in reply to Re^6: dmake and bash on Win?
in thread let Makefile.PL to do the Readme file for me -- new target?

I was referring to skip "rm -rf cover_db"

Okay, that comes from my distro's Makefile.PL lines 124-129:

# addition to realclean: also delete coverage database, if testcover w +as run realclean :: $(NOECHO) ( $(TEST_D) cover_db && $(RM_RF) cover_db ) || $(ECHO) r +ealclean:: skip "rm -rf cover_db" $(NOECHO) ( $(TEST_F) LICENSE.pod && $(RM_RF) LICENSE.pod ) || $(E +CHO) realclean:: LICENSE.pod intermediary $(NOECHO) ( $(TEST_F) README.pod && $(RM_RF) README.pod ) || $(ECH +O) realclean:: README.pod intermediary $(NOECHO) ( $(TEST_F) MYMETA.json.lock && $(RM_RF) MYMETA.json.loc +k ) || $(ECHO) realclean:: MYMETA.json.lock temp file

It is using a variety of macros:

By using the macro variables that MakeMaker defines in the Makefile instead of literally using common UNIX-style commands, you can write rules that are much less dependent upon the system you are on, the availability of bash, etc., and will work cross-platform but will still look and behave like the linux commands that many are used to using.

Again, I recommend looking through the Makefile that perl Makefile.PL generates in your build environment; using that, you can see what macros are available to you, and how they are implemented on your system.

Replies are listed 'Best First'.
Re^8: dmake and bash on Win?
by LanX (Saint) on Jan 21, 2021 at 18:14 UTC
    Are you saying that the literal "rm -rf cover_db" is never executed and just used as comment or message?

    Cheers Rolf
    (addicted to the Perl Programming Language :)
    Wikisyntax for the Monastery

      Are you saying that the literal "rm -rf cover_db" is never executed and just used as comment or message?

      Yes, that is what I am saying.

      When the Makefile runs the command $(ECHO) realclean:: skip "rm -rf cover_db", it effectively prints all the arguments after the $(ECHO) as text to the console/terminal/equivalent. I am just using that as a debug print to tell me that it didn't delete cover_db (probably because it didn't exist).