Should we broaden the topicality of this group?

[jacob navia]

Pierre Asselin said:


The installation instructions did not specify this as a requirement. If
they had done so, I would have expected them also to mention why this was
necessary, what potential harm it could do to my system, and how I can
back out of the change if I later decide I don't want to use the software.

Your Honour:

You were expressing doubts about the very existence of the
linux version. I was unable to port anything outside
windows you said.

Then, I rush to publish a version that is not totally
finished, to prove that I am working since MONTHS to port
this to linux. And NO, the work of porting weren't the details of
standard or not standard C (those issues were completely
minor) but the issues were the different executable
formats and the different calling conventions of course!
THOSE were the *real* issues. I had to rewrite the
debug information in a format suitable to gdb, and that
was a lot of work, much more than rewriting the few places where I
used a function that wasn't there under linux.

Now you complain that the documentation of the software
is not finished. RIGHT!

Personally, I prefer the "I won't write an installer for you - if you want
to run it, *you* install it - here are the steps you need" school of
software installation, because it gives you more control over what's going
on, and lets you decide intelligently whether the installation game is
worth the security candle.

A linux wizard like you will find not a problem to do

ldd lcc

and see what is using and where you have to put it.

When I am writing software for "normal" people - i.e. non-programmers,
non-propeller-heads, typically Windows users, not interested in anything
but "please just make it work" kind of people, I normally just give them
the binary. Nothing else. And no instructions. This works fine, because
they manage to work out all by themselves that there are no installation
steps to take (except, possibly, copying onto the hard disk), and
uninstallation is as simple as deleting the binary. About the only thing I
do tell them is "this doesn't write anything in your registry" (even
though I just get a blank look in return).

I do the same under linux. Isn't it NICE? What do you want then?
I do NOT write under the registry under linux, and please don't stare
with that blank look!

 

[Mark L Pappin]

/* ObC */

Your Honour:

You were expressing doubts about the very existence of the
linux version.
assert(Richard_Heathfield_claims_no_Linux_version);

Then, I rush to publish a version that is not totally
finished,

assert(jacob_navia_confirms_no_Linux_version_at_time_of_claim);

Good to hear.

mlp

 

[jacob navia]

/* ObC */


assert(jacob_navia_confirms_no_Linux_version_at_time_of_claim);

Good to hear.

mlp

Very comic.
Of course. I ported lcc-win, wrote some thousands lines
of code in 3 days.

The linux version has been there for months, but I do not have an
installer. To avoid complaints like R.H. I did not publish
the URL.

You just want to read what you want to read.
Not what I am saying

 

[Richard Heathfield]

jacob navia said:

You were expressing doubts about the very existence of the
linux version.

Yes, I did have doubts, and I still have doubts. There is no observable
difference in behaviour between linux-lcc (as installed according to the
instructions) and:

#include <stdio.h>

int main(void)
{
fputs("lcc: error while loading shared libr", stdout);
fputs("aries: libbfd-2.11.92.0.12.so: canno", stdout);
puts("t open shared object file: No such file or directory");
return 0;
}

which doesn't look like a compiler to me.

I was unable to port anything outside windows you said.

I said no such thing.

Now you complain that the documentation of the software
is not finished. RIGHT!

No, Wrong. I have not complained that the documentation is not finished. I
have pointed out that the installation instructions don't work.

 

[=?iso-2022-kr?q?=1B=24=29CHarald_van_D=0E=29=26=0F]

jacob navia said:



Yes, I did have doubts, and I still have doubts. There is no observable
difference in behaviour between linux-lcc (as installed according to the
instructions) and:

#include <stdio.h>

int main(void)
{
fputs("lcc: error while loading shared libr", stdout); fputs("aries:
libbfd-2.11.92.0.12.so: canno", stdout); puts("t open shared object
file: No such file or directory"); return 0;
}

which doesn't look like a compiler to me.

This is entirely OT, but hopefully it'll stop the complaints here. To get
a working lcc linux installation in one directory (and have absolutely no
lcc files outside of that directory), using /usr/local/lcc as a default:

LCCPATH=/usr/local/lcc
tar -xzf /path/to/lccdist.tar.gz
mv lccdist "${LCCPATH}"
cd "${LCCPATH}"
mv bin/lcc bin/lcc.bin
cat >bin/lcc <<EOF
#!/bin/sh
LD_LIBRARY_PATH=\${LD_LIBRARY_PATH+"\${LD_LIBRARY_PATH}:"}'${LCCPATH}/bin'
export LD_LIBRARY_PATH
exec '${LCCPATH}/bin/lcc.bin' -I'${LCCPATH}/include' "\$@"
EOF
chmod +x bin/lcc

Now, make sure $LCCPATH/bin is in your PATH, and you can compile.

