On Sat, Sep 17, 2016 at 05:38:26PM +0800, Alex Vong wrote:
     Hi,
     
     John Darrington <john@darrington.wattle.id.au> writes:
     
     > They need more than that.  They need the binary to be installed
     > setuid.  So these
     > cannot be run as package tests.  They must be tested after the
     > as-yet-to-be-written
     > service is installed.  But I'll update the comment to make it more clear.
     >
     I see, so this is a circular dependency problem here. The service
     depends on this package, but the test depends on the service. Am I
     right?

More or less :)  
     
     >      > +       #:phases
     >      > +       (modify-phases %standard-phases
     >      > +         (add-before 'build 'replace-/bin/sh
     >      > +           (lambda _
     >      > +             (substitute*
     >      > +                 (append
     >      > + (list "smrsh/smrsh.c" "sendmail/conf.c" "contrib/mailprio"
     >      > +                        "contrib/mmuegel" "devtools/bin/configure.sh")
     >      > +                  (find-files "." ".*\\.m4")
     >      > +                  (find-files "." ".*\\.cf"))
     >      I think this can be simplified to:
     >      `("smrsh/smrsh.c" "sendmail/conf.c"
     >         "contrib/mailprio" "contrib/mmuegel"
     >         "devtools/bin/configure.sh"
     >         .@(find-files "." ".*\\.m4")
     >         ,@(find-files "." ".*\\.cf"))
     >      using the quasi-quote quasi-unquote-splicing notation, which is similar
     >      to string interpolation in shell "foo bar $(CAR) $(TAR)".
     >
     > Is that simpler?  It has more characters?
     >
     Hmmm, it is more of a style thing. You can ask how others think about
     it.

I don't have any strong opinions on either style.
     
     >      > +                        "contrib/mmuegel" "devtools/bin/configure.sh")
     >      > +               (("/bin/sh") (which "bash")))
     >      > +
     >      > +             (substitute* "devtools/bin/Build"
     >      > + (("SHELL=/bin/sh") (string-append "SHELL=" (which "bash"))))
     >      > +             #t))
     >      I think the `#t' is not neccessary here, since `substitute*' uses
     >      `substitute', which will either return #t or throw an exception.
     >
     > WTF??  Didn't you complain earlier this week when I *didn't* put #t in
     > exactly this
     > scenario??
     >
     Yes, I am a different Alex :)
     Also, it seems we are not being consistent here, sometimes we put `#t'
     after `substitute*', sometimes we don't. Anyone has an idea?

I did raise some suggestions in my earlier posts.  But again I don't have any strong
opinion.
     
     > Would that make any difference? Since only the last expression would
     > get returned from lambda ?
     >      
     You are right. The following should chain things up correctly:
     
     (and (zero? (system* "sh" "Build"))
          (with-directory-excursion "cf/cf"
            (begin
              (copy-file "generic-linux.mc" "sendmail.mc")
              (zero? (system* "sh" "Build" "sendmail.cf")))))

You're right.  I'll do it that way.
     

Thanks,

J'

-- 
Avoid eavesdropping.  Send strong encrypted email.
PGP Public key ID: 1024D/2DE827B3 
fingerprint = 8797 A26D 0854 2EAB 0285  A290 8A67 719C 2DE8 27B3
See http://sks-keyservers.net or any PGP keyserver for public key.