I'm trying to get an INTERCAL compiler running in order to be able, at least one day in my life, to feel what is a really terrific powerful computing tool.
I feel that it could harm someone orother if I was really about to -say- use it in order to hack a mainframe in my company, so I promise I won't say it's me^W^W^W.
By the way, it looks like I can't hack anything at all, as i^HIck allways claims there's "NO SKELETON IN MY CLOSET, WOE IS ME".
What^WPlease, does anyone figure out what it could be about ? A wrong or missing compilation option in bison ?
On 2009-11-16, Herve Autret <aut...@alussinan.org> wrote:
> By the way, it looks like I can't hack anything at all, as i^HIck allways > claims there's "NO SKELETON IN MY CLOSET, WOE IS ME".
That means exactly what it says. Ick needs a skeleton[1] to compile your program, and after looking all over the place (including in the closet[0]) it still cannot find it.
If you built ick but did not install it, you probably need to tell ick where the closet it by setting ICKDATADIR to some suitable directory[-1], preferably one which contains "ick-wrap.c".
I'm sure AIS will have a better answer, but hope this helps a bit.
CLC
[-2] this footnote is unreferenced [-1] how you do that depends on your operating system / shell / philosophy [0] that would be the dark hidden place where C-INTERCAL keeps its secrets [1] a file containing bits to feed to your C compiler. C compilers are known to eat all sorts of weird things.
... Install ? I'll consider installing after 2-3 things else fail.
> you probably need to [set] ICKDATADIR to some suitable directory[-1], > preferably one which contains "ick-wrap.c".
It didn't do anything. So I installed it.
Wow ! Now I'm told "A SOURCE IS A SOURCE, OF COURSE, OF COURSE". Okay okay, I had not to try writing my own INTERCAL source codes before I knew anything about it.
> hope this helps a bit.
Thanks a lot. I'll be back as soon as I feel I've a clever question.
> C compilers are > known to eat all sorts of weird things.
> I'm trying to get an INTERCAL compiler running in order to be able, at > least one day in my life, to feel what is a really terrific powerful > computing tool.
> I feel that it could harm someone orother if I was really about to -say- > use it in order to hack a mainframe in my company, so I promise I won't > say it's me^W^W^W.
> By the way, it looks like I can't hack anything at all, as i^HIck allways > claims there's "NO SKELETON IN MY CLOSET, WOE IS ME".
> What^WPlease, does anyone figure out what it could be about ? A wrong or > missing compilation option in bison ?
> Cheers
This error means that the skeleton file ick-wrap.c couldn't be found. The compiler will try looking in various places for it; you can see where it's looking by using ick -u rather than just ick (which will print a little list of the locations in question). You'll definitely be able to make it work by copying ick-wrap.c to the current directory; there are a few other places it will be accepted from, too.
Note that if you're having the skeleton error, you're likely to have a problem finding syslib.i when you try to compile a program that needs it; the error message for this is "SAYING 'ABRACADABRA' WITHOUT A MAGIC WAND WON'T DO YOU ANY GOOD" (error 127). It can be fixed the same way as the ick-wrap error, by moving syslib.i (which should be in pit/lib/ syslib.i in the distribution tarball) somewhere that ick can find it. (Likewise, if you want to be able to compile TriINTERCAL programs that use its version of the system library, you'll want to move syslib.3i, etc, in the same way.)
Incidentally, what platform are you on? Issues where finding the skeleton and syslib files don't "just work" are something I'd like to fix between the beta and actual release of 0.29, but in order to fix them I need to know what's causing them first.
ais523 wrote: > "SAYING 'ABRACADABRA' WITHOUT A MAGIC WAND WON'T DO YOU ANY GOOD"
Phew ... frightening computing as I love it !
> Incidentally, what platform are you on?
It's a flat plate with 5 rolling feet, a back and 2 armrests. I think it's called an armchair. I've installed it just in front of a linux box with a Slackware-12.0.0 running in it.
> Issues where finding the skeleton and syslib files don't "just work" > are something I'd like to fix between the beta and actual release of > 0.29,
> You can download one from http://overload.intercal.org.uk/#c > Please let me know if there is any problem with this site and I'll go > and find my big clue bat :-)
Thank you. To make tests in pit/, it lacks count.3i, iogerunds.i, sink.i and hello.i. I had such-called source files in intercal-0.24 and I copied them in pit. Everething compiled and some of the programs even seem to work.
I've been impressed by the clarity of count's output, for instance. Conversely, I did not catch up what does bubble mean when, say : $ ./bubble 5 6 4 7 3 8 ICL579I WHAT BASE AND/OR LANGUAGE INCLUDES 5? ON ...
I suppose that I didn't spell "5" correctly. How would you type it ?
Herve Autret wrote: > I've been impressed by the clarity of count's output, for instance. > Conversely, I did not catch up what does bubble mean when, say : > $ ./bubble > 5 6 4 7 3 8 > ICL579I WHAT BASE AND/OR LANGUAGE INCLUDES 5? > ON ...
> I suppose that I didn't spell "5" correctly. How would you type it ?
You want to type each of the digits of a number separately, and one number per line. Additionally, the numbers need to be written out in words in a choice of languages (recent C-INTERCAL accepts English, Sanskrit, Basque, Tagalog, Classical Nahuatl, Georgian, Kwakiutl, and Latin).
