libclc - from Boost thread

[Marco]

Rouben Rostamian said:
An attempt toward creating a set of library tools was made
by some participants of this newsgroup a few years ago but
it fell by the wayside. See:

http://libclc.sourceforge.net/

The failure of libclc to take root is not surprising.
Minimalism has been a characteristic of C from the very
beginning.

 

[Richard Heathfield]

Marco said:

What were the goals of [libclc]?

Initially, it was intended to be a set of fully portable routines which
built on the standard library.

Then some people decided it would be neat to add some stuff which wasn't
actually portable, some other people (myself included) expressed surprise
and concern that a comp.lang.c library would not be portable, and that's
pretty much when it died.

 

[jacob navia]

Richard Heathfield a écrit :

Marco said:

What were the goals of [libclc]?

Initially, it was intended to be a set of fully portable routines which
built on the standard library.

Then some people decided it would be neat to add some stuff which wasn't
actually portable, some other people (myself included) expressed surprise
and concern that a comp.lang.c library would not be portable, and that's
pretty much when it died.

Mmm what stuff wanted those people to add?

And why wasn't it "portable" ???

jacob

 

[Richard Heathfield]

jacob navia said:

Richard Heathfield a écrit :

Marco said:

What were the goals of [libclc]?

Initially, it was intended to be a set of fully portable routines which
built on the standard library.

Then some people decided it would be neat to add some stuff which wasn't
actually portable, some other people (myself included) expressed surprise
and concern that a comp.lang.c library would not be portable, and that's
pretty much when it died.

Mmm what stuff wanted those people to add?

And why wasn't it "portable" ???

I don't remember. If you care enough, you should be able to find it in the
archives.

There's so much you can do portably that it seems silly for such a
worthwhile project to have foundered for such a reason.

 

[Denis Kasak]

There's so much you can do portably that it seems silly for such a
worthwhile project to have foundered for such a reason.

Indeed. I must admit this is the first time I have encountered this
project. Seeing the huge potential such a library would have it's really
dissapointing it didn't succeed because of such an obstacle.

 

[Rod Pemberton]

jacob navia said:

Richard Heathfield a écrit :

Marco said:


What were the goals of [libclc]?


Initially, it was intended to be a set of fully portable routines which
built on the standard library.

Then some people decided it would be neat to add some stuff which wasn't
actually portable, some other people (myself included) expressed surprise
and concern that a comp.lang.c library would not be portable, and that's
pretty much when it died.

Mmm what stuff wanted those people to add?

And why wasn't it "portable" ???

I don't remember. If you care enough, you should be able to find it in the
archives.

There's so much you can do portably that it seems silly for such a
worthwhile project to have foundered for such a reason.

I remember deleting libclc. That usually means there is an issue with the
code's licensing...


Rod Pemberton

 

[William Ahern]

Rouben Rostamian said:
An attempt toward creating a set of library tools was made
by some participants of this newsgroup a few years ago but
it fell by the wayside. See:

http://libclc.sourceforge.net/

The failure of libclc to take root is not surprising.
Minimalism has been a characteristic of C from the very
beginning.

And, also, a reliance on established practice. I keep a "compat"
library of useful routines. My rule is that they must have been vetted on
an existing platform, and of course fit in with the spirit of C (or Unix).
I include strlcpy()/strlcat(), and the BSD sys/queue.h and sys/tree.h
macro interfaces.

I keep a separate collection of other homegrown or salvaged routines. And
as expected the churn in that library is much, much higher.

 

[pete]

jacob navia said:

Richard Heathfield a écrit :

Marco said:
What were the goals of [libclc]?

I think the whole idea was fundamentally flawed.
I didn't see that libclc was worth learning.
Learning libclc doesn't excuse you
from learning the standard library.
And if you know the standard library, then you can write libclc.

 

[Jordan Abel]

And, also, a reliance on established practice. I keep a "compat"
library of useful routines. My rule is that they must have been vetted on
an existing platform, and of course fit in with the spirit of C (or Unix).
I include strlcpy()/strlcat(), and the BSD sys/queue.h and sys/tree.h
macro interfaces.

I keep a separate collection of other homegrown or salvaged routines. And
as expected the churn in that library is much, much higher.

Any chance you might publish said libraries?

 

[boa]

* jacob navia wrote, On 10.05.2006 17:48:

Richard Heathfield a écrit :

Marco said:

What were the goals of [libclc]?

