummy"; then > > func_append modules " dummy" > > Huh? The existing code seems correct - if there were no other > lib_SOURCES (the variable is empty), then we need to add dummy so that > we have at least one lib source. Your patch would change it so that if > there are lib_SOURCES, then we also include dummy. I think I need a bit > more context why your patch would be correct.
Mmmhh, so I must have misunderstood the code. I first thought that its role was to add a kind of required “dummy” lib if any library code is included, otherwise building no library. It is not the case, on the contrary, it is there so that there always is some library code. Sorry for my mistake. My goal is to avoid building any library at all when I am only using gnulib for some utils like gitlog-to-changelog or git-version-gen (furthermore when I don't even check for a compiler at all; my project is shell scripting only). I thought that this was the purpose of this test, but i was mistaken. Is there a way to do what I want? I don't think so. Do you think this would be a worthwhile possibility? Or should I just import those modules' code into my VCS and drop lib/ & m4/ entirely? I juste liked the way gnulib can import these script without them being version-controlled in my project. Regards, -- Benjamin Cama <benjamin.c...@telecom-bretagne.eu>