 

[jacob navia]

This is entirely OT, but hopefully it'll stop the complaints here. To get
a working lcc linux installation in one directory (and have absolutely no
lcc files outside of that directory), using /usr/local/lcc as a default:

LCCPATH=/usr/local/lcc
tar -xzf /path/to/lccdist.tar.gz
mv lccdist "${LCCPATH}"
cd "${LCCPATH}"
mv bin/lcc bin/lcc.bin
cat >bin/lcc <<EOF
#!/bin/sh
LD_LIBRARY_PATH=\${LD_LIBRARY_PATH+"\${LD_LIBRARY_PATH}:"}'${LCCPATH}/bin'
export LD_LIBRARY_PATH
exec '${LCCPATH}/bin/lcc.bin' -I'${LCCPATH}/include' "\$@"
EOF
chmod +x bin/lcc

Now, make sure $LCCPATH/bin is in your PATH, and you can compile.

HEY HARALD!

THANKS A LOT!

REALLY!

I am a ZERO in shell programming. I will put your code in the
web site, if you do not mind.

The problem for me is that I am doing this evenings and weekends
and I can't do a lot.

 

[=?iso-2022-kr?q?=1B=24=29CHarald_van_D=0E=29=26=0F]

This is entirely OT, but hopefully it'll stop the complaints here. To
get a working lcc linux installation in one directory (and have
absolutely no lcc files outside of that directory), using
/usr/local/lcc as a default:
[...]

HEY HARALD!

THANKS A LOT!

REALLY!

I am a ZERO in shell programming. I will put your code in the web site,
if you do not mind.

Sure, I don't mind.

As an aside, since linking is supposed to be using gcc, and there is no
matching gcc wrapper script (nor should there be), libraries in
/usr/local/lcc/lib won't be directly available when linking. Since the
system default libc will be available and linked in by default, this
isn't a problem for standard C, but it might cause problems for code
using lcc-specific extensions.

 

[Richard Heathfield]

$)CHarald van D)&k said:

This is entirely OT, but hopefully it'll stop the complaints here. To get
a working lcc linux installation in one directory (and have absolutely no
lcc files outside of that directory), [...]

Now, make sure $LCCPATH/bin is in your PATH, and you can compile.

Well, you're right about it being OT, but it was a kind thought. Alas, it
hasn't fixed the problem here (which hasn't changed in any way).

Anyway, it *is* OT, so I'll shut up about it.

 

[jacob navia]

$)CHarald van D)&k said:

This is entirely OT, but hopefully it'll stop the complaints here. To get
a working lcc linux installation in one directory (and have absolutely no
lcc files outside of that directory), [...]

Now, make sure $LCCPATH/bin is in your PATH, and you can compile.

Well, you're right about it being OT, but it was a kind thought. Alas, it
hasn't fixed the problem here (which hasn't changed in any way).

Anyway, it *is* OT, so I'll shut up about it.

You can complain in comp.compilers.lcc where it is on topic if something
doesn't work in your setup

 

[Richard Heathfield]

jacob navia said:

You can complain in comp.compilers.lcc where it is on topic if something
doesn't work in your setup

I've done my complaining already. As far as I'm concerned, the
implementation is broken. Since no fix has been forthcoming, I will
continue to consider it broken until such time as I happen to learn of a
fix.

 

[jacob navia]

jacob navia said:



I've done my complaining already. As far as I'm concerned, the
implementation is broken. Since no fix has been forthcoming, I will
continue to consider it broken until such time as I happen to learn of a
fix.

That's a very good idea.

 

[Richard]

jacob navia said:



I've done my complaining already. As far as I'm concerned, the
implementation is broken. Since no fix has been forthcoming, I will
continue to consider it broken until such time as I happen to learn of a
fix.

Your bitterness towards Jacob reflects poorly on yourself. It is OT here
to discuss compiler specific bugs/issues and you should take it to the
relevant newsgroup where I am sure Jacob will help you. As it is, it is
blatantly obvious that you have set out to do a hatchet job on Jacob's
work.

 

[Richard Heathfield]

jacob navia said:

That's a very good idea.

So the official line on lcc-linux from its developer is "consider it
broken"? Fine. I'll pass that information on when occasion arises.

 

[Richard Heathfield]

Richard said:

Your bitterness towards Jacob reflects poorly on yourself.

Did I mention Jacob? I think not. I mentioned that a recently-released
implementation is broken.

As it is, it is
blatantly obvious that you have set out to do a hatchet job on Jacob's
work.

He has demonstrated that he does not need my help in that area.

 

[Mark McIntyre]

jacob navia said:


So the official line on lcc-linux from its developer is "consider it
broken"? Fine. I'll pass that information on when occasion arises.

You're being unduly agressive and deliberately misinterpreting JN's
words - in fact two of hte things you frequently complain about him
doing. Perhaps you need to review your position.