Initially, it was intended to be a set of fully portable routines
which built on the standard library.

Then some people decided it would be neat to add some stuff which
wasn't actually portable, some other people (myself included)
expressed surprise and concern that a comp.lang.c library would not be
portable, and that's pretty much when it died.

Mmm what stuff wanted those people to add?

And why wasn't it "portable" ???

Of course libclc was portable. The only "portability" issue I can
recall, was that some participants wanted to add concepts which would
have made the library usable in both ST and MT environments, others did
not.

The project failed for other reasons, which we probably don't want to
get into, at least not in public.

boa

 

[pete]

ST and MT environments

What are those?

 

[Ian Collins]

boa wrote:




What are those?

Singe Thread and Multi Threaded.

 

[Skarmander]

jacob navia said:

Richard Heathfield a écrit :
Marco said:
What were the goals of [libclc]?

I think the whole idea was fundamentally flawed.
I didn't see that libclc was worth learning.
Learning libclc doesn't excuse you
from learning the standard library.
And if you know the standard library, then you can write libclc.

In other words, classic case of Not Invented Here syndrome.

There seems to be a taboo against C programmers using libraries that do not
strictly provide functionality you could *not* have written
yourself--because you either don't understand how to implement it in the
first place or because it's just too tedious for you to bother with.

More a psychological barrier than anything else, really.

S.

 

[Richard Heathfield]

pete said:

jacob navia said:

Richard Heathfield a écrit :
Marco said:
What were the goals of [libclc]?

I think the whole idea was fundamentally flawed.

Well, I wasn't involved with it myself, but I think there might have been
some value there.

I didn't see that libclc was worth learning.
Learning libclc doesn't excuse you
from learning the standard library.
And if you know the standard library, then you can write libclc.

Yeah, but if someone's already written it then you don't have to. A great
time-saver.

Also, if libclc were truly portable, it could have been accepted into the
"canon" here in clc, and thus used as a sort of code-based version of the
FAQ. "Q: How do I <foo>?" "A: See the source for <foo>.c in libclc" - or it
could be used within answers to shortcut one's way through semi-boilerplate
stuff.

 

[Richard Heathfield]

boa said:

The project failed for other reasons, which we probably don't want to
get into, at least not in public.

That's news to me. If the project failed for non-public reasons, then
presumably some stuff was going on outside the clc newsgroup (which is
public). So libclc would have been a bit of a misnomer, wouldn't it?

 

[boa]

* Richard Heathfield wrote, On 11.05.2006 08:14:

boa said:


That's news to me. If the project failed for non-public reasons, then
presumably some stuff was going on outside the clc newsgroup (which is
public). So libclc would have been a bit of a misnomer, wouldn't it?

Why would it be a misnomer?
boa

 

[Richard Heathfield]

boa said:

* Richard Heathfield wrote, On 11.05.2006 08:14:

Why would it be a misnomer?

If it was happening outside clc, it should have been called libnotclc, or
libnothingtodowithclc, or libwedoallourfightinginemail.

 

[boa]

* Richard Heathfield wrote, On 11.05.2006 08:37:

boa said:


If it was happening outside clc, it should have been called libnotclc, or
libnothingtodowithclc, or libwedoallourfightinginemail.

If *what* was happening outside clc? Exactly what part of a project like
libclc could not happen outside clc?

Let me try with an analogy: If UK politicians discuss stuff outside the
House of Commons, UK is not a democracy?


boa

 

[Richard Heathfield]

boa said:

* Richard Heathfield wrote, On 11.05.2006 08:37:

If *what* was happening outside clc?

The stuff that made the project fail. Obviously I don't know what stuff that
is, because it happened outside clc.

Exactly what part of a project like libclc could not happen outside clc?

I don't know, because it happened outside clc. But whatever it was, it
caused the project to fail.

Let me try with an analogy: If UK politicians discuss stuff outside the
House of Commons, UK is not a democracy?

If UK politicians discuss stuff outside the House of Commons and, as a
consequence, decide amongst themselves to cede sovereignty to a foreign
despot (e.g. the EU), then yes, the UK is no longer a democracy. (In fact,
this is basically what happened in real life.)

 

[boa]

* Richard Heathfield wrote, On 11.05.2006 09:10:

boa said:


The stuff that made the project fail. Obviously I don't know what stuff that
is, because it happened outside clc.

Nothing happened outside clc that made the project fail, at least not as
far as I'm concerned. Quite the opposite, actually.

boa

[snip]