In general I tend to agree with your posts, but you seem increasingly
unable to respond appropriately to JN - which is part of why I now
tend to threadplonk any thread that you replied to him in, or
vice-versa.

--
Mark McIntyre

"Debugging is twice as hard as writing the code in the first place.
Therefore, if you write the code as cleverly as possible, you are,
by definition, not smart enough to debug it."
--Brian Kernighan

 

[Mark McIntyre]

He has demonstrated that he does not need my help in that area.

What he's demonstrated is that he doesn't know much about unix/linux,
Your style of posting is not going to either help him learn, or
persuade him that he needs to.
--
Mark McIntyre

"Debugging is twice as hard as writing the code in the first place.
Therefore, if you write the code as cleverly as possible, you are,
by definition, not smart enough to debug it."
--Brian Kernighan

 

[Richard Heathfield]

Mark McIntyre said:

What he's demonstrated is that he doesn't know much about unix/linux,
Your style of posting is not going to either help him learn, or
persuade him that he needs to.

Mark, he has demonstrated a great many times that he is not in the
slightest bit interested in learning. He is only interested in pushing
that compiler of his, and locking people into it. Every other article he
writes is an advert for that blessed compiler. And you know the astounding
thing? Even though his default behaviour is massively anti-social and his
advice not only biased but often utterly erroneous, on the very few
occasions when his behaviour is /not/ like that, his respondents reply as
if hoping that *this time* he's finally seen the light and will behave
rationally and altruistically - in the teeth of many years' evidence. He
habitually insults his interlocutors, he owes many people in this group a
great many apologies which he has no intention of delivering, and he
continues to pollute the group with ill-informed and/or self-serving
"advice". There is no point trying to teach this guy. But there *is* a
point in warning newbies away from him, at least until he learns C.
Fortunately, though, at least there appears to be little danger of
Linux-using newbies using his "compiler" for the time being.

 

[John J. Smith]

jacob navia wrote:

[big snip]

I've cross-posted this to comp.compilers.lcc and set follow-ups to it.
Hope you don't mind.

[f'ups temporarily ignored but again set to comp.compilers.lcc]

It's not a mistake.

IMO it is a mistake, but for a different reason.

Since `libbfd' is licenced under the terms of the GNU General
Public License (GPL) [1], any program linking[2] to it must be
licenced under the GPL or compatible license as well. To
comply with this license, distributed binaries of said programs
must have their source code available. Since, AFAICT, the linux
version of lcc-win32 is still closed source, linking with libbfd
would be a violation of libbsd's license terms.

-----
Footnotes
[1] see `COPYING' file in libbfd's sources for it's exact terms
[2] technically speaking, linking--either statically or
dynamically--constitutes the generation of a derived work
which means the linking executable is a derived work of libbdf

 

[Philip Potter]

Mark McIntyre said:


Mark, he has demonstrated a great many times that he is not in the
slightest bit interested in learning. He is only interested in pushing

Just recently, when people pointed out to him a problem in the wording
of the warnings of his compiler, he (after some argument) agreed and
changed lcc-win32.

that compiler of his, and locking people into it. Every other article he
writes is an advert for that blessed compiler. And you know the astounding
thing? Even though his default behaviour is massively anti-social and his

He may seem anti-social to you, but you often provoke him.

advice not only biased but often utterly erroneous, on the very few
occasions when his behaviour is /not/ like that, his respondents reply as
if hoping that *this time* he's finally seen the light and will behave
rationally and altruistically - in the teeth of many years' evidence. He
habitually insults his interlocutors, he owes many people in this group a

Let's not get into who started the insults. In the last month you called
him a frog, completely unprovoked. I'm guessing you didn't mean anything
other than simply "he is french", but it is difficult to detect tone on
Usenet so quite understandably Jacob felt insulted.

great many apologies which he has no intention of delivering, and he
continues to pollute the group with ill-informed and/or self-serving

There's no denying that he sometimes (neither never nor always) gives
erroneous advice; but when this is pointed out to him, I generally see
him accept it, and thank the person who educated him. He is particularly
responsive to Harald, but then Harald is particularly calm when dealing
with him.

"advice". There is no point trying to teach this guy. But there *is* a
point in warning newbies away from him, at least until he learns C.

Even if you think that's what's worth doing with respect to JN, that
certainly isn't what you're doing here.

Fortunately, though, at least there appears to be little danger of
Linux-using newbies using his "compiler" for the time being.

I agree that this is a good thing.

 

[Keith Thompson]

Richard Heathfield wrote: [snip]
Let's not get into who started the insults. In the last month you
called him a frog, completely unprovoked. I'm guessing you didn't mean
anything other than simply "he is french", but it is difficult to
detect tone on Usenet so quite understandably Jacob felt insulted.

Richard Heathfield did not call him a frog; that was somebody else.
(I won't add to the flames by identifying the offender.)

[snip